69345 ARAB REPUBLIC OF EGYPT MANAGEMENT AND SERVICE QUALITY IN PRIMARY HEALTH CARE FACILITIES IN THE ALEXANDRIA AND MENOUFIA GOVERNORATES REPORT MIDDLE EAST AND NORTH AFRICA HUMAN DEVELOPMENT GROUP (MNSHD) Document of the World Bank i TABLE OF CONTENTS ACKNOWLEDGEMENTS 5 EXECUTIVE SUMMARY 6 1. INTRODUCTION AND MOTIVATION 15 2. CONCEPTUAL FRAMEWORK 20 3. THE DEMAND FOR PRIMARY HEALTH CARE 23 4. THE SUPPLY OF PUBLIC PRIMARY HEALTH CARE: AVAILABILITY AND QUALITY OF CARE 34 5. PAYMENTS 76 6. INSTITUTIONS OF QUALITY SUPERVISION AND GOVERNANCE 91 7. CONCLUSION AND IMPLICATIONS FOR HEALTH POLICY 96 ANNEX 1: GLOSSARY OF PRIMARY HEALTH SERVICE DELIVERY IN ALEXANDRIA AND MENOUFIA100 ANNEX 2: DATA COLLECTION 104 ANNEX 4: CALCULATION OF PER-CAPITA CONSUMPTION INDEX 110 REFERENCES 116 ii LIST OF FIGURES FIGURE 1: PRIMARY HEALTH CARE PROVISION IN EGYPT ............................................................................................ 15 FIGURE 2: SUPPLY- AND DEMAND SIDE MODEL OF HEALTH SERVICE DELIVERY......................................................... 21 FIGURE 3: MAIN REASONS GIVEN FOR UTILIZING HEALTH CARE IN THE PAST SIX MONTHS, BY GENDER ....................... 28 FIGURE 4: TYPE OF FACILITY FOR LAST VISIT, IN PERCENTAGE OF USERS ..................................................................... 31 FIGURE 5: PERCENTAGE OF CLIENTS "EXTREMELY SATISFIED" WITH LAST VISIT, BY TYPE OF CLINIC AND GOVERNORATE .................................................................................................................................................... 32 FIGURE 6: SIZE OF CATCHMENT AREA, IN PERCENTAGE OF FACILITIES ......................................................................... 34 FIGURE 7: AVAILABILITY OF NON MEDICAL INFRASTRUCTURE..................................................................................... 36 FIGURE 8 DAILY VISITS FOR CHRONIC DISEASES, NUMBER OF EMPLOYEES IN FACILITY................................................ 50 FIGURE 9: DISTRIBUTION OF FACILITIES ALONG QUALITY INDEX, BY GOVERNORATE ................................................... 71 FIGURE 10: DISTRIBUTION OF FACILITIES ALONG QUALITY INDEX, BY ACCREDITATION STATUS .................................. 72 FIGURE 11: CONSTRAINTS TO IMPROVEMENT OF FACILITY, BY PERCENTAGE OF FACILITY DIRECTORS THAT ASSESSED ISSUE AS “IMPORTANT� OR “VERY IMPORTANT� CONSTRAINT TO FACILITY IMPROVEMENT ................................ 75 FIGURE 12: HEALTH SERVICE DELIVERY IN ALEXANDRIA AND MENOUFIA ............................................................... 100 FIGURE 13: TYPOLOGY OF HEALTH CLINICS ACCORDING TO THE HEALTH SECTOR REFORM PROJECT ....................... 103 LIST OF TABLES TABLE 1: KEY SOCIO-ECONOMIC CHARACTERISTICS OF ADULTS .................................................................................. 23 TABLE 2: SHARE OF ADULTS THAT COMPLETED SECONDARY EDUCATION AND ABOVE, BY CONSUMPTION QUINTILE .. 24 TABLE 3: ADULTS WITH ILLNESS OR INJURY IN PAST 6 MONTHS ................................................................................... 24 TABLE 4. PREVALENCE OF ILLNESS OR INJURY, BY CONSUMPTION QUINTILE ............................................................... 25 TABLE 5: DIAGNOSES OF KEY CHRONIC ILLNESSES....................................................................................................... 26 TABLE 6. RELATIONSHIP BETWEEN CHRONIC ILLNESS PREVALENCE AND CONSUMPTION QUINTILE ............................. 26 TABLE 7: TIME BETWEEN PROBLEM AND DECISION TO GO TO FACILITY ........................................................................ 26 TABLE 8. HEALTH FACILITY VISITATION, BY CONSUMPTION QUINTILE ......................................................................... 27 TABLE 9: FINANCIAL BARRIERS TO SEEKING HEALTH CARE, BY CONSUMPTION QUINTILE ............................................ 28 TABLE 10: ILLNESSES FOR WHICH NEAREST HEALTH FACILITY OR OTHER HEALTH FACILITY IS CHOSEN ...................... 30 TABLE 11: CLIENT SATISFACTION WITH HEALTH CARE SERVICES AT FACILITY LAST VISITED ....................................... 32 TABLE 12: PERCENTAGE OF ADULTS WITH INSURANCE, BY CONSUMPTION QUINTILE ................................................... 33 TABLE 13: HEALTH SERVICES OFFERED: PERCENTAGE OF FACILITIES THAT TREAT SELECTED HEALTH PROBLEMS................................... 35 TABLE 14: RATING OF QUALITY OF INFRASTRUCTURE, IN PERCENTAGE OF INTERVIEWERS .......................................... 37 TABLE 15: FACILITIES WITH PATIENTS FOR SELECTED CATEGORIES OF SERVICES......................................................... 38 TABLE 16: AVAILABILITY OF SUPPLIES FOR DIABETES MELLITUS TREATMENT ............................................................ 39 TABLE 17 AVAILABILITY OF SUPPLIES FOR HYPERTENSION/ CORONARY HEART DISEASE TREATMENT ....................... 41 TABLE 18: AVAILABILITY OF SUPPLIES FOR SICK CHILD TREATMENT.......................................................................... 43 TABLE 19: AVAILABILITY OF SUPPLIES FOR ANTENATAL CARE .................................................................................... 44 TABLE 20: AVAILABILITY OF DRUGS (CONDENSED LIST) .............................................................................................. 46 TABLE 21: MEAN NUMBER OF EMPLOYEES .................................................................................................................. 49 TABLE 22 SHARE OF FACILITIES BY NUMBER OF EMPLOYEES, IN PERCENT OF TOTAL FACILITIES ................................. 49 TABLE 23: TOTAL NUMBER OF STAFF, BY CATEGORY ................................................................................................... 51 TABLE 24 COMPOSITION OF MEDICAL PROFESSIONALS ................................................................................................ 51 TABLE 25 MAIN CHARACTERISTICS OF STAFF .............................................................................................................. 52 TABLE 26: ECONOMETRIC ANALYSIS OF STAFF ABSENCES ........................................................................................... 56 TABLE 27: STAFF ABSENCES BY MAIN FUNDING SOURCE OF FACILITY.......................................................................... 57 TABLE 28: STAFF ABSENCES BY SELF-REPORTED PRESENCE OF INFRASTRUCTURE ....................................................... 57 TABLE 29: STAFF ABSENCES ACCORDING TO PRESENCE OF AN AUDIT SYSTEM (“DOES THE FACILITY HAVE A SYSTEM TO AUDIT COSTS?�) ................................................................................................................................................... 58 TABLE 30: STAFF ABSENCES ACCORDING TO THE CONSEQUENCES WHICH ARE POSSIBLE AFTER A PERFORMANCE REVIEW (MULTIPLES POSSIBLE) ........................................................................................................................... 58 TABLE 31: STAFF ABSENCES ACCORDING TO PUBLIC OUTSIDE SUPERVISION OF THE FACILITY ..................................... 59 TABLE 32: STAFF ABSENCES ACCORDING TO THE ROLE OF LOCAL MEDIA IN THEIR FACILITY’S CATCHMENT AREA ...... 59 TABLE 33 OBSERVATION PROTOCOL FOR DIABETES MELLITUS TREATMENT ................................................................ 61 TABLE 34: OBSERVATION PROTOCOL FOR CHD/HYPERTENSION TREATMENT ............................................................. 63 TABLE 35: OBSERVATION PROTOCOL FOR ANTENATAL TREATMENT ............................................................................ 66 iii TABLE 36 OBSERVATION PROTOCOL FOR SICK CHILD TREATMENT............................................................................... 68 TABLE 37 QUALITY INDEX OF OBSERVATION OF PROTOCOLS ....................................................................................... 70 TABLE 38: HYGIENE PRACTICES OBSERVED, BY FACILITY............................................................................................ 74 TABLE 39: DID YOU PAY A FEE AT YOUR LAST VISIT AT THE HEALTH FACILITY, AND HOW MUCH? BY GENDER AND AGE ........................................................................................................................................................................... 78 TABLE 40: DID YOU PAY A FEE AT YOUR LAST VISIT AT THE HEALTH FACILITY AND HOW MUCH? BY POVERTY QUINTILE (PREDICTED CONSUMPTION) ................................................................................................................ 79 TABLE 41: DID YOU PAY A FEE AT YOUR LAST VISIT AT THE HEALTH FACILITY AND HOW MUCH? BY TYPE OF FACILITY ........................................................................................................................................................................... 80 TABLE 42: DID YOU PAY A FEE AT YOUR LAST VISIT AT THE HEALTH FACILITY? BY SHIFT ........................................... 80 TABLE 43: HOW MUCH DID YOU PAY AT YOUR LAST VISIT TO THE HEALTH FACILITY? BY REGION AND FH AFFILIATION. ........................................................................................................................................................................... 81 TABLE 45: HOW MUCH DID THE DOCTOR’S EXAMINATION COST LAST TIME? BY GENDER AND AGE. ........................... 81 TABLE 46: HOW MUCH DID THE DOCTOR’S EXAMINATION COST LAST TIME? BY POVERTY QUINTILE (PREDICTED CONSUMPTION) ................................................................................................................................................... 81 TABLE 47: HOW MUCH DID THE DOCTOR’S EXAMINATION COST LAST TIME? BY TYPE OF FACILITY. ........................... 82 TABLE 49: HOW MUCH DID THE DOCTOR’S EXAMINATION COST LAST TIME? BY REGION AND FH AFFILIATION. .......... 82 TABLE 50: HOW MUCH DID THE DOCTOR’S EXAMINATION COST LAST TIME? ALL FACILITIES, BY SHIFT. ..................... 82 TABLE 51: HOW MUCH DID YOU PAY THE LAST TIME YOU NEEDED TESTING/ LAB/ X-RAYS? BY GENDER AND AGE..... 83 TABLE 52: HOW MUCH DID YOU PAY THE LAST TIME YOU NEEDED TESTING/ LAB/ X-RAYS? BY FACILITY TYPE. ........ 83 TABLE 55: HOW MUCH DID YOU PAY THE LAST TIME FOR DRUGS? BY GENDER AND AGE. ........................................... 84 TABLE 56: HOW MUCH DID YOU PAY THE LAST TIME FOR DRUGS? BY POVERTY QUINTILE (PREDICTED CONSUMPTION) ........................................................................................................................................................................... 84 TABLE 57: HOW MUCH DID YOU PAY THE LAST TIME FOR DRUGS? BY TYPE OF FACILITY. ........................................... 84 TABLE 58: HOW MUCH DID YOU PAY THE LAST TIME FOR DRUGS? BY TYPE OF FACILITY. ........................................... 85 TABLE 59: ARE YOU AWARE OF THE PAYMENT EXEMPTION FOR POOR PEOPLE? BY AGE AND GENDER. ........................ 85 TABLE 60: ARE YOU AWARE OF THE PAYMENT EXEMPTION FOR POOR PEOPLE? BY PROVIDER AT LAST VISIT. ............. 86 TABLE 61: ARE YOU AWARE OF THE PAYMENT EXEMPTION FOR POOR PEOPLE? BY REGION AND AFFILIATION. ........... 86 TABLE 62: ARE YOU AWARE OF THE PAYMENT EXEMPTION FOR POOR PEOPLE? BY SHIFT VISITED............................... 87 TABLE 63: WHY DID YOU NOT PAY A FEE AT YOUR LAST VISIT? ................................................................................... 88 TABLE 64: WHY DID YOU NOT PAY A FEE AT YOUR LAST VISIT? BY TYPE OF PROVIDER AT LAST VISIT. ....................... 89 TABLE 65: FACILITIES OFFERING EXEMPTIONS FOR THE POOR ...................................................................................... 89 TABLE 66: FACILITY HAS A BOOK WHERE EXEMPTIONS AND DISCOUNTS ARE COLLECTED ........................................... 90 TABLE 67: PERSON THAT MAKES THE FINAL DECISION ON EXEMPTION IN FACILITIES ................................................... 90 TABLE 68: DID YOU PAY EXTRA FOR A HOME VISIT? BY REGION AND AFFILIATION...................................................... 91 TABLE 69: HOW MUCH DO YOU PAY, ON AVERAGE, FOR A HOME VISIT? BY AGE AND GENDER. ................................... 91 TABLE 70: COLLECTION OF CLIENT FEEDBACK, IN PERCENTAGE OF FACILITIES............................................................ 93 TABLE 71: CONTACT WITH MAYOR/TOWN ADMINISTRATION ....................................................................................... 94 TABLE 72: CONTACT WITH NGOS ................................................................................................................................ 94 TABLE 73: CONTACT WITH RELIGIOUS LEADERS .......................................................................................................... 95 TABLE 74: SAMPLE SUMMARY: TYPES OF FACILITIES ................................................................................................ 105 TABLE 75: SAMPLE SUMMARY: HOUSEHOLDS ........................................................................................................... 105 TABLE 76: SAMPLE FOR THE IN-DEPTH INTERVIEWS .................................................................................................. 106 TABLE 77: SUMMARY OF FOCUS GROUP DISCUSSIONS ................................................................................................ 107 TABLE 78: AVAILABILITY OF DRUGS .......................................................................................................................... 108 TABLE 79: DESCRIPTION OF POSSESSIONS .................................................................................................................. 112 TABLE 80: HOUSING CHARACTERISTICS AS THEY APPEAR IN HIECS 08/09 ................................................................ 113 TABLE 81: DUMMY VARIABLES DESCRIBING HOUSING CHARACTERISTICS ................................................................. 115 iv Acronyms and Abbreviations ANC Ante-natal care CHD Coronary Heart Disease DCO Direct Clinician Observation DHS Demographic and Health Survey DPO District Provider Organization EC European Commission ECG Electro-Cardiogram EDHS Egypt Demographic and Health Survey ESPA Egypt Service Provision Assessment FHC Family Health Center FHF Family Health Funds FHM Family Health Model FHU Family Health Unit GOE Government of Egypt GP General Practitioner HIO Health Insurance Organization HSRP Health Sector Reform Project IMCI Integrated management of childhood illness LE Egyptian Pounds MCH Mother and Child Health Center MD Ministerial Decree MOH Ministry of Health MOSS Ministry of Social Solidarity NCSCR National Center for Social and Crime Research NGO Non-governmental Organization OBGYN Specialist physician in obstetrics and gynecology PETS Public Expenditure Tracking Survey PTES Program of Treatment at the Expense of State QIP Quality Improvement Program QSDS Quantitative Service Delivery Survey TSO Technical Support Office 5 ACKNOWLEDGEMENTS This report was prepared by a team comprising Rebekka Grun (TTL and senior economist, MNSSP), Yoonyoung Cho (Economist, HDNSP), Bjorn Ekman (Senior Economist, MNSHH), Luca Etter (Junior Professional Associate, MNSSP), Kimie Tanabe (Economist, MNSHH), Harsha Thirumurthy (Economist, MNSSP), and Xiao Yu and Irene Jillson (consultants). We cannot thank our consultants enough for their outstanding work. The quantitative data for this report has been collected by El Zanaty & Associates (Cairo), and the qualitative data by the Social Research Center at the American University in Cairo. For the follow-up of the field work, the Bank team was locally represented by Dina Kamel (consultant). The qualitative data were further analyzed by Prof. Irene Jillson, Kareen Shabaclo, and Rachel Pittluck (Georgetown University). The team would like to also thank the management team in MNSHD for their guidance and support. The team further benefited from advice and information provided by Sami Ali (Senior Operations Officer, MNSHH), Andreas Seiter (Senior Health Specialist, HDNHE), and Margaret Koziol (HDNCE). The peer reviewers for this report are Kai Kaiser (Senior Economist, Public Sector Governance), Kathleen Beegle (Senior Economist, DECRG) and Magnus Lindelow (Senior Economist, EASHH). The task team would also like to express its gratitude towards the cooperation of the Technical Support Office at the Ministry of Health in Cairo. The task especially benefited from the contributions of Dr Mohamed Abdel Rahman, Dr Mohamed Nouh, Dr Omaima Metwally, Dr Laila Moustafa and Dr Osama Ahmed. 6 EXECUTIVE SUMMARY PUBLIC HEALTH CARE IN EGYPT The public health coverage for the Egyptian population is provided through a combination of social health insurance and subsidized government health services. Social health insurance provided through the Health Insurance Organization (HIO) covers about 48% of the population, which includes one-third of the active labor force. Adults without a formal job cannot affiliate with HIO. The Ministry of Health (MOH) and other government agencies also operate a nationwide network of government health care providers, primary, secondary and tertiary; and these function as an “insurer of last resort� by providing free or substantially subsidized health services to the citizens not covered under HIO. The HIO also operates its own primary care facilities. The Health Sector Reform Program In 1997, the MOH launched the Health Sector Reform Program (HSRP) which addressed both the delivery and the financing of Ministry-provided primary health services and came to a close over 2006. The service delivery component included interventions regarding the renewal of infrastructure and equipment; human resource development centered on family health training; and quality assurance, through a system of accreditation standards and a regular inspection schedule for facilities. Facilities that were included in the service delivery interventions are referred to as “reformed� facilities. If they subsequently pass the survey of accreditation they are called “accredited� 1. The financing component envisaged the re-channeling of funds from direct financing to contracted financing through so-called Family Health Funds (FHF) at the governorate level. The financing component also envisaged affiliating the uninsured with a non-linear price system at the point of delivery, requiring a one-off co-payment for opening a file and a co-payment for each visit.2 Poor people would be exempt from the co-payments. Facilities that contract with the FHF to participate in the financing component are called “contracted� facilities. Also facilities outside ministerial provision, such as private and NGOs and HIO, can contract with the FHF but there are very few. A logical next step after the HSRP is the new national health insurance program, announced in 2005, which aims to serve as a catalyst to effect a transition from a system driven by budget inputs to a “money follows the patient�- demand-based system. MOTIVATION FOR THIS STUDY Despite the reform efforts, evidence suggests that issues remain in the quality of service and management in both reformed and non-reformed public primary care facilities, including 1 The HSRP introduced an accreditation mechanism which involves assessing the primary facilities with survey and observation tools similar to those used for this report and calculating a % score of compliance. Facilities that score above 50% are accredited for 1 year and for 2 years if above 80%. 2 For details of the official co-payment regime, see chapter 5, and for the institutional details of the Health Sector Reform, Annex 1. 7 availability of supplies, correct co-payment exemptions for the poor, and consequently, utilization through the population.3 There is also increasing evidence that the demand-side empowerment of beneficiaries could improve the governance of health care, which would lead to a quality increase and higher utilization of health care. This suggests the need to explore the potential for demand-side mechanisms to improve service delivery and help ensure improvements in individual and population health. To that end, the Ministry of Health and the World Bank signed an agreement in 2009 that called for three self-contained but linked activities whose objective it is to increase awareness of beneficiaries and empower the local community, and ultimately to improve quality and utilization of public health services. The three activities aim at (i) diagnosing service and management quality in health care facilities; (ii) using participatory methods to design and pilot two interventions to empower citizens locally to become educated and demanding patients; and (iii) comprehensively evaluating the effect of the interventions. In more detail, the activities included or will include: i. A thorough diagnosis of service and management quality in public primary care facilities, and of perception and utilization of these services by households in their catchment area (completed). To this end, a quantitative facility survey was carried out as a census of all 362 public primary health care facilities in Menoufia and Alexandria, as well as a quantitative household survey of a total of 5,471 households in their catchment area. In each of the households, interviews were conducted with all members of the randomly selected nuclear family, amounting to a total of 21,703 individual respondents. Qualitative in-depth interviews were conducted with 20 users and non-users and 20 providers of four facilities. Eight focus groups were held, four with providers and four with users, in both governorates. ii. Pilot of two interventions to empower health beneficiaries locally to understand public health care services and claim their patient rights (in planning). These interventions, aiming at making primary care providers more accountable, are currently being planned and financed by a World Bank team, after qualitative consultation with local stakeholders, and in close cooperation with the MOH. iii. A mixed-method (qualitative and quantitative) evaluation with a repeated diagnosis of service and management quality is planned two years after the inception of the pilot in order to generate a genuine panel and a quantitative and qualitative assessment of beneficiary impact of the piloted interventions. The output of the first of these activities is this present baseline report. Preliminary findings were presented in a workshop held in Egypt in January 2010. MAIN FINDINGS This report presents the findings of the first activity named above: the multi-method diagnosis of service and management quality. Based on the study objectives, the report: 3 E.g. Egyptian Service Provision Assessment 2004, Health Insurance Survey 2006, HSRP quarterly monitoring data, HSRP Implementation Supervision Reports. 8 (1) provides an objective, unbiased assessment of the performance of public primary facilities in the Alexandria and Menoufia governorates. The performance will be evaluated against the standards introduced with the Health Sector Reform Program; (2) analyzes the quality perceptions, health situation, utilization and economic situation of households living in the catchment areas of the facilities; and (3) examines the management processes of different institutions involved in primary care. The findings are presented by study objective. 1. Performance of Public Primary Facilities in the Alexandria and Menoufia Governorates Availability of Non medical Infrastructure Regarding non-medical infrastructure, almost all (98%) public primary health care facilities in Alexandria and Menoufia have access to electricity. Working phones were available in 76.2% of the facilities and a water outlet is generally available in 93% of all facilities. Only two thirds of the facilities, however, reported that water is always available. Comparing different providers of facilities shows that Ministry of Health (MOH) ranks below the Health Insurance Organization (HIO) in the availability of all non medical supplies except for overnight beds. Further, within MOH facilities there is a difference between reformed and non reformed facilities. Many of the supplies were found more regularly in reformed than in non- reformed facilities and in urban facilities compared to rural facilities. Availability of medical infrastructure Many facilities lack basic supplies to conduct Diabetes Mellitus treatment. Overall, only 31.2% of all facilities in Alexandria and Menoufia possess a working ECG, 40.5% a machine to measure blood pressure ad hoc, 48.4% could find a working reflex hammer, 30.6% have insulin ampoules on stock, and 34.8% have a minimum of five 2 or 3ml disposable syringes. Almost all supplies for Diabetes Mellitus treatment are more likely to be found in reformed facilities than in non reformed facilities. Supplies specifically needed for coronary heart disease CHD/hypertension treatment can be found only in very few facilities; only 4.1% of the facilities had fibrates, 2.4% statins and 23.3% blood thinning medication. While still on an overall low level, the availability of supplies for CHD/hypertension treatment is significantly higher in accredited facilities as compared to non accredited facilities, as well as in reformed facilities compared to non reformed ones. Almost 9 out of 10 facilities have basic measurement instruments for treating children such as infant scale (90.1% of all facilities), child scale (88.4%) or a functioning thermometer (93.3 %). Materials to educate mothers about child health issues, on the other hand, could be found in only half of the facilities and just 34.2% of facilities could show the interviewers IMCI4 mother cards. Most of the facilities seem well equipped to conduct basic antenatal care (ANC) services. A table for ANC exam could be found in 93.5% of all facilities. Furthermore, over two thirds of facilities 4 Integrated management of childhood illness 9 had available supplies such as a spotlight source (70.7 %), clean gloves (80.4 %), safety box for needles (80.1 %), or decontamination solution for clinical equipment (78.3 %). Drugs The HSRP established an ‘essential list of drugs’ that has to be available at each primary care facility. However, only a limited number of drugs of the essential list were found to be regularly available. In particular, the availability of drugs for maternal health (less than 50% of facilities have folic acid, ergometrine/ oxytocin) is alarming, as it means many units cannot offer ante- or perinatal care. Absence of drugs for treating chronic non-communicable diseases is another problem. There are few anti-hypertensives, anti-diabetics, and drugs for high cholesterol available and none for stomach/duodenal ulcer and reflux disease. The procurement of drugs has been raised as a major issue of concern by several facility managers. In particular, many have complained that usually only a part of the drugs ordered by the facility are actually delivered. On the user side, most people reported buying medicines outside the health facility that they visited. 88 % of respondents reported buying them at a pharmacy and 13% of respondents reported the health facility itself as being one of the places where they bought medications. Quality of Care: Structural Observations of Doctors We conducted structural observations of diabetes mellitus, hypertension/CHD, sick child and antenatal consultations through direct clinician observations (DCO), whereby a trained doctor observes actual medical consultations. During the observation, the qualified surveyor compares a checklist reflecting the guidelines introduced by the HSRP with the actions actually performed by the doctor. The average Diabetes consultation is missing many of the elements that are part of the ministerial Family Health Model (FHM) guidelines5 for Diabetes Mellitus treatment. In less than half of the observed consultations the doctor examined the patient for sensations or reflexes (33.9 % of observations), examined arms and hands or feet and legs for pulsation (20.2 % and 23.7 %, respectively), or examined the back of the thorax with a stethoscope (43.8 %). Only 12.6 % of doctors asked about the patient’s smoking habits. When consulting and examining CHD/hypertension patients, the basic procedures appeared to be carried out to a large extent. For all components of the treatment, the guidelines were followed at a higher rate by MOH-provided facilities compared to HIO facilities. Furthermore, facilities in Alexandria have a significantly higher rate of carrying out standard procedures compared to facilities in Menoufia. There is no clear trend when comparing reformed facilities with non reformed facilities, as well as accredited facilities with non accredited facilities. The observation of antenatal treatments showed that very basic procedures are carried out often; more advanced but essential procedures are missing in many cases. There is no distinct difference in the performance of antenatal care between HIO and MOH facilities, and between facilities in Alexandria and Menoufia. Of the different examinations required by the guidelines of sick child treatment, very few were observed across the board. When comparing sick child consultations in Alexandria and 5 These guidelines were introduced with the HSRP. 10 Menoufia or in urban and rural clinics, there are no clear differences. Most examinations are more likely conducted in reformed than in non reformed facilities. As an additional analysis, a quality index was constructed for each of the four observation categories, which allows for a comparison between the different treatment categories. The index compiles all components of the observation checklist into a score between 0 and 1, where 0 means that none of the elements of a checklist was observed and 1 means that the provider conducted every single aspect of the consultation according to the checklist. Overall, the average value of the index is the highest for antenatal consultations, meaning that providers across all facilities adhered more to the guidelines when consulting pregnant women, than in consultations for sick children, diabetes or CHD/hypertension patients. Furthermore, the index shows that for all consultations the highest value is reached in Alexandria. Moreover, the value is higher for reformed facilities compared to non reformed facilities and for accredited facilities compared to not accredited facilities in all four observation subjects. The observance of basic hygiene practices by doctors throughout the consultations is alarming; in only 22.4 % of all observations did the provider wash their hands with soap and water prior to engaging with the patient. The numbers are on a similar low level for the usage of disposable gloves and the change paper or sheet on the examination table. Constraints to Improvement In the eyes of facility manager, the three most severe constraints to improving the quality of services at the facility level are the low motivation of staff (viewed as a constraint by 43% of all managers), general lack of supplies (39%) and the (non) availability of qualified staff (37.5%). Payments Official payments differ by type of facility. In all reformed primary facilities in Menoufia and contracted facilities in Alexandria, patients will co-pay: 10 LE to register in the family health unit and to open a family folder, 5 LE in Alexandria and 10 LE in Menoufia for annual renewal of the folder, 3 LE per examination, 35% of the medical treatment (drugs and other therapy), and 50% upon repeat treatment. Home visits are officially not part of the package provided by contracted facilities. Non-reformed Primary Health Care Units charge 1LE for the examination and nothing for inscription or drugs, and HIO units charge 50Piasters per examination and 25- 33% of drug costs. Both latter types of facilities also apply an official fee of 10LE per home visit. Some people are officially exempt from payment at the point of service. For people who are already insured in the payroll-based HIO but seek treatment in reformed facilities, the HIO reimburses for treatment. Poor people, as identified through household enumeration and poverty criteria, are supposed to be exempt from any fees in reformed units. Nearly 80% of the people in our sample paid something at their last doctor’s visit. The share of people charged is actually not much lower at reformed public facilities (71-79%) than in private facilities (85%). Patients have paid on average 27LE for just the doctor’s examination at their last visit. And the average de facto examination fee at public units is above the official co- payment for reformed units. Many facilities charge for home visits. Out of all patients, 65% report having paid extra for a home visit, i.e. exceeding the usual visit fee. 11 The vast majority of respondents, 97%, have never heard of the payment exemption for poor people in reformed facilities. When enrolling for the co-payment regime, 61% of the enrollees report a status research into their wealth and income situation, 39% did not receive any enquiry whatsoever. At their last visit to a facility, 24% of those registered in the co-payment regime reported being exempted. Unfortunately, this percentage does not vary with the poverty status (consumption quintile). Of these, 34% were already covered by HIO, 30% reported to have paid at another visit, and 17% believed the service was free for everyone. Only 12% believed themselves exempt due to either co-payment enrollment or status research. 