54105 A Tale of Two Countries: Contracting for Health Services in Afghanistan and Congo (DRC) BY MICHAEL H. C. McDOWELL War-torn, devastated Afghanistan, ravaged by the Soviets, damaged by the Mujahadeen, divided by the Taliban, still riven by strife. And yet, a success in health delivery. The vast Democratic Republic of Congo (DRC), scourged by a cruel civil war, genocide, disease, but with a surviving health infrastructure and a surplus of health workers. And yet a much tougher challenge in health delivery, initially. Why did contracting delivery of health services to the non-profit sector work well in one country--but sluggishly in the other? Five health professionals give their opinions, with lessons for other states hit by conflict and poverty. Those lessons show that focusing above all on results, not process, is key to success. The World Bank made grants to Afghanistan and the DRC to support delivery of health services for the poorest, using local and international non-governmental organizations (NGOs). The World Bank expected major challenges in pre- dominantly illiterate Afghanistan especially. But, to many development experts' surprise, the Afghan experience exceeded many of the World Bank's highest expectations. In the DRC, where many World Bank officials expected this much more literate nation to comfortably implement improvements, this proved more difficult. French-trained Afghan health professional, Dr. Ghulam Sayed, has no doubt why health contracting-out worked in his country: "The actual implementation of health services was contracted out to well qualified, young, mostly local NGOs with some experience of competent project management. This was the decision of our Ministry of Health. The government did not have control over the whole country and vast areas, 80 percent, were untouched by health services. We had a huge shortage of health personnel, particularly women, and especially in remote areas, but the NGO contracting produced quick results. Photo © Homira Nassery/World Bank "Independent evaluation shows that quality of care improved 21 percentage points from 2004 to 2008 and the number of patients served has more than tripled. An independently conducted household survey in 2006 indicates that the * Michael H.C. McDowell, is an international public health communications consultant, former foreign affairs journalist, and think-tank senior fellow in Washington. 1 www.rbfhealth.org feature under-5 mortality rate in Afghanistan declined from an estimated 257 per Figure 1: Change in the utilization rate for curative care during the first year of the contracts (per 100 population per year) 1,000 live births in 2001 to about 191 per 1,000 in 2006. This means 70 that about 80,000 fewer children 60 are dying each year now, compared to during the Taliban rule." 50 40 Benjamin Loevinsohn, at the World Bank, is a Canadian physician who 30 worked on both of these programs. 20 He confirms Ghulam Sayed's view: "In Afghanistan, the Ministry 10 of Health focused on results, not 0 process. They said `We are not Congo (DRC) Afghanistan going to micromanage the delivery of health services by the NGOs. We are going to set priorities and quality standards but we are going to free the NGOs to figure out how best to do it. As long as they produce good results, we are happy; if results are not good, we will take action, including replacing the NGO. How the NGOs achieve results is their business.' The results were good. "The state focused more "In the DRC, government officials emphasized central control and focused on process issues on how things were done, including approving work plans and procurement plan. This tended to slow down delivery rather than what was done, of services and limited innovation. The state focused more on how things were done, rather prioritizing inputs than what was done, prioritizing inputs more than outputs. In the DRC, the system was also more risk-averse; there was more resistance to innovation, to trying something different." more than outputs." Patrick Mullen, who works on the DRC for the Bank, says things are better now. "The problem earlier was that the Ministry of Finance, which had responsibility for managing the funds, and the Ministry of Health (MoH), which was re- sponsible for overseeing the work, did not collaborate together effectively. Secondly, the MoH initially decided to take back control of procurement of drugs, equipment, and vehicles. Delays in this centralized procurement of key inputs paralyzed the NGOs for two years. Divided responsibility for various parts of the project led to a situation where no one actor could be held responsible for the results. "The government has learned from this and is currently revising the NGO contracts to provide them with more au- tonomy--and will be able to hold them accountable for results. The government had a previous good experience with contracting NGOs, but lessons from that were not effectively applied during the start-up of this current project. However, over the past year, the government's effort to improve the management of the NGO contracts has started to show some results. The government and NGOs are also working together to improve management of personnel, through the use of performance-based incentive payments to health facilities. Measuring the results of these efforts, with household and health facility surveys, during the first part of 2010, will be very important for the government to assess both its own performance and the performance of the NGOs." Professor Emile Okitolonda in the DRC, a World Bank staff member working on the project, agrees with Patrick Mul- len: "After the first two years, it was evident the procedural mechanisms in place were too cumbersome to be efficient but, given the fragility of governance within our post-conflict country, many checks and balances were necessary in the early stages. The new reforms are starting now to produce concrete movement at all levels, shortening monitoring 2 www.rbfhealth.org feature procedures, improving procurement, delivery, and quality. Now there is better participation of and