The Long Journey to Health Equity

A passionate public debate regarding the best way to conduct clinical trials ethically in low-income countries developed after studies to prevent mother-to-child transmission of human immunodeficiency virus (HIV) were initiated in Africa and Asia. Some of these studies had placebo-only treatment groups and generated concern after publication of the first US trial showing an effective way to prevent mother-to-child transmission.¹ Even though this debate focused on research ethics, the underlying problem was and remains differences in available levels of resources.² The debate over medical research ethics in the context of vastly unequal health care resources served in important ways to focus public attention on research that uses placebos or unproven treatments when effective interventions exist.³ However, a focus on the best way to conduct research in the midst of gross disparities in health care misses the more fundamental ethical imperative: how to achieve a more equitable distribution of health care and ensure that effective health care is brought to resource-poor countries as rapidly as possible.

In this issue of JAMA, articles by Kent and colleagues⁴ and Bulterys and colleagues⁵ include findings that are important to any effort to increase the speed and extent of implementation of effective life-saving and health-improving programs in low-income countries. Kent et al reviewed clinical trials of HIV treatment, tuberculosis (TB) treatment, and malaria prevention conducted in sub-Saharan Africa and published between 1998 and 2003 to see if the trials met "best current" standards of care. In particular, they examined whether the studies met recommendations supported by the research literature, global standards when guidelines had been promulgated by the World Health Organization (WHO), and national practice in high-income countries such as the United States. All of the TB treatment trials reviewed met best current standards; however, only a small percentage of HIV trials provided optimal care even in the intervention groups, and 90% of the malaria prevention trials failed to offer acceptable prevention strategies in the control groups.

More disturbing than the failure to meet the best global standard is the failure of many studies to provide interventions that should be readily obtainable locally. A striking example is that the malaria studies Kent et al reviewed did not include clearly feasible prevention strategies such as bednets. Though some of the reviewed malaria prevention trials were conducted before 1998, even by 1995 substantial research supported the use of insecticide-treated bednets for malaria prevention.⁶ With average costs as low as $3.50 for a bednet lasting 5 years and $0.30 for insecticide treatment per year,⁷ it is hard to argue that providing bednets was not affordable. Implementation is clearly feasible; in fact, insecticide-treated bednets have been successfully introduced into remote, low-resource areas such as the Kilombero and Ulanga districts in southern Tanzania using subsidies and social marketing techniques.⁸ Why has this evidence still not been translated into widespread programs to reduce malaria morbidity and mortality, particularly in sub-Saharan Africa?

Hypothetical concerns about diminished partial immunity to malaria may have contributed to the slow uptake of insecticide-treated bednets, even though the benefits convincingly outweigh the risks.⁹ No such concerns exist for numerous other proven interventions for the prevention and treatment of infectious diseases, yet millions continue to develop illness and die for lack of implementation. Worldwide, approximately 600 000 individuals each year contract hepatitis B virus (HBV) infection, making it one of the most important causes of morbidity and mortality from an infectious agent worldwide.¹⁰ Unlike HIV infection, TB, and malaria, a highly effective vaccine for HBV exists.¹¹ In 1992, the WHO recommended that HBV vaccine be integrated into universal childhood vaccination programs by 1997 with the goal of vaccinating 90% of eligible children by 2010.¹⁰ Despite having among the highest prevalence rates of HBV worldwide, only about 25% of the estimated 26.7 million children born in the African region in 2003 received HBV vaccine.¹²

Similarly, the cost-effectiveness of TB treatment in even the world's poorest countries has been known for almost 15 years.¹³ Despite this knowledge, approximately 8.3 million new TB cases and 1.8 million deaths from TB occurred in 2000.¹⁴ In 2001, only about 26% of new TB cases were successfully treated in recommended treatment programs.¹⁵ Studies conclusively demonstrate the effectiveness of isoniazid to reduce morbidity and mortality in HIV-infected, tuberculin-positive individuals.^{16 - 17} Yet more than 600 000 new TB cases and 246 000 deaths occurred in persons coinfected with HIV in 2000.¹⁴