6% believe they have been exempted because they know the personnel at the facility. 2. Quality Perceptions, Health Situation, Utilization and Economic Situation of Households To assess the perception of quality and use of the public facilities, we examined self-reported prevalence of illnesses and use of health care services, as well as respondents’ experiences with them. Nearly 40 % of the sample report having been ill or injured in the past six months. Those with lower levels of schooling were more likely to report having been ill or injured. Cancer prevalence rates were low, and 5.41 % of the sample report having been diagnosed with diabetes. About 7.5 % of the sample reported they have had high blood pressure in the past 12 months. In general, a very large fraction of individuals sought care when they were ill. 95 % of adults reportedly sought care when they were ill in the six months prior to the interview. Utilization in general is high by international standards. Among all adults, nearly 45% visited a health facility in the past six months, 10% visited it twice. Older citizens report higher utilization, as can be expected. Access to healthcare appears to be no problem in terms of geographical availability, but possibly in terms of finance. The average travel time to the main health facilities visited was slightly less than one hour. However, 33 % of non-users said there were occasions when lack of money kept them from going to seek care. Only 50 % of household heads reported they have payroll-based health insurance (HIO). 43 % reported they were registered in the co-payment regime introduced through the HSRP. By far the most frequently chosen providers of care were private doctors. Among all health care users, nearly 47 % had seen a private doctor in the past six months. However, people who did seek care at public health facilities reported high subjective satisfaction rates. 3. The Management Processes Related to the Public Primary Care Facilities Human Resources Allocation of Human Resources Most of the health facilities in Alexandria and Menoufia serve a catchment area of 30,000 people or less. The median facility has 50 health workers which each carry out – on average – 12 90 consultations per year. The ratio of health workers to beneficiaries is much higher in the more rural Menoufia, with one health worker for 336 inhabitants, than in Alexandria, where there is a health worker for 1193 inhabitants. Furthermore, there are fewer health workers per beneficiary in reformed facilities than in non reformed facilities and in accredited facilities compared with non accredited facilities. About 20% of the 18,253 staff working in the surveyed public primary health care facilities are trained medical professionals, of whom approximately half are general practitioners; 25% each are specialists or pharmacists. There are major differences in the socio-demographic composition of the different categories of employees. In particular, doctors and pharmacists are much younger, less experienced and have been at the facility in which they currently work for a much shorter period of time compared to nurses and, in particular, administrative staff. Overall, primary health care in Alexandria and Menoufia has a female dominated workforce: 64.8% of doctors and pharmacists, 72% of administrative staff and almost 98 % of nurses are female. Presence and Absence of Staff Officially, primary care facilities are available 24 hours each day. De facto, while all offer a morning shift, only 24% offer an afternoon shift and 4% offer a shift after 8pm. The survey included three surprise visits during the morning shift to monitor staff presence. The researchers found that on average across all three visits, a visitor would find 21% of full-time staff absent, and 32% of all staff. 42% of full time staff were absent at least once, 14% were absent twice or more during the shifts on which they were scheduled to be present. 6 The highest presence in all three visits was observed in administrative staff (66%), nurses (64%), social workers (64%), lab technicians (62%), and other staff (62%). The staff categories with absences of 2 or 3 times were highest among ambulance workers (39%, most of these reported the reason of ‘stopped working’ or ‘runs a personal errand’); gynecologists (25%); Family Health Doctors (24%); and Health educators (24%). General practitioners, pediatricians, midwives, pharmacists, Raeda Rifiya (female outreach workers), health inspectors and -workers showed an intermediate absence profile, with about 50% of these categories present in all three visits. In the econometric analysis, professional categories appear to explain absences to a stronger extent than other variables. Other correlations with individual characteristics show that female staff members are less absent than men, and a higher education is consistently and significantly linked with higher absence rates. People born within 5 km of the facility are less frequently absent than those who were born further away. A similar relation holds for those who live near the facility. Further, staff from cities above 100k are significantly more absent and staff in rural areas shows a better presence pattern. Staff whose relatives visit the facility, can be found present more often. Facility management, infrastructure, a cost audit system on site and positive personnel incentives all matter for better staff presence. Facilities that judge the lack of major infrastructure, such as medical equipment, clean water or electricity an important issue, report somewhat higher staff absences. The more a facility is making use of a cost audit system, the better the presence record of staff. As for the consequences of performance reviews, it is 6 The presence of part time staff was not included in the analysis. 13 possible that positive incentives work best, with facilities using salary increases and promotions being associated with better average presence records. Facilities experience some outside quality supervision from both formal (government supervision) and informal (mayor, media, mosques) institutions. Many of these have a positive association with presence. Most facilities are regularly visited by supervisors from the Health District, and in the case of contracted facilities, also from the Family Health Fund. Those who report not receiving these visits show higher absences. Facilities that cooperate with the religious institutions to announce their vaccination campaign show somewhat better staff presence than those who did not. Staff of (the very few) facilities where local media are reportedly critical show a better presence record than where local media is very positive or does not play a role. A relationship with or visits by the mayor show no influence. Perhaps surprisingly, the presence of, and relationships with NGOs do not seem to be associated with the staff presence pattern, at least not positively. Institutions of Quality Supervision and Governance The study looks at both formal and informal quality supervision and governance in public primary health facilities in Alexandria and Menoufia. As formal mechanisms we understand administrative supervision executed by different layers of the MOH, as well as the FHF; and institutional collection of client feedback at the facility. Informal mechanisms include media, mayors, religious organizations or NGOs. Administrative supervision consists of direct supervision by the district and the governorate, as well as – in reformed facilities - supervision in the form of the accreditation process of the health sector reform carried out by the quality assurance group. District level supervision takes place in the form of routine inspections and follow-up on compliance with standards. The Family health fund (FHF) has an autonomous supervision system which focuses on the administrative and financial arrangements of the facilities. The central FHF is supervising the adherence to the HSRP strategies on the regional level. Moreover, the central FHF carries out monitoring and evaluation of regional FHFs. Further, over 85% of all facilities in Alexandria and Menoufia have a system for determining client opinion about the facility or services. There is some traction and follow-up after client feedback; a bit over one third of the 85.6% of the facilities that collect client information reported they have made changes as a result of client opinion. Accredited facilities have a higher rate of collecting feedback compared to non accredited facilities, the same can be observed when comparing reformed with non reformed facilities. During the focus group discussions, however, few users said they were aware of feedback mechanisms, and none reported having ever used them. There are several informal governance institutions that can potentially influence the facilities’ work, such as media, mayors, religious organizations or NGOs. In many public primary health care facilities in Alexandria and Menoufia, the town administration is involved in one way or another with the facility. Further, more than a third of primary care facilities appear to have a fruitful relationship with the religious institutions in their neighborhood. Over one-third of facilities, 37%, reported that the mosques and churches in their neighborhood helped them announce their vaccination campaign. Press and media however had rarely any relationship to 14 the facility. Finally, almost two thirds (64.4 %) of all facilities do not know of any NGO in the area where they operate. CONCLUSION The examination of public primary health care in Alexandria and Menoufia, and of its responsiveness to the needs of the population yields a series of interesting results. Utilization of health care in general is rather high by international standards. People do use health care and go to the doctor very quickly after developing an ailment. But the provider of choice is usually private, leading to relatively low utilization rates for public primary care in spite of its officially lower cost and physical availability. This is reflected in the low number of consultations per health worker – 90.1 on average per year - in public facilities. The analysis of public primary care offers various possible explanations for this pattern, among which we want to underline three. First, the quality of care shows substantial scope for improvement, especially regarding the availability of drugs and the compliance with consultation guidelines and hygiene practices. The availability of basic medications is erratic and a major complaint of users and non-users, who describe bypassing public facilities in order to obtain medications even if they have to pay for them. Regarding compliance with protocols, the gap to the standards promoted by the HSRP is particularly big in the area of chronic diseases, such as CHD/hypertension and diabetes, which does not bode well for a country such as Egypt, which has passed the epidemiological transition. Relatively little attention is paid to prevention of chronic diseases and to improved health habits. For example, while tobacco consumption is endemic in Egypt, and increasing among youth7, it is scarcely discussed during consultations, and diet and exercise in far less than half of the visits for CHD/hypertension. Second, the de facto opening times of facilities run contrary to expressed patient needs. Most facilities offer a morning shift (8am-2pm), but very few offer an afternoon, let alone an evening shift, although they are officially required to do so. This makes it difficult for working patients to use these facilities. In addition, even during the morning shift about 21% of full-time facility staff is on average absent. – In the private sector, according to the complementary qualitative analysis, the presence times are significantly more reflective of patient needs. Finally, the mechanisms to exempt the poor from the new co-payment regime cannot be considered functional. About the same share of patients pay full fees at a public facility as do at a private facility, and the share that is exempted is not higher among the poor than among the rich. Also, in some public facility types, the actually charged average fees exceed official co- payments. Moreover, one of the primary avenues for access to care, the co-payment exemption for the poor, is essentially not known to the population. The vast majority – 97% - of household respondents have never heard of this exemption, while however 84% of facilities report offering an exemption for the poor. 7 The 2005 Demographic and Household Survey (DHS) found that 14% of males < 20 years of age reported using tobacco products (cigarettes and waterpipe) in Cairo, and 4% and 15% of boys, aged 12 and 17 years, respectively, were smokers. A study by Gadalla et al. 2003 showed that among Egyptian youth ages 15-19, smoking prevalence was significantly different by gender: 13% for males and 3% for females; the median age for onset of smoking was 15. 15 1. INTRODUCTION AND MOTIVATION PUBLIC HEALTH CARE IN EGYPT The public health coverage for the Egyptian population is provided through a combination of social health insurance and subsidized government health services. Social health insurance coverage provided through the Health Insurance Organization (HIO) covers about 48 % of the population, which includes one-third of the active labor force. The bulk of the population under HIO coverage (80 %8) are schoolchildren and infants, affiliated through their fathers. Adults without a formal job, including housewives, cannot affiliate with HIO. Regarding subsidized services, the Ministry of Health (MOH) and other government agencies operate a nationwide network of government health care providers, primary, secondary and tertiary; and these function as an “insurer of last resort� by providing free or substantially subsidized health services to the citizens not covered under HIO. The HIO also operates its own primary care units. The following graph gives an overview over the different types of primary care available, their providers, and the user’s options for choice between them. Figure 1: Primary Health Care Provision in Egypt Provider MOH HIO Private (public) (public) Reform-status Reformed Non-reformed Non-reformed Non-reformed (with few •Family Health exceptions) Type of facility Centers (FHC) – Primary urban, larger Health Care HIO •General •Family Health Units (PHCU) polyclinics Practitioners Units (FHU) – (urban or (with rural, smaller rural) specialists) •Specialists User must be Must use facility Must use facility HIO member. User choice in catchment. In in catchment. Members can Free choice, if urban areas, Sometimes, & eligibility use FHU/C. user can pay catchments PHCU and Catchments overlap. FHC/U overlap. overlap. Users often not aware of reform and new nomenclature. Often think in old categories of ‘urban’ and ‘rural’ units. 8 Figures from the Public Expenditure Review 2006 16 THE HEALTH SECTOR REFORM PROGRAM In 1997, the Government of Egypt (GoE) launched the Health Sector Reform Program (HSRP). The HSRP was introduced in a first phase in the pilot governorates of Alexandria, Menoufia and Sohag (1998-2004) and subsequently extended to Qena and Suez (2004-2005). The reform included both the delivery and the financing of primary health services. The service delivery reform of primary care was the first step in the reform process and addressed persistent needs in maternal and child health especially through the introduction of the family health model (FHM) as the principle of provision. The overarching goal of the HSRP was to achieve the coverage of basic health services for each citizen. This was articulated in five guiding principles:  Universality: Covering the entire population with a basic package of priority services.  Quality: Improving and assuring the standard of health care and facilities, enhancing diagnostic and clinic effectiveness, updating medical and nursing education and training.  Equity: Financing for health services is based on the ability to pay, while the provision of services is based on needs.  Efficiency: Allocating and mobilizing human, financial, and infrastructure resources for health care based on population needs and cost-effectiveness.  Sustainability: Ensuring the continuity, self-sufficiency and lasting establishment of the health care system reforms. Service Delivery Component The service delivery component included interventions regarding the (1) renewal of infrastructure and equipment; (2) human resource development, especially family health training; and (3) quality assurance, through a system of accreditation standards and a regular inspection schedule for facilities. More specifically, these comprised (1) A standard catalogue of equipment a facility needs to possess in order to be able to contract with the Family Health Fund (FHF, see explanation later). These guidelines have been developed to ensure facilities have the equipment necessary to perform the basic benefits package defined under the reform. The reform invested in the equipment to bring facilities in line with the catalogue. A precondition to receiving these investments was accreditation (see below). (2) Guidelines regarding staff numbers and qualifications needed at each facility. The recommended staff roster for any facility follows three criteria, namely (i) the population of the catchment area (in urban areas the MOH aims at covering 50% of the population, in rural areas 100%)9, (ii) the size of the facility (a facility can have 1 to 3 clinics, depending on the catchment areas), and (iii) the expected number of daily visits, which is estimated by multiplying the catchment population by 1.9 (number of visits per family member). Per physician, 700 families can be registered10. - The implementation of these guidelines aimed at improving staff quality through training, and at making the existing primary healthcare units (which were often overstaffed) more efficient. 9 Family Health Manual, Vol. 1, p.67 10 Family Health Manual, Vol. 1, p.67 17 The guidelines further regulate working hours of the facilities, which guarantee 24 hours services in both rural and urban facilities. However, over 70% of visits are expected during the morning shift (8am to 2pm) which is reflected in the staffing guidelines. (3) A system of quality assurance. The MOH established the Quality Improvement and Accreditation Program (QIP) in 1998. The program developed (i) a standardized process of accreditation and (ii) a system for continuous improvements where facilities play an active role in monitoring their performance. (i) Is based on a facility survey composed of eight dimensions, which are patient rights, patient care, management of the facility, management of human resources, management of support services and information, quality improvement program and infection control program. The purpose of the survey is to collect data to verify whether the facility has met the established standards. Those who meet 60% or more will be accredited by the Quality Department in MOH. The final accreditation score is based on the three elements interview, record review and observation of patient treatments. – From the facility’s point of view, accreditation is a seal of quality and allows it to benefit from the infrastructure and HR interventions under the reform. In terms of timing, the service delivery interventions would be implemented as follows. (1) Prior to any reform interventions, a facility is ‘non-reformed’. (2) As a first step under the reform, the facility receives new infrastructure (equipment and renovation) and HR training. A facility that has received this HSRP package is ‘reformed’. (3) Once acquired reform status, the facility has its quality evaluated and receives accreditation if satisfactory. The facility is ‘accredited’. (4) Once accredited, it can contract with the FHF, it is ‘contracted’. Financing of primary health services The financing component envisaged (i) the re-channeling of funds from direct financing to contracted financing; and (ii) affiliating the uninsured with a co-payment system at the point of delivery. Regarding (i) the Family Health Funds (FHF) were established in 1999 at the governorate level to act as agents and contractors that purchase family health services for the insured as well as the uninsured from different provider types (Public, Private and NGO). A FHF receives funds from various sources. The main costs are covered by the MOH. The FHF also receives revenues from patient co-payments, contracting services to the Health Insurance Organization (HIO), and on-going donations from the European Commission (EC). These funds are used to cover incentives to contracted health service providers, as well as administrative costs including salaries for term contracted personnel. The bulk of financing at the primary level is however still direct. Regarding (ii), this type of insurance paid at the point of delivery is also called ‘enrolment in the Family Health Model (FHM).’ The co-payments include a one-off payment for opening a family file and another co-payment for each visit to the facility. Poor people are exempt from any co- payments. The incentives of a facility contracted with the FHF, as opposed to a non-contracted facility, mainly differ along the personnel bonuses. The bonuses (‘incentives’) paid out by the FHF are related to the month by month performance of the facility and are higher with a higher performance. At average performance levels, they have recently become more stringent than 18 the bonuses received at non-reformed facilities (which are not performance related), making the latter facilities more attractive for staff. The FHF organization in general and the co-payment regime in particular are described in more detail in Annex 2 and the chapter ‘Payments’, respectively. MOTIVATION FOR THIS STUDY Despite the reform efforts, evidence suggests that issues remain in the quality of service and management in both reformed and non-reformed public primary care facilities, including availability of supplies, correct co-payment exemptions for the poor, and consequently, utilization through the population.11 There is also increasing evidence that the demand-side empowerment of beneficiaries could improve the governance of health care, which would lead to a quality increase and higher utilization of health care. This suggests the need to explore the potential for demand-side mechanisms to improve service delivery and help ensure improvements in individual and population health. To that end, the Ministry of Health and the World Bank signed an agreement in 2009 that called for three self-contained but linked activities whose objective it is to increase awareness of beneficiaries and empower the local community, and ultimately to improve quality and utilization of public health services. The three activities aim at (i) diagnosing service and management quality in health care facilities; (ii) using participatory methods to design and pilot two interventions to empower citizens locally to become educated and demanding patients; and (iii) comprehensively evaluating the effect of the interventions. In more detail, the activities included or will include: iv. A thorough diagnosis of service and management quality in public primary care facilities, and of perception and utilization of these services by households in their catchment area (completed). To this end, a quantitative facility survey was carried out as a census of all 362 public primary health care facilities in Menoufia and Alexandria, as well as a quantitative household survey of a total of 5,471 households in their catchment area. In each of the households, interviews were conducted with all members of the randomly selected nuclear family, amounting to a total of 21,703 individual respondents. Qualitative in-depth interviews were conducted with 20 users and non-users and 20 providers of four facilities. Eight focus groups were held, four with providers and four with users, in both governorates. v. Pilot of two interventions to empower health beneficiaries locally to understand public health care services and claim their patient rights (in planning). These interventions, aiming at making primary care providers more accountable, are currently being planned and financed by a World Bank team, after qualitative consultation with local stakeholders, and in close cooperation with the MOH. vi. A mixed-method (qualitative and quantitative) evaluation with a repeated diagnosis of service and management quality is planned two years after the inception of the pilot in order to generate a genuine panel and a quantitative and qualitative assessment of beneficiary impact of the piloted interventions. 11 E.g. Egyptian Service Provision Assessment 2004, Health Insurance Survey 2006, HSRP quarterly monitoring data, HSRP Implementation Supervision Reports. 19 The output of the first of these activities is this present baseline report. Preliminary findings were presented in a workshop held in Egypt in January 2010. OBJECTIVE OF THIS REPORT The present report presents the results of four types of surveys conducted in the governorates of Alexandria and Menoufia, since July 2009, (i) a quantitative in-depth census of the 362 public primary health care facilities and their staff, (ii) a quantitative survey of 5,417 households in their catchment area, (iii) qualitative interviews and focus groups with a subset of these facilities and households and (iv) qualitative institutional expert interviews. The surveys were designed in cooperation between the TSO of the MOH and the World Bank, and are described in more detail in Annex 1. The objective of this report is threefold. It attempts to  Give an overview over the performance of primary care facilities in perspective of the objectives of the HSRP;  Account for the basic health needs of the population in the catchment areas of these facilities, and how these needs are met by the primary care facilities in our sample; and  Explore existing formal and informal governance institutions that might play a role in approaching service delivery performance to the population’s health needs. With this in mind, the remainder of the report is structured as follows: chapter 2 provides a conceptual framework for our analysis; chapter 3 analyzes the demand side of primary health care, including health needs and access to health services; chapter 4 analyzes the supply of public primary health care, including the availability of medical and non-medical equipment and human resources; chapter 5 analyzes co-payments and payment exemptions for medical treatment; and chapter 6 examines the formal and informal governance structures that work in the two governorates. 20 2. CONCEPTUAL FRAMEWORK Publicly provided services, such as basic education and health care, are essential in maintaining and improving individual welfare. There is a wide range of services traditionally provided by public agencies and organizations (and some by private or semi-private actors) that can play a critical role for economic growth and poverty reduction. Public services contribute to the building of production factors for the economy, such as human capital (education and health services), infrastructure (roads, transport systems), innovation and technology (research and development, higher education) and to the total productivity of all these factors (mobility and employment services). While many of these services determine the performance of an economy, the ones that are particularly critical to the poor are the provision of basic education and health services. The quality of public services will determine the extent to which citizens are able to prosper from them and this especially for the poor and vulnerable groups. The importance of understanding the determinants of quality has been acknowledged by policymakers and researchers alike. In social sectors in particular, international evidence has shown that increased spending does not necessarily translate into enhanced quality and ultimately in improved outcomes (World Bank, 2003). Many factors intervene between the input of spending and the outcome of individual welfare, including the composition of spending, the capacity to effectively manage resources, and the ability to effectively target scarce resources to those in most need. Some of these influential factors can be measured more accurately than others, and obtaining an understanding of what determines quality of public services is essential for effective policy development12. This report focuses on transforming scarce inputs into outcomes in the context of the delivery of publically provided primary health care. The overall question of concern is: Do citizens receive the services they are entitled to? The approach taken looks at citizens as the customer of public services and measures the quality of these services at the point of interaction between government and citizen, in our case a public health care facility. A SUPPLY AND DEMAND SIDE MODEL OF HEALTH SERVICE DELIVERY Delivering health service involves a complex system of actors and their relationships. And quality of health care is difficult to define and measure. This study adopts the general definition of quality as health care that meet the health needs and expectations of its clients, both from a clinical and a governance perspective. To facilitate our analysis of this system, we follow a simplified conceptual framework that identifies the actors’ roles, incentives and accountability structures. It builds on previous work on the role of enhancing accountability and governance in public service delivery (World Development Report 2004). Note that this framework is generic and does therefore not represent a specific health system in a specific country. Rather, it tries to condense the theoretical literature in an illustrative way. The figure below provides the framework. 12 World Development Report 2004 or Filmer, Hammer and Pritchett (2000) provide a broad overview of how these intervening factors shape the relationship between inputs and welfare outcomes. 21 Figure 2: Supply- and Demand side Model of Health Service Delivery The core of the model is the main object of the current study: the delivery of primary health services. Specifically, this study examines the interaction between the doctor (or nurse) and the patient in a given facility and for a given health issue. They both are the frontline representatives of a larger system that we divide into supply and demand side of service delivery. On the supply side, the doctor is an employee of a specific health facility and a civil servant that acts within the framework of the health provision agency or the facility itself, depending on the degree of autonomy of the facility. Within the supply side, therefore, there is a principal-agent relationship between the employee and the facility and also between the facility and the ministry (or agency). As an individual the doctor/nurse might have intrinsic ambitions to help the client as much as possible, as an employee, however, the provider faces incentives and constraints from the facility management, such as keeping the budget straight and focus on key services. The facility as an entity, then, is on one hand the sum of all employees working at a given facility but on the other hand constitutes a unit in the government’s line of health service delivery. Decisions on all three levels – government, facility, and individual health worker - contribute to the quality of service received by the beneficiary. However, it is important to disentangle the different layers as they each have a very different role in providing the health services. On the demand side of the model, the recipient has a dual role as both client and citizen. As citizen they participate in political processes, depending on the form of governance. As direct clients of service providers, households hope to get clean water, have their children educated 22 and – in this particular case – protect the health of their family. Clients can exercise their power if they can choose between providers (referred to as ‘Exit’) and if they can complain about provision from the same provider and the complaint has an effect (also called ‘Voice’). Voice can be exercised in a private setting, directly complaining to the provider, or a in a public setting, if citizens have a right to participate in governing the service provider. An important issue at the point of service is the information asymmetry between provider and client, i.e. the extent to which access to relevant information differs between the provider and the patient. (Economists call this a ‘market failure’.) This problem has two dimensions: first, there is a natural technical information asymmetry between the trained provider and the laymen client and, second, there is an administrative asymmetry between provider and client. With regard to the last issue, the client may not always know about his or her rights as a patient and is therefore subject to receiving suboptimal treatment, paying higher fees and, in the worst case, staying away from using health services. The government (i.e. health ministry) can attempt to bridge this information gap. But its representatives also have to take other incentives into account, like maintaining power and thereby needing a good perception of service quality, irrespective of the underlying realities. Given the information asymmetry, and the potential incentive conflict for the government to bridge it, there is a role for independent observers. These ‘quality supervisors’, as we called them in the figure, are expert observers which are independent from the public administration and can interpret quality for clients. Quality supervisors can be specific independent agencies charged with quality assurance, or the media, or consumer protection associations, or other experts in the field that publish their views. They are able to perceive the actual quality of services and inform clients in a manner that is useful to them. The rest of this report is laid out to logically follow the elements of the conceptual framework: chapter 3 analyzes the demand side of primary health care; chapter 4 analyzes the supply of public primary health care; chapter 5 analyzes payments for public primary care; and chapter 6 examines the formal and informal governance structures that can potentially supervise quality of service delivery. 