2
An Overview of Quality of Care in Low- and Middle-Income Countries
Key Points Made by Individual Speakers
- Unsafe medical care is a leading cause of death and disability around the world. It will not be possible to improve health in low- and middle-income countries without improving quality, but the topic gets considerably less attention than improving access. (Jha)
- The six strategies under consideration at the workshop are more similar than different. All can work in some contexts, and understanding the contextual factors that favor one over another is part of the challenge of implementation. (Jha)
- Like most other quality improvement methods, the six strategies put an emphasis on changing provider behavior, which is still several steps removed from changing patient outcomes. (Jha)
- USAID is a major supporter of quality improvement work. To do its work well, the agency needs to understand where different strategies are most suitable, how they work, and what the main gaps in the evidence are. (Heiby)
- It is difficult to glean impartial evidence about the different methods as long as the people closest to the work are asked to evaluate its effectiveness and perceive that they must minimize their failures and promote their successes to the wider audience. (Rowe)
After brief welcoming remarks from Victor Dzau, the IOM president, and Sheila Leatherman of the University of North Carolina at Chapel Hill, Ashish Jha of Harvard University gave a keynote address describ-
ing the current research on quality of care in low- and middle-income countries.
KEYNOTE REMARKS
Quality is a topic that has been neglected in global health, but that is changing. As countries grow wealthier, universal health coverage has become a common policy goal. The combination of growing wealth and increased access to services translates into more demand for health care. The number of doctor’s visits, nurse’s appointments, and hospital stays is going to grow rapidly in the next decade, and most of this growth will come from in low- and middle-income countries. At the same time, Jha pointed out, simply increasing use of services, or even demand for them, is not the goal of expanding coverage. The goal is to improve health, and that will not happen without changes to the quality of the services offered.
Jha then summarized his 2013 study that found the vast majority, about two-thirds, of the world’s 43 million adverse events1 occur in low- and middle-income countries (Jha et al., 2013). He qualified these numbers further, explaining that the calculations were based on a small group of in-hospital adverse events. Given what a small piece of health care hospitalization accounts for, Jha reckoned that unsafe care is probably 1 of the top 10 global causes of death and disability.
Further understanding of the problem comes from reviewing the work of Jishnu Das and similar researchers. Their work suggests that in low- and middle-income countries the probability of a patient receiving the correct diagnosis is, depending on other factors, in the range of 30 to 50 percent. Similar studies that have attempted to estimate the probability of a patient receiving non-harmful treatment found a likelihood of about 45 percent. While the probabilities vary widely depending on country and setting, Jha concluded that quality of care is a serious obstacle between expanded access and improved health.
Another important barrier to good quality care is lack of trust. The recent Ebola outbreak in West Africa has highlighted problems with public trust in the health system. Jha recounted a December meeting he attended on Ebola response. The participants included the health ministers of Guinea, Liberia, and Sierra Leone and many other global health leaders. One major theme from their discussion was that a lack of trust in the health system prevents people from using services. As long as people feel that they cannot trust their health system or that they are not treated respectfully,
_________________
1 An adverse event refers to injury or harm to a patient as a result of medical care, rather than the underlying disease or condition.
the best quality services will still be useless, and the population’s health will not improve.
Jha then briefly introduced the six strategies being discussed at the workshop (see Box 1-1). He explained that the literature suggests that each of these strategies works sometimes, in some contexts, but no strategy seems to work consistently in all contexts. He also observed that the six strategies have far more in common than they have differences. One example is a shared emphasis on measurement and iterative feedback.
All of the strategies rely on buy-in from leaders and adapting services to the local context. Quality improvement is generally a complex social intervention, and even seemingly universal tools like surgical checklists need to be tailored to the local environment. Adapting a tool requires consideration of the social relationships in the clinic. For example, if nurses are not empowered to correct doctors when they breach hygienic protocol, the organization will not be able to improve.
