EVIDENCE November, 2010 from to POLICY a note series on learning what works, from the Human Development Network 59367 The Challenge of Ensuring Adequate Stocks of Essential Drugs in Rural Health Clinics Health experts and policymakers want people to have in Zambia, the World Bank supported a project exploring access to affordable and high-quality medical care. But how to guarantee the availability of essential medicines in some developing countries, making quality health- in often-remote health facilities. The 12-month study, care available may first necessitate ensuring that essen- which covered almost 22 percent of Zambia's rural popu- tial medicines are available, such as anti-malaria pills lation, found that streamlining the delivery of medicines and antibiotics. The challenge to guaranteeing a steady directly to health centers and introducing a dedicated supply is not only related to the financial side of paying staff member to help facilitate and track orders cut down for medicines. Poor roads, limited communications and on the rate at which clinics ran out of basic medicines. storage problems can make it difficult to keep medical The focus on just one aspect of good healthcare--making facilities stocked with what they need to provide children certain necessary supplies are in stock in medical clin- and adults with regular and lifesaving care. ics--does not answer all the questions that experts face in The World Bank is working to help countries provide building or supporting functioning health systems. But it quality medical care, a key part of many of the United may help them as they work towards creating the quality Nations Millennium Development eight goals. Recently, healthcare that all people deserve. Case Study Zambia Zambian officials were frustrated by shortages of essential plies from district stores to local health facilities. Dis- medicines in public health facilities in rural and semi- trict Health Management Teams, which reported to the rural parts of the country. Clinics frequently reported Ministry of Health, were responsible for sending supplies that they had run out of basic lifesaving drugs, such as from the district stores to some 1,500 health facilities. antibiotics and anti-malarial drugs, leaving children and Reaching the facilities could be challenging. The facilities adults at greater risk of illness or death. The problem was were spread out and some roads required special, off-road not the availability of the necessary drugs in the country vehicles. Communications between the facilities and the overall. It was the distribution system. district stores were intermittent--usually via two-way ra- While the government had a working system for mov- dio. Local facilities were running out of medicines at a ing medical supplies from the capital Lusaka to district rate that was double or more that reported by district stores and hospitals, it was less efficient at getting sup- stores, a clear sign that drugs were not getting to where they were needed. Did You Know... A pilot program, consisting of two different models for distribution, was initiated to test the best way to over- Life expectancy in Zambia is 45 years. come the bottleneck at the district level. In Model A, a And the under-5 mortality rate is 141 per 1,000... commodity planner was put in place at the district level. Compared with an average of 129 for sub-Saharan Africa. The person was responsible for making certain that health clinics submitted accurate monthly orders to the district worked with health facilities to make sure they got their store, for collating the clinic orders, and then placing the orders in properly to the central stores and also delivered bulk order with the central medical stores with the national the packages as they arrived. stock of drugs, called Medical Stores Limited. The goal was The pilot began in April 2009 and ran for 12 months. to ensure that the district stockroom had a ready supply The two models were each implemented in 8 rural and of items for the clinics. The planner would put together semi-rural districts, randomly selected out of a total of packages for the clinics and arrange transport. 50 possible districts. An additional eight districts were se- In Model B, health facilities submitted orders di- lected for observation as controls. A baseline survey was rectly to the central Medical Stores Limited. The district conducted prior to the start of the pilot, and then again at store, instead of stocking drugs, acted as a point of tran- the end. Information was collected on the inventory and sit, receiving and forwarding-on already packaged orders stock-out rates of 15 drugs, in addition to interviews with with the help of a new commodity planner. The planner the community planners. The Findings Hiring a special facilitator to work directly drugs. At the end of the 12-month pilot, malaria treatments for with health facilities, while minimizing the adults were out-of-stock in 6 percent of facilities, while pediatric role of district stores for storing and delivering treatment was out of stock in 12 percent. This compared with a medicines, helped clinics stay better stocked. previous rate of 48 percent for adult treatments and 43 percent for pediatric treatment. In facilities that relied on the regular In Model A, which introduced a commodity planner to co- system, but now had a commodity planner to help coordinate ordinate orders, the stock-out rates for needed drugs ranged orders, the stock-out rate for adult malaria treatment was 22 from 17 percent to 46 percent after one year, depending on percent (compared with 43 percent before the study) and for the particular drug, compared with the baseline rate ranging pediatric treatments, 30 percent, down from 34 percent. from 34 percent to 74 percent. In Model B, in which the role of the district office