23 3. THE DEMAND FOR PRIMARY HEALTH CARE This chapter assesses the demand side of health care in Alexandria and Menoufia. We examine the care needs of the population, access to care, and actual utilization of services. PROFILE OF THE POPULATION Socio-demographic profile of population The basic characteristics of adults aged 15 years and above in Alexandria and Menoufia are reported in Table 1. The average age of the population is 35.5 years, and 80 % of them report having attended at least some school. The highest level of schooling completed, for those who did attend some school, was primary for 18.5 %, secondary for 43.9 % and university for 17.8 %. Nearly two thirds of the adults are married (64.9 %) while 28.4 % of them report being never married. Among the 5,417 households included in the survey, the average household size was 4.5. Table 1: Key socio-economic characteristics of adults Mean Std. Dev. Female 50% Age 35.5 15.29 Attended any school 80% Highest level of school completed, among those who attended Primary 19% Preparatory 16% Secondary 44% Upper intermediate 4% University 18% Marital status Married 65% Widowed 6% Divorced 1% Separated 0% Never married 28% Most household heads work, but a minority has pension coverage. The average number of days that the heads of household reported working in the past week is 4.4 days; approximately 10 % of the heads of household had retired. Among those who were employed, over 60 % were paid daily or hourly. Approximately 75 % of those employed responded that their employer was registered or authorized. The five sectors in which most heads of household were employed were: agriculture, public administration and defense, community or social service, manufacturing and transportation/communications. Only 44 % of employed heads of household reported that their job included a pension. Women and men have about the same education, but much more men are employed. The data collected on educational attainment do not indicate large differences between men and women, with 33 % of women and 31 % of men having completed no more than primary school. 24 Roughly 50 % of both men and women completed secondary school or higher. The largest differences between men and women emerge in the employment data, as men are much more likely to have been employed in the past week than women (69 % for men compared to only 11 % for women) and men also worked more days in the past week in the labor market (4.04 days for men compared to 0.65 days for women on average). As expected, wealthier individuals have a higher education level. An index of economic status was constructed by using measures of asset holdings and housing characteristics in the survey to predict per-capita consumption.13 This predicted consumption value was used to calculate the consumption quintile in which households belonged. Using this information on the consumption quintile of the households in which individuals reside, we find that individuals in the richest households were significantly more likely to have completed secondary school or above (78.6 % of individuals in the highest consumption quintile had completed secondary school or above, compared to 58.68 % in the lowest consumption quintile, see Table 2). This relationship between education and consumption is important to keep in mind when interpreting the tabulations further below. Table 2: Share of adults that completed secondary education and above, by consumption quintile Consumption quintile First Second Third Fourth Fifth 58.68% 62.58% 64.22% 67.48% 78.60% User health profile Nearly 40 % of the sample report having been ill or injured in the past 6 months. Women were more likely to report having been ill or injured than men (46.4 % compared to 33.5 %). Not surprisingly, the likelihood of illness and injury rises with age as 30 % of those between the ages of 15-24 years reported an illness, compared to 58 % of those above the age of 50 years. Self- reported health appeared to be slightly worse in urban areas than in rural areas, but there was no difference between Alexandria and Menoufia. Those with lower levels of schooling were more likely to report having been ill or injured. Table 3: Adults with illness or injury in past 6 months If educated, highest level of schooling Went Whole to Upper inter- More than Sample Male Female school Primary Preparatory Secondary mediate University university 33.51 39.78% % 46.35% 37.63% 45.95% 36.42% 37.16% 36.03% 32.11% 33.33% The prevalence of illness and injury increases with lower economic status. As Table 4 shows, individuals from poorer households are more likely to report having been ill or injured in the 13 Since the survey did not collect consumption information, the index was constructed based on the observed relationship between various housing and wealth indicators and per-capita consumption in the HIECS 2008/2009 (restricted to the governorates of Alexandria and Menoufia). This observed relationship was used to predict consumption in the health survey households using this survey’s values of the same housing and wealth indicators. The predicted consumption values were then used to calculate the consumption quintiles. 25 past six months, but the difference between individuals from the richest and poorest households is relatively small (4 percentage points). As such, self-reported illness does appear to be driven primarily by this measure of economic status. Table 4. Prevalence of illness or injury, by consumption quintile Consumption quintile First Second Third Fourth Fifth 43.67% 40.39% 37.85% 37.71% 39.65% Turning to major accidents or catastrophic illnesses, the fraction of people who reported experiencing these in the past 12 months was 1 % or below. The only group with considerably higher likelihood of having experienced a catastrophic illness in the past 12 months was the elderly, for whom the rates tended to be near 5 %. Table 5 reports the prevalence for several chronic diseases. Cancer prevalence is 0.2 %, but the rate among women is three times higher than among men (0.24 to 0.08 %). Cancer prevalence is higher in Alexandria than Menoufia (0.28 to 0.09 %), and similarly it is higher in urban areas as well. As expected, prevalence is higher in older age groups. It is also higher among the unemployed, a result that is not entirely to be expected. A relatively small proportion of the respondents, 5.4%, report having been diagnosed with diabetes.14 Prevalence rises considerably with age, with those over the age of 50 having a reported prevalence of 20.9 %. In Alexandria, prevalence is almost double that in Menoufia. Prevalence is also higher among those whose highest level of schooling is primary school. The nature of diabetes – a chronic illness that requires life-long treatment that is relatively expensive – makes it an important tracer condition for the extent to which the health system responds to the needs of the population. In Egypt, around 19,000 men and 40,000 women die every year from complications related to diabetes. The economic impacts of diabetes are found both at the individual and the aggregate levels. Egypt spends approximately US$116 per capita per year on diabetes treatment, equivalent to around 16 % of total health spending. However, Egypt does not have a national diabetes program. The prevalence of diabetes in the survey areas, it should be noted, is slightly higher than the national prevalence reported in the 2008 Demographic and Health Survey, in which 2.3 and 3.6 % of men and women, respectively, reported having been diagnosed by diabetes. About 7.5 % of the sample report high blood pressure in the past 12 months. Again, prevalence was higher among women (9.4%) than among men (5.8%). Older individuals were also much more likely to have had high blood pressure, with nearly 30 % of those above the age of 50 reporting the problem. Prevalence was also higher in Alexandria than in Menoufia (10 % compared to 6 %). Lower levels of schooling are also associated with higher prevalence. Prevalence of high blood pressure was lower in the survey areas than the national prevalence reported in the 2008 Demographic and Health Survey (5.6 and 12.5 % for men and women, respectively). 14 Total national prevalence of diabetes mellitus for Egypt is 10.4% in the 20-79 age group (estimated for 2010; International Diabetes Federation Atlas, accessed on December 9, 2009). Furthermore, 75% of all cases are in the 40-79 age group. 26 Table 5: Diagnoses of key chronic illnesses Whole Male Female Sample Cancer 0.2% 0.1% 0.2% Diabetes 5.4% 5.0% 5.9% High blood pressure 7.6% 5.8% 9.4% Chronic diseases are more prevalent among those with higher economic status. Table 6 reports the prevalence of cancer, diabetes and high blood pressure among individuals in each of the five consumption quintiles. In general, we find higher prevalence among individuals in the highest consumption quintiles (i.e. the richest households). The reported prevalence rates in the highest consumption quintile are anywhere from 2 times higher (high blood pressure) to 10 times higher (cancer) than in the lowest consumption quintile. The pattern should be interpreted with caution, however, as it could be due to more frequent contact with medical practitioners among richer individuals. Table 6. Relationship between chronic illness prevalence and consumption quintile Consumption quintile First Second Third Fourth Fifth Cancer 0.04% 0.04% 0.21% 0.10% 0.40% Diabetes 2.62% 3.26% 5.89% 6.23% 8.69% High blood pressure 4.54% 5.65% 6.68% 9.08% 11.44% HEALTH CARE UTILIZATION, HEALTH EXPENDITURE, AND INSURANCE STATUS Given the self-reported health status of respondents, it is worth considering how much they utilized health care services and what type of care they sought. Utilization patterns In general, a very large fraction of individuals sought care when they were ill. 95 % of adults sought care when they were ill in the six months prior to the interview. This was similar for men and women, but on average adults in Menoufia were slightly more likely to have sought care when ill compared to adults in Alexandria (96% versus 93%). As Table 7 indicates, people also sought care very soon after they had a health problem. 85 % of those who went to a health facility or health care provider said they went within one day of having a health problem. Table 7: Time between problem and decision to go to facility Entire Male Female Sample Less Than 1 Day 85.0% 85.0% 85.1% 1 Day 3.0% 3.0% 3.0% 2-3 Days 6.5% 7.0% 6.1% Week 2.2% 2.6% 1.9% More Than Week 3.2% 2.3% 3.9% 27 Don't Know 0.1% 0.2% 0.0% Among all adults, nearly 45% visited a health facility in the past six months. 8 % of adults visited a health facility just once and 10 % just two times; the rest visited a health facility more frequently. Women were considerably more likely to visit health facilities. While 64 % of men did not visit a health facility at all in the past six months, this was the case for only 44 % of women. There are also strong age patterns in that older adults had a greater number of visits than younger adults. As Table 8 shows, there was not any significant variation in utilization by consumption quintile. This suggests that utilization of at least some type of health facility does not depend on economic status. Our results on usage of health care are somewhat consistent with the results in the 2008 Demographic and Health Survey (DHS). In the DHS, an average 8.1 % of Egyptian adults reported visiting a health care provider in the past 4 weeks, which is roughly consistent with 45 % visitation in the past six months. Table 8. Health facility visitation, by consumption quintile Consumption quintile First Second Third Fourth Fifth Visited health facility in past 6 mths 49.6% 47.3% 42.7% 43.8% 45.0% Times visited health facility 4.2 4.6 4.7 4.8 5.2 The main reasons for seeking care were fever, sore throat, cough, hypertension, diabetes, and to a lesser degree, dental issues. It is striking that some of these conditions are chronic while others are acute. For women, while the main reasons above remain relevant, other major reasons were pregnancy and reproductive health. Around 10 % of women had medical consultations for pregnancy, and 2-3 % of women had consultations for reproductive health and family planning issues. Only 0.2 % of men and women had consultations for a routine check-up, which suggests that the system is not providing sufficient levels of preventive interventions and health promotion. Figure 3 shows the distribution of main reasons given for utilizing health care in the past six months. 28 Figure 3: Main reasons given for utilizing health care in the past six months, by gender 16 14 12 10 8 6 Total 4 Male Female 2 0 Users and non-users The data allow us to explore what are the characteristics of those who use care vs. those who do not use any health care at all. First, it does not appear that a lack of health insurance is the driving factor. The fraction of adults who have health insurance is almost identical (35 %) in the group of health care users and non-users. As noted, males and younger adults are more likely to be non-users. Consistent with the relationship between age and usage, the data also indicate that those never married are more likely to be non-users. Somewhat surprisingly, there is no strong relationship between education and care usage, contrary to international experience. One likely factor influencing whether or not adults use health care is health status. It is likely that a larger fraction of non-users are in better health than health care users. The absence of objective clinical health data at the individual level prohibits a further analysis of this issue. In addition, those who do not use care may be unable to afford it. 33 % of non-users said there were occasions when money kept them from going to see a doctor. As Table 9 shows, there was a strong pattern in the relationship between one’s economic status and the likelihood that money was a barrier to seeking care. Those with lower economic status, as indicated by lowest consumption quintile, were much more likely to report money as being a barrier to utilization of health care services (49.89 %) compared to those in the highest consumption quintile (17.88 %). Men and women reported money as a barrier at roughly the same rate, as did individuals from different age groups. Table 9: Financial barriers to seeking health care, by consumption quintile Consumption quintile First Second Third Fourth Fifth Money was a barrier to seeing a doctor 49.89% 40.97% 30.93% 26.13% 17.88% 29 Drugs When people seek care at health facilities, they rarely receive medication. Only 20.5 % of adults who visited a health facility received medications. Medications were given more often in Menoufia than in Alexandria, and surprisingly, younger patients were more likely to have received medications than older patients. The most common reason why people do not receive medications, as reported by respondents, iss that there was no need for them. 45 % of adults reported this as the reason. However, 15 % of adults did report that the medications were out of stock. Many people bought medicines outside the health facility that they visited. 88 % of respondents reported buying them at a pharmacy. 13 % of respondents reported the health facility itself as being one of the places where they bought medications. Choice of facility The nature of one’s health problem plays a central role in determining the facility used. 30 Table 10 shows the breakdown of illnesses for which people sought care at the relevant facility. When the nearest health facility is visited, it is most often for family planning issues, followed by fever, hypertension and pregnancy. When the respondent reported visiting other facilities, the three most common reasons for the visit were hypertension, fever and sore throat. 31 Table 10: Illnesses for which nearest health facility or other health facility is chosen Facility type chosen Nearest facility Other facility Number who visited facility 1024 5494 Other 23.8% 41.9% Family planning 17.0% 3.5% Fever 12.3% 9.5% Hypertension 10.5% 9.9% Pregnancy 9.1% 5.2% Sore throat 8.5% 8.4% Diabetes 8.4% 7.4% Dentist 7.1% 6.2% Cough 6.7% 7.4% Vaccination 3.1% 0.5% Reproductive Health 3.0% 3.2% Difficulty breathing 2.3% 3.7% Skin issues 2.1% 2.5% Ear problem 1.2% 1.8% Eye problems 1.2% 4.1% Injury 1.0% 1.6% Diarrhea/Dysentery 0.9% 1.0% Routine check up 0.2% 0.4% Disability 0.1% 0.1% Cancer 0.0% 0.2% Hepatitis c 0.0% 1.1% The most frequently reported reason for choice of facility was perceived quality and type of service offered. Other factors that people reported as important were distance, trust of staff, and coverage by insurance. Interestingly, issues such as cheaper drugs and lower fees were not reported important. It is worth noting that people did not necessarily choose the health facility that was closest to them. In the entire sample, only 15.7 % went to the closest facility. However, those with worse economic status, as indicated by their employment status, were more likely to go to the closest facility. The most common providers of care were private doctors. Among all health care users, nearly 47 % had seen a private doctor in the past 6 months, as illustrated in figure 3. Surprisingly, the likelihood of visiting a private doctor was very similar across consumption quintiles. No other health care providers or health facilities received nearly as much visitation. The next most 32 visited health care providers were urban hospitals and rural health units, which were each visited by about 8 % of health care users. Other providers of health care – such as NGO clinics and hospitals, private hospitals, and pharmacies – were visited by at most 5 % of users. The biggest difference between men and women were in their usage of HIO clinics and hospitals and of private doctors. 8-9 % of men reported having visited an HIO clinics and HIO hospitals in the past 6 months, but only 1.5 % of women visited these facilities (likely reflecting the fact that housewives cannot be insured under HIO). Men were also more likely to visit private doctors than were women. Participants of the focus group discussions mentioned that the main advantages of private providers are availability of drugs, the better quality of care provided, and the options for after- hours care. Furthermore, users stated continuity of care as an important factor to prefer private over public providers, as the latter tend to change frequently, and few providers stay with one facility in the long run. This finding was confirmed in the facility survey and is analyzed in more detail in chapter 4 below. Figure 4: Type of facility for last visit, in percentage of users 0 0 0 0 0 All Users Alexandria 0 Menoufia 0 0 Most people travelled to the health facility or health care provider by walking (42 %) or using public transportation (45 %). The average travel time was slightly less than one hour. In Alexandria walking was slightly more common than public transportation, whereas the opposite was true in Menoufia. People generally went for health care during the mornings or evenings. Only 15.5 % of adult users went to the clinic during the afternoon shift. 33 Preferences and satisfaction about health services Over 95 % of respondents indicated a “positive� opinion (extremely satisfied or somewhat satisfied) on friendliness and availability of staff, qualifications of staff, cleanliness and comfort level and location of facilities, see Table 11. Cost and waiting time were not as highly rated, although even for these, 89-92 % of respondents had a “positive� opinion. There were no major differences in opinion between men and women or people of difference age groupsThere was also no significant difference in satisfaction rates between individuals from richer and poorer households. Table 11: Client satisfaction with health care services at facility last visited Whole How satisfied were you with the facility? Sample Extremely satisfied 75.5% Somewhat satisfied 20.3% Neutral 2.6% Somewhat dissatisfied 1.2% Extremely dissatisfied 0.4% Overall, people who visited a private doctor were more likely to be extremely satisfied with their last visit, at 84 %, see Figure 5. Furthermore, patients in Menoufia tended to be more satisfied with their health services than patients in Alexandria. Figure 5: Percentage of clients "extremely satisfied" with last visit, by type of clinic and governorate 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Private Mon Family Family All HIO clinic Alex Primary doctor Health Health Health Unit Center Care Unit 34 Enrollment in insurance and the Family Health Model Only 50% of household heads reported that they have payroll-based health insurance (HIO). 43% reported that they were registered in the co-payment based Family Health Model, although the fraction enrolled was considerably lower in Alexandria than in Menoufia. This is higher than the health insurance coverage of other family members, which stands at an average 35.6%. The coverage rate is slightly higher than the national average for Egypt, 27.6%, as reported in the 2008 DHS. This may reflect that Alexandria and Menoufia, in Lower Egypt, are economically more developed than several other governorates, notably in Upper Egypt. The main reason given for not having health insurance was lack of affordability (66.6% of uninsured), followed by lack of knowledge about how to enroll (21.7%) and lack of perceived need (11.2%). Table 12 supports the importance of affordability and shows that the percentage of individuals with health insurance is considerably lower in the poorest consumption quintiles (29.2% in the lowest consumption quintile, compared to 45.5% in the highest). As should be the case, however, registration in the Family Health Model appeared to be inversely related to household economic status – as individuals in poorer households were more likely to be registered. The fact that more than one-fifth of the sample report they lack sufficient information on how to enroll in a health insurance program suggests a weakness in the communication and outreach about available services. Table 12: Percentage of adults with insurance, by consumption quintile Consumption quintile First Second Third Fourth Fifth Has health insurance 29.2% 31.2% 34.1% 37.1% 45.5% Registered in FHM 49.8% 45.9% 39.8% 39.1% 30.8% 35 4. THE SUPPLY OF PUBLIC PRIMARY HEALTH CARE: AVAILABILITY AND QUALITY OF CARE This chapter examines the public primary care facilities in Alexandria and Menoufia with respect to availability and the quality of (i) infrastructure and equipment, (ii) human resources, and (iii) care. We look at the number of facilities per population, the medical and non-medical equipment of the facilities including drugs, and the staff working in these facilities. Differences are highlighted between reformed and non-reformed, as well as between accredited and non accredited facilities, in order to gauge the influence of the HSRP. AVAILABILITY OF INFRASTRUCTURE AND EQUIPMENT AT THE FACILITY LEVEL Coverage with primary health care The population of Alexandria and Menoufia – 4.1 million and 3.3 million people respectively15 – are served by 362 public primary health care facilities16 at the time of the survey. Facilities per population Over a third of all facilities in Alexandria and Menoufia and almost half of the facilities in rural areas have a catchment area of 10,000 people or less. As can be seen in figure 5, another 40 % of facilities are serving a catchment area between 10,000 and 30,000 people. Approximately 20 % of all facilities in the two observed governorates have a catchment area comprising more than 40,000 people, out of which 10 % serve more than 100,000 people. These are mainly clinics in urban areas, predominantly in urban Alexandria. Figure 6: Size of catchment area, in percentage of facilities 15 Egypt Central Agency for Mobilization and Statistics 16 This number does not include hospitals and private clinics. 36 Services Offered 9 out of 10 facilities offer services for sick children, post- and antenatal, diabetes and CHD services, amongst others. In facilities run by the MOH, these services are offered at around 95% of all facilities, and near universally at reformed facilities, see Table 13. Table 13: Health services offered: percentage of facilities that treat selected health problems All MOH Re- Non Re- Accredited Not Alexandria Menoufia Urban Rural only formed formed accredited Child 95.3 99.7 99.6 100 99.6 100 99.1 100 98.8 100 health services Maternal 91.7 96.7 98.2 89.1 98.3 91.6 91.6 99.1 92.9 97.8 Health Services Diabetes 92 94.8 98.5 76.4 99.2 82.8 86.9 98.6 91.7 95.9 services CHD 93.7 94.8 98.9 74.5 99.6 81.6 86.9 98.6 91.7 95.9 services The availability of services does not mean that these are taken up by the population. For example, the number of deliveries carried out in family health facilities is very small. In the 12 months prior to the survey only 27 facilities carried out deliveries, amounting to a total of 2259 between all facilities. This accounts for only a minimal share of births in Alexandria and Menoufia. Most households prefer to go to a private doctor for childbirth (33%), followed by an urban hospital (21%), private hospital (16%) and a provider home visit (14%). Rural health units (2%) and urban health units (0.5%) rank even below Traditional Healers (3%). When discussing priority services in the focus groups, over a quarter of respondents indicated satisfaction with the health services available at their public health facility. Those who commented on the need for more services specified dentists, orthopedists, otolaryngologists, surgeons, emergency care, and more emphasis on diagnostics. Infrastructure, medical and non medical supplies While facilities may officially offer several services, the availability of basic infrastructure and medical supplies will determine when and whether these services can actually be provided. This section explores therefore  the availability of non-medical infrastructure, such as electricity, sanitary infrastructure and waiting areas for clients; and second,  the availability of basic medical supplies to conduct treatments for diabetes, CHD/hypertension, antenatal and children’s health care services. Non medical infrastructure The facility survey collected information on basic infrastructure observed on the day of the visit in all 362 facilities. Almost all facilities (98 %) have access to electricity. Working phones were available in 76.2 % of the facilities and a water outlet is generally available in 93 % of all facilities. Only two thirds of the facilities, however, reported that water is always available. While almost all facilities have a toilet which can be used by patients, the interviewers found 37 that in only 30.2 % of the observed facilities there is soap and only 3.5 % of the facilities offer toilet paper in their bathrooms, see Figure 7. Figure 7: Availability of non medical infrastructure Electricity Working phone to make calls outside 100 100 95 90 90 80 85 70 80 60 75 50 70 40 65 30 60 20 55 10 50 0 Water always available Toilet paper in the toilet 100 100 90 90 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 Note: In % of 368 facilities Many of the supplies were found more regularly in reformed than in non reformed facilities; the same notion can be made when comparing accredited with non accredited MOH clinics. While 83.4% of accredited facilities possess a phone with which they can make outside calls, this was only found in 47.1% of non accredited facilities. Similarly, while in 30.6% of accredited facilities soap was found in the toilet, this was the case in only 25.7% of non accredited facilities. There is no distinct difference when comparing facilities in Alexandria with those in Menoufia. While facilities in Alexandria seem to be better equipped in terms of sanitary supplies, the reverse is true for communication devices. When looking at the urban/rural strata, however, facilities in urban areas are better supplied with non medical infrastructure than facilities in rural areas. Only about half (53%) of all facility managers in MOH clinics reported that they have all heavy infrastructure they currently need. The number was higher in Menoufia (55.2%) than in Alexandria (47.7%), and much higher in accredited facilities (59%) than in non accredited facilities (35.6%). Apart from accounting for the availability of non medical infrastructure, the surveyors, all trained doctors, were asked to rate the quality of the building, electrical, 38 plumbing/water/wastewater and medical equipment. Some of the results of these ratings are summarized in Table 14. In terms of quality, two thirds or more of buildings were rated good whereas this applies to less than half of the water infrastructure and medical infrastructure. There are no significant differences in the perception of quality of infrastructure by the interviewers between facilities in Alexandria and Menoufia. MOH facilities had the highest rate of those in the sample including all facilities, including the HIO facilities. Table 14: Rating of quality of infrastructure, in percentage of interviewers (% of 362 facilities) Good Satisfactory Poor Very poor Building 69.9 19.3 10.2 0.8 Electrical 69.9 20.7 9.1 0.3 Plumbing/water/wastewater 47.2 27.9 21.3 3.6 Medical equipment 48.9 38.1 12.1 0.8 MOH only Good Satisfactory Poor Very poor Building 71 18.9 9.15 0.91 Electrical 71.3 18.9 9.45 0.3 Plumbing/water/wastewater 49.1 25.6 21.65 3.66 Medical equipment 49.7 36.6 12.8 0.9 Alexandria Good Satisfactory Poor Very poor Building 65.6 25.8 7.8 0.8 Electrical 72.7 20.3 6.3 0.8 Plumbing/water/wastewater 49.2 34.4 13.3 3.1 Medical equipment 50 32.8 16.4 0.8 Menoufia Good Satisfactory Poor Very poor Building 71.8 15.8 11.5 0.9 Electrical 68.4 21 10.7 0 Plumbing/water/wastewater 46.2 24.4 25.7 3.9 Medical equipment 48.3 41 9.8 0.9 Medical Infrastructure and Supplies This section describes the availability of medical supplies to provide treatment for  Diabetes Mellitus,  CHD/hypertension,  Integrated Management of Child Illness (IMCI), and 39  Antenatal care Only facilities that currently have patients seeking the respective services were surveyed. Table 15 summarizes the number of facilities that offer each of the services at the time of data collection. Table 15: Facilities with patients for selected categories of services Facilities that Total MOH HIO Alexandria Menoufia Reformed Non Accredited Not provide (MOH only) (MOH reformed accredited only) Antenatal Health 332 317 12 98 219 268 49 237 80 Services Diabetes 333 311 19 93 218 269 42 239 72 CHD/Hypertension 339 311 25 93 218 270 41 240 71 Child Health 345 327 13 106 221 272 55 240 87 Services Supplies for Diabetes Mellitus treatment 40 Table 16 summarizes the availability of supplies needed for Diabetes Mellitus treatment. Only items that could be observed by the interviewer are counted for as available. The list includes equipment that is, according to international standards, considered essential for a satisfactory treatment of Diabetes Mellitus patients. The list also features supplies that are specific to Diabetes Mellitus treatment, such as insulin ampoules, as well as supplies that are imperative to conduct a basic medical check-up, for instance a stethoscope or hand washing items. Many facilities lack basic supplies to conduct Diabetes treatments. Overall, only 31.2 % of all facilities in Alexandria and Menoufia possess a working ECG, 40.5 % a machine to measure blood pressure ad hoc, 48.4 % could find a working reflex hammer, 30.6 % have insulin ampoules on stock, and 34.8 % have a minimum of 5 2 or 3ml disposable syringes. The numbers are similarly low, or even lower, for aides to educate patients, such as MOH guidelines (observed in 33 % of all facilities) or leaflets for patients (15.3 %) and other visual aids for teaching patients (7.8%). Half or more of the facilities can show at least one functioning item or unit of glucophage (53.8 %), clean gloves (79.9 %), a safety box for needles (76 %), decontamination solution for clinical equipment (78.7 %), waste receptacle (82.6 %) and hand washing items (67.9 %). Items that were present at almost every facility are stethoscope (99.7 %) and a blood pressure apparatus (99.4 %). Almost all supplies for Diabetes Mellitus treatment are more likely to be found in reformed facilities than in non reformed facilities. The differences are particularly significant when comparing availability of ECG (33.8 versus 14.3 %), machine to measure blood sugar ad hoc (46.5 versus 9.5 %), metformin (41.3 versus 19.1 %), insulin ampoules (33.1 versus 14.3 %) and clean gloves (84.8 versus 54.8 %). The same pattern holds when comparing currently accredited with currently not accredited facilities, however, the differences are less distinct. 