understanding by the communities and this has led to better access to health services." How does the DRC Government itself feel about how mat- ters have worked over the past two years? Dr. Jacques Wangata is Health Coordinator for the project at the Ministry of Health and acknowledges that, looking back, split- ting responsibility between his Ministry and the Ministry of Finance was problematic. However, he says NGOs also had some responsibility for delays and mistakes: "But after two years we have learned to give a greater margin of freedom to the NGOs, insisting on them being responsible, carrying out their specific tasks and adhering to a clear timetable, and evaluating the quality of the deliverables. Now there are Photo © Caryn Bredenkamp/World Bank fewer problems with cash flow and availability of medica- tions, and there are better results overall." Results-Based Financing Performance-based contracting is a prime example of what is called Results-Based Financing (RBF), i.e. programs which transfer money or goods to either patients (when they take health-related actions, such as having their children immunized), or transfers of funds to health care providers (when they achieve performance targets, such as immuniz- ing a certain percentage of children in a given area). The Bank managers, when the projects began, were concerned that Afghanistan had been so dam- Bank managers were aged that it would be an uphill battle to improve health delivery substantially. In the DRC, the conflict concerned that it would had cooled, and a roughly functioning health delivery system was in place but very weak. be an uphill battle to improve health NGOs had done important work in both Afghanistan and DRC, says Dr. Loevinsohn, but Afghan delivery substantially officials were highly focused on results and open to risk; the DRC officials, in contrast, were more in Afghanistan. process-driven and more risk-averse. Quality of care in Afghanistan went up every year. Knowledge of health workers increased, counseling improved, interaction between patient and health provider got better. In the DRC it was not clear at the beginning there had been much improvement, with the important excep- tions of malaria control and immunization services. There was greater success in DRC with contracting as time went on, though. In Afghanistan, where larger health centers were simply too spread out and inaccessible to a large fraction of the people because of the difficult terrain and long distances, NGOs set up small sub-centers covering 3,000­7,000 people: "The NGOs staffed these with a male and a female health worker and it worked really well. They just did it. They didn't ask permission of Kabul and now it is government policy. To have attempted something similar in the DRC would have required more approvals." In Afghanistan, enterprising NGOs were sufficiently autonomous to hire and fire staff. So where there were staff short- ages within the country, the NGO managers were able to recruit well qualified female physicians and midwives outside the country, for example, in neighboring Tajikistan, which has a common language, Dari. Says Ghulam Sayed: "Huge rural areas were just not reached by government health services which had been dev- astated by both the Mujahadeen and the Taliban. So each NGO we partnered with took a slightly different approach, 3 www.rbfhealth.org feature depending on what worked or what did not. And the Ministry of Health of course monitored things but it did not micromanage our NGO partners. For example, in troubled Helmand province, one NGO expanded health services by opening and closing health facilities depending on the level of security or danger. They were nimble. The government could never have done this because the Taliban would have stopped them. But the local NGOs were flexible and, criti- cally, worked with local leaders in small villages and rural districts. In some cases they were able to negotiate with the local Taliban." Dr. Loevinsohn enthusiastically agrees: "The Afghan NGOs had to deal with a real shortage of not just trained health personnel but women especially. And obviously because of the culture, men could not be alone with women patients, so they arranged to hire the husbands and brothers of women health workers, trained the men in basic health delivery methods, and solved the problem." Today, the picture in the DRC is far more positive, says Patrick Mullen: "The government has learned and has removed the bottlenecks to centralized procurement, at the same time as reforming management of the NGO contracts. For ex- ample, the project has now distributed millions of bed nets for control of malaria, the number one cause of illness and death in this country, and the NGOs are now getting the drugs and equipment that they need to do their job effectively. "There are reasons for why the project evolved as it did. Unlike in Afghanistan, there is an established health system in the DRC--NGOs were not contracted to provide services but to channel support to the existing health system. For example, NGOs never had direct control over health personnel in the DRC. What the project is now moving towards is a compromise, where the NGOs need to work closely with the existing health system but also have more autonomy and accountability for results." Lessons Learned from a Tale of Two Countries In summary, says Dr. Loevinsohn, "Lack of trust is corrosive. By all means trust but verify, however, if you stymie people they won't do a good job. Don't over manage. We will not serve poor people well if we have no stomach for risk. Good results can still be obtained even in the most difficult places like Haiti, Bangladesh, Afghanistan, or the DRC. "The way in which people deal with high-risk situations is all too often to focus on process and `getting it right.' But what they really should be doing is focusing like a laser beam on getting results. We have proven that a major suc- cess for Results Based Financing is contracting health services to NGOs. Focus on performance, focus on results, not process. That's key." 4 www.rbfhealth.org feature