Another feature common to all six methods is a reliance on changing provider behavior, which is still several steps removed from the final goal of changing outcomes for patients. Measuring changes in the way care affects patients is complicated. Program evaluators usually have to be content with surrogate measures, such as changes in provider behavior. Often the success or failure of the program hinges on the choice of these surrogates. Successful programs rely on meaningful targets and set compelling goals. Jha felt that a goal such as eliminating hospital-based infection would motivate people more than a target of reducing hospital-based infections by 10 percentage points. Pay for performance programs may be particularly vulnerable to problems in this area. He explained that these programs, which have the power to improve efficiency, often choose targets that are not meaningful to clinicians or patients.
Quality programs that focus on provider behavior risk losing sight of the larger systemic obstacles that prevent change from taking root. Jha emphasized the need for commitment to change from the top levels of an organization. For example, a common goal of quality improvement programs in the United States is to reduce waste. To this end, doctors are encouraged to order fewer tests. But the organization’s payments depend on the number of tests they order. This kind of obstacle makes it difficult to sustain any reduction in the number of tests ordered.
Jha concluded that improving quality is as important to global health as increasing coverage, although it gets much less attention. He suggested that quality improvement works best when the program prescribes a clear, meaningful goal, but not how to get there, which will vary in different settings. At the same time, the methods for quality improvement do matter. Accreditation, for example, may work better in places where patients use insurance or prepayment plans; in places where the workforce capability is
low, strategies such as supervision or collaboratives might not work well. Unfortunately, the literature is not clear on what the facilitating factors are for different strategies.
USAID’S WORK IN QUALITY OF CARE
James Heiby of USAID built on this point in his presentation about the agency’s involvement in quality improvement. USAID has been active in quality improvement since around 1990, but initially most of that activity centered around the agency’s headquarters. Over the past 10 years, the missions have expanded their efforts in this area, and the agency is now a major supporter of quality improvement.
Heiby framed his comments on quality in relation to the Donabedian model of a health system with inputs, processes, and outcomes. Quality improvement, especially at USAID, has attempted to change health care processes, the neglected piece of the triad. Quality improvement is essentially process improvement, and the agency has a stake in identifying the best strategies for process improvement. There are a range of strategies in use, exemplified by the six methods that dominate USAID’s quality investments (shown in Box 1-1). Some strategies, like training and supervision, are so widely used that spending on them dwarfs any other global investment in quality, though many training and supervision programs are not primarily intended to improve quality. Other strategies, accreditation for example, are discrete interventions with the aim of changing the quality of services. These are different still from a set of discrete interventions that use management and incentive techniques to change provider behavior.
For USAID, it is important to understand how and where these different strategies are most suitable and what the main gaps in the evidence are. There are many possible strategies beyond the six discussed at the workshop, and Heiby encouraged the participants to think about the larger context of quality improvement. The first contextual question is whether or not everyone in the field understands quality to mean the same thing. Quality improvement is usually a complicated social change. It is hard to articulate the piece of the program that is the quality improvement, so people are more comfortable talking about measurable indicators. Researching and measuring quality tends to accompany a whole range of research projects—projects that produce a great deal of data and knowledge. Managing and synthesizing this information is a daunting responsibility. To complicate the matter, most data come from self-report and are difficult to validate. In discussing his concerns about the evidence base, Heiby reminded the audience that finding weak evidence for any or all of the target methods should not be seen as a failure in the discussion: understanding where the data are weak helps scientists form a clearer research agenda.
As an added benefit, increasing attention to quality will likely drive improvements in health information technology, an area of health systems strengthening that is often neglected. In a larger sense, all quality improvement can effect meaningful changes in low- and middle-income country health systems. Progress depends, however, on transferring the skills to the regular employees already working in country. When foreign consultants run these programs, the likelihood of sustaining changes becomes very low. Therefore, Heiby cautioned against blindly supporting quality improvement as a movement, saying that the programs should be held accountable for how they spend and how they influence their partner countries. The best way to do that may be by demanding rigorous evaluations of quality improvement programs.
The subsequent discussion illuminated some of the barriers to open sharing of program data and impact evaluations. Alexander Rowe of the Centers for Disease Control and Prevention (CDC) pointed out that the organizations charged with doing projects generally feel that they will be penalized if they report anything less than success. It will be difficult to glean impartial evidence about the pros and cons of different methods as long as the people closest to the work have the perception that they should be promoting their successes and downplaying their failures in the wider field.
This page intentionally left blank.