41 Table 16: Availability of supplies for Diabetes Mellitus treatment (% of 333 facilities) Observed availability of items All MOH Alexandria Menoufia Urban Rural Reformed Non Accredited Not only reformed Accredited ECG 31.2 31.2 23.7 34.4 29.9 31.6 33.8 14.3 35.6 16.7 Machine to 40.5 41.5 19.4 50.9 31.2 44.9 46.5 9.5 45.2 29.2 measure blood sugar at hoc Stethoscope 99.7 99.7 98.9 100 98.7 100 99.6 100 99.6 100 Blood pressure 99.4 99.4 98.9 99.6 98.7 99.6 99.3 100 99.2 100 apparatus Reflex hammer 48.4 50.2 65.6 43.6 70.1 43.6 53.2 31 55.6 31.9 Metformin 37.9 38.3 31.2 41.3 37.7 38.5 41.3 19.1 43.1 22.2 Insulin ampoules 30.6 30.6 26.9 32.1 35.1 29.1 33.1 14.3 35.6 13.9 Glucophage 53.8 55.3 30.1 66.1 41.6 59.8 55.8 52.4 58.2 45.8 Clean gloves 79.9 80.7 80.7 80.7 84.4 79.5 84.8 54.8 85.8 63.9 Safety Box for 76 79.4 71 83 79.2 79.5 79.6 78.6 80.8 75 Needles 5 or more 2 or 3 34.8 35.7 22.6 41.3 22.1 40.2 32.3 57.1 33.1 44.4 ml disposable syringes Decontamination 78.7 80.4 75.3 82.6 79.2 80.8 81 76.2 81.2 77.8 solution for clinical equipment Waste receptacle 82.6 85.2 90.3 83 85.7 85 85.5 83.3 88.3 75 with plastic liner and working foot control Hand washing 67.9 70.7 74.2 69.3 71.4 70.5 71.7 64.3 74.5 58.3 items (soap and towels) Water for hand 85.3 87.5 93.5 84.9 92.1 85.9 89.7 71.4 93.7 66.7 washing A copy of the 33 35.4 39.8 33.5 46.8 31.6 39 11.9 41 16.7 MOH Guidelines on Management for Diabetes Mellitus Leaflet for 15.3 15.4 17.2 14.7 20.8 13.7 17.1 4.8 18 7 patients with basic education on DM Other visual aids 7.8 7.7 8.6 7.3 6.5 8.2 7.8 7.1 8.4 5.6 for teaching patients Supplies for Hypertension/CHD treatment 42 Table 17 summarizes the availability of supplies for Coronary heart disease/hypertension treatment in a total of 339 facilities that currently treat patients for these issues. The items observed are checked against the requirements of the essential drug list for hypertension/CHD enacted with the HSRP. Supplies specifically needed for CHD/hypertension treatment can be found in very few facilities; only 4.1% of the facilities had fibrates, 2.4% statins and 23.3% blood thinning medication. Two thirds of all facilities treating patients for hypertension/CHD had antihypertensive drugs on stock. The numbers are also very low for information material on CHD/hypertension; only 26.8% of facilities could show the interviewer a copy of the MOH guidelines on hypertension, 21.1% the guidelines on CHD, the leaflets for patients on hypertension (11.2% of facilities) and CHD (9.1%) were available in even fewer of the facilities. While still on an overall low level, the availability of supplies for CHD/hypertension treatment is significantly higher in accredited facilities as compared to non accredited facilities, as well as in reformed facilities compared to non reformed ones. While at least one third of accredited facilities possess some information materials on both hypertension and CHD, these were available in less than 10% of non-accredited facilities. The same could be observed for blood thinning medication which was present in 26.3% of accredited facilities compared to 9.9% in non accredited facilities, and in 23.7% of reformed facilities but only 14.6% of non reformed facilities. There is no significant difference in the availability of supplies for CHD/hypertension treatment between facilities in Alexandria and Menoufia or between urban and rural facilities, except for the availability of blood thinning medications, which seem to be available on a much broader scale in Menoufia compared to Alexandria, and, subsequently, in rural compared to urban clinics. 43 Table 17 Availability of supplies for Hypertension/ Coronary Heart Disease treatment (% of 339 Facilities) All MOH Alex M’fia Urban Rural Reformed Non Accredite Not only reformed d accredited Blood pressure 99.7 99.7 100 99.6 100 99.6 99.6 100 99.6 100 manometer Stethoscope 99.4 99.4 100 99.1 100 99.1 99.3 100 99.6 98.6 Anti-hypertensive 62.5 65 45.2 73.4 59.7 66.7 66.7 53.7 68.8 52.1 Drugs Fibrates 4.1 2.3 1.1 2.8 1.3 2.6 2.6 0 2.9 0 Statins 2.4 0.32 1.1 0 0 0.4 0.4 0 0.4 0 Blood thinning 23.3 22.5 2.2 31.2 3.9 28.6 23.7 14.6 26.3 9.9 medication Clean gloves 77.3 78.5 77.4 78.9 83.1 76.9 82.2 53.7 83.3 62 Safety Box for needles 74.9 79.1 75.3 80.7 84.4 77.4 79.3 78 80.4 74.7 34.5 35.7 24.7 40.4 24.7 39.3 32.2 58.5 32.9 45.1 5 or more 2 or 3 ml disposable syringes (w/ 21 gauge needles) Decontamination 79.4 81.7 77.4 83.5 83.1 81.2 81.6 80.5 81.7 81.7 solution for clinical equipment Waste receptacle with 81.4 85.2 91.4 82.6 87 84.6 85.6 82.9 87.9 76.1 lid and plastic liner Hand-washing items 67.9 70.4 73.1 69.3 70.1 70.5 71.5 63.4 74.2 57.8 (soap and water) Water for hand 85.8 87.8 95.7 84.4 94.8 85.5 90 73.2 93.8 67.6 washing Copy of the MOH 26.8 29.3 28 29.8 32.5 28.2 32.6 7.3 35 9.9 Guidelines on Hypertension Copy of the MOH 21.2 23.2 23.7 22.9 27.3 21.8 25.9 4.9 27.5 8.5 Guidelines on CHD Leaflet for patient with 11.2 11.9 10.8 12.4 10.4 12.4 12.6 7.3 13.3 7 basic education on Hypertension 9.1 9.7 9.7 9.6 10.4 9.4 10.4 2.4 11.3 4.2 Leaflet for patient with basic education on CHD/ heart pain Other visual aids for 4.7 4.8 6.5 4.1 6.5 4.3 4.8 4.9 5.4 2.8 teaching patient 44 Supplies for Integrated Management of Child Illnesses (IMCI) 45 Table 18 summarizes the availability of essential supplies to conduct basic child care services. A total of 347 of the 362 facilities in the sample reported that they do currently offer sick child services. Supplies were counted as available only if the interviewer could observe at least one functioning item. The benchmark list of supplies was again extracted from the guidelines enacted with the HSRP. Almost 9 out of 10 facilities are in possession of basic measurement instruments for treating children such as an infant scale (90.1% of all facilities), child scale (88.4%) or a functioning thermometer (93.3%). Other basic supplies needed for sick child treatment are less frequently available. Only 71.6% of all facilities have an oxygen cylinder, 69.9% a nebulizer, 54.2% do have a functioning light to look into the child’s throat and 51.6% of all facilities have a cup and spoon ready for use. Materials to educate mothers about child health issues could only be found in half of the facilities, or less (34.2% of facilities could show to the interviewers an IMCI mother card). Supplies for sick child treatment are more frequently available in MOH clinics compared to HIO clinics. Furthermore, accredited facilities are more likely to be in possession of supplies for child health services than non accredited facilities, in particular regarding specific child care instruments such as infant child scale (available in 91.7% of accredited facilities versus 82.8% of non accredited ones), or thermometer (96.7% versus 87.4%). There is a similar trend that can be noted when comparing reformed and non reformed facilities; however, the difference is less distinct. When comparing facilities in Alexandria and Menoufia the pattern becomes less clear. Many of the supplies needed to provide basic child care services are more frequently available in Menoufia than in Alexandria. There is a 10% or more difference in the availability of timer/watch (69.2% in Menoufia versus 55.6% in Alexandria), oxygen cylinder (91.9 versus 35.9%), Nebulizer (90.5 versus 33%) and cup and spoon (59.3 versus 41.5%). On the other hand, clinics in Alexandria are better equipped with supplies such as hand washing items, as well as documentation for both providers and patients. 46 Table 18: Availability of Supplies for Sick Child treatment (% of 347 Facilities) All Facilities MOH Alexandria Menoufia Urban Rural Reformed Non Accredited only Reformed Infant Scale 90.1 91.4 90.6 91.9 90.4 91.8 91.2 92.7 91.3 Infant Scale if 98.8 98.7 98.1 99.1 97.9 99.1 99.2 96.2 99.1 functioning (%of those available) Child Scale available 88.4 89.3 89.6 89.1 96.4 86.9 90.1 85.5 91.7 Child Scale is 98.7 99 98.1 99 97.9 99.1 99.6 95.5 99.6 functioning (% of those available) Availability of Thermo- 93.9 94.2 93.4 94.6 95.2 93.9 95.2 89.1 96.7 meter Timer/ Watch with 63.5 64.8 55.6 69.2 59 66.8 68 49.1 69.2 second hand is available Oxygen cylinder and 71.6 73.7 35.9 91.9 51.8 81.4 75 67.3 75.8 regulator is available Nebulizer is available 69.9 71.9 33 90.5 45.8 80.7 72.8 67.3 73.4 Light for looking in 54.2 54.4 65.1 49.3 71.1 48.8 56.3 55.5 57.5 throat is available Wooden tongue 93.9 94.1 99.1 91.9 100 92.2 94.9 90.9 95 depressor is available Jar for ORS is available 33.3 34.3 37.7 32.6 41 32 36 25.5 37.5 Cup and spoon are 51.6 53.5 41.5 59.3 45.8 56.1 55.1 45.5 57.1 available Height measuring tool 77.4 80.4 83 79.2 86.7 78.3 80.9 78.2 81.3 is available Waste Receptacle is 82 84.7 86.8 83.7 81.9 85.7 85.7 80 88.3 available Water receptacle is 94.4 94.2 94.6 94.2 96 93.8 93.6 97.7 92.9 functioning (% of those available) Availability of Hand 68.1 70.3 78.3 66 72.3 69.3 71 65.5 73.3 washing items Availability of Water 88.1 89.6 98.1 85.5 96.4 87.3 91.9 78.2 95.4 for hand-washing Medical Protocols for 46.4 48.9 50 48.4 63.9 43.9 54.8 20 56.3 treating child illnesses is available IMCI Chart Booklet is 54.2 56.6 63.2 53.4 67.5 52.9 57 54.5 58.8 available IMCI counseling cards 53.3 56 49.1 59.3 56.6 55.7 57.7 47.3 58.3 are available IMCI mother cards (to 34.2 35.5 46.2 30.3 48.2 31.1 34.2 41.8 34.2 give to caretaker) Other visual aids for 31.1 32.4 33 32.1 38.6 30.3 33.8 25.5 34.6 teaching caretaker Supplies for Antenatal Care Most of the 332 facilities in the sample offer antenatal services at the time of the survey. The essential supplies for antenatal care correspond to basic items that are essential for the treatment of other basic health needs as well, and therefore were already accounted for in the paragraphs above. 47 Most of the facilities seem well equipped to conduct basic antenatal services. A table for ANC exam could be found in 93.5% of all facilities. Further, over two thirds of facilities had available supplies such as a spotlight source (70.7%), clean gloves (80.4%), a safety box for needles (80.1%), or decontamination solution for clinical equipment (78.3%). Surprisingly, a functioning table for ANC exam was more widely available in non accredited and non reformed facilities (95% versus 91.5% non-accredited versus accredited, and 95.9% non- reformed versus 91.8% reformed facilities). Furthermore, tables for ANC exams are available in all facilities surveyed in Alexandria and in 89% of facilities in Menoufia. Table 19: Availability of supplies for antenatal care (% of 332 Facilities) Observed availability of items All MOH Alexandria Menoufia Urban Rural Reformed Non Accredited Non- Facilities only Reformed accredited Spotlight 70.7 69.9 78.4 66.2 83.1 65.7 72.7 55.1 73.3 60 source (flashlight or examination light accepted) Table for ANC 93.5 92.4 100 89 98.7 90.4 91.8 95.9 91.5 95 exam Clean gloves 80.4 80.6 80.4 79.9 85.7 78.2 82.7 65.3 82.6 72.5 Safety box for 80.1 81.3 72.2 85.4 76.6 82.9 81.3 81.6 81.8 80 needles 5 or more 2 or 45 45.3 26.8 53.4 24.7 51.9 41.2 67.4 41.1 57.5 3 ml disposable syringes (w/21 gauge needles) Decontaminati 78.3 78.5 74.2 80.4 79.2 78.2 79.8 71.4 80.8 73.8 on solution for clinical equipment Waste 84.9 86.1 90.7 84 89.6 84.9 86.9 81.6 89.4 76.3 receptacle with plastic liner and working foot control Hand washing 70.4 71.5 79.4 68 77.9 69.5 72.7 65.3 74.6 62.5 items (soap and towel) Water for 90.3 90.5 97.9 87.2 94.8 89.1 92.5 79.6 95.3 76.3 hand washing Availability of Drugs Table 20 shows the availability of essential drugs at 360 facilities in Alexandria and Menoufia. The table covers only the most essential drugs necessary to conduct primary health care services, based on the essential drug list of the MOH, enacted with the HSRP. Only drugs that could be observed by the interviewers were accounted for as available. 48 The availability of drugs shows a worrying pattern. Only a limited number of drugs of this compacted essential drug list for primary care are regularly available - many others are not. In particular, the availability of drugs for maternal health (folic acid, ergometrine/ oxytocin under 50%) is alarming, as it means many FHUs cannot offer ante- or peri-natal care. Absence of drugs for treating chronic non-communicable diseases is another problem. There are few anti- hypertensives, anti-diabetics, and drugs for high cholesterol and nothing for stomach/duodenal ulcer and reflux disease. Over two thirds (66 %) of facility managers reported that they use generics, with 3 % using original brand products, and 30 % a combination of the two. Interestingly, the decision on whether a facility is using brand or generic drugs is made by different people in different facilities: 27 % of facilities report the decision is made by the pharmacist, 24 % by the nurse and 14 % of facilities report that the physician is making the call. One third of all facilities replied that the decision of whether to use generics or brands is entirely made at the district level i.e. at the health directorate. Several facility managers have mentioned that the procurement of essential drugs is a major issue of concern. In particular, many have complained that the quantity ordered by the facility seldom reaches the facility but that, in many cases, only a part of the drugs demanded by the facility are delivered. This is a major risk as facilities may not be able to guarantee a continued supply of medication to their patients, and patients may need to look for their drugs elsewhere or have to substitute their medication with a different drug. During the focus group discussions and the in-depth interviews, providers and clients alike have mentioned that after the HSRP, every clinic now has a pharmacy operated by a pharmacists, which, in theory, improved service quality. However, it has been repeatedly mentioned that doctors prescribe medications that are not available in the pharmacy of the clinic – and patients are forced to purchase their drugs at a different, mostly private, pharmacy. 49 Table 20: Availability of drugs (condensed list)17 All MOH Alexan Menoufia Urban Rural Re- Non Accredited Not -dria formed reformed accredited Amoxicillin Oral 95.6 97.6 98.1 97.3 98.8 97.1 97.4 98.2 97.5 97.7 Quantity of 90.2 89.7 81 94 95.2 87.8 92.5 75.9 94 77.7 Amoxicillin Oral in stock matches quantity in books Aspirin Oral 74.4 73.2 88.8 65.6 88.1 68 73.3 72.7 75.1 67.8 Doxycycline 7.8 4.9 11.2 1.8 8.3 3.7 3.3 12.7 3.7 8.1 Quantity of 85.7 75 75 75 100 55.6 88.9 57.1 88.9 57.1 Doxycycline in stock matches quantity in books Ergometrine/ 3 3 1.9 3.6 2.4 3.3 2.6 5.5 2.9 3.5 methergine Erythromycin 67.5 68.3 62.6 71 69.1 68 73.6 41.8 75.1 49.4 Folic acid 49.4 45.7 58.9 39.4 66.7 38.5 50.2 23.6 52.3 27.6 Iron 82.5 81.7 92.5 76.5 89.3 79.1 81.7 81.8 83.4 77 Multivitamins 91.1 91.5 86.9 93.7 88.1 92.6 93.8 80 94.6 82.8 Paracetamol Oral 92.8 93 80.4 99.1 86.9 95.1 96 78.2 96.3 83.9 Paracetamol Syrup 85 89.9 81.3 94.1 88.1 90.6 93 74.6 93.8 79.3 120 mg. Paracetamol Drops 3.3 3.7 4.7 3.2 4.8 3.3 4 1.8 2.9 5.8 100mg Ibuprofen Syrup 34.4 34.2 25.2 38.5 21.4 38.5 33 40 34 34.5 100 mg Ibuprofen tablets 8.9 7.9 11.2 6.3 13.1 6.2 8.8 3.6 7.9 8.1 400 mg Penicilin Oral 9.2 9.5 21.5 3.6 11.9 8.6 9.5 9.1 7.9 13.8 Tatracycline Oral 62.5 62.5 65.4 61.1 63.1 62.3 61.9 65.5 61.8 64.4 Vitamin A low dose 80.3 85.4 89.7 83.3 89.3 84 85.4 85.5 87.6 79.3 (25k or 30k iu) Oral Rehydration 62.2 61 91.6 46.2 82.1 53.7 62.6 52.7 63.9 52.9 Salts Quantity of Oral 91.1 90 82.7 97.1 94.2 87.8 94.7 62.1 96.1 69.6 Rehydration Salts in stock matches quantity in books Metformin 850mg 93.3 95.1 95.3 95 95.2 95.1 96.7 87.3 96.7 90.8 Bezafibrate 80mg 6.4 0.9 2.8 13.1 3.6 11.5 1.1 0 1.2 100 Warfarin 2g 8 1.5 4.7 13.1 4.8 0.4 1.5 1.8 1.7 1.2 Warfarin 5mg 12.5 6.4 16.8 1.4 17.9 2.5 7.3 1.8 8.3 1.2 Ergometrine/ 46.7 48.2 40.2 52 38.1 51.6 46.5 56.4 46.1 54 oxytocine Antibiotic Eye 44.7 44.8 72 31.7 69 36.5 45.1 43.6 46.1 41.4 Ointment Benzyl Penicilin 1.4 0.6 1.9 13.1 1.2 0.8 10.6 3.6 12 2.3 (Procaine) Diazepam 6.7 7.3 1.8 10 3.6 8.6 8 5.5 7.5 6.9 Neparin Na 5000u 70.6 74.1 61.7 80.1 66.7 76.6 75.8 65.5 77.2 66.5 Saline 83.3 88.7 73.8 95.9 67.9 95.9 88.6 89.1 90 85.1 Dextrose and Water 17.2 18 13 20.4 20.2 17.2 19.1 12.7 19.5 13.8 17 For a table with results on the full essential drug list, see Annex 3. 50 Procurement of medical and non-medical equipment According to 61% of facilities, there is no standard procedure to purchase small medical supplies, including drugs. On the one hand, as part of the accreditation process, every facility contracts with the department of engineering in the health district which is responsible for repairing small equipment such as pressure cuffs or stethoscopes. However, the department of engineering takes action only when notified by the facility. In addition, the Family Health Fund (FHF) conducts a monthly evaluation of the existence of a maintenance committee in the unit. In reformed facilities, small and medium equipment is supposed to be financed by the facilities’ so-called ‘Service Improvement Fund’ (SIF), which is funded through patient co-payments18. However, due to the low number of visits, the resources in the SIF are often limited and thereby delay repairs and procurement. In qualitative interviews, facility managers have stated that private funds of staff are often used to pre-finance missing or dysfunctional infrastructure and equipment. Furthermore, in reimbursing any expenditure for equipment, the facilities are bound to accounting procedures of the Ministry of Finance, which can be cumbersome at times. For heavier and more expensive equipment, procurement decisions are made at the higher level, and differ between Alexandria and Menoufia. In Menoufia, a regional repair center is in charge of repairing and replacing heavy infrastructure. The request by the facility needs to be approved by the financial director at the district level. When asked about the average waiting time for heavy equipment, 16% of facilities reported they wait 2 weeks or less; while 40% noted they wait over one year until a requested piece of equipment is delivered. Moreover, a fourth of all facilities reported they do not request heavy infrastructure. During the focus group discussions, several providers mentioned that they rely on donations from the community or staff to procure supplies. HUMAN RESOURCES Health workers and other staff form the backbone of service delivery. Employees in health facilities in Alexandria and Menoufia are therefore the principal element in the delivery of health services to people and the quality of health care. At the same time, health workers are an important factor in the budget of the MOH, and their allocation affects the sustainability of health services. This chapter examines the number of staff allocated to health facilities, their qualifications, experience and other socio-demographic characteristics. It also reports how many staff are regularly present at the facility, and which factors may play a role in staff presence and absence patterns. Allocation of Human Resources Institutional responsibilities Decisions on staffing are made at an intermediate level of the administration, with no decision- making authority at the facility level. In both Alexandria and Menoufia decisions regarding the target number of health workers by district (hiring and dismissal of staff) are made at the governorate level. This target number is then given to decision-makers at the district level who 18 36% of visit co-payments and 40% of enrolment fees go into the Service Improvement Fund. 51 allocate the staff to facilities in their respective districts. According to the guidelines of the Family Health Model (introduced with the HSRP), reformed facilities are informed about any changes in workforce levels, in order to guide the systematic distribution of staff to different facilities. No decisions about staffing are made at the facility level. When asked about which administrative level was the most important to decide on the number of staff at each facility, two thirds of facility managers report that any decision on hiring involves an agreement from the Ministry. Almost 20% of all facilities believe that decisions on staffing are made by the MOH alone, while 18% report decisions are made by the MOH in conjunction with the directorate and 19% that decisions can be made at all three; the Ministry, directorate, or district level. 20% of all facilities find that staffing decisions are made at the district level alone. The district decides if and what training staff receive. In addition, depending on the level of training and the grade of staff, the Ministry and the Governorate levels also have to approve of training for employees. For some training the facility can decide whether or not it allows its staff to participate in it. Staff numbers and capacity utilization On average, there are 50 employees per facility in all clinics in Alexandria and Menoufia. This number is almost the same whether or not HIO clinics are included in the sample. The average is lower in Menoufia, where the mean facility has 45 employees compared to just over 50 in Alexandria. It is also significantly lower in rural facilities, with a bit over 19 employees on average, than in urban areas, where the typical facility has almost 66 employees. There is also quite a significant difference in the number of employees between reformed facilities and non reformed facilities (50 employees compared to 34, respectively) and between accredited facilities and not currently accredited facilities (50 versus 38). There is one public primary health worker for 665 inhabitants in Alexandria and Menoufia. This ratio includes only public primary health facilities, which cover about 50% of all visits to health units. Comparing the numbers between Alexandria and Menoufia shows that there are almost four times as many health workers available to the population of Menoufia than for Alexandria. This is particularly striking as the difference between facilities serving urban areas and those in rural areas is less distinct. Furthermore, reformed and accredited facilities serve more beneficiaries per health worker than non accredited facilities. Most staff are currently under-utilized. For each public primary health worker there are on average only 90 consultations per year for sick children, antenatal, postpartum, diabetes and hypertension consultations combined. This number is lower, at 87, for MOH run clinics. There are also fewer consultations per health worker in Menoufia than there are in Alexandria. There are almost twice as many consultations per health worker in facilities in rural clinics compared to facilities in urban clinics. Furthermore, the ratio is higher in reformed clinics than in non reformed clinics and in accredited clinics compared to not accredited clinics. 52 Table 21: Mean number of Employees All MOH Alexandria Menoufia Urban Rural Reformed Non Accredited Not only reformed accredited Total Number of 48.8 46.8 50.2 45.2 65.7 19.4 49.4 33.9 50 37.9 Employees by facility Inhabitants 1170. per Health 665.6 607.6 1193.4 336.0 695.6 622.1 454.1 618.1 556.1 3 Worker Consul- tations per 90.1 87.3 96.6 79.6 93.4 168.2 85.3 78.7 85.5 81.5 Health worker Most facilities have between 20 and 40 employees (41.7%) or between 40 and 60 employees (24.9%). There are more very small facilities (less than 20 employees) than there are very large ones (more than 140 staff). There is more variation in facility size in Alexandria than in Menoufia, and in urban compared to rural facilities. Furthermore, many non reformed facilities are very small (over 20% have less than 20 staff). Table 22 Share of facilities by number of Employees, in percent of total facilities Total Alexand Menou Urban Rural Reform Non- All HIO Accredit Not ria fia ed Reform MOH ed accredit ed ed <20 8.29 16.41 3.85 6.84 8.98 4.01 21.59 8.56 3.45 3.72 17.50 20-40 41.71 28.91 48.72 19.66 52.24 44.16 34.09 44.65 17.24 42.98 39.17 40-60 24.86 21.09 26.92 24.79 24.90 27.37 17.05 25.38 24.14 28.51 17.50 60-80 12.98 12.50 13.25 19.66 9.80 13.50 11.36 11.62 31.03 13.64 11.67 80-100 4.14 7.03 2.56 9.40 1.63 4.01 4.55 3.67 10.34 3.72 5.00 100- 3.87 7.03 2.14 7.69 2.04 3.65 4.55 3.06 10.34 3.72 4.17 120 120- 1.38 2.34 0.85 3.42 0.41 1.46 1.14 1.53 0.00 1.65 0.83 140 >140 2.76 4.69 1.71 8.55 0.00 1.82 5.68 1.53 3.45 2.07 4.17 The number of visits per day in different facilities varies a lot, and is not at all proportional to the number of employees. As a proxy for each facility’s caseload, Figure 8 looks at the number of daily visits for chronic diseases in each facility and compares the number of visits with the number of employees in each facility. The graph shows the concentration of facilities with staff numbers between 20 and 60, and a few facilities with more staff. There are a number of facilities that have to cope with a rather high volume of visits per day and, on the other hand, there are facilities where 60 or more staff deal with less than 50 patients for chronic diseases on a given day. 53 Figure 8 Daily visits for chronic diseases, number of employees in facility 200 150 100 50 0 0 20 40 60 80 Number of employees in this facility When asked about their opinion on the overall level of staff in the in-depth interviews and focus group discussions, most facility managers reported satisfaction with the number of employees. However, answers tended to distinguish between medical personnel and administrative employees. A greater proportion of respondents were satisfied with the number of doctors than with any other type of employee. The absolute number of staff does not reveal whether or not human resources are allocated efficiently. Two other factors are equally important:  First, is the composition of staff, the mix between doctors, nurses, administrative workers and other staff adequate?  And second, are the correct number of staff available at the facility when people are seeking services, i.e. are staff present during peak hours? Composition of staff Overall, approximately 20% of the 18,253 health care workers employed in the public primary health care facilities in Alexandria and Menoufia are trained doctors ( Table 23. The biggest part, 13% of total employees, are general practitioners while 7% are specialists. A quarter of all employees are nurses and 10% pharmacists or other medical technicians. 15% of all employees are administrative staff. Other categories represented are health inspectors (4%), Raeda Rifya (health outreach worker, 5%) or other general non-medical staff. There is a difference in composition of staff between facilities in Alexandria and Menoufia with a higher share of specialists and doctors in general, compared to non medical staff in Alexandria. In Alexandria, over 10% of total staff are medical specialists while this category comprises less than 5% of all staff in Menoufia. Furthermore, general practitioners compose 18% of all employees in Alexandria, compared to 11.3% in Menoufia. On the other side, nurses 54 and administrative staff account for a higher share of employees in Menoufia compared to Alexandria. Table 23: Total number of staff, by category (362 Facilities) Total Staff Percentage All MOH Alexandria Menoufia All MOH Alexandria Menoufia Ob/gyn Physician 93 61 31 30 0.51 0.39 0.58 0.3 Family Physician 130 128 97 31 0.71 0.83 1.8 0.31 Pediatrician 53 38 23 15 0.3 0.24 0.43 0.15 General Practitioner 2374 2111 970 1141 13.01 13.61 18.05 11.25 Other Specialist 1368 829 478 351 7.49 5.34 8.89 3.46 Pharmacist 1366 1038 490 548 7.48 6.69 9.12 5.41 Lab Technician 762 586 301 285 4.17 3.78 5.6 2.81 Nurse with Midwifery 201 191 68 123 1.1 1.23 1.27 1.21 Nurse 4381 3959 1072 2887 24 25.52 19.94 28.48 Administrative Staff 2872 2444 751 1693 15.73 15.75 13.97 16.7 Health Inspector 746 739 244 495 4.09 4.76 4.54 4.88 Ambulance Worker 23 17 9 8 0.13 0.11 0.17 0.08 Reada Rifya 913 892 68 824 5 5.75 1.27 8.13 Health Educator 108 104 79 25 0.59 0.67 1.47 0.25 Worker 1820 1551 421 1130 9.97 10 7.83 11.15 Social Worker 148 94 57 37 0.81 0.61 1.06 0.36 Other 895 730 215 515 4.9 4.71 4 5.08 Total 18253 15512 5374 10138 When looking specifically at the composition of trained medical specialists (excluding nurses) there is a consistent pattern that approximately half are general practitioners, and 25% specialists and pharmacists. The ratio of pharmacists is very consistent among different facilities while the ratio between specialists and general practitioners varies. The biggest share of specialists can be found in urban clinics (28.7% compared to 21% in rural clinics) and in Alexandria (30.1% compared to 20.2% in Menoufia). The share of specialists is slightly higher in non-reformed facilities compared to reformed facilities. This echoes some of the statements made by stakeholders in the focus group discussions. Many patients argued that after the reform they had the feeling there were more general practitioners in facilities than before, and some commented on the lack of specialists. Table 24 Composition of medical Professionals MOH Alexandria Menoufia Urban Rural Reformed Non reformed General Practitioners 50.2 46.4 53.9 47 53.8 50 51.9 Specialists 25 30.1 20.2 28.7 21 24.9 26.7 Pharmacists 24.6 23.4 25.9 24.3 25.1 25.1 21.4 55 There are major differences in the socio-demographic composition of the different categories of employees. In particular, doctors and pharmacists are much younger; less experienced and have been at the facility they currently work in for a much shorter period of time. While almost a third of physicians and pharmacists are 25 or younger, only 11% of nurses and 2.5% of administrative staff fall into this category. Administrative staff is significantly older than medical staff with almost half of the employees over the age of 45. The same observation can be made with regards to years of experience in a particular facility: 54% of administrative staff has been working at the same facility for over 10 years, compared with 41% of nurses and 12% of doctors and pharmacists. Public primary health care has a female dominated workforce. Nearly two-thirds, 65%, of doctors and pharmacists; 72% of administrative staff; and almost 98% of nurses are female. Furthermore, staff working at facilities is very much recruited in the close vicinity of the facility. Almost 60% of administrative staff, 55% of nurses and 33% of doctors and pharmacists were born within 5 Km of the facility in which they work. The education levels are very clearly distributed between the different staff categories. Almost all doctors and pharmacists have a university degree. Most administrative staff, 81%, have competed secondary school and 13% have also completed pre-university. Most nurses, 83%, have completed nursing school. Table 25 Main Characteristics of Staff Specialized medical Staff Nurses Administrative Staff Gender Male 35.2 2.2 28 Female 64.8 97.8 72 Age 18-25 31.9 11.4 2.5 25-35 26.7 37.9 11.3 35-45 10.51 30.2 37.4 >45 30.93 20.57 48.8 Years of experience in particular facility 0-1 54.3 9.9 8.73 2-5 24.2 23.8 18.8 5-10 9.5 24.9 18.4 >10 12 41.3 54.12 Years of experience on the job 0-1 33.3 0.6 2 2-5 20 7.45 3.9 5-10 6.33 18.4 11.1 >10 40.33 73.6 83 Geographical proximity to Facility Born <5km 32.5 54.6 59.2 Live <5km 46.5 74.1 74.7 Education Level Completed Secondary School 0.3 8.4 81.2 Pre-university 0.5 3.2 12.9 56 Specialized medical Staff Nurses Administrative Staff College/University 86.9 5.4 3 Postgraduate Degree 12.4 0 0 Nursing School 0 82.6 0 Presence and Absence of Staff Official Presence of Staff The limited shift hours of most public facilities are probably one of the most important barriers to utilization. Nearly three-fourths of facilities, 71%, are open during the morning shift only, from 8-2pm; 24% of facilities also are open during the afternoon shift, from 2pm to 8pm. Only 4% of facilities are open after 8pm. This does not correspond to the 24h availability policy of the HSRP. In the qualitative data collection, several healthcare users commented on this. The shift pattern contrasts sharply with household’s preferences: 45% of respondents would prefer to go to an evening shift. This may explain why many choose the private sector. Only 31% of our surveyed sample went to their nearest public primary facility for their last health visit. In the focus groups, many users commented on the lack of an evening shift at the public reformed clinics. There was also confusion as to the true availability of afternoon shifts. In the case of some units, the users reported a morning shift only, while the providers claimed there was also an afternoon shift. Even among providers, there was little consistency in the afternoon hours they reported. The lack of afternoon or evening shifts is one factor that prompts users to switch to private providers. First, morning shifts do not accommodate employees. Second, there is a need for emergency care in the evening. One story from the qualitative in-depth interviews illustrates how the hours in public facilities lead people to choose private: “When was the last time you went to a private clinic? - The last time, for my daughter's arm so I took her to see an orthopedist. - Why didn't you think to take her to the public unit? - Because it was hurting her at night. - You went to the private doctor because you trust them more? - No. because they were available then. Honestly, sometimes the treatment at the private doctor isn't good. - But if she had gotten sick during the day would you prefer to take to her to the private doctor? - I went to the lab analysis at the unit, but I haven't tried the doctors. We get off work at 3 pm, and by then they were finished at the unit. - So because of the hours, you as employees, prefer to go to the private doctor? - Yes, because the private doctor is available at night.