Embargoed Until:
March 27, 2007
12:01 a.m. Eastern Time


Christopher Fleming

Going Beyond Health Care Availability In Lower-Income Countries: Researchers Focus On Measuring And Improving Quality Of Care In The Developing World

Health Affairs Package Examines Quality Of Care
In India, Indonesia, Mexico, Paraguay, And Tanzania

Bethesda, MD -- For a quarter-century, development experts have concentrated primarily on increasing the availability of health care in poorer countries. Now, in light of evidence that the “use of health care is high, even in countries with low per capita incomes and even among the poor,” it is time to focus more on measuring and improving the low quality of the care that many patients in the developing world receive, Jishnu Das and Paul Gertler say in a paper published today on the Health Affairs Web site.

Das and Gertler provide an overview of six studies written by experts from the World Bank and elsewhere and published in Health Affairs that look at the quality of health care in five different countries: India, Indonesia, Mexico, Paraguay, and Tanzania. These countries stretch across three continents and span a wide range of developmental stages: Per capita gross national income in the poorest nation, Tanzania, is only 7 percent of that of the richest nation, Mexico. Health care delivery systems range from widespread reliance on the private sector in India, mixed systems including semiprivate facilities operated through social security contributions in Paraguay and Mexico, and the use of the nongovernmental organization (NGO) sector in Tanzania. You can find Das and Gertler’s overview and the six individual-country studies at http://content.healthaffairs.org/cgi/content/full/hlthaff.26.3.w296/DC2

“The overall quality of care documented in these studies is low, although there is considerable variation across countries and even within countries over time,” say Das, an economist with the Human Development and Public Services Group in the Development Research Group of the World Bank, and Gertler, a University of California, Berkeley, professor and former chief economist of the World Bank’s Human Development Network. In India, for example, doctors identified only 26 percent of the tasks that were medically required for a patient presenting with tuberculosis, an affliction that kills more Indians each year than every other infectious disease combined. In Tanzania, doctors hit less than a quarter of the essential checklist for patients presenting with malaria, which kills 63,000-96,000 Tanzanians each year.

“Not surprisingly, the urgency with which countries increased the availability of health resources led to a proliferation of medical care providers with very different levels of expertise. This implies that the quality, rather than the quantity, of medical care should now form an important focus of health policy in low-income countries,” Das and Gertler say. “Given the observed poor performance of medical care providers and the large disparities within countries, the gains from increasing performance in these countries are likely higher than in high-income settings.”

Studies In Health Affairs Address Gap In Information
On Practice Quality In Lower-Income Countries

Unfortunately, “questions relating to practice quality remain unanswered in the literature, because the quality of care in low-income countries is difficult to measure. Traditional measures of quality have focused on ‘structural’ quality, such as the availability of electricity, the physical condition of the clinic, or the stock of medicines.” The problem is that “these measures tell us little about the actual quality of medical advice patients receive when they visit a doctor,” say Das and Gertler. “Variations in practice quality rather than variations in availability and structural quality are more likely to explain a large fraction of variance in health outcomes, even in low-income settings.”

The seven papers address how the gap in information affects practice quality through several different kinds of quality measures. In two countries -- India and Tanzania -- for instance, medical competence and knowledge were measured by using “vignettes.” Researchers visited practitioners (who knew they were researchers) and imitated patients with specified symptoms. Practitioners’ actions were then measured against a set of essential questions and exams.

Medical competence, as determined through these vignettes, “measures the quality of medical advice patients would receive if health care providers completed all tasks commensurate with their knowledge,” Das and Gertler observe. To measure the quality of care patients actually received, researchers in Mexico, Tanzania, and Indonesia used household surveys or direct clinical observation. Finally, researchers in Paraguay used an index of physician effort that measured time spent with patients, questions asked, and examinations performed.

The differences between these measures are illustrated by the case of Tanzania. In that country, “of the checklist items that doctors knew to do (from the vignettes), only 53 percent were actually completed when these doctors faced similar patients in practice.” The drop-off in quality from vignette to clinical practice was larger in the public sector than it was in the private/NGO sector. Similarly, in India, public-sector doctors treating a child with diarrhea scored 47 percent in vignettes versus only 9 percent in clinical practice, while private fee-for-service doctors scored a relatively modest 29 percent in vignettes but fell only four percentage points, to 25 percent, when observed in clinical practice.

Improving Practice Quality: Using Pay-For-Performance,
Beefing Up The Public Sector In Poor Areas, And Licensing Providers
What steps can be taken to improve health care quality in lower-income countries? “Additional training, in general, seems to be an expensive proposition,” Das and Gertler say. As revealed in the Tanzania and India examples, the problem is often not that doctors “do not know what to do; it’s that they don’t do it.” It would be better, the authors argue, to offer targeted training for specific illnesses or diseases, or to attempt to get public-sector doctors to exert more effort, possibly through financial rewards for good performance. “Although such pay-for-performance schemes are still new in the United States, they might be more feasible in low-income settings where there are fewer players and organized vested interests in the provision of health care.”

Another quality improvement strategy put forth by Das and Gertler is using the public sector to balance private-sector providers’ market-driven decisions on where to locate. In all the countries studied, “households in poor areas have access to, and receive, low-quality care from the private sector,” the authors observe. However, “the evidence from Indonesia and Paraguay suggests that policies can be designed that equalize the competence of public-sector doctors in poor and rich areas. In countries such as India or Tanzania, an urgent imperative is to correct the imbalances inherent in the private sector by reexamining the systems of transfers and postings of public-sector doctors.”