was further minimized, the out-of- When looking at the number of days that stock rate for drugs ranged from 1 percent to 33 percent, essential drugs were unavailable, health compared with a baseline rate of 40 percent to 72 percent. facilities in Model A did only marginally better than the control group. Life-saving malarial drugs were among those more likely to be in stock, especially in health Pediatric malaria drugs, for example, were out of stock an facililties that directly ordered their own average of 29 days, out of a maximum of 92 days, in the supplies from the central stores. control facilities. For facilities in Model A, the stock-out du- ration averaged 18 days; in districts in Model B, the average Health facilities that had more control over ordered medical number of days this essential medication was out of stock supplies were better able to maintain needed supplies of essential dropped to five days. This bulletin summarizes the results of the research paper "Enhancing Public Supply Chain Management in Zambia" by Monique Vledder, Jed Friedman, Mirja Sjoblom and Prashant Yadav. The paper is based on the results of the World Bank-supported Essential Drug Public Pilot Program in Zambia, which was funded by the World Bank, U.S. Government and DFiD. Researchers estimated that if Model B were im- plemented nationwide, malaria-related deaths could drop by more than 20 percent. In children under the age of five, the number of deaths annually from malaria may decline to an estimated 12,218, from an estimated 15,538 currently. In the over-five age group, deaths from malaria would drop to 1,318 from 1,766 annually. This is a result solely due to the increased availability of life-saving drugs at the clinic. And when essential drugs are available, house- holds are less likely to lose income because a working adult is ill or has to stop working to care for an ill child. Introducing a commodity planner into the system helped facilities better coordinate their Researchers estimated that if Model B were applied orders with the central agency. throughout Zambia, more than $1.6 million annually in expected household income loss would be avoided. In The rates at which health facilities placed regular orders turn, this means that families would be less likely to have with the central agency, a key part of ensuring the avail- to sell assets, pull children out of school or lose crops be- ability of drugs at the district level, rose to 95 percent or cause of the often-devastating economic-related effects of better. This compared with a pre-pilot rate of 72 percent to severe illness. 79 percent, depending on the district. The cost of taking steps to boost the availability "The program has worked very well. Now people have access of essential drugs in rural clinics is not cheap. to medicines. Despite the hardships, I would consider continu- But it is still worthwhile. ing as a commodity planner," another planner concluded. Model B cost an additional $3,971 per district per month, while Model A, which was not as effective, cost $3,479. Nonetheless, commodity planners still faced More than half went for the commodity planner's salary, some of the same transportation, storage and with the remainder covering expenses associated with staff- communications problems that district stores related expenses, office supplies and training. When weighed faced in getting medicines to remote clinics. against stock out rates, Model B is almost four times more cost-effective than Model A. Commodity planners had to rely on vehicles and fuel pro- vided by the district health office to get the supplies to the clinics. This meant they were "competing" with other programs for use of vehicles. They also did not always have sufficient storage space for medicines shipped in from the The Zambia National Malaria Indicator Survey (2008) reported central agency. Communicating with the facilities was an- that just seven percent of children in rural areas under the age other problem. Most health facilities have to rely on two- of five received pediatric ACT (Artemisinin-Based Combination Therapy)--the most effective first-line medicine for malaria-- way radio, which made it difficult for community planners within 24 hours of starting a fever. to be in contact on a regular basis. Conclusion Making policy from evidence Model B, in which district stores were reduced to transit right amount of stocks in district or national centers. At points for shipments to health facilities, was so successful the same time, the study does not consider what happens that it has now been extended to all districts that were in after essential medicines get to health facilities. Are peo- the pilot study. Further expansion throughout the coun- ple able to access what they need? Is the quality of care, in try will require additional funding, which donors and other words, on par with the new-level of available drugs? Zambian officials are now considering. These are questions that policymakers and health experts But as the pilot underscores, successful distribution need to consider next. of drugs is about more than just money--or having the The Human Development Network, part of the World Bank Group, supports and disseminates research evaluating the impact of development projects to help alleviate poverty. The goal is to collect and build empirical evidence that can help govern- ments and development organizations design and implement the most appropriate and effective policies for better educational, health and job opportunities for people in developing countries. For more information about who we are and what we do, go to: http://www.worldbank.org/hdchiefeconomist THE WORLD BANK, HUMAN DEVELOPMENT NETWORK 1818 H STREET, NW WASHINGTON, DC 20433 Produced by Office of the Chief Economist, Human Development Network, Communications/Aliza Marcus