� (Healthcare user) Quite consistently with the above findings, some users report that the same public unit doctor offers private care at night, in the unit: “There isn’t anything in the evening except the private examination of the unit doctor.� 57 Staff Absences The facility survey included three unannounced visits to the facilities, during which presence of staff was monitored. Given the shift structure above, we restricted our analysis to the first shift, and staff that was officially assigned to full-time work in this shift.19. On average across all three visits, a visitor would find 21% of full-time staff absent, and 32% of full-time and part-time staff. 42% of staff were absent at least once, 15% were absent twice or more. The most frequently cited reason was “these are not official working hours� (although the staff was assigned to that shift), 38%, and “running a personal errand�, 29%. Comparing these absenteeism numbers with other countries where similar studies have been carried out, the overall absenteeism rate in Alexandria and Menoufia is below average. Across the developing world, absenteeism of health workers has been estimated at roughly 35%, with particular high rates in South and Southeast Asia (in India, Bangladesh, or Indonesia, an estimated 40% of health workers are absent on average) and Africa (in Uganda, 37% of health workers were found absent).20 Absence and individual characteristics Female staff are less absent than men, with 40% absent at least once (against 52% of men), and 13% absent twice or more (19% men). This result also holds in the econometric analysis, which considers all individual and facility characteristics together, see 19 We undertook two types of analysis (i) descriptive analysis, which calculates percentages looking at one or two categories at once, such as the overall % of people absent, or the % of women with a certain absence rate; and (ii) econometric analysis, which looks at a multitude of categories at once to determine which ones are significantly linked to absences, and which ones are not. In the following text, we discuss various factors which might be linked to absences, and make use of both types of analysis. 20 Chaudhury, Nazmul et al. (2006). 58 Table 26. Female staff are significantly less absent than male staff. The highest share of presence in all three visits was observed in administrative staff (66%), nurses (64%), social workers (64%), lab technicians (62%), and other staff (62%), while absences of 2 or 3 times were highest among ambulance workers (39% in a small sample, most of these for the reason of ‘stopped working’, 31% or ‘runs a personal errand’, 23%), OBGYN (25%), Family Health Doctors (24%) and Health educators (24%). - General practitioners, pediatricians, midwives, pharmacists, Raeda Rifiya (female outreach workers), Health inspectors and workers showed an intermediate absence profile, with about 50% of these categories present in all three visits. In the econometric analysis, professional categories appear to explain absences to a large extent, certainly more than any other variables that we can measure. Compared with the default category OBGYN, the following categories are significantly and much less absent, in this order: administrative staff, social workers, nurses, other job categories, lab technicians, pharmacists, and general workers. No job category shows a significantly higher absence rate than OBGYN specialists, and family physicians, pediatricians, other specialists and health educators did not show a significant difference to the OBGYN. Staff contracts appear to matter. People on term contracts show significantly lower absence rates than people on open-ended contracts, while staff assigned from other units or on civil service assignment does not differ significantly from open-ended staff, according to the econometric analysis. Staff education shows some association with absence, with Pre-university graduates, college graduates and nursing school graduates most frequently absent twice or more (18%, 18% and 20% respectively). Postgraduate degree holders appear to be the least absent, 64% could be found on all three unannounced visits, and only 10% were missing twice or more. In the econometric analysis, a higher education is consistently and significantly linked with higher absence rates, unless we control for professional categories, in which case education becomes insignificant. People born within 5 km of the facility are less frequently absent (64% present all three visits) than those who were born further away (51%). A similar relation (63% vs. 48%) holds for those who live near the facility, respectively further away. This is fully consistent with the econometric analysis. Different sizes of the home city do not make a difference until 100,000 inhabitants (whether below 5000, below 10,000 or below 100,000, 40% of people are absent at least once, and 11% are absent twice or more). But 50% of inhabitants of cities above 100,000 inhabitants are absent at least once, and 20% twice or more. This is fully consistent with the econometric analysis: staff from cities above 100k are significantly more absent. An urban or rural setting of the facility does not appear to make a difference in the descriptive analysis, but the econometric analysis shows that, other things equal, staff in rural areas shows a better presence pattern, see the 59 Table 26. Also, staff in Menoufia can be found more often all three times (64%) than in Alexandria (47%). Also, in Alexandria 21% of staff are missing twice or more, while the figure is 11% in Menoufia. Staff whose relatives visit the facility, can be found present more often (64% present each time vs. 50%). This is fully consistent with the econometric analysis, which underlines the significance of this variable, see 60 Table 26. 61 Table 26: Econometric analysis of staff absences Dependent Variable: Absent (1, 2 or 3) Sample: all workers of shift 1, working full time OLS, robust SE, clustered at the facility level (362 clusters) Coeff t Coeff t Coeff t Coeff t Coeff t Coeff t age -0.001 -1.47 -0.001 -0.61 0.001 0.76 0.003 2.37 -0.001 -0.82 -0.003 -2.81 female -0.170 -8.31 -0.129 -6.23 -0.175 -8.32 -0.101 -4.64 -0.145 -7.39 -0.119 -5.99 salary level 4.81E-07 0.04 -4.23E-06 -0.34 -0.0002 -2.04 -9.62E-06 -0.72 -3.64E-06 -0.31 -0.00001 -1.08 salary level squared 1.62E-08 2.13 education level (ordinal) 0.028 4.74 0.038 5.05 0.006 0.52 -0.00004 -0.01 primary completed -0.047 -0.67 -0.065 -0.95 preparatory completed -0.064 -0.94 -0.073 -1.13 secondary completed -0.091 -2.03 -0.096 -2.36 pre-university education 0.052 1.1 0.024 0.53 college/university attended -0.003 -0.01 college/university completed 0.117 2.71 0.090 2.23 postgraduate degree completed 0.056 0.61 nursing school completed -0.077 -1.83 -0.084 -2.16 other education 0.142 2.08 0.035 0.36 family physician -0.194 -0.97 pediatrician -0.320 -1.47 general practitioner -0.384 -2.08 other specialist -0.304 -1.67 pharmacist -0.510 -2.81 lab technician -0.563 -3.17 nurse midwife -0.425 -2.11 nurse -0.603 -3.31 admin staff -0.655 -3.67 health inspector -0.383 -2.06 raeda rifya (outreach worker) -0.349 -1.87 health educator -0.238 -1.22 worker -0.500 -2.76 social worker -0.621 -3.2 other job category -0.599 -3.31 years in this facility -0.007 -4.15 assigned civil service -0.046 -0.7 term contract -0.080 -1.81 assigned from other unit 0.112 1.29 works part-time 0.643 8.13 Family Health Unit 0.134 2.39 Primary HC Unit 0.138 1.24 HIO -0.344 -2.95 other type of unit -0.265 -2.53 Provider NGO -0.133 -1.03 Provider HIO -0.058 -0.6 non-reformed health unit 0.071 1 rural area -0.224 -5.44 staff doesn't live within 5km 0.159 4.84 village bigger than 5000, < 10,000 -0.0001 0 village 10,000 - under 100,000 0.012 0.22 more than 100,000 inhabitants 0.189 3.27 relative doesn't go to facility 0.087 3.18 constant 0.634 10.63 0.726 11.49 0.565 8.44 1.132 5.95 0.827 10.06 0.637 8.59 default variables no formal school obgyn, open- FHC, MOH provider, staff lives within ended, ft reformed, urban 5km, village smaller than 5000, staff's relative goes to facility observations 14335 14877 14335 14909 14280 13873 R-squared 0.01 0.02 0.01 0.07 0.04 0.04 F-test 22.82 13.4 18.52 18.78 15.53 24.13 62 Absence and facility characteristics Interestingly, staff in non-reformed facilities can be found more often on all three visits (64% vs. 56% in reformed facilities). However, when controlling econometrically for various staff and facility characteristics at once, this difference does no longer appear significant. One reason for better presence in non-reformed facilities may be the financial incentives, which staff complained had worsened in the reformed units: “Yes, there was a reasonable system (of incentives) that people worked to get. But since the funding stopped, we barely get incentives which has affected the staff here.� However, when compared to the default category FHC, staff at FHU are significantly more absent, and staff at HIO and other facility types significantly less absent according to the econometrics. When considering different provider types, staff in HIO run facilities is most present on all three visits (73%, and only 8% absent twice or more), while MOH or NGO management does not seem to matter for staff presence (56% present all three visits, and 15% and 18% respectively missing twice or more). When analyzing absence data according to the main funding source of the facility, staff in facilities funded by HIO is found most often: 73% were present during all three unannounced visits, see Table 27. Facilities financed mainly by Private Donors and Other sources also showed high presence rates, but the sample of staff reported in these categories (below 100 each time) is too low to draw conclusions. Table 27: Staff absences by main funding source of facility Annual MOH, not Service Staff absent X budget through Private Patient FHF Improveme HIO Other times during 3 from annual Donors revenues nt Fund visits MOH budget 0 52% 61% 67% 37% 50% 33% 73% 80% 1 30% 27% 26% 36% 29% 54% 19% 17% 2 13% 9% 5% 19% 17% 13% 5% 3% 3 5% 4% 2% 7% 4% 0% 3% 0% Not surprisingly, facilities in which the lack of major infrastructure, such as medical equipment, clean water or electricity are an important issue, also have somewhat higher staff absences, see Table 28. However, this logic does not apply to the reported lack of financial means. Table 28: Staff absences by self-reported presence of infrastructure Lack of equipment Lack of clean water Lack of electricity Lack of financial means Not Not Not Not staff Important important/ Important important/ Important important/ Important important/ absent X mentioned mentioned mentioned mentioned times 0 54% 59% 55% 58% 52% 58% 59% 58% 1 29% 27% 30% 28% 30% 28% 26% 28% 2 12% 10% 11% 10% 14% 10% 10% 10% 3 5% 4% 4% 4% 3% 4% 5% 4% 63 Absence and accountability mechanisms When examining the use of accountability mechanisms for example for staff presence, interesting associations emerge. For example, the more a facility is making use of a cost audit system, the better the presence record of staff, see Table 29. Table 29: Staff absences according to presence of an audit system (“Does the facility have a system to audit costs?�) staff absent this many Yes, installed here Yes, installed Yes, sporadic times during 3 visits at facility elsewhere analysis No 0 65% 62% 56% 44% 1 23% 26% 24% 33% 2 9% 8% 11% 17% 3 3% 4% 9% 6% However, facilities that measure client opinion through suggestion boxes or survey forms show a marginally worse presence pattern than those who do not have such a system (56%-59% present all times vs 61% present all times). Note that this may be linked to the overall management model of the facility (FHU vs HIO, reformed vs non-reformed, etc) some of which include customer satisfaction surveys by definition. Whether or not a facility uses regular performance reviews does not seem to make a difference for presence; in any case the vast majority of facilities use them for all workers. As for the frequency of reviews, annual reviews appear to work best, with 70% of staff subject to them present each time, while semi-annual reviews seem to be less effective, with only 8% affected staff present each time, and 40% absent twice or more. As for the consequences of performance reviews, it is possible that positive incentives work best, with salary increases and promotions being associated with better average presence records. Please note that no facility in the sample can take the decision to dismiss a person as a consequence of bad performance. Table 30: Staff absences according to the consequences which are possible after a performance review (multiples possible) staff Salary Training Bonus Salary Pro- Training Bonus De- absent X Gift reduc- Obliga- Nothing increase increase motion reward decrease motion times tion tion 0 64% 70% 72% 61% 58% 56% 56% 66% 59% 60% 1 33% 19% 21% 27% 31% 29% 28% 26% 27% 27% 2 13% 8% 4% 8% 7% 11% 13% 5% 10% 9% 3 5% 2% 3% 4% 4% 5% 3% 3% 4% 4% Most facilities are regularly visited by supervisors from the Health District. Contracted facilities also receive tri-monthly visits from the FHF. Those who do not receive such visits show higher absences, see Table 31. Similarly, regular feedback from the FHF may show a mild positive association with presence, while feedback from the Health District does not. 64 Table 31: Staff absences according to public outside supervision of the facility A supervisor from outside FHF gives regular Health district gives staff absent visits regularly feedback regular feedback X times Yes No Yes No Yes No 0 58% 47% 57% 52% 56% 57% 1 28% 22% 28% 32% 28% 29% 2 10% 25% 11% 12% 11% 10% 3 4% 6% 4% 4% 5% 3% The facility’s relationship to the local town administration, mayor/ mayor and local council, does not seem to be associated with high or low absences. Facilities that are regularly visited by the mayor or council members show about the same scores as facilities that do not report a relationship to the town administration. However, local media activity seems to play a role. Staff of facilities where local media are reportedly critical show a better presence record than where local media is very positive or does not play a role, see Table 32.21 Table 32: Staff absences according to the role of local media in their facility’s catchment area Local media do not Local media report staff absent X Local media are No relation to talk much about about us very times frequently critical local media us positively 0 67% 52% 73% 58% 1 25% 33% 18% 27% 2 5% 10% 7% 10% 3 3% 5% 2% 4% The facility’s relationship to the religious institutions (mosques and churches) in the catchment area does not seem to be important at first sight: staff in facilities with a close relationship (e.g. including regular visits by religious officials, and mentions in services) shows the same absence pattern as staff in facilities without any relationship. However, there is one notable exception: the facilities that cooperated with the religious institution to announce their vaccination campaign show somewhat better staff presence than those who did not. Perhaps surprisingly, the presence of, and relationships with NGOs do not seem to be associated with the staff absence pattern, at least not positively. The facilities that reported no presence of NGOs in the area showed marginally better presence (59% present all three times, 14% twice or more) than those who did. No facility reported a difficult relationship with NGOs. 21 Please note that as this question was asked in the quantitative survey, no further detail is available on the way in which the media are ‘critical’. 65 QUALITY OF CARE: OBSERVATION OF CARE PROTOCOLS, QUALITY INDEX AND HYGIENE PRACTICES This chapter describes specifically the interaction between physicians and patients and tries to analyze the quality of care received. As the main instrument used in this survey to measure and compare quality of care, 5,040 structural observations were conducted, where interviewers, all trained physicians, observed patient consultations and noted on a checklist of actions whether doctors carried them out. The checklists were derived from the treatment protocols of the Egyptian Family Health Guidelines, instituted with the HSRP. Structural Observations of Doctors’ Consultations In order to measure the quality of care, the survey team has conducted so-called ‘structural observations’ (also called direct clinician observations DCO) of doctors’ consultations in four areas of treatment:  Diabetes Mellitus,  CHD/Hypertension,  Antenatal care and  Sick child treatment. Structural observations are obtained by trained doctors sitting in as surveyors on actual medical consultations of doctors. During the observation, the qualified surveyor compares a checklist with the actions actually performed by the doctor and ticks the accomplished items. The checklist used for our survey was derived from the examination protocols in the official Family Health (HSRP) guidelines and condensed by World Bank health experts. Only observations where a clear statement could be made about whether or not a given action was carried out by the doctor are included. Structural observation of Diabetes Mellitus Treatment The average Diabetes consultation is missing many of the elements that are part of the ministerial guidelines for Diabetes Mellitus treatment, see Table 33.22 In less than half of the observed consultations did the doctor examine the patient for sensations or reflexes (33.9% of observations), examine arms and hands or feet and legs for pulsation (20.2% and 23.7%, respectively), or examine the back of the thorax with a stethoscope (43.8%). For other standard procedures the observations rates are higher: in 58.2 % of all observations the doctor did weigh the patient, in 88.5% providers took the blood pressure in the arm, in 51.6 % feet or any wound or pressure points were observed, 62.3% of observations the abdomen was examined and 66.5 % of practitioners examined the patient’s chest. Only blood glucose tests are carried out on a regular basis, such an exam could be observed in 77% of the observed observations. Other exams were conducted on a less regular basis, such as urine (29% of observations), blood cholesterol (18%), electrocardiogram (19%) or consultations with ophthalmologist (26%). 22 Only patients who have been diagnosed with diabetes at least 12 months prior to the observed consultation were eligible. Furthermore, all patients observed here have been to the same facility prior to the observed consultation. Patients currently in renal dialysis were excluded from the survey. 66 In discussing the patient’s health, doctors stress the importance or ask the patient about the way they are taking their medications in 89% of all observed observations and talk about the patient’s diet in 80% of all consultations. However, other health behavior that is key to improve the condition of diabetes patients were discussed on a less regular basis: Only 13% of doctors asked about the patient’s smoking habits and 37% of all consultations mentioned the need for the patient to exercise. MOH facilities have a higher rate of following guidelines of Diabetes Mellitus treatment than facilities run by the HIO. In particular, standard examinations of arms, feet, wounds or chest, that should be part of any consultation for patients suffering from Diabetes Mellitus, are conducted significantly less often in consultations in HIO facilities compared to MOH facilities. Providers in facilities in Alexandria and in urban facilities have a higher rate of conducting standard procedures than facilities in Menoufia and facilities in rural areas. When it comes to informing the patient about his or her illness or answering the patient’s questions, however, this observation does not hold, and indeed, providers in facilities in Menoufia are more likely to talk about health behavior with the client than their colleagues in Alexandria. Reformed and accredited facilities have a higher rate of carrying out standard procedures such as weighing the patient, examine feet, arms and hands, or the abdomen. When it comes to consultation and information, however, the differences are less clear. However, fewer providers ask their patients about smoking habits or diet patterns in reformed facilities compared to non reformed facilities. Table 33 Observation protocol for Diabetes mellitus treatment (991 Observations) All MOH Alexandria Menoufia Urban Rural Reformed Non reformed Accredited Not accredited Were any of the following client examinations observed? Weigh the patient 58.2 61 83.9 49.7 57.4 52.9 65.7 37.7 64.4 49.5 Measure the 4.7 5.4 5.9 5.1 10 3.7 6.5 0 5.8 4 circumference of the waist Measure the 1 1 1 0.8 1.7 0.8 1.7 0 1.4 0 circumference of the hips Take blood 88.5 92.1 95.5 90.4 93.5 91.6 91.2 96.6 92.1 92.1 pressure in the arm Examine feet or 51.6 57.2 71.3 50.3 67.4 53.5 60.4 41.1 59.3 50 any wound or pressure points Examine the 33.9 36.9 57.3 26.8 58.3 29.2 40.1 20.5 40.5 24.8 patient for sensations and reflexes Examine arms and 20.2 22.3 36 15.5 23 22 20.3 32.2 19.3 32.2 hands for pulsation Examine feet and 23.7 24.9 44.1 15.5 41.3 19 25.5 21.9 24.2 27.2 legs for pulsation Examine the chest 66.5 70.6 81.1 65.5 74.8 69.1 73.8 74.8 73.2 61.9 Examine the back 43.8 46.4 54.9 42.2 50 45.1 48.8 34.2 48.7 38.6 of the thorax with a stethoscope Examine the 62.3 66.7 71.6 64.6 69.3 66 70.9 45.9 70.1 55.4 abdomen while the patient is in 67 (991 Observations) All MOH Alexandria Menoufia Urban Rural Reformed Non reformed Accredited Not accredited supine position Conduct an eye 1.7 2 0.7 2.6 0.9 2.3 2.5 0.7 2.5 0 exam using an ophthalmoscope Measure the 9.3 9.6 0.4 14.2 3 12 11.5 0.7 10.8 5.9 blood sugar with stick Did the provider order, require, or said he/she received the results of any of the following exams? Urine Test 28.8 28 25.9 29 23.5 29.6 27.7 29.5 27.5 29.7 Blood glucose 77.2 77 78.7 76.2 81.3 75.5 77.7 73.9 77.7 74.5 Blood cholesterol 18.2 18.5 30.4 12.7 32.2 13.6 19.9 11.6 20.3 13.4 Vascular Doppler 0.4 0.5 0 0.7 0 0.6 0.6 0 0.6 0 Electrocardiogram 18.7 18.9 19.6 18.6 21.7 17.9 21.5 6.2 20.8 12.9 Consultation with 26 28.5 30.8 27.4 27.8 28.7 30.8 17.1 29.9 23.8 Ophthalmologist Did the doctor ask or the patient mention information about the following issues Is patient taking 91.4 92.5 83.6 96.9 80.7 96.7 91.6 97.3 91.6 95.5 medications as prescribed Is patient smoking 12.6 13.4 13.3 13.5 10.9 14.4 13.8 11.6 12.3 17.3 tobacco products Is patient smoking 4.3 4.4 4.2 4.5 3.5 4.7 4 6.1 3.9 5.9 shisha Need for patient 74.4 75.7 74.5 76.2 70.9 77.4 75.2 77.4 75.8 75.3 to follow diet Need for patient 25.8 27.3 30.8 25.6 29.1 17.8 29.2 17.8 28.3 24.3 to exercise Attending or had 7.8 9 6.3 10.3 6.5 9.8 10.3 2.1 10.3 4.5 attended any health education session or consultation Did the doctor inform the patient about any of the following? Is patient taking 88.8 88.6 70.4 93.2 74.3 93.8 87 97.3 85.7 98.5 medication as prescribed Harms of smoking 7.7 8.2 10.8 6.8 7 8.6 7.6 11 7.3 10.9 tobacco products Harms of smoking 3.3 3.4 3.8 3.2 2.6 3.7 3 5.5 2.8 5.5 shisha Need for patient 79.5 80.8 75.2 83.6 71.7 84.1 81.2 78.8 80 83.7 to follow diet Need for patient 36.5 39.2 38.8 39.3 41.7 38.2 43.4 17.8 42.3 28.7 to exercise Getting 14.4 16.1 18.5 14.9 22.6 13.7 18.9 2.1 18.2 8.9 information about changing diet, quit smoking or being physically active 68 Structural observation of CHD/Hypertension treatment Basic procedures of CHD/hypertension consultations appeared to be carried out to a large extent, see Table 34.23 Apart from checking the patient’s blood pressure, was observed in almost all consultations, only two thirds of all providers examined the patient with a stethoscope and checked the abdomen; in half of the consultations the provider weighed the patient or examined the back of the thorax with a stethoscope. In approximately 1 out of 4 consultations the provider ordered an exam, most frequently with the electrocardiogram (27.3% of all observations), followed by blood glucose tests (25.8%), blood cholesterol tests (24.1%) and eye examination (16.4%). Nine out of 10 providers talked about medications the patient was currently taking, and in 61.7 % of all observations the doctor talked about the diet of the patient. Other issues, such as tobacco consumption or the need for the patient to exercise were addressed in fewer consultations (12.6 and 23.5%, respectively). For all components of the treatment, the guidelines were followed at a higher rate by providers in MOH facilities compared to HIO facilities. Furthermore, facilities in Alexandria have a significantly higher rate of carrying out standard procedures compared to facilities in Menoufia. In particular, providers in Alexandria are more likely to weigh the patient (83.5 versus 42.7%), examine arms and hands for pulsations (35.5 versus 19.1%), examine the chest with a stethoscope (85.3% versus 67.1 %) or to examine the back of the thorax with a stethoscope (85.3% versus 67.1%). There are no clear differences when comparing reformed facilities with non reformed facilities, as well as accredited facilities with non accredited facilities. Table 34: Observation protocol for CHD/Hypertension treatment (1011 Observations) All MOH Alex M’fia Urban Rural Reformed Non Accredited Not reformed accredited Were any of the following client examinations observed? Weigh the patient 51.3 55.5 83.5 42.7 80.5 47.1 59.6 34.8 57.9 48.1 Measure the 5.2 6.1 7.7 5.4 11.8 4.2 7.3 0 6.7 4.3 circumference of the waist Measure the 1.2 1.2 1.1 1.2 1.4 1.1 1.2 0.7 1.2 1 circumference of the hips Take blood pressure 98.6 99.2 99.6 99 99 99.2 99 100 99.1 99.5 in the arm Examine arms and 22 24.3 35.5 19.1 22.3 24.9 23 30.5 21.8 31.9 hands for pulsation Examine feet and 17.5 18.9 31.1 13.3 25.9 16.6 18.5 21.3 17.7 22.9 legs for pulsation Examine the chest 67.6 72.9 85.3 67.1 80 70.4 75.7 58.2 75.2 65.7 with a stethoscope Examine the back of 46.4 50.4 64.8 43.7 60 47.1 52.7 38.3 53.2 41.4 the thorax with a stethoscope Examine the 60.6 65.8 73.6 62.3 71.8 63.8 69.5 46.8 68.8 56.7 abdomen while the patient is in supine position 23 Only patients who have been diagnosed with CHD at least 12 months prior to the observed consultation were eligible. Furthermore, all patients observed here have been to the same facility prior to the observed consultation. 69 (1011 Observations) All MOH Alex M’fia Urban Rural Reformed Non Accredited Not reformed accredited Conduct an eye 1.3 1.5 0.7 1.9 1 1.7 1.8 0 1.8 0.5 exam using an ophthalmoscope Did the provider order, require, or said he/she received the results of any of the following exams? Blood glucose 25.8 26.9 30 25.5 31.4 25.4 29.4 14.2 28.2 22.9 Blood cholesterol 24.1 22.8 38.5 15.5 39.5 17 24.8 12.1 23.5 20.5 Electrocardiogram 27.3 27.4 27.4 26.1 31.4 26 30.1 13.5 28.4 24.3 Consultation with 16.4 18.4 18 18.6 20 17.8 20.3 8.5 19.7 14.3 Ophthalmologist Did the doctor ask or the patient mention information about the following issues Is patient taking 91.4 93.7 87.6 96.5 90 96.7 92.3 99.3 92.4 97.6 medications as prescribed Is patient smoking 12.6 13.4 16.9 11.8 17.3 12.1 13.5 12.8 13 14.8 tobacco products Is patient smoking 4.7 4.7 4.4 4.9 3.2 5.3 4.4 6.4 4.7 4.8 shisha Need for patient to 61.7 64.9 70.7 62.2 67.7 63.9 67.3 52.5 66.9 58.6 follow diet Need for patient to 23.5 25.4 31.1 22.8 27.7 24.6 26.5 19.9 25.8 24.3 exercise Attending or had 5.9 6.7 3.7 8.1 5 7.3 7.9 0.7 7.3 4.8 attended any health education session or consultation Did the doctor inform the patient about any of the following? Is patient taking 88.2 89.4 78.4 94.4 73.2 94.9 87.9 97.2 86.4 98.6 medication as prescribed Harms of smoking 7.6 8 11 6.6 9.1 7.6 7.5 10.6 7.6 9.1 tobacco products Harms of smoking 3.2 3.2 3.7 3 1.4 3.9 2.8 5.7 3.1 3.9 shisha Need for patient to 68.7 71.8 74 70.8 71.4 72 76.4 48.2 74.6 63.3 follow diet Need for patient to 33.5 37 39.6 35.6 39.6 36.1 40.4 19.2 39 30.5 exercise Possibility to get 13.4 15.1 16.1 14.7 20.5 13.3 17.9 0.7 17.3 8.6 more information about changing diet, quit smoking or being physically active 70 Structural observation of antenatal care Very basic procedures of antenatal care are carried out often; more advanced but essential procedures are missing in many cases, see 71 Table 35.24 Almost all providers conducting antenatal observations measure the pregnant woman’s blood pressure, 9 out of 10 weigh the patient, and in more than two third of all observations visited by the survey team the doctor did take a urine sample to test the client’s sugar, protein and anemia level. However, only in 50.1 % of all consultations did the doctor palpate or measure the women’s abdomen to check for the fetal position or the fetal height; even fewer doctors did listen to the client’s abdomen for fetal heartbeat (30.7 % of observations) or examine the abdomen by sonar (28.4 %). There is no distinct difference in the performance of antenatal care between HIO and MOH facilities, and between facilities in Alexandria and facilities in Menoufia. While providers in Alexandria could be observed at a higher rate to measure the weight of the client (93.7 versus 88.2 % of observations), examine the abdomen by sonar (41.1 versus 20.9), or take a urine sample (77.5 versus 63.4 %), providers in Menoufia are more likely to palpate the abdomen (51.6 versus 45 % of observations) and ask questions to their patients about her pregnancy (mentioned in 53.9 % of all consultations versus 38.7 %) than their colleagues in Alexandria. The comparison between accredited and non-accredited units is inconclusive. Providers working in accredited facilities are more likely to palpate or measure abdomen (52.9 % of providers in accredited facilities versus 39.9 % in not currently accredited facilities) or measure the client’s weight (91.4 % versus 85.8 %). On the other hand, there were more consultations in non accredited facilities in which providers ordered a urine sample (72.2 % versus 64.6 %) than in currently accredited facilities. 