Das and Gertler also note that one “striking” result of their analysis was “the relatively poor performance . . . of private providers without an MD or equivalent qualification, especially in poor areas.” They suggest that governments “might want to impose licensing of providers to guarantee a minimum quality standard” or to give patients better information about the quality of medical care providers.

Individual-Country Studies

Variations In Prenatal Care Quality For The Rural Poor In Mexico. Sarah Barber, Gertler, and Stefano Bertozzi found “significant variation” in prenatal care quality by provider qualifications, income, indigenous status, and setting of care. Women who sought care from providers with a medical degree generally received better care than women who went to nonphysicians such as trained nurses, midwives, or nonallopathic practitioners, although the authors warn that this could reflect the poor professional training that non-MDs receive in Mexico, rather than the importance of a medical degree. On average, the poorest quartile of women in the sample received worse care than the least poor quartile, and indigenous women received worse care than their nonindigenous counterparts. Despite the widespread belief in Mexico that private providers outperform public providers, the private sector actually achieved the lowest quality scores, leading Barber and her coauthors to recommend efforts to better inform consumers through public quality reporting. (Barber: University of California, Berkeley; Gertler: Cal Berkeley; Bertozzi: National Institute of Public Health, Cuernavaca, Mexico)

Variations In Doctor Effort: Evidence From Paraguay. “The key finding from our study is that doctors treat patients from different backgrounds similarly and that large differences in doctor effort arise from differences in physician and institutional characteristics,” say Das and Thomas Pave Sohnesen. “That is, doctors develop a certain ‘style’ of treatment, which does not vary depending on the specific patient treated. This suggests that to get better-quality care to the poor (at least in the public sector), policymakers need to ensure that the poor have access to, and use, high quality doctors, but they need not worry that these doctors would treat their rich and poor patients very differently.” (Das and Sohnesen: World Bank, Washington, D.C.)

Location, Location, Location: Residence, Wealth, And The Quality Of Medical Care In Delhi, India. “The poor clearly had access to worse providers than the rich,” say Das and Jeffrey Hammer. Health care providers in poorer neighborhoods were less well educated than those in richer areas, and within every educational category, providers in poorer areas were less competent. The public health care sector “should balance out inequalities arising from the location choices of the private sector,” say Das and Hammer. However, this did not occur. Within the same neighborhood, rich and poor households visited private-sector providers with similar levels of competence, but rich households visited public providers who were more competent than those visited by their poorer counterparts. (Das and Hammer: World Bank, Washington and South Asia)

Differences In Access To High-Quality Care Outpatient Care In Indonesia. “It is remarkable that we found no wealth differences in access to available quality,” say Barber, Gertler, and Pandu Harimurti. “However, in comparing the prenatal care available with what was actually received, the poorest households received significantly lower levels of quality in both” densely populated Java-Bali and the more rural Outer Java-Bali. The authors say this suggests “that facility-level discrimination against the poor might exist,” which could stem from reasons such as “formal or informal user fees” or “social and educational differences between providers and poor patients.”

Barber and her coauthors also found that public facilities offered above-average care in Java-Bali but below average care in Outer Java-Bali, suggesting that “quality differences lie within the health resource allocation systems.”  (Barber and Gertler: Cal Berkeley; Harimurti: World Bank, Jakarta)

The Contribution Of Human Resources For Health To The Quality Of Care In Indonesia. “Deployment of physicians to [public-sector regional] health centers has remained the prevalent health policy focus” in staffing Indonesian public health facilities, say Barber, Gertler, and Harimurti. And indeed, the authors found that, “in particular, moving from no physicians to one appeared to provide … significantly increased quality, separate from the presence and number of other staff.” However, given Indonesia’s inability to attract adequate numbers of physicians to rural areas, “expanding the managerial roles and responsibilities of nurses, midwives, and other professionals in the primary-level public system” is a promising long-term strategy for increasing access to high-quality providers in remote regions.

The authors break down the effects on quality of increased staffing by provider type and patient condition, finding, for example, that increasing the number of nurses resulted in larger quality gains for curative care than increasing the number of physicians or midwives. “By expanding on these types of analyses,” optimum staffing mixes could be identified for each region’s particular health problems and priorities, they say.  (Barber and Gertler: Cal Berkeley; Harimurti: World Bank, Jakarta)

Variations In The Quality Of Care Accessible To Rural Communities In Tanzania. Patients at government health facilities in rural Tanzania receive worse care than their urban counterparts, but this gap can not be explained solely by physical and fiscal limitations, say Kenneth Leonard Melkiory Masatu. Leonard and Masatu found that “although doctors in the rural areas knew less than other doctors, they were less likely to do what they knew they were supposed to do,” a quality gap that existed only in government facilities. Facilities run by NGOs in rural areas “were staffed by doctors with lesser qualifications and competence, but these doctors managed to deliver reasonable care despite these limitations,” largely due to the NGOs’ superior management structure. The authors suggest having NGOs manage government facilities, and they encourage government facilities to adopt protocols such as the Integrated Management of Childhood Illnesses protocol, which appears to improve adherence to known best practices. (Leonard: University of Maryland; Masatu: Centre for Educational Health in Development, Arusha, Tanzania)


Health Affairs, published by Project HOPE, is the leading journal of health policy. The peer-reviewed journal appears bimonthly in print with additional online-only papers published weekly as Health Affairs Web Exclusives at www.healthaffairs.org.


©2007 Project HOPE–The People-to-People Health Foundation, Inc.