New advances in clinical research may lead to either a decrease or an increase in health inequities between rich and poor countries, depending on where and how the results are translated into practice. The Mother-Infant Rapid Intervention At Delivery (MIRIAD) study⁵ in this issue of JAMA is an example of one such advance. This study, and an earlier report,¹⁸ demonstrate the feasibility of rapidly screening women of undocumented HIV status in labor for infection at the point of care. In the current study, the median time from blood collection to patient notification of her HIV status decreased from 28 hours to 66 minutes. Knowing a mother's HIV status is essential for providing antiretroviral therapy to prevent transmission of HIV and to treat the mother.^{19 - 20}

In the MIRIAD study, women in 8% of all visits recorded were eligible for rapid testing. Of the 4849 women in the MIRIAD study who consented to participate, 34 were found to be HIV-infected.⁵ These results have important implications for individual US patients, but their real potential for a substantial public health impact on mother-to-child HIV transmission depends on being able to translate these results into effective point-of-care screening programs for women in sub-Saharan Africa. Approximately 1 in 5 pregnant women in southern Africa is HIV-infected.²¹ Among pregnant women between the ages of 15 and 24 years, HIV prevalence rates range from approximately 13% in Mozambique to 39% in Swaziland.²² About 29% of pregnant women in sub-Saharan Africa have no antenatal visits; unlike many other world regions, most African women who present for antenatal care do so in the second trimester or later.²³ These factors contributed to, but are not solely responsible for, the estimated 700 000 HIV infections in infants in 2003, the vast majority of which were due to mother-to-child transmission in Africa.^{19 ,24} The ability to screen women rapidly for HIV infection and offer antiretroviral therapy has the potential to prevent HIV transmission to hundreds of thousands of infants that otherwise might occur.²⁵

The ability to effectively prevent, diagnose, and treat major global public health threats continues to improve rapidly, as highlighted in the study by Bulterys et al.⁵ Yet, much of the world is being left further and further behind. There needs to be increased attention, both publicly and in the medical and research communities, to the extent of health care disparities that exist and the availability of proven prevention and treatment programs. Public awareness surrounding the Global Fund²⁶ and, more recently, the "3 x 5" initiative²⁷ are examples of how public consciousness may be raised and resources mobilized. Nongovernmental organizations and funders also play a crucial role. The Global Alliance for Vaccines and Immunization and the Vaccine Fund have expanded HBV vaccination programs,¹⁰ but large gaps in coverage remain.

Global consensus concerning the effectiveness and importance of health interventions is helpful but not sufficient. It is not coincidental that all of the TB treatment trials reviewed by Kent et al met best current standards, whereas only a minority of the HIV treatment trials and the malaria prevention trials did. Tuberculosis treatment guidelines have been widely promulgated since 1993 and do not significantly differ between industrialized and developing countries.^{28 - 29} Consensus recommendations for the expanded use of insecticide-treated bednets were not promulgated until 2000,⁹ and there is still disagreement over what may constitute acceptable antiretroviral therapy to prevent mother-to-child transmission of HIV in Africa.^{20 ,25} Even with global consensus, though, most individuals with active TB worldwide never receive treatment.

The WHO and others publish information on progress toward global health goals, but these data are often buried in official reports. Periodic succinct summaries or maps of progress toward the prevention or treatment of major health problems, or lack thereof, should be made available for publication in medical journals. These updates could go a long way toward improving awareness of global health inequities.

Governments, foundations, and academic institutions that support original research must also help to ensure that research findings move into clinical practice in low-resource areas where they are needed. Every institution should reexamine what it does to promote translation of research into action to reduce health inequalities. It is time to shift the debate and the concern from finding ways to conduct studies ethically in regions where unacceptably wide differences in health care exist to focusing on eliminating these health disparities. If this is successful, the ethical issues raised concerning the conduct of research amidst grave disparities will resolve themselves.