24 Only mothers who have made prior visits to the facility they were examined were included in the sample. Furthermore, all patients in the sample did at least give birth once before the current pregnancy. 72 Table 35: Observation protocol for antenatal treatment (1,279 Observations) Were any of the following client examinations observed? All MOH Alex M’fia Urban Rural Reformed Non Accredited Not reformed accredited Measure blood pressure 99.1 99 99.5 98.8 99.3 98.9 98.8 100 98.7 100 Palpate or measure 49.8 49.5 45 51.6 55.4 49.2 53.8 29 52.9 39.9 abdomen for fetal presentation/position Palpate or measure 50.1 49.2 47.4 50 54.1 48 53.5 28.6 53.5 37 abdomen for fetal height Listen to the client's 30.7 31.2 35.9 29.1 37.2 29.3 33.5 20.5 33.9 23.7 abdomen for fetal heartbeat Measure weight of 90.1 89.9 93.7 88.2 95.3 88.2 89.4 92.4 91.4 85.8 client Examine abdomen by 28.4 27.3 41.1 20.9 35.5 24.6 28.2 22.9 26.7 28.8 sonar Was a urine sample 66.2 66.5 77.5 61.5 66.9 66.4 62.3 86.7 64.6 72.2 taken or laboratory examination ordered for the client to test sugar Was a urine sample 67.6 67.9 77.5 63.4 67.6 67.9 63.9 86.7 66.2 72.5 taken or laboratory examination ordered for the client to test protein Was a blood sample 69.5 70.3 82.2 64.7 73.7 86.2 66.9 86.2 69.5 72.5 taken or laboratory examination ordered for the client to test for anemia Was a blood sample 3 3.2 2.9 3.4 5.4 2.5 3 4.3 3.1 3.5 taken or laboratory examination ordered for the client to test for the patient's blood group Was a blood sample 7.6 8 16.8 4 12.8 6.5 7.7 9.5 8.2 7.6 taken or laboratory examination ordered for the client to test for the patient's RH factor? Did the provider talk 49.4 49.1 38.7 53.9 44.3 50.7 53 30.5 53.8 35.8 about the number of pregnancy weeks with the patient Did the provider ask if 50.9 51.1 48.4 52.3 46.3 52.6 51.7 48.1 50.4 52.9 she had any questions about her pregnancy Did the provider answer 99.2 99.2 98.9 99.3 99.3 99.2 99 100 99.1 99.4 all the client's questions Did the provider ask the 70.7 70.2 74.3 68.2 77.4 67.8 71.4 64.3 72.3 64.2 client about her health generally If the provider identified 26.4 23.7 17 27.1 20.3 25.1 26 13.7 22.9 25.7 any non-pregnancy related health problems did he/she treat them or refer the client to a specialist 73 Structural observation of sick child consultation Table 36 scrutinizes the consultation of children under 5 years suffering from any non chronic health issues.25 Of the different examinations required by the guidelines of sick child treatment, very few were observed across the board for every observation, such as taking the child’s temperature (84% of all observations), the use of a stethoscope to hear chest and back (80.8%), check the throat (69.4% without the use of light and 18.6% using a light) or weigh the child. Other procedures foreseen in the manual for sick child treatment were carried out less frequently: 21.7% of all providers felt the child for fever or hotness, 31.2% measured skin turgor and dehydration and 22.6% looked at the ear of the children. In the dialogue with the caretaker doctors examined the child’s health status by asking about cough or breathing difficulty (89.9%), diarrhea (79.1%), fever or body hotness (94.4%) in at least 8 out of 10 observed consultations. In fewer cases providers enquired whether or not the child is suffering from ear pain (44.2%), throat problems (50.4%), child’s vomiting (57.2%), or ask about the child’s feeding practices (33.3 %) or growth (51.7%). The observance of most examinations and consultations is slightly higher in MOH facilities compared to HIO facilities. When comparing sick child consultations in Alexandria and Menoufia or in urban and rural clinics, there are no clear differences. Most examinations are more likely conducted in reformed than in non reformed facilities, however, this was not observed when it comes to using the stethoscope (85.3% of non reformed observations versus 79.9% of observations in reformed facilities) or the doctor enquiring about general health status of the child (54.7% versus 42.6). There is no clear trend in comparing accredited with not accredited facilities. 25 A total of 1382 observations were visited by the survey team. Only consultations for children that have been to the respective health facility at a prior occasion were taken into account. 74 Table 36 Observation protocol for sick child treatment (1382 All MOH Alex M’fia Urban Rural Reformed Non Accredited Not Observations) reformed accredited Were any of the following client examinations observed Take 84 84.7 89.7 82.2 89.4 83.1 85.6 80.8 88.3 75.1 temperature using thermometer Feel the child 21.7 21.8 7.8 28.8 13.6 24.6 23.2 15.9 23.3 17.9 for fever or hotness Count 35.6 36.2 40 34.3 50.2 31.4 42.3 9.8 44.1 15.4 respiration (breath) Use 80.8 80.9 78.9 81.9 77 82.2 79.9 85.3 78.9 86.3 stethoscope on chest and back Check skin 31.2 31.7 28.3 33.5 23.6 34.5 31.7 31.8 30.5 34.9 turgor for dehydration Check for 23.3 23.8 21.2 25.2 28.4 22.2 26.5 12.3 27.3 14.5 pallor by looking at lower lip Check throat 69.4 69.4 78.2 64.9 69.8 69.2 68 75.1 68.2 72.6 with tongue compressor using no light Check throat 18.6 19.2 17 20.3 25.1 17.2 21.3 9.8 22.2 11.2 with tongue compressor and light Look in ear 22.8 23.3 27.8 21 29.3 21.2 23.3 22.9 23.4 22.9 and feel behind ear Press both 12.6 12.9 10.6 14.1 16.9 11.6 14.1 7.8 13.8 10.6 feet (check for edema) Check arms 10 10 9.9 10.1 15.1 8.3 11.9 2 12.7 3.1 and shoulders Weigh the 75.1 77 80.7 75.2 82.5 75.1 78.6 70.2 80.8 67 child Plot weight on 37.8 38.2 42.7 35.7 52.4 32.9 39.3 33.1 39.1 35.4 a grow chart (% of those who did weigh the child) Did the doctor ask or the patient mention information about the following issues Cough of 89.9 90.6 91.3 90.3 89.4 91 91.4 87.3 91.4 88.6 difficulty breathing Diarrhea 79.1 80 79.6 80.2 75.8 81.4 81.5 73.5 81 77.4 Fever or body 94.4 94.6 91.7 96.1 94 94.8 95.3 91.8 95.7 91.6 hotness Ear pain or 44.2 44.5 46 43.8 57.4 40.8 47.5 31.4 47.7 36 discharge 75 (1382 All MOH Alex M’fia Urban Rural Reformed Non Accredited Not Observations) reformed accredited Throat 50.4 50.7 52 50.1 55 49.3 52 45.3 52 47.5 problems Child is unable 34 34.5 29.2 37.2 37.8 33.4 36.5 25.7 35.8 31 to drink or breastfeed at all Child is 57.2 58.5 55.2 60.2 58.6 58.5 58.4 58.8 60.6 53.1 vomiting The child has 34.3 34.9 37.2 33.7 47.1 30.7 39.2 16.3 39.6 22.3 convulsions with its sickness Provider 44.9 44.9 45.8 44.4 39 46.9 42.6 54.7 41.1 54.8 asked about other problems Did the doctor ask about or perform other assessments of the child's health Offer the child 4.6 4.9 2.8 6 4.5 5 5.9 0.8 5.8 2.2 something to drink or put the child to the breast (to verify if child can drink) Ask about 33.3 34.1 34.9 33.5 40 33.5 35.6 27.8 35.7 29.9 normal feeding practices when the child is not ill Ask about 46.8 48.3 58.8 43.6 55 45.9 48.3 48.7 57.2 65.3 normal breast feeding practices when the child is not ill Discuss the 51.7 53.8 55 53.2 58.3 52.3 56.2 43.7 57.3 44.7 growth with caretaker Ask caretaker 57.3 59 69.4 53.9 70.4 55 59.2 58 60.2 55.6 about the child's vaccination history Look at child 59.6 61.2 53.3 59.8 65.3 59.8 64.7 46.1 63.8 54.5 health card 76 Quality Index In order to measure differences in adherence to protocols between different types of consultations, all structural observations have been normalized to a quality index. The index compiles all components of the observation checklist into a score between 0 and 1, where 0 means that none of the elements of a checklist was observed and 1 means that the provider conducted every single aspect of the consultation according to the checklist. Overall, the average value of the index is highest for antenatal consultations, meaning that providers across all facilities adhered more to the guidelines when consulting pregnant women, than in consultations for sick children, diabetes or hypertension patients. While the average provider performed 63% of standard procedures that constitute a routine antenatal examination, only 38% of all standard procedures were performed in the average consultation for CHD/hypertension patients. For all four categories of consultations, the highest value on the index is reached in facilities in Alexandria. Moreover, the value is higher for reformed facilities compared to non reformed facilities and for accredited facilities compared to not accredited facilities. Table 37 Quality Index of observation of protocols All MOH Alexandria Menoufia Mean SD Mean SD Mean SD Mean SD Antenatal 0.632 0.126 0.632 0.124 0.654* 0.112 0.623 0.129 Sick Child 0.582 0.147 0.586 0.148 0.596 0.142 0.582 0.15 Diabetes 0.392 0.151 0.407 0.147 0.448* 0.111 0.389 0.159 CHD/Hypertension 0.375 0.164 0.392 0.161 0.421* 0.12 0.341 0.155 *= Significantly different at 95% confidence level Reformed Non reformed Accredited Not accredited Mean SD Mean SD Mean SD Mean SD Antenatal 0.634 0.126 0.622 0.119 0.641* 0.124 0.61 0.124 Sick Child 0.592* 0.145 0.513 0.158 0.593* 0.145 0.544 0.158 Diabetes 0.419* 0.146 0.359 0.143 0.417* 0.146 0.38 0.148 CHD/Hypertension 0.376* 0.147 0.314 0.154 0.368 0.154 0.357 0.159 *= Significantly different at 95% confidence level Quality of care is more uniform and predictable in antenatal and sick child consultations than in diabetes and hypertension consultations. Figure 9 shows the distribution of facilities along the quality index. The differences between facilities are closer to a normal distribution for antenatal and sick child treatments, compared to hypertension and diabetes treatments, where there is more variety within the sample. Further, providers in Alexandria conduct on average more aspects required by the guidelines of a given consultation than providers in Menoufia. The graphs show that for all four consultations the quality index distribution in Alexandria is further to the right than the distribution for Menoufia. 77 Figure 9: Distribution of facilities along quality index, by governorate (a) Antenatal (b) Diabetes 25 18 16 20 14 Percentage of Faclities Percentage of Faclities 12 15 10 8 10 6 5 4 2 0 Quality Index 0 Quality Index 0 1 0 1 All Alexandria Menoufia All Alexandria Menoufia (c) Hypertension (d) IMCI The overall picture shows a similar, if milder, trend when looking at the distribution according to facilities’ accreditation status, see Figure 10. Further, in the average consultation for sick child and antenatal care, the patient can expect the provider to perform more of the standard procedures than in hypertension and diabetes consultations. The peak of the curve is further to the right for the IMCI and antenatal consultations, compared to diabetes and hypertension consultations. The distribution on the index for hypertension and diabetes shows near random variation. 78 Figure 10: Distribution of facilities along quality index, by accreditation status (a) Antenatal (b) Diabetes 30 25 Percentage of Faclities 20 15 10 5 0 Quality Index 0 1 All Accredited Not accredited (c) Hypertension (d) IMCI Despite these limitations, providers and beneficiaries alike spoke highly of the competence levels of personnel at the reformed facilities in the in-depth interviews and the focus groups discussions. With notable exceptions, most beneficiaries reported good treatment by medical providers and staff at the reformed facilities. Participants in the focus groups noted that the services are good; highlighting that they felt the doctors spent a sufficient amount of time with each and every patient. Interestingly, the providers who were interviewed expressed a greater degree of dis-satisfaction than either users or non-users. Although they value the progress made with the reform and the FHM, a lack of resources prevents them from fulfilling its potential. The lack of specialists and technicians limits the usefulness of existing equipment, such as x-ray machines. Many providers noted during the interviews that they had to refer patients on a regular basis because they 79 were unequipped to handle the machinery. These services are also important for income- generation, which has implications for the funding of the facilities. Hygiene Practices 80 Table 38 presents the adherence to selected standard hygiene practices in the structural observations of diabetes, CHD, antenatal and sick child consultations. The results are based on a total of 4,665 structural observations and the percentage represents the observations in which the provider did perform the respective hygiene procedure. The numbers are rather stunning: In only 22.4% of all observations did the provider wash his/her hands with soap and water prior to engaging with the patient. The numbers are on a similar low level for the usage of disposable gloves and the paper sheet on the examination table, at 23.1 and 19.7%, respectively. Somewhat better are the usage of disposable wooden tongue compressor for each patient, at 60.8%, and the usage of waste receptacles with lid and plastic liner to dispose of waste material, at 70%. While still at an alarming level, the numbers for MOH clinics are marginally better than HIO clinics, at 24.5, 24.7 and 21.1% for washing hands, using gloves, and change paper or sheet on the examination table, respectively. HIO clinics actually show rates at very low levels. No provider in HIO clinics could be observed washing their hands before observing a patient. Only 6.6% of providers use disposable gloves prior to engaging with the patient and less than 4% change the paper or sheet on the examination table after each patient. Reformed facilities do better overall than non-reformed facilities. 28.2% of doctors in reformed facilities wash their hands before touching a patient compared to only 7.1% in non-reformed clinics. The differences are similarly severe for the usage of disposable gloves or changing of sheets after every patient, 28% versus 9.2% and 24% versus 7%, respectively. The same can be observed when looking at differences between accredited and not accredited facilities; accredited facilities are showing better numbers in all hygiene categories than non-accredited facilities. There are no overall differences between rural and urban clinics; while clinics in urban areas show better rates of hand washing of providers, the opposite can be observed when looking at the usage of disposable gloves or changing the sheet on the examination table. This observation is reflected in the differences between clinics in Alexandria (more urban) and Menoufia. 81 Table 38: Hygiene Practices observed, by facility % of providers that did All Alex Menoufia MOH HIO Urban Rural Reformed reformed Not Accredited accredited Not (19,754 Observations) Wash their hands with soap 22.4 24.4 24.6 24.5 0.0 29.6 22.8 28.2 7.1 29.0 11.6 and water after each patient Use disposable wooden tongue 60.8 63.8 62.6 63.0 33.1 63.1 63.0 64.4 56.8 64.1 59.8 depressor for each patient Use disposable gloves when 23.1 12.4 35.4 24.7 6.6 16.6 27.5 28.0 9.2 29.3 11.5 appropriate Change paper or sheet on the 19.7 8.7 26.7 21.1 3.9 13.7 23.7 24.1 7.1 24.7 10.6 examination table after each patient Use waste receptacle with lid 70.0 74.0 73.9 77.5 24.5 78.7 72.3 77.6 56.5 78.3 61.0 and plastic liner to dispose of waste material Constraints to improvement As part of the provider interviews, facility managers were asked to list the major constraints they experience in improving the quality of service delivery at their facility. The interviewers did not prompt any suggestion but listened to the answer of the manager and ranked 18 predefined potential constraints on a scale from 1 to 5 with 1 being “very important constraint to facility improvement� and 5 “not important to facility improvement�. Figure 11 summarizes the constraints that were mentioned most frequently as important or very important. In the view of facility management the three most severe constraints to improving the quality of services at the facility level are (i) the low motivation of staff (viewed as a constraint by 43% of all managers), (ii) general lack of supplies (39%) and (iii) the (non) availability of qualified staff (37.5%). Other important issues mentioned by the interviewees were lack of drugs (29.3%), and general problems with the quality of buildings (21%), plumbing infrastructure (14.4%) and non medical supplies, i.e. furniture (13%). Managers of facilities run by the HIO seem to rank several aspects as less important to quality improvement compared to their colleagues at MOH clinics. Only the lack of supervision from the MOH and the disrespect from patients towards doctor’s advice is more of an issue to HIO facility managers than it is to MOH facility managers. Other potential issues, such as absence of staff, staff having a second job or staff that do not respect working hours were mentioned less frequently. 82 Figure 11: Constraints to improvement of facility, by percentage of facility directors that assessed issue as “important� or “very important� constraint to facility improvement 50 45 40 35 All 30 MOHP 25 20 HIO 15 10 5 0 83 5. PAYMENTS One of the HSRP’s goals is to grant access to basic health care services to all citizens. In order to allow this sustainably, it instituted a co-payment regime for the uninsured, while granting the exemption of the poor from any co-payments. - This chapter first describes the official co- payment fees in detail, and then looks at their actual application at the point of service. This allows an assessment of the effectiveness of the fee regime in the light of its goals, and discerning any potential improvements. OFFICIAL FEES Public primary care Family Health Units and Family Health Centers (reformed units) As stated in the introduction, the Health Sector Reform Program introduced a type of health insurance that is provided at the point of service: people who are not covered by the HIO, due to their having informal or no work, can obtain primary health services at the facility through minimal co-payments. The official payments are regulated in Decree 147 of 2003 for Alexandria, and decree 231 of 2006 for Menoufia. They stipulate that patients will provide the following co-payments:  10 LE to inscribe in the family health unit and to open a family folder,  5 LE in Alexandria and 10 LE in Menoufia for annual renewal of the family health folder,  3 LE per examination, and  35% of the medical treatment (drugs and other therapy), and 50% upon repeat treatment prescribed by specialist.  Home visits are officially not part of the package provided by contracted facilities. These fees apply to all MOH public primary care units in Menoufia, and the contracted public primary care units in Alexandria. For individuals who are already insured in the HIO but seek treatment in a FHU, the HIO reimburses the FHU for the treatment; it pays the FHF (in Alexandria), or pays the FHU directly for folder opening and examinations (in Menoufia), according to the above fee scale. Total revenues are transferred to the FHF (Alexandria) or to the Service Improvement Fund of the unit (Menoufia). The HIO insured patient pays as he would at an HIO facility: a symbolic 50 piasters per visit, and 25-33% of the drugs. According to the Health Sector Reform, poor people are supposed to be exempt from any fees in both governorates. The rules for eligibility for exemption currently differ de facto by district and, as of July 2009, include the following:  Household enumeration through the National Center for Social and Criminal Research (NCSCR) in both Alexandria and Menoufia; 84  Accreditation of the NCSCR’s household list through Ministry of Social Solidarity (MOSS) in both Alexandria and Menoufia;  In Alexandria, under Ministerial Decree 147, o The following population categories are also exempted: Sadat pension and social solidarity beneficiaries; orphans under 18 years without a supporter; mentally, physically or socially disabled persons without supporter; families whose breadwinner is conducting military service or is imprisoned; divorced women, widows, low-income families, unemployed, senior citizens without a pension; patients of endemic diseases such as bilharzias; those referred by the police or medical commission to have a medical report; antenatal and postnatal care, as well as those with other conditions decided by the board of directors and approved by the board of secretariats. o The individual exemption is applied within each district, depending on a household enumeration tool prepared in the Central Department of Technical Support and Projects, and conducted by the social worker in each FHU for those requesting exemption.  In Menoufia, under Ministerial Decree 231, o Eligibility for exemption also follows the criteria under Ministerial Decree 147. o Exemption applications can be prompted by the individual, by the social worker during filling in the family folder, or through the rural leader. o The final decision on exemption is taken after a proxy means test, a tool similar to that applied by the Ministry of Social Solidarity. The facility is reimbursed by the FHF on a monthly basis for services rendered to exempted patients, with 5 LE for a GP examination and 7 LE for a specialist examination. More than two- thirds, 36%, of the monthly reimbursement is paid into the Service Improvement Fund and 64% as a check to pay out incentives. User’s perception In the Focus Groups, users were generally satisfied with the official co-payments for reformed units, and understood their rationale. Users generally found the FHM co-payments low, even the 35% drug costs, because they compared them to the cost of private providers. A number of users accepted the co-payment as a necessity that helps provide quality and availability of services. Beneficiaries also felt the co-payment enabled them to ask for more services. Primary Health Care Units (non-reformed units) Users of primary health care units pay minimal out-of-pocket costs: inscription and treatment or therapy including drugs are officially free of charge. The examination fee is 1 LE. Home visits are made by clinicians of the units after 2pm; the charge for each visit is 10 LE, according to Ministerial Decree 239. 85 User’s perception Sometimes, reformed (FHU, FHC) and non-reformed (PHCU) units are situated close to each other. This puts their different co-pay regimes into competition, to the disadvantage of the more expensive reformed units. However, in the Focus Groups the participants rarely commented on the public facilities with and without co-payment, except for those located in the same building. In those cases, users recognized that they would save money by going to the non-reformed unit. Providers commented on this trend and asked for a unification of all public co-pay regimes independent of the type of facility. Facilities of the Health Insurance Organization (HIO facilities) Patients insured by HIO do not pay for inscription or renewal of folder. They pay 50 piasters for an examination and 25-33% of drug costs. They usually do not pay for any non-drug treatment/therapy at the point of service, as they are already covered through the formal insurance contributions via their payroll. Hospitals In the Governorate of Menoufia, if the patient is formally referred from a FHU or FHC, he/she pays 2 LE for treatment at the hospital. If the patient is not referred, he/she pays 10 LE. There is no such regime for Alexandria. Private health care Fees levied by private primary practitioners or private hospitals are not subject to official price caps. APPLICATION OF FEES IN PRACTICE This section analyzes which fees are paid by patients in practice. As can be seen from the Demand chapter, the households in our sample visit the whole range of facilities available in Egypt: primary, secondary and tertiary, in both the public and private sectors. The range of facilities visited by the household therefore exceeds the categories that were surveyed in our companion facility survey, and allows for wider comparison. Overview Nearly 80% of the people in our sample paid some fee at their last doctor’s visit. The proportion is slightly higher for women than for men, and higher for younger than older adults. Patients paid on average 47 LE during their last visit, women slightly more (48 LE) than men (45 LE). People aged over 65 years paid the highest average fees, 60 LE, compared with 42.3 LE for those aged 40-64 years, see table 41. Table 39: Did you pay a fee at your last visit at the health facility, and how much? By gender and age Payment? Total Male Female 15-39 yrs 40-64 yrs 65+ yrs Yes 77% 70% 81% 80% 73% 64% No 23% 30% 19% 20% 27% 36% Average 46.6 44.9 47.6 48.0 42.3 60.1 payment in LE 86 There are virtually no differences by poverty quintile with respect to payment of any fee. However – disturbingly – the average overall amount paid at the last visit appears to regress slightly with wealth. This means that, in practice, patient payments are not lower for the poor, but rather higher. Table 40: Did you pay a fee at your last visit at the health facility and how much? By poverty quintile (predicted consumption) Payment? First (poorest) Second Third Fourth Fifth Yes 77% 76% 75% 78% 77% No 23% 24% 25% 22% 23% Average payment in LE 47.7 50.1 49.3 44.3 41.3 There were substantial variations in patient payments by type and location of facility. Most people were charged when seeking treatment in public primary facilities, 84% and 76% respectively in urban and rural health units. These shares remain broadly similar when we consider reformed units only; 79% were charged at FHC and 71% at FHU. The share of people charged is actually not significantly lower in public facilities than in private facilities: 85% of those seeing a private physician paid, as did 84% of those receiving services in a hospital. This was true even though the reformed public facilities are supposed to exempt the poor, in order to provide financial access to health services. Of those who used an HIO clinic or hospital, just 16% and 15% respectively paid a fee, as did just 20% of those receiving health care in a mobile clinic. (See 87 Table 41.) The average amount paid at the last visit was highest at private (197 LE) and university hospitals (95 LE) and laboratories (69 LE), and lowest at Urban Health Units (4.3 LE), Mother and Child Health (MCH) Centers (6 LE), Mobile Clinics (7LE) and Rural facilities (8 LE). 88 Table 41: Did you pay a fee at your last visit at the health facility and how much? By type of facility Average payment in Facility Yes No Sample LE Sample Urban Hospital 80% 20% 456 55.2 362 Urban Health Unit 84% 16% 344 4.3 290 Rural hospital 85% 15% 34 8.4 29 Rural health unit 76% 24% 643 8.4 491 MHC Center 69% 31% 29 6 20 Mobile clinic 20% 80% 20 7 4 Other MOH clinic 84% 16% 25 13.2 21 HIO clinic 16% 84% 225 17.7 36 HIO hospital 15% 85% 304 13.5 46 NGO clinic 87% 13% 240 25.4 208 NGO hospital 86% 14% 84 67.5 72 Private doctor 85% 15% 3077 43 2623 Private hospital 84% 16% 403 197.1 336 School clinic 0% 100% 9 10 1 Traditional healer 100% 0% 1 ? ? Pharmacy 90% 10% 311 12.8 279 University Hospital 57% 43% 141 94.7 81 Laboratory 93% 7% 46 68.5 43 Provider home visit 50% 50% 2 40 1 Medical convoy 0% 100% 16 Na na Other government 24% 76% 46 14.1 11 Other 8% 92% 26 10.5 2 The share of people paying a fee is highest in the evening shift (86%), unsurprisingly, as this shift is usually only served by the private sector. The share is lowest in the morning shift (66%). Table 42: Did you pay a fee at your last visit at the health facility? By shift Morning shift Afternoon shift Evening shift Yes 66% 77% 86% No 34% 23% 14% Sample 2,611 993 2,878 The average amount paid did not differ substantially by governorate or urban/rural location. However, people registered in the Family Health Model paid somewhat less on average (40 LE) than those not registered (52 LE). 89 Table 43: How much did you pay at your last visit to the health facility? By region and FH affiliation. Alexandria Menoufia Urban Rural Registered in Not registered FH Model Average payment 50.4 44.4 48.7 45.4 39.7 51.9 in LE Sample 1817 3139 1748 3208 2166 2790 Family Folder Nearly all, 99.52% of households report not having been charged at all for opening their Family Folder. Six report paying the correct fee of 10 LE and only 2 out of 5000 people report having been charged more than 10 LE. Examination fees Patients have paid on average 27 LE for the doctor’s examination at their last visit. Patients over 65 years of age paid an average 40 LE. Please note that these figures include the private sector. Table 44: How much did the doctor’s examination cost last time? By gender and age. Total Male Female 15-39 yrs 40-64 yrs 65+ yrs Average payment 27.4 23.7 29.7 29.0 23.1 39.7 in LE Unfortunately, the poorest quintile appears to pay most and the richest least for examination fees. Table 45: How much did the doctor’s examination cost last time? By poverty quintile (predicted consumption) First (poorest) Second Third Fourth Fifth Average payment 33.6 23.1 29.0 27.5 21.9 in LE Several public facilities charge examination fees above their official fee scale. Public primary care units are reported to have charged an average examination fee above their official fee of 1 LE, ranging in fact from 4 LE (Urban Health Unit) to 8 LE (Rural Health Unit). HIO clinics charged on average 18 LE for the examination, although their official fee is only 50 piasters. 90 Table 46: How much did the doctor’s examination cost last time? By type of facility. Average payment in LE Sample Urban Hospital 77.3 358 Urban Health Unit 4.3 290 Rural hospital 9.7 29 Rural health unit 7.7 490 MHC Center 6 20 Mobile clinic 7 4 Other MOH clinic 13.2 21 HIO clinic 17.7 36 HIO hospital 13.5 46 NGO clinic 18.6 206 NGO hospital 7.5 72 Private doctor 42.5 2620 Private hospital 159.8 331 School clinic 10 1 University Hospital 2.4 79 Laboratory 68.5 43 Provider home visit 0 1 Medical convoy Na na Other government 14.1 11 Other 10.5 2 Patients have paid on average slightly more for an examination in Alexandria (31 LE) than in Menoufia (26 LE). Further, people registered in the Family Health Model have paid a bit less (24 LE) on average than those not registered (30 LE on average). Table 47: How much did the doctor’s examination cost last time? By region and FH affiliation. Alexandria Menoufia Urban Rural Registered Not in FH Model registered Average payment 30.7 25.6 25.3 28.6 23.6 30.4 in LE Sample 1804 3137 1736 3205 2162 2779 The average charge for a doctor’s examination appears to differ substantially by the shift. Patients using the morning shift paid on average only 5 LE while those who used the afternoon shift paid 43 LE, and patients of the evening shift 30 LE. Table 48: How much did the doctor’s examination cost last time? All facilities, by shift. Morning shift Afternoon shift Evening shift Average payment in LE 5.3 42.8 30.0 91 Sample 1705 761 2475 Testing/ Lab/ X-rays Patients paid moderately for testing and labs, the average being 4.7 LE. The number is higher for males (6.7 LE) and patients over 65 years (7.4 LE). Table 49: How much did you pay the last time you needed Testing/ Lab/ X-rays? By gender and age. Total Male Female 15-39 yrs 40-64 yrs 65+ yrs Average payment in LE 4.7 6.7 3.5 3.7 5.9 7.4 Primary health facilities, including even private doctors, were reported to have charged very moderately for testing, the charges ranging from 0.8LE on average (Urban and Rural Health Units) to 3.5LE (private doctors). The reason may be that more advanced testing is left to free- standing laboratories and university hospitals, which charge 66 LE and 25.3 LE on average, respectively. Fees paid to NGO hospitals and urban hospitals were approximately the same— 10.5 LE and 9.8 LE, respectively. They may have facilities for more sophisticated tests than rural hospitals. Table 50: How much did you pay the last time you needed Testing/ Lab/ X-rays? By facility type. Average payment in LE Sample Urban Hospital 9.8 359 Urban Health Unit 0.8 289 Rural hospital 0.1 29 Rural health unit 0.8 490 MHC Center - 20 Mobile clinic - 4 Other MOH clinic 6.2 21 HIO clinic 0.8 36 HIO hospital 4.8 46 NGO clinic 1.6 206 NGO hospital 10.5 72 Private doctor 3.5 2616 Private hospital 10 327 School clinic - 1 Pharmacy 0.2 281 University Hospital 25.3 79 Laboratory 66 43 Provider home visit - 1 Medical convoy na na Other government 1.8 11 Other - 2 92 Drugs The co-payments for drugs are somewhat higher than those for testing, with 7.6 LE on average. Table 51: How much did you pay the last time for Drugs? By gender and age. Total Male Female 15-39 yrs 40-64 yrs 65+ yrs Average payment in LE 7.6 8.5 7.0 7.6 7.7 6.0 Apart from the poorest quintile, poorer people pay less for drugs. Table 52: How much did you pay the last time for Drugs? By poverty quintile (predicted consumption) First (poorest) Second Third Fourth Fifth Average payment in LE 7.3 5.8 6.6 6.7 10.1 Primary care facilities are reported to charge very moderate co-payments for drugs that they dispense. The average co-payment ranges from 1.1LE in Urban Health Units to 7LE and 7.1LE respectively for HIO clinics and private doctors. Charges in hospitals are considerably higher, undoubtedly reflecting the type of drugs dispensed and the conditions for which they are dispensed. These charges ranged from 3.0LE in rural hospitals to 19.4LE in private hospitals. The highest average payment for drugs, 25.0LE, was made for a provider home visit. Table 53: How much did you pay the last time for Drugs? By type of facility. Average payment in LE Sample Urban Hospital 9.6 350 Urban Health Unit 1.1 284 Rural hospital 3.0 29 Rural health unit 3.1 472 MHC Center 1.4 20 Mobile clinic 6.3 4 Other MOH clinic 1.6 21 HIO clinic 7.0 36 HIO hospital 4.5 46 NGO clinic 2.7 206 NGO hospital 13.3 72 Private doctor 7.1 2,620 Private hospital 19.4 326 School clinic 10.0 1 Pharmacy 12.3 279 University Hospital 16.2 78 Laboratory - 43 Provider home visit 25.0 1 Medical convoy na na Other government 5.9 11 Other 10.0 2 93 When examining the distribution of drug co-payments in the primary care facilities in our sample, it appears that most people were in fact not charged for drugs. Table 54: How much did you pay the last time for Drugs? By type of facility. FHC FHU Primary HC Unit HIO Other 0 LE 74% 43% 84% 42% 60% 1-3 LE 14% 25% 3% 33% 20% 4-10 LE 6% 24% 3% 25% 20% >10 LE 3% 6% 2% 0% 0% However, during the qualitative focus groups, many users reported they avoided public facilities because they could not get the medicine they needed, or not enough of it. As a consequence, they had to go to a private pharmacy to obtain, and pay for, medications. Generally, they would be happy to pay for drugs if only they found them all at the facility right away. Sometimes the drug-copayment keeps people from using Family Health Units: “I go all the time to the family center but sometimes I don’t have money, I go to the ordinary unit because for the family center, you have to pay 1/3 of the medication but the other unit is for free.� (Healthcare User) EXEMPTION Knowledge about exemption Awareness of the payment exemption for poor people The vast majority of people, 97%, have never heard of the payment exemption for poor people in reformed facilities, see Table 55. This percentage barely varies by age and gender, or even poverty. Table 55: Are you aware of the payment exemption for poor people? By age and gender. Total Male Female 15-39 yrs 40-64 yrs 65+ yrs Yes 3% 3% 4% 3% 4% 2% No 97% 97% 96% 97% 96% 98% People who had been to a university hospital, rural health hospital, rural health unit, or other MOH clinic for their most recent visit were somewhat more aware of the payment exemption: 19%, 12%, 12% and 11%, respectively. There is also a payment exemption at university hospitals, known by 19% of patients who used that provider last time. 94 Table 56: Are you aware of the payment exemption for poor people? By provider at last visit. Yes No Sample Urban Hospital 7% 93% 456 Urban Health Unit 4% 96% 344 Rural hospital 12% 88% 34 Rural health unit 11% 89% 643 MHC Center 7% 93% 29 Mobile clinic 0% 100% 20 Other MOH clinic 12% 88% 25 HIO clinic 0% 100% 225 HIO hospital 4% 96% 304 NGO clinic 0% 100% 240 NGO hospital 5% 95% 84 Private doctor 1% 99% 3077 Private hospital 1% 99% 403 School clinic 0% 100% 9 Traditional healer 0% 100% 1 University Hospital 19% 81% 141 Laboratory 0% 100% 46 Provider home visit 0% 100% 2 Medical convoy 0% 100% 16 Other government 0% 100% 46 Other 0% 100% 26 The substantial lack of awareness of the exemption is marginally less severe in rural areas and for people who are registered in the Family Health Model. Still, in all analyzed groups the percentage of people aware did not exceed 5%. Table 57: Are you aware of the payment exemption for poor people? By region and affiliation. Alexandria Menoufia Urban Rural Registered in Not FH Model registered Yes 1% 4% 2% 4% 5% 2% No 99% 96% 98% 96% 95% 98% Sample 2,393 4,089 2,373 4,109 2,853 3,629 95 People frequenting the morning shift (who are also usually those frequenting public facilities) are marginally more aware, at 6%, than those frequenting other shifts. Table 58: Are you aware of the payment exemption for poor people? By shift visited. Morning Afternoon Evening shift shift shift Yes 6% 2% 1% No 94% 98% 99% Sample 2,611 993 2,878 Given the lack of awareness, 98.6% of heads of household do not know how to describe the exemption policies available. Around 1% are aware of studies into the social situation of the household, and exemptions granted based on the study results, and 0.4% believe that exemption is granted upon the submission of documents by the patient. Another 0.4% reported a different procedure. In the qualitative focus groups, some beneficiaries indicated that they are aware of exemption possibilities, but expressed dissatisfaction at its low coverage, both in terms of number of people and number of services covered. Application of exemption At enrollment When enrolling for the Family Health Model (FHM), 77% of those who enrolled paid a fee. Disturbingly, this proportion rises to 79-80% for the two poorest quintiles. That means that the poorest 40% of the population have a lower chance of being exempted than richer groups. Nearly one-fourth, 23%, was exempted in total—but only 21% in the poorest quintile. Of those who did not pay, 27% reported they had been exempted after a status research, 7% had the HIO pay their enrollment and 66% thought the enrollment was free of charge in any case. These percentages vary little with the respective household’s wealth. Nearly two-thirds, 61%, of the FHM enrollees report any status research into their wealth and income situation. 35% were asked questions at enrollment and received a home visit, 13% and 14% received either of these, respectively. 39% did not receive any enquiry whatsoever. Of those who enrolled in the FHM, 33% believed they should have been exempted; that is, 10% more than those who actually have been exempted. The reasons quoted were poverty (77%), illness (21%), ‘this service should be free’ (17%), age (10%) and ‘the doctor knows me’ (0.7%). During visits At their last visit to a facility, 76% of those registered in the FHM paid a fee, 24% did not. These percentages do not change with poverty status. Of those who did not pay, 34% were covered by HIO, 30% reported to have paid at another visit, and 17% believed the service was free for everyone. Only 12% believed themselves exempt due to either FHM enrollment or status research. 6% believe they have been exempted because they know the personnel at the facility. 96 Table 59: Why did you not pay a fee at your last visit? All patients Patients registered in FHM The service is free for everyone 14% 17% I'm covered by HIO 40% 34% I know the personnel at the facility 6% 6% I had already paid at a prior visit 32% 30% I am FHC/U enrolled and therefore exempt 3% 7% I was exempt after a status research 4% 5% Other 1% 1% Don't know 0% 0% Interestingly, the exemption policy appears to work better in practice at NGO, private facilities and rural hospitals. 84% of cases exempted at an NGO clinic and 81% exempted at a private doctor note FHM membership or status research as reason for exemption. This is also the case for all five cases that have been exempt at a rural hospital. Further, 42% of exempted patients frequenting a private hospital report exemption because of FHM membership. 97 Table 60: Why did you not pay a fee at your last visit? By type of provider at last visit. I know I am FHC/U I was The I had I'm the enrolled exempt service is already Don't covered personnel and after a Other Sample free for paid at a know by HIO at the therefore status everyone prior visit facility exempt research Urban Hospital 33% 22% 5% 0% 9% 4% 20% 6% 93 Urban Health Unit 31% 33% 2% 0% 13% 7% 11% 2% 54 Rural hospital 0% 0% 0% 0% 40% 60% 0% 0% 5 Rural health unit 51% 7% 7% 0% 5% 21% 8% 1% 152 MHC Center 67% 0% 11% 0% 11% 11% 0% 0% 9 Mobile clinic 100% 0% 0% 0% 0% 0% 0% 0% 16 Other MOH clinic 25% 50% 0% 0% 0% 0% 25% 0% 4 HIO clinic 1% 98% 1% 0% 0% 0% 0% 0% 189 HIO hospital 1% 97% 2% 0% 0% 0% 0% 0% 258 NGO clinic 3% 6% 6% 0% 84% 0% 0% 0% 32 NGO hospital 0% 25% 25% 0% 42% 0% 8% 0% 12 Private doctor 1% 5% 11% 1% 81% 0% 0% 0% 454 Private hospital 3% 41% 9% 2% 42% 0% 2% 2% 66 School clinic 33% 67% 0% 0% 0% 0% 0% 0% 9 Pharmacy 30% 0% 20% 10% 40% 0% 0% 0% 30 University Hospital 27% 18% 3% 0% 12% 2% 27% 12% 60 Laboratory 0% 67% 33% 0% 0% 0% 0% 0% 3 Home visit 0% 0% 100% 0% 0% 0% 0% 0% 1 Medical convoy 100% 0% 0% 0% 0% 0% 0% 0% 16 Other government 11% 89% 0% 0% 0% 0% 0% 0% 35 Other 8% 88% 0% 4% 0% 0% 0% 0% 24 The facility view While 84% of the primary facilities in our sample report offering an exemption for the poor, the percentage is slightly higher in reformed units, 86%, and much lower in HIO facilities, 29%. Table 61: Facilities offering exemptions for the poor FHC FHU Primary HC HIO Other Total Exemption 86% 86% 80% 29% 100% 84% Discount 3% 8% 10% 0% 0% 7% No 10% 7% 10% 71% 0% 9% Sample 29 225 10 7 2 273 98 81% of the facilities showed the surveyor a register where exemptions and discounts are noted. Table 62: Facility has a book where exemptions and discounts are collected FHC FHU Primary HC HIO Other Total Yes, register seen 90% 83% 60% 29% 50% 81% Yes, register not seen 7% 13% 20% 14% 0% 13% No register 3% 3% 20% 57% 50% 5% Sample 29 225 10 7 2 273 The decision-making process on exemption differs noticeably between facilities. The person to make the final decision for exemption in 41% of facilities is the Social Worker; 27% use another person, and 27% do not have an official person to make this decision. Table 63: Person that makes the final decision on exemption in facilities FHC FHU Primary HC HIO Other Total Doctor 3% 3% 0% 0% 0% 3% Social worker 41% 43% 10% 29% 100% 41% HIO official 0% 2% 0% 0% 0% 1% Other 24% 26% 90% 0% 0% 27% Does not exist 31% 27% 0% 71% 0% 27% Sample 29 225 10 7 2 273 In the qualitative research, reformed facility staff reported that the household income in an area might not be sufficient to maintain a public facility via co-payments. This would not be a problem if the poor were exempt and the reimbursement of the money arrived at the facility fast enough. UNOFFICIAL FEES There is currently no official fee scale for public facilities to charge for appointments, or tips. Reformed primary care units are also not allowed to charge for home visits (while non- reformed units can charge 10 LE). Nonetheless, some incidences of such fees are reported in all types of facilities. Appointment fees Nearly all respondents, 99.1% did not pay anything to get an appointment, only 40 people in the overall sample of around 10,000 adults reported having paid for an appointment. Home visits Of all patients, 65% report having paid extra for a home visit; that is, exceeding the usual visit fee. The percentage varies little by age or gender; however people over 65 years reported more frequent extra pay, at 70%. People registered in the FH Model paid less frequently (59%) than the non-registered (71%). The average amount per home visit was reported substantially higher in Alexandria (45LE) than Menoufia (19LE), and substantially lower (19LE) for those registered in the FHM, vs. those not registered (35LE). 99 Table 64: Did you pay extra for a home visit? By region and affiliation. Payment? Alexandria Menoufia Urban Rural Registered in FH Model Not registered Yes 68% 63% 69% 63% 59% 71% No 32% 37% 31% 38% 41% 29% Average LE 44.5 19.3 44.0 18.0 19.3 35.3 The average amount paid per home visit was 28LE; it was considerably higher for older patients (47LE) than for those in younger age groups. Table 65: How much do you pay, on average, for a home visit? By age and gender. Total 15-39 yrs 40-64 yrs 65+ yrs Average LE 28.0 23.5 29.9 47.0 Tips 99.1% of patients reported not having paid any tips, neither cash nor in kind. Only 43 of around 10,000 adults reported paying a tip, ranging from 1 to 300 LE. 6. INSTITUTIONS OF QUALITY SUPERVISION AND GOVERNANCE Primary health care facilities have the quality of their services and management supervised by a variety of institutions;  formally: through the MOH, its representatives at the governorate and district levels, as well as self-evaluation by the health facility; and  informally: through local governments and mayors, the media, religious institutions and NGOs. This chapter examines how each of these institutions function and how the health providers perceive their impact. FORMAL GOVERNANCE MECHANISMS Formal governance mechanisms cover essentially (i) administrative supervision and (ii) self- evaluation by health facilities. Administrative Supervision Administrative supervision consists of direct supervision by the district, the governorate and the MOH, as well as – in reformed facilities - supervision in the form of the accreditation process carried out by the quality assurance group of the MOH.26 26 The accreditation process is explained in detail in Annex 2. 100 District level supervision takes place in the form of routine inspections and follow-up on compliance with standards. At the governorate level, the Family Health Fund (FHF) has an autonomous supervision system which focuses on the administrative and financial arrangements of the facilities. At the central level, the central FHF is supervising the adherence to the HSRP at the regional level. Moreover, the central FHF carries out monitoring and evaluation of regional FHFs. Further direct supervision by the MOH focuses on the areas of population control and maternal and child health. The accreditation program complements this supervision by covering a wider range of services that are part of the HSRP/Family Health Model’s basic benefits package. Client Feedback at the facility Being responsive to the needs of the citizens requires knowing what the community expects from “their� facility and how the service at the facility is perceived by its clients. 101 Table 66 presents data concerning ways that facilities collect feedback from their clients and if and if not actions are taken as a result of this feedback process. Over 85 % of all facilities in Alexandria and Menoufia have a system for determining client opinion about the facility or services in place. Amongst those, a suggestion box to collect client opinion is the most frequently used instrument (67%) while 62% of this group of facilities carry out patient surveys. 25% of the facilities that collect client opinion use patient interviews to do so. Meetings between facility staff and community members happen more frequently in Alexandria than in Menoufia (71% versus 37.6%) and in urban areas compared to rural areas (66.6 % versus 42.2%). Accredited facilities have a higher rate of collecting feedback compared to non accredited facilities, the same can be observed when comparing reformed with non reformed facilities. The degree to which facilities follow up with client feedback varies. Almost half (46.5 %) of the facilities that collect client information have been able to show a report that shows how they are collecting patient information, and a bit over one third of the 85.6 % of the facilities that collect client information reported they have made changes as a result of client opinion. With respect to clients, many users and non-users of public health care reported unfamiliarity with feedback mechanisms at the facility level. During the focus group discussions many beneficiaries stated they did not know about the complaint box or had never used it. One user commented that they had seen the box but never knew its purpose. Providers, however, claimed that patients always provided feedback either through the box or verbally, the latter being particularly pronounced in rural settings. 102 Table 66: Collection of client feedback, in percentage of facilities Alexandria MOH only Reformed Menoufia Accredited accredited Reformed Urban Rural Non- All Not Hold routine meetings (at least every 45.3 48.5 51.8 25 50.4 35 71 37.6 66.6 42.2 2 mo) about facility activities of mgt issues that include both facility managers and community members Have system for determining client 85.6 87.4 95.6 55.5 97.1 62.2 86 88.3 94.1 85.3 opinion about the health facility or services. Use suggestion box (of those that 67.1 67.7 77 36.4 80.6 40 54.2 74.2 67.9 67.6 have a system for determining client opinion) Use patient surveys (of those that 61.6 52.4 60.2 20.5 65.7 20 57.9 54.6 69 47.1 have a system for determining client opinion) Use patient interviews (of those that 24.6 24.7 32.5 28.4 31.4 32.5 30.8 33 31 32.8 have a system for determining client opinion) Were able to show a report that they 46.5 47.7 50.2 23.1 53.4 22.6 55.4 44.1 62 42.3 are collecting and reporting on patient opinion (of those that have a system for determining client opinion) Have made changes in the past 3 36.2 36.3 36.2 37.5 36.9 29.4 33.3 37.9 34.8 37 months as a result of client opinion (of those that collect and report feedback) INFORMAL GOVERNANCE MECHANISMS This section examines the relationship between the facility and its outside stakeholders, such as media, political institutions, religious institutions, and NGOs. These can play an important role in supporting the community in the communication with the facility and thus help improve the quality of service delivery. Furthermore, the facility can work closely with these partners and ask for their support, for example in engaging the community. Media Press and media play a marginal role in helping the public primary health care facilities deliver their services to the citizens. Less than 4 % of all facility reported to have a relationship with the local media, including that the media would report about them. Town Administration In many public primary care facilities, the town administration is involved in one way or another. In 19.1 % of all facilities, the mayor/mayor visits the facility on a regular basis, and in 34.5 %, members of the local council visit the facility regularly. Only 35.4 % of all facilities report they do not have regular contact with politicians. 103 Table 67: Contact with mayor/town administration % of facilities that MOH Non agree with All only Reformed reformed Alexandria Menoufia We know the Mayor/Mayor 16.3 17.68 18.7 12.7 13.1 19.9 The Mayor/Mayor talks to us regularly 12.2 15.2 16.1 10.9 11.2 18.5 The Mayor/Mayor is visiting us regularly 19.1 18.9 21.2 7.2 0.9 18.6 We know some local council members and they talk to us regularly 22.93 24.4 28.9 20 18.7 27.1 Local council members visit us regularly 34.25 35.7 39.9 20 34.6 37.5 NGOs Almost two thirds (64.4 %) of all facilities do not have an NGO in the area where they operate, see Table 68.27 This number is significantly higher in Menoufia than in Alexandria (70.1% compared to 44.9%). There seems to be more of a relationship between reformed facilities and NGOs than there is between non-reformed facilities and NGOs. Whereas 8.8% of reformed facilities claim that they talk regularly to NGO representatives and 13.2% of reformed facilities cooperate with NGOs when conducting their outreach work, this is only the case for very few non reformed facilities. Table 68: Contact with NGOs % of facilities that agree All MOH Reformed Non Alexandria Menoufia with only reformed We talk to representatives 7.2 7.9 8.8 0.2 16.8 3.62 of NGOs regularly We rarely talk to 2.2 2.4 2.9 0 4.7 5 representatives of NGOs We have a positive 18.8 11.6 28.6 20 21.5 19 relationship with NGOs Local NGOs help us with our 10.2 11 13.2 5.6 17.8 7.7 community/outreach work There is no NGO in the area 64.4 61.9 59 76.4 44.9 70.1 27 Facilities that are run by NGOs were excluded from the analysis. 104 Religious leaders Many primary care facilities appear to have a collaborative and fruitful relationship with the religious institutions in their neighborhood. More than two-thirds of facility managers, 37%, reported that the mosques and churches in their neighborhood helped them announce their vaccination campaign, while 27% of facilities knew and talked to the Imam and 22% were confident he mentioned the facility in Friday prayer. Only 4% of facilities reported no contact between the facility and local religious institutions. Table 69: Contact with religious leaders % of facilities that agree with All Mosques/churches announce vaccination campaign 36.8 Facility knows/talks to Imam 27.4 Imam mentions facility in Friday prayer 22.4 Facility knows/talks to Priest 5.83 Priest mentions facility during services 5.6 No contact between facility and mosques and churches 4.4 Other When raising the issue of informal governance mechanisms with the focus group participants, most participants stressed the role of donations. As facilities struggle with budgets and are increasingly depended on donations, contributions from both organizations and individuals were emphasized as an important aspect of the interaction between the facilities and the community. 105 7. CONCLUSION AND IMPLICATIONS FOR HEALTH POLICY This chapter interprets the main findings of the report and points to possible policy measures to improve quality and utilization of public primary health care. The measures have been jointly elaborated with stakeholders from the Egyptian public health provision, from central government, the governorate level administration and health facilities, at the occasion of a stakeholder workshop in Cairo on January 21st, 2010. They are ranked in the order of their feasibility, starting with the short term (‘low hanging fruit’), followed by medium term policy measures and concluding with the long term and most far-reaching measures. POTENTIAL SHORT TERM POLICY MEASURES (‘LOW-HANGING FRUIT’) 1. Create awareness of exemption rights and procedures of enrolment in the Family Health Model As shown in chapter 3, only 3% of the population of Alexandria and Menoufia are familiar with the exemption policy for co-payments. Moreover, the application of the exemption by the providers, as shown in chapter 5, reveals the lack of a clear process that ensures that patients are exempted as intended by the reform – by having clear criteria by which the social worker at the facility can make an assessment regarding eligibility. This results in inequities in application of the exemption policy and in under-coverage through this policy. There is therefore an urgent need to (1) simplify and unify the exemption process across facilities and districts; and (2) create awareness amongst providers, users and non users. Clarification and unification of the process require the implementation of the existing exemption guidelines of the Ministry of Social Solidarity (MOSS) through training at all involved levels. Creation of awareness calls for an intensive dialogue with providers and with users and non-users of the facilities. Beneficiaries need to be made aware of their rights as well as duties, and criteria need to be transparent, clear, and equitably applied. 2. Review facility shift patterns and align with the demands of the community According to the manual laid out by the HSRP, all facilities should be operating 24 hours, and have a physician on call during all this period. In reality, many facilities do not operate a second or third shift, or do so without a trained doctor at the facility. Moreover, when matching the household and the facility surveys, there is a discrepancy between the times when beneficiaries prefer to visit the health facility and the actual opening times. Most facilities are opened during the morning, whereas a majority of the people prefer to visit the facility at night. It would be better to use resources more efficiently by aligning the opening times of the facility with the demands of the population. A possible policy recommendation could therefore include (1) an analysis at the facility level of the community needs and; (2) aligning the opening times of the facilities with the community demands. There are also options of working within the current shift regimes. At the moment, the morning shift is covering 8am to 2pm in most facilities. This may be too early for beneficiaries who have to come to the facility in the afternoon and who, if there is no operating second or third shirt, have no option other than seeking care in a different (usually private) facility. Here, changing the shift regime slightly, say from 10am to 4pm, might be more suitable and would serve a larger share of the population without increasing the number of staff hours. 106 3. Increase training efforts for operation of existing medical equipment One issue raised by providers is that training content under the HSRP does not always match the existing medical equipment at the facilities, especially with regards to costly medical technology, such as x-ray machines. Sometimes facilities are equipped with this technology, but the staff lack the necessary qualifications and experience to use them and to interpret the findings for clinical care. To ensure that the available equipment can be used appropriately, it is advised that the training plan should be aligned with the specific needs of each facility. Training modules on how to use certain equipment for diagnosis and treatment planning should be based on the availability of equipment at the facility. It is a loss of expensive infrastructure investments if equipment cannot be used by the medical staff and a potential threat to the health of beneficiaries if they are not applied properly. 4. Improve adherence to hygiene practices/ standards As pointed out in chapter 4, the observance of basic hygiene practices by doctors is alarming. In only 22.4 % of structural observations of client consultations did the provider wash his/her hands with soap and water prior to engaging with the patient. The numbers are on a similarly low level for the usage of disposable gloves and the change of paper on the examination table. While the survey did not probe into the hygiene practices of beneficiaries, the qualitative survey revealed that there is a major information gap between providers and beneficiaries, in part exacerbated by real or perceived hierarchical differences. Most patients reported they do not dare to speak up to a doctor who is neglecting his/her duty to carry out basic hygiene practices. This is an essential issue and it is in the MOH’s strategic interest to react quickly and deliberately to improve the situation. Among the possibilities for short term policy measures is an awareness / advocacy campaign to remind doctors of the importance of basic hygiene procedures. Using visual aides, ideally located in areas where they can also be seen by the users, can further increase the effect of this short term measure. Further, according to discussions in the focus groups, theft of soap is a common issue in many facilities and hence many providers came up with the suggestion to use liquid soap in dispensers that are attached to the wall. PROPOSED MEDIUM TERM POLICY MEASURES 5. Overhaul training and guidelines for CHD and diabetes treatment Chapter 4 shows that throughout all structural observations, the quality score was higher for sick child and antenatal consultations than for CHD and diabetes consultations. The insufficient treatment for diabetes may have consequences for the awareness of diabetes in the population. The household survey found that only 5.41% of respondents reported having diabetes. This contrasts with the WHO’s estimation according to which Egypt has an 11% prevalence of diabetes in 2008. There is a need to overhaul existing diabetes practices and guidelines for General Practitioners, in cooperation with faculties for medicine in universities. Moreover, there seems to be a rather urgent need to create awareness of non communicable diseases among the population, with a particular focus on prevention. Another policy recommendation would be to use adherence to consultation protocols in the supervision and incentive systems. 107 6. Improve supply chain system for essential drugs The assessment of the availability of drugs (see chapter 4) suggests that many facilities still lack basic drugs and that many beneficiaries are not able to get their medications at the facility. As a consequence, they are required to visit another pharmacy, mostly in the private sector. This poses a significant constraint to the quality of care, as people may have to switch between different variations of the same drug, or use another one due to unavailability. This can have severe consequences on health status of beneficiaries. From the point of view of facility managers, the problem lies with the procurement system for essential drugs. Many facility managers noted both during the qualitative and quantitative survey that often times they are not receiving the amount of drugs they had ordered, and are constantly out of stock. The main policy action should therefore focus on improving the supply chain for drugs and make sure that facilities do receive the drugs they need and order. This will likely have to be accompanied with an increased drug budget in the primary health sector. Another, less costly, policy option would be to involve the community in monitoring the system of drug stocks, and implement a community based feedback system to the district. PROPOSED LONG TERM POLICY MEASURES 7. Ensure continuity of care The chapter on human resources (chapter 4) points out a clear pattern of staff turnover. Most nurses and administrative staff seem to stay at a certain facility over a long period of time. However, the picture is very different for specialized medical staff—doctors and pharmacists. Over 50 % of surveyed doctors have spent less than 12 months at the facility where they are currently employed. Moreover, doctors tend to be much younger and much less experienced than other staff categories at the facility level. While the quantitative survey did not probe into the reasons for the differences in turnover, the qualitative discussions with providers offered an insight: human resource regulations oblige physicians to take up a position at a public facility for a certain amount of time. Once at the facilities, many doctors prefer to move to the private sector – where pay is slightly better – either permanently, or while on leave from the public sector. The incentive payment regime introduced by the HSRP has tried to ease this concern. However, over the last year, incentive payments have been decreased dramatically as a consequence of funding issues of the FHFs. This problem can be addressed in a sustainable way only in the long term. Policy options could include devising and implementing a plan to move facilities to fully efficient rosters, with a pay scale for personnel that competes with the risk-return profile of the private sector. It will also be necessary to supervise doctors’ taking off time to work in the private sector, and to eliminate any ‘double harvesting’ of public sector security and benefits and private sector pay. In the short term, the following measures may improve the situation. (1) Improve the filing system to allow for a smooth transition between two physicians. Here, focusing on improving the information technology of the facilities, as well as having a centrally managed information system may help. There should be no loss of information if a patient switches from one facility to another or, even more importantly, between two physicians at the same facility. (2) Limit the time lag between departing and arriving physicians at the facility to allow for a structured transition. According to feedback received mostly in the qualitative survey, there is a 108 substantial time lag between a departing and an arriving physician at some facilities that leads to undesired absences at certain facilities. (3) In collaboration with universities, improve training and status of “Family Physician� by awarding the profession the status of a specialist. This is likely the most sustainable of the short term measures. If universities are able to improve the standing of family physicians, by introducing improved curricula, more young doctors may be interested in taking up the position of family physicians, and, more importantly, may be willing to stay for a longer time at a given facility. 8. Unify fee regimes across governorates The facility survey reveals that the co-payment regime introduced by the HSRP has not yet been applied in each facility throughout the two governorates. This leads to an unsatisfying situation whereby beneficiaries have incentives to avoid reformed facilities because they can get the treatment at the non reformed facilities at a lower price. Making things worse, there have been situations where two different fee regimes are applied in two facilities that are located in the same building. For the reform to work, it is essential that the regime is uniformly applied in all public facilities in a reformed governorate. The MOH should therefore urge to create a situation where all facilities are operating according to the same standards, and that the regime is clearly communicated to beneficiaries. This will be also essential for financial sustainability. 109 ANNEX 1: GLOSSARY OF PRIMARY HEALTH SERVICE DELIVERY IN ALEXANDRIA AND MENOUFIA At the Central government level, the Ministries of Finance, of Health (MOH), of Education and of Social Solidarity (MOSS) are all involved in healthcare organization and financing, see Figure 12. At the Governorate Level, the Family Health Fund (FHF) and the Health Insurance Organization (HIO) both provide financing, for different types of clinics. The Health Directorate is the governorate-arm of the MOH. At the Health District level, the District Provider Organization provides a de-centralized management layer. Different primary, secondary and tertiary clinics provide care. Please note that from the governorate level down, aspects of the described system may be specific to Alexandria and Menoufia, as these two are ‘reform’ governorates where the Health Sector Reform Project (HSRP) is already implemented. Figure 12: Health Service Delivery in Alexandria and Menoufia Below we explain the main institutions specific to the Egyptian health system in detail. Family Health Fund The Family Health Fund (FHF) was developed under the Health Sector Reform Project (HSRP) as the main contracting and purchasing agency for primary health care services. It is the entity which was supposed to put into effect the separation between service provision and financing in health services. The role of the FHF is to purchase primary health care (and potentially 110 secondary care in the future) by contracting health service providers. The FHF was funded by ministerial decree in 1999 and put in place in 2001. The FHF plays an important role in primary health delivery on two fronts. First, as the contracting agency, the FHF establishes and supervises rules and eligibility criteria for primary health care providers. The FHF has the right to contract with public, private, and NGO providers if they apply the Funds’ guidelines and if they were accredited by the MOH accreditation unit. The accreditation process is a four step procedure including (1) a primary assessment of the unit to establish an action plan for accreditation; (2) staff training in the areas of quality improvement, infection control, leadership, the accreditation program and clinical guidelines; (3) pre-accreditation through on-job training to evaluate the current performance and to establish an improvement plan and finally (4) accreditation based on an assessment of patient rights, patient care, support services, compliance with safety standards, quality improvement program, family practice, and quality of facility management. The accreditation unit awards a score to each facility and facilities can be accredited for two years if their score is above 80%, and for one year if they score between 50% and 80%. Facilities that score below 50% cannot be accredited by the FHF. Second, as the main financing entity in the health sector in the reformed governorates the FHF plays an important role in channeling the funds allocated to primary health services. The FHF is financed by a mix of sources including the contracting fees paid by the HIO to the health directorate, contributions from the Ministry of Finance, the budget guarantee from the MOH to cover any financing gaps in the HSRP, revenues from the co-payment system put in place by the Ministerial Decree 147, as well as other funds approved by the Minister of Health. On the spending side, the FHF is responsible for paying salaries of staff, as well as monthly incentive payments to contracted facilities based on pre-defined performance criteria. The FHF is instrumental in the administration of co-payments at the facility level. On the level and structure of co-payments in different primary care units in Alexandria and Menoufia, kindly see chapter 5 “Payments�. These revenues generated at the facility level are transferred to the FHF which keeps 50% of the income. 36% of the co-payment revenues are transferred back to the facility where they are kept in a service improvement fund to administer small procurement at the local level; another 4% goes to the MOH. The remaining 10% are incentive payments that may or may not be transferred back to the facility, depending on their monthly evaluation by the FHF. Furthermore, the FHF reimburses the facilities for treatments of the poor which are exempted from the co-payment regime described in chapter 5. Health Insurance Organization (HIO) The HIO was established in 1963 with the historic mandate to cover all Egyptians with social health insurance. However, four decades later, the HIO manages several separate compulsory health insurance programs only for formal sector workers, pensioners, widows, and schoolchildren, and for infants – but not for informal workers and housewives. The HIO has 13 regional branches and operates a nationwide system of health facilities for its beneficiaries. It also contracts with public and private providers to extend services to beneficiaries that it is unable to provide within its own network. More specifically, HIO contracts with individual doctors to work in its facilities (about 25 % of all staff) as well as with public and private 111 providers and pharmacies to serve the health care needs of the HIO’s beneficiaries. Therefore, the HIO can be characterized as both a purchaser and a provider of health care services. The HIO offers its services with a network of approximately 9,000 primary care facilities (mostly small clinics in schools) and approximately 40 secondary care facilities (hospitals) throughout Egypt. Despite the fact that almost half of the population is covered by the HIO, only 5-10% of all health care visits in Egypt are made to HIO facilities/providers, pointing to quality issues with HIO delivered services. This vacuum is taken over by the private insurers. In Cairo alone, 48 prepaid health care plans cover two million enrollees. About 70 % of private coverage arises through employer groups (including, notably, the Ministry of Finance as well as other government agencies). The growth of prepaid plans indicates that employers are increasingly “opting out� of enrolment in the HIO. Likewise, there are about 5,000 private primary care facilities and 1,200 private secondary care facilities throughout Egypt. District Provider Organization A key institutional layer at the district level is the District Provider Organization (DPO). The DPO work across lines of different agencies and entities and include all relevant health care points of service in the district, a traditionally defined geographic entity, including those provided by the MOH, other government agencies, private providers and NGOs. A district health system thus consists of a variety of different elements that contribute directly to the health of the population in schools, work places, and communities. It comprises all health care workers and facilities up to and including the hospitals, as well as the respective support services, such as laboratories, and diagnostic instruments, and logistical support. The DPOs are instrumental to the HSRP in that they plan and coordinate the health services provided by different players in a district. In particular, the DPO is in charge of the following: (1) Managerial authority: The DPO will maintain the long term implementation of the HSRP. Strengthening the managerial capacity of the DPO will improve the performance of service delivery at the facility level. (2) Integration of Primary Health Care services. This is particularly relevant for the establishment of standardized health services and quality assurance. (3) Establishment of family physician roster. (4) Guaranteeing quality standards as well as compliance with district’s health needs of private providers. (5) Implementing a referral system at the district level. (6) Streamlining quality standards across all facility and all providers (7) Training practice model and the establishment of a training center in each district. The rationale behind the DPO system is that there is substantial managerial authority at a local level and the DPO can function as an interface between individual facilities and decision makers at the central level. This should enable the DPO to have an open dialogue with the population and to cater to the beneficiaries’ needs. At the same time the DPO is supervised by the MOH and has to comply with some requirements of the MOH, such as the implementation of the HSRP, completing the training plans for District health authority staff, completing the district health business plan, implementing the FHF, completing the accreditation process for at least 112 60% of the facilities, and signing service level agreements between district level authorities and health facilities. Typology of different clinics The health sector reform defines that the public and private service delivery at governorate level is organized in a system based on the family health model, where preventive, promotive and curative health services are provided through family health units (FHU), family health centers (FHC) and district hospitals (DH). These facilities constitute the first 3 levels of a 5 level system (Figure 13). FHUs, FHCs, district hospitals, as well as general and specialized hospitals can be managed both publicly and privately, using standardized service guidelines, as well as quality indicators. Furthermore, a substantial number of clinics are run by NGOs. Figure 13: Typology of health clinics according to the Health Sector Reform Project Source: MOH, 2004 113 ANNEX 2: DATA COLLECTION QUANTITATIVE DATA COLLECTION Quantitative data was collected from both primary care facilities and households, in the Governorates of Alexandria and Menoufia. Facility Survey The quantitative facility survey was carried out as a census in all 362 public primary health care facilities in Menoufia and Alexandria. This survey follows some key elements used for public expenditure tracking surveys (PETS) and quantitative service delivery surveys (QSDS). According to Dehn, Reinikka and Svensson (2003) PETS and QSDS typically collect data about the characteristics of the facility, inputs, outputs, quality, financing and institutional mechanisms and accountability. The facility survey combined interviews with facility management and the collection of data from the accounting registers at the point of delivery. Moreover, the questionnaire developed for this survey included a thorough staff roster that looked at the presence of each staff on three different days during unannounced visits. In addition, in each of the surveyed facilities, at least two consultations for the treatment of sick child, antenatal care, coronary heart disease (CHD) and diabetes were tracked with a real-time structural observation (no vignette). In the case of diabetes and CHD, short patient interviews were conducted after the consultation as well. The quantitative data collection was carried out between April and June 2009. Facilities were instructed in advance about a possible visit of the survey team, through the MOH and the HIO respectively. However, only the period of the potential visits was announced to them. On the day of the visit itself, no notice was given to facility staff prior to the arrival of the survey team. Subsequently, each facility was visited two additional times unannounced. If the number of consultation observations during the visits were too low (minimum of two consultation per facility per observation), the survey team arranged for an additional, announced visit to ensure the required minimum number of consultations were observed. Facility Sample The total sample of the quantitative survey includes 362 health facilities out of which 128 facilities are in Alexandria and 234 in Menoufia. The major clinic type, accounting for over 70% of the clinics in the sample is Family Health Units (FHU). Furthermore there are 33 Family health centers (FHC) and 37 Primary Health Care (PHC) clinics in the sample. Combined 322 of the 362 facilities (88%) of the clinics are run by the Ministry of Health (MOH). 30 facilities (19 in Alexandria, 11 in Menoufia) are run by the Health Insurance Organization (HIO), with the share of HIO clinics as part of the total sample being substantially higher in Alexandria than in Menoufia, 14% compared to 4% respectively. 114 Table 70: Sample Summary: Types of Facilities Type of Facility Alexandria Menoufia Total Family Health Centre 13 20 33 Family Health Unit 71 187 258 Primary Health Care 23 14 37 HIO Clinic 19 10 29 Other 2 3 5 MOH Reformed 78 196 274 MOH Non Reformed 31 28 59 Total 128 234 362 Household Survey The household survey data was collected in each of the surveyed facilities’ catchment areas. A total of 15 households were selected around each facility on a random walk approach by the survey firm and asked about their knowledge of health services, the decision process that affect the choice of health services, the overall health expenditure patterns, service experience and satisfaction. This way of conducting interviews was chosen in lieu of exit interviews at the point of service. In particular, this methodology should create a sample with a representative number of users and non-users of health services, and users of health services at other points of delivery. This should lead to conclusions that are valuable to examine the health pattern of the population on a representative scale. A total of 5,471 households have been surveyed; 1,952 in Alexandria and 3,465 in Menoufia. In each of the households, interviews were conducted with all members of the randomly selected nuclear family, amounting to a total of 21,703 individual respondents. The mean family has 4.8 members and the average age of the respondents is 25. Furthermore, there are slightly more men in the sample than women (11,153 male and 10,550 female respondents). The urban/rural setting between the two governorates is significant: In Menoufia, 90% of the households live in rural settings and only 10% in urban areas. In Alexandria, the pattern is quite different with 1/4 of the population living in rural areas and over 75% in urban areas. Table 71: Sample Summary: Households Alexandria Menoufia Total Households Total 1,952 3,465 5,417 Urban 75% 10% 34% Rural 25% 90% 66% Individuals Total 7,847 13,856 21,703 Male 51% 52% 51% Female 49% 48% 49% 115 Qualitative Data Collection In-depth Interviews In-depth interviews with both providers and clients were carried out in July and August 2009 in a semi-structured, qualitative design with a purposive respondent sample aimed at contributing to the findings of the quantitative survey instruments and validating and explaining the quantitative data. The facilities were chosen based on stratification variables in the quantitative data, most prominently urban/rural and well performing/poor performing facilities. In each of the selected facilities’ catchment areas, users and non-users were selected for the client interviews. Table 82 gives an overview of the interviewees, as well as the location of the selected facilities. Table 72: Sample for the In-depth Interviews Clinic Governorate Clients Providers Deberky Menoufia 3 users Director/Financial Manager/Social 2 non users Worker/Pharmacist/Head of Nurses El Bagour Menoufia 2 users Director/Financial Manager/Social 3 non users Worker/Pharmacist/Head of Nurses El Suef Alexandria 3 users Deputy Director/Financial Manager/Social 2 non users Worker/Pharmacist/Head of Nurses/ Abees Alexandria 2 users Director/Financial Manager/Social 3 non users Worker/Pharmacist/Head of Nurses Focus Group Discussions After termination of the quantitative data collection and the in-depth interviews, focus group discussions with both providers and clients were organized to probe into the findings of the other instruments. The participants of the focus group discussions were the same as in the in- depth interviews. The focus group discussions were carried out based on a facilitation guide including 5 to 8 questions per discussion. The format was open; a facilitator only guaranteed the main questions were followed. A total of eight focus group discussions were held, four each in Alexandria and Menoufia, as well as four with providers and four with beneficiaries, as summarized in Table 73. The focus groups were structured by professions and gender: two focus groups included physicians, pharmacists, and nurses; the other two provider focus group included administrative staff; the beneficiaries’ focus groups included both users and non-users, and they were divided into a 116 male and a female focus group. In both governorates, the provider and beneficiaries focus groups were with providers and (potential) clients of the same facility. Table 73: Summary of focus group discussions Menoufia Alexandria Health Providers 1 1 Administrative and Social Staff 1 1 Female Beneficiaries (Users) 1 0 Male Beneficiaries (Users) 0 1 Female Beneficiaries (Non-Users) 0 1 Male Beneficiaries (Non- Users) 1 0 Total 4 4 Institutional Interviews Additional information was collected at the institutional level through semi-structured interviews with different levels of management of the MOH and the FHF. The guidelines for these interviews were developed jointly between the bank team and the MOH. The interviews were conducted by staff from the technical support office (TSO) to the HSRP. The goal of these interviews were to understand the processes of flows of resources between the central level, the health governorates and the facilities, which can then be compared with the perception of the process at the point of delivery. This should add value of the findings of the quantitative interviews at the facility level. To deepen the understanding of the survey results, in-depth interviews with providers have been carried out to discuss some of the issues raised. 117 ANNEX 3: FULL TABLE OF DRUG AVAILABILITY Table 74: Availability of drugs All MOH Alexa Meno Urban Rural Refor Non Accre Not ndria ufia med refor dited accred med ited Amoxicillin Oral 95.6 97.6 98.1 97.3 98.8 97.1 97.4 98.2 97.5 97.7 Quantity of Amoxicillin Oral in 90.2 89.7 81 94 95.2 87.8 92.5 75.9 94 77.7 stock matches quantity in books Aspirin Oral 74.4 73.2 88.8 65.6 88.1 68 73.3 72.7 75.1 67.8 Ciprofloxin PO 28.9 24.4 11.2 30.8 11.9 28.7 24.5 23.6 27 17.2 Quantity of Ciprofloxin PO in stock 97.1 96.3 75 100 100 95.7 100 76.9 100 80 matches quantity in books Cotromoxazole Oral 77.8 77.1 62.6 84.2 70.2 79.5 82.8 49.1 84.2 57.5 Quantity of Cotromoxazole Oral in 95.4 94.9 88.1 97.3 94.9 94.9 96 85.2 97 86 stock matches quantity in books Doxycycline 7.8 4.9 11.2 1.8 8.3 3.7 3.3 12.7 3.7 8.1 Quantity of Doxycycline in stock 85.7 75 75 75 100 55.6 88.9 57.1 88.9 57.1 matches quantity in books Ergometrine/methergine 3 3 1.9 3.6 2.4 3.3 2.6 5.5 2.9 3.5 Erythromycin 67.5 68.3 62.6 71 69.1 68 73.6 41.8 75.1 49.4 Folic acid 49.4 45.7 58.9 39.4 66.7 38.5 50.2 23.6 52.3 27.6 Iron 82.5 81.7 92.5 76.5 89.3 79.1 81.7 81.8 83.4 77 Mebendazole Oral 55 53.4 55.1 52.5 45.2 56.2 52.8 56.4 54.4 50.6 Methyldopa 16.4 11.9 28 4 32.1 4.9 11.7 12.7 11.6 12.6 Metronitiazole (Flagyl) 97.5 97.9 96.3 98.6 95.2 98.8 97.8 98.2 98.3 96.6 Multivitamins 91.1 91.5 86.9 93.7 88.1 92.6 93.8 80 94.6 82.8 Paracetamol Oral 92.8 93 80.4 99.1 86.9 95.1 96 78.2 96.3 83.9 Paracetamol Syrup 120 mg. 85 89.9 81.3 94.1 88.1 90.6 93 74.6 93.8 79.3 Paracetamol Drops 100mg 3.3 3.7 4.7 3.2 4.8 3.3 4 1.8 2.9 5.8 Ibuprofen Syrup 100 mg 34.4 34.2 25.2 38.5 21.4 38.5 33 40 34 34.5 Ibuprofen tablets 400 mg 8.9 7.9 11.2 6.3 13.1 6.2 8.8 3.6 7.9 8.1 Penicilin Oral 9.2 9.5 21.5 3.6 11.9 8.6 9.5 9.1 7.9 13.8 Tatracycline Oral 62.5 62.5 65.4 61.1 63.1 62.3 61.9 65.5 61.8 64.4 Vitamin A low dose (25,000 or 80.3 85.4 89.7 83.3 89.3 84 85.4 85.5 87.6 79.3 30,000 iu) Oral Rehydration Salts 62.2 61 91.6 46.2 82.1 53.7 62.6 52.7 63.9 52.9 Quantity of Oral Rehydration Salts 91.1 90 82.7 97.1 94.2 87.8 94.7 62.1 96.1 69.6 in stock matches quantity in books Metformin 850mg 93.3 95.1 95.3 95 95.2 95.1 96.7 87.3 96.7 90.8 Glibenclamide 5mg 41.9 39 28 44.3 39.3 38.9 40.3 32.7 43.6 26.4 Gliclazide 80mg 12.8 11 6.5 13.1 7.1 12.3 11.4 9.1 12.5 6.9 Glimepiride 1mg 5 1.5 4.7 13.1 6 11.5 1.8 0 2 0 Bezafibrate 80mg 6.4 0.9 2.8 13.1 3.6 11.5 1.1 0 1.2 100 Warfarin 2g 8 1.5 4.7 13.1 4.8 0.4 1.5 1.8 1.7 1.2 118 All MOH Alexa Meno Urban Rural Refor Non Accre Not ndria ufia med refor dited accred med ited Warfarin 5mg 12.5 6.4 16.8 1.4 17.9 2.5 7.3 1.8 8.3 1.2 Propranolol 10mg 7.8 2.1 3.7 1.4 3.6 1.6 2.2 1.8 2.5 1.2 Propranolol 40mg 50.6 47.9 14 64.3 20.2 57.4 50.6 34.6 52.7 34.5 Captopril 50mg 9.4 3.4 5.6 2.3 6 2.5 3.3 3.6 2.9 4.6 Indapamide 2.5mg 17.8 11.9 26.2 5 26.2 7 12.1 10.9 12.9 9.2 Spironolactone 25mg 15.6 9.8 28 0.9 34.5 1.2 11.7 0 12.9 1.2 Nifedipine 20mg 4.7 13.1 4.8 0.4 4.8 0.4 1.5 1.8 1.7 1.2 Diltiazem 60mg 8.6 3.7 11.2 13.1 14.3 11.5 4.4 0 5 0 Bisoprolol 5mg 13.3 7.9 18.7 2.7 22.6 2.9 9.5 0 9.5 3.45 Ergometrine/oxytoxin 46.7 48.2 40.2 52 38.1 51.6 46.5 56.4 46.1 54 Isosorbide dinitrate 5mg 4.4 1.2 3.7 13.12 4.8 11.5 1.5 0 1.7 0 Isosorbide dinitrate 10mg 5.8 1.2 3.7 13.1 4.8 11.5 1.8 0 1.7 0 Isosorbide dinitrate 20mg 24.7 19.8 58 1.4 65.5 4.1 20.9 14.6 22.4 12.6 Glyceryl trinitrate (long-acting) 18.3 13.4 35.5 2.7 44.1 2.9 16.1 0 17 3.45 2.5mg Digoxin 2.5mg 86.7 87.8 82.2 90.5 85.7 88.5 88.3 85.5 88.4 86.2 Antibiotic Eye Ointment 44.7 44.8 72 31.7 69 36.5 45.1 43.6 46.1 41.4 Benzyl Penicilin (Procaine) 1.4 0.6 1.9 13.1 1.2 0.8 10.6 3.6 12 2.3 Diazepam 6.7 7.3 1.8 10 3.6 8.6 8 5.5 7.5 6.9 Ergometrine/oxytoxin 46.7 48.2 40.2 52 38.1 51.6 46.5 56.4 46.1 54 Gentamycin 11.4 11.3 27.1 3.6 16.7 9.4 8 27.3 7.9 20.7 Xylocaine 1.7 0.6 0 0.9 1.2 0.8 0.4 1.8 0.4 1.2 Neparin Na 5000u 70.6 74.1 61.7 80.1 66.7 76.6 75.8 65.5 77.2 66.5 Furosemide 40g 86.4 91.5 81.3 96.4 79.8 95.5 90.8 94.6 90.9 93.1 Saline 83.3 88.7 73.8 95.9 67.9 95.9 88.6 89.1 90 85.1 Dextrose and Water 17.2 18 13 20.4 20.2 17.2 19.1 12.7 19.5 13.8 Ringers Lactate 3.3 3.7 1.9 4.5 1.2 4.9 4 1.8 4.2 2.3 D5NS 46.4 48.5 32.7 56.1 42.9 50.4 54.2 20 56.4 26.4 119 ANNEX 4: CALCULATION OF PER-CAPITA CONSUMPTION INDEX THE DATA The data used are from the national Egyptian Household Income, Expenditure and Consumption Survey (HIECS) 2008/2009 restricted to the governorates of Alexandria and Menoufia, in order to correspond to the regions of the survey documented in this report (hereafter “QSDS 2009�). This restriction resulted in the use of a subsample of 4989 households. THE VARIABLES Possessions A sub list of possession was used in the regression in order to correspond to the list in the QSDS 2008. The HIECS full list contains 32 possessions listed in Table 75. The table also includes the description and descriptive statistics of the possessions. The two lists of possessions were made identical through the following actions:  Possessions in12yn and in13yn were grouped in order to correspond to the question HH101K “Own automatic washing machines� in the QSDS 2008  Possessions in29yn and in30yn were grouped in order to correspond to HH101Y “Own standard camera/video camera� in the QSDS 2008  Possessions in29yn and in30yn were not used because they do not have an equivalent in the QSDS 2008. Housing Characteristics While running the regression, two different sets of housing characteristic were used. The first set corresponds to the full categories of the housing characteristics available in the HIECS, see Table 76. As for the second set a list of dummy variables was created following the literature and in order to correspond to the QSDS 2008 in order to best describe housing characteristics. Details of this list are enclosed in Table 77. CALCULATION OF PREDICTED CONSUMPTION Following Filmer and Scott (2008), who consider the merits of different types of asset and consumption indexes, a predicted consumption index was constructed with information on housing characteristics and household assets. The predicted consumption index represents the linear combination of assets and housing characteristics that best predicts per capita consumption expenditures. This regression-based prediction is consistent with what is sometimes used in proxy means tested approaches to targeting populations for social programs. It is also considered the best possible “economic� asset index that can be constructed from information about asset holdings and housing characteristics. The form used to calculate the indexes takes the following shape: 120 Predicted Consumptioni = b1 a1i + b2 a2i + …+ bk aki where Predicted Consumptioni is the predicted consumption for household i, a1i , a2i, …, aki are k indicators of asset ownership and housing quality variables described in Tables 1-3, and b1, b2, …, bk are weights used to aggregate the indicators into an index. The weights were derived from ordinary least square regressions (OLS) of per capita consumption on the households’ asset and housing indicators, that is, the estimates of the �’s in the following equation: pc_cons = �1 a1i + �2 a2i + …+ �k aki + �i The final step in the process was to divide households into consumption quintiles based on the distribution of the predicted consumption measures. This classification of households into one of five quintiles is used in the report to compare households of varying socio-economic status. 121 Table 75: Description of possessions Variable Description Minimum Maximum Mean name in01yn Own private car 0 1 0.062 in02yn Own bicycles 0 1 0.165 in03yn Own motorcycle 0 1 0.028 in04yn Own telephone 0 1 0.593 in05yn Own cellular phone 0 1 0.713 in06yn Own Internet lines 0 1 0.068 in07yn Own Refrigerator 0 1 0.916 in08yn Own Deep Freezer 0 1 0.050 in09yn Own Gas bottles/Electricity/ microwave 0 1 0.979 in10yn Own Microwave 0 1 0.027 in11yn Own normal electric washing machine 0 1 0.708 in12yn Own Half Automatic washing machine 0 1 0.073 in13yn Own Full Automatic washing machine 0 1 0.225 in14yn Own electric dish machine 0 1 0.010 in15yn Own Gas/electric water heater 0 1 0.522 in16yn Own vacuum cleaner 0 1 0.231 in17yn Own Air condition 0 1 0.031 in18yn Own Electric Fan 0 1 0.825 in19yn Own Electric/ gas bottle/ Kerosene heater 0 1 0.035 in20yn Own electric Iron 0 1 0.828 in21yn Own colored Television 0 1 0.871 in22yn Own Black/white Television 0 1 0.082 in23yn Own Video/DVD 0 1 0.063 in24yn Own Cassette (Normal/stereo/Radio) 0 1 0.665 in25yn Own MP3/ MP4/ MP5 0 1 0.019 in26yn Own Dish/Dish Cable 0 1 0.685 in27yn Own Personal Computer/ Laptop 0 1 0.208 in28yn Own Blender/mixer 0 1 0.892 in29yn Own Normal Camera 0 1 0.036 in30yn Own Digital / Video Camera 0 1 0.014 in31yn Own Sewing Machine 0 1 0.056 in32yn Own Kitchen machine/mixer 0 1 0.022 Own half/full automatic washing machines wash_mach 0 1 0.296 (is the sum of in12yn and in13yn) Own standard/digital camera/video camera Cameras 0 1 0.047 (is the sum of in29yn and in30yn) 122 Table 76: Housing characteristics as they appear in HIECS 08/09 Variable Variable Min Max Mean Category Category label Frequency % name description code h01 House type 1 7 1.655 1 Apartment 3994 80.05 more than one 74 1.48 2 apartment 3 Villa 9 0.19 4 Rural house 546 10.95 Rooms in shared 295 5.91 5 apartment 6 Separate rooms 69 1.37 7 others 2 0.05 number of 1 20 3.601 h02 rooms h03 house area 10 560 82.099 h04 house tenure 1 8 3.485 1 rented 1187 23.80 2 rented by new law 282 5.65 3 furnished 11 0.21 4 owned 2409 48.29 5 ownership 674 13.51 6 gift 396 7.94 7 work offered 19 0.38 8 others 11 0.23 h05 water source 1 4 1.045 1 public network 4799 96.188 2 pump 173 3.4637 3 well 1 0.0206 4 others 16 0.3274 connected to 1 3 1.107 Tap inside dwelling 4675 93.7 h06 water network 1 2 Tap outside 93 1.9 3 No connection 221 4.4 connected to sewerage 1 5 1.747 public network 3134 62.8 h07 system 1 2 local network 46 0.9 3 well 1755 35.2 4 others 45 0.9 5 No connection 9 0.2 main source of 1 4 1.003 Electricity 4980 99.8 h08 lighting 1 2 Kerosene 6 0.1 3 Gas bottels 1 0 4 Others 2 0 main source of 1 6 1.170 Gas bottles 4179 83.8 h09 fuel 1 2 Natural Gas 780 15.6 3 Kerosene 25 0.5 4 Electricity 2 0 5 Wood 1 0 123 Variable Variable Min Max Mean Category Category label Frequency % name description code 6 others 1 0 h10 wall material 1 8 1.281 1 Bricks & cement 4514 90.5 2 Concrete 6 0.1 3 Stones 20 0.4 4 Mud bricks 445 8.9 5 Wood 2 0 6 Tin 0 0 7 Aspistos 0 0 8 Others 2 0 h11 Separate kitchen 1 3 1.221 1 Private 4356 87.3 2 Shared 163 3.3 3 Does not exist 470 9.4 Separate 1 3 2.908 h12 BATHROOM 1 Private 215 4.3 2 Shared 27 0.5 3 Does not exist 4747 95.2 Bathroom and toilet with 1 3 1.723 h13a flusher 1 Private 3146 63.1 2 Shared 80 1.6 3 Does not exist 1763 35.3 Bathroom and toilet without 1 3 2.467 h13b flusher 1 Private 1235 24.8 2 Shared 188 3.8 3 Does not exist 3566 71.5 Seperate toilet 1 3 2.968 h14a with flusher 1 Private 78 1.6 2 Shared 2 0 3 Does not exist 4909 98.4 Seperate toilet 1 3 2.883 h14b without flusher 1 Private 260 5.2 2 Shared 63 1.3 3 Does not exist 4666 93.5 means of Garbage 1 5 2.725 h15 disposals 1 Garbage collector 935 18.7 Garbage collecting 2 company 1198 24 3 public containers 1869 37.5 4 Street 282 5.7 5 others 705 14.1 124 Table 77: Dummy variables describing housing characteristics Variable Variable description Min Max Mean name apartment = 1 if House type apartment or more than one apartment 0 1 0.814 rented = 1 if House rented under old or new law or furnished 0 1 0.297 Owned = 1 if House owned by the household 0 1 0.618 Waternet = 1 if House connected to a public water network 0 1 0.962 Tap = 1 if House connected to a tap inside 0 1 0.937 Outtap = 1 if House is connected to an outside tap 0 1 0.019 sewage = 1 if House connected to public sewage network 0 1 0.627 electric = 1 if Electricity is the main source of lighting 0 1 0.998 LPG = 1 if Gas bottles as the main source of fuel 0 1 0.842 NG = 1 if Natural Gas as the main source of fuel 0 1 0.152 bricks = 1 if Bricks and cement as a wall material 0 1 0.905 kitchen = 1 if House has a private kitchen 0 1 0.872 bathroom = 1 if House has a private bathroom 0 1 0.043 BflushT = 1 if House has a private bathroom with flush toilet 0 1 0.63 BnoflusT = 1 if House has a private bathroom without flush toilet 0 1 0.247 Tflush = 1 if House has a private Toilet with flush 0 1 0.015 Tnoflush = 1 if House has a private Toilet without flush 0 1 0.052 GCollect = 1 if Garbage is disposed by means of a garbage collector 0 1 0.189 GColCo = 1 if Garbage is disposed by means of a garbage collecting 0 1 0.238 company GPubCon = 1 if Garbage is disposed by means of a public container 0 1 0.373 125 REFERENCES Ahmad, Junaid, Shantayanan Devarajan, Stuti Khemani and Shekkar Shah (2005). Decentralazation and Service Delivery. World Bank Policy Research Paper 3603. World Bank, Washington, DC. Antos, Joseph; Hafez, Nihal and Sameh El-Saharty (2004): « Egypt’s Health Sector Reform and Financing Review », World Bank Report Besley, Timaothy and Maitreesh Ghatak (2003). Incentives, Choice, and accountability in the provision of public services. Oxford Review of Economic Policy, 2003, Vol.19, No.2. London: Oxford university Press. Besley, Timothy and Robin Burgess (2001). Political agency, government responsiveness and the role of media. Typescript, the London School of Economics and Political Science (LSE). Chaudhury, Nazmul et al. (2006). “Missing in action: Teacher and health worker absenteeism in developing countries.� Journal of Economic Perspectives, Volume 20, Number 1, January 2006. Dehn, J., R. Reinikka and J. Svensson (2003). Survey Tools for Assessing Performance in Seervice Delivery. World Bank, Washington, DC. El Saharty, Sameh and Maeda, Akiko (2006): “Egypt Public Expenditure Review�, World Bank Policy Note, January 2006 El Zanaty Associates, Ann Way and Macro International Inc (2009): “Egypt: Demographic and Health Survey 2008�. Cairo: Ministry of Health, National Population Council. El Zanaty Associates and Ann Way (2006): “Egypt: Demographic and Health Survey 2005�. Cairo: Ministry of Health, national Population Council. El Zanaty Associates, USAID and MOH (2004): “Service Provision Assessment Survey 2004� (ESPA 2004) Filmer, Deon and Kinnon Scott (2008). “Assessing Asset Indices� World Bank Policy research paper, No. 4605. Grun, Rebekka, Ayala Javier et al. (2006): “Impact Evaluation of the Egyptian Health Sector Reform Project�, HNP Discussion Paper (in process), World Bank 2006 Reinikka, Ritva and Jakob Svensson (2003). Working for God? Evaluating service delivery of religious not-for-profit health care providers in Uganda. World Bank Policy research paper, No. 3058. World Bank (2008). Are you being Served? New Tools for Measuring Service Delivery. World Bank, Washington, DC. World Bank (2003). World Development Report 2004: Making Services Work for Poor People. Washington, DC: World Bank; New York: Oxford University Press.