Drivers of the progress achieved by Peru in reducing childhood diarrhoea mortality: a country case study.

BACKGROUND: Childhood diarrhoea mortality has declined substantially in Peru in recent decades. We documented trends in childhood diarrhoea mortality from 1980 to 2015, along with trends in coverage of diarrhoea-related interventions and risk factors, to identify the main drivers of mortality reduction. METHODS: We conducted desk reviews on social determinants, policies and programmes, and diarrhoea-related interventions implemented during the study period. We reviewed different datasets on child mortality, and on coverage of diarrhoea-related interventions. We received input from individuals familiar with implementation of diarrhoea-related policies and programmes. We used the Lives Saved Tool (LiST) to help explain the reasons for the decline in diarrhoea mortality from 1980 to 2015 and to predict additional reduction with further scale up of diarrhoea-related interventions by 2030. RESULTS: In Peru under-five diarrhoea mortality declined from 23.3 in 1980 to 0.8 per 1000 livebirths in 2015. The percentage of under-five diarrhoea deaths as related to total under-five deaths was reduced from 17.8% in 1980 to 4.9% in 2015. Gross domestic product increased and poverty declined from 1990 to 2015. Access to improved water increased from 56% in 1986 to 79.3% in 2015. Oral rehydrating salts (ORS) use during an episode of diarrhoea increased from 3.6% in 1986 to 32% in 2015. Vertical programmes focused on diarrhoea management with ORS were implemented successfully in the 1980s and 1990s, and were replaced by integrated crosscutting interventions since the early 2000s. LiST analyses showed that about half (53.9%) of the reduction in diarrhoea mortality could be attributed to improved water, sanitation and hygiene, 25.0% to direct diarrhoea interventions and 21.1% to nutrition. The remaining mortality could be reduced by three-quarters by 2030 with improved diarrhoea treatment and further with enhanced breastfeeding practices and reduction in stunting. LiST does not take into account the role of social determinants. CONCLUSIONS: The reduction of diarrhoeal under-five mortality in Peru can be explained by a combination of factors, including improvement of social determinants, child nutrition, diarrhoea treatment with ORS and prevention with rotavirus vaccine and increased access to water and sanitation. The already low rate of diarrhoea mortality could be further reduced by a number of interventions, especially additional use of ORS and zinc for diarrhoea treatment. Peru is a remarkable example of a country that was able to reduce childhood diarrhoea mortality by implementing interventions through vertical programmes initially, and afterwards through implementation of integrated multisectoral packages targeting prevalent illnesses and multi-causal problems like stunting.

Estimating resource needs for HIV/AIDS health care services in low-income and middle-income countries.

As funding mechanisms like the Global Fund for HIV/AIDS, Tuberculosis and Malaria increasingly make funding decisions on the basis of burden of disease estimates and financial need calculations, the importance of reliable and comparable estimating methods is growing. This paper presents a model for estimating HIV/AIDS health care resource needs in low- and middle-income countries. The model presented was the basis for the United Nations' call for US dollars 9.2 billion to address HIV/AIDS in developing countries by 2005 with US dollars 4.4 billion to address HIV/AIDS health care and the rest to deal with HIV/AIDS prevention. The model has since been updated and extended to produce estimates for 2007. This paper details the methods and assumptions used to estimate HIV/AIDS health care financial needs and it discusses the limitations and data needs for this model.

Estimation of "needs" and "probable uptake" for HIV/AIDS preventive vaccines based on possible policies and likely acceptance (a WHO/UNAIDS/IAVI study).

Once an effective HIV vaccine is discovered, a major challenge will be to ensure its world wide access. A preventive vaccine with low or moderate efficacy (30-50%) could be a valuable prevention tool, especially if targeted to populations at higher risk of HIV infection. High efficacy vaccines (80-90%) could be used in larger segments of the population. Estimated "needs" for future HIV vaccines were based on anticipated policies regarding target populations. Estimated "needs" were adjusted for "accessibility" and "acceptability" in the target populations, to arrive at an estimate of "probable uptake", i.e. courses of vaccine likely to be delivered. With a high efficacy vaccine, global needs are in the order of 690 million full immunization courses, targeting 22 and 69%, respectively, of the 15-49 years old, world wide and in sub-Saharan Africa, respectively. With a low/moderate efficacy vaccine targeted to populations at higher risk of HIV infection, the global needs were estimated to be 260 million full immunization courses, targeting 8 and 41%, respectively, of the world and sub-Saharan African population aged 15-49 years. The current estimate of probable uptake for hypothetical HIV vaccines, using existing health services and delivery systems, was 38% of the estimated need for a high efficacy vaccine, and 19% for a low/moderate efficacy vaccine. Bridging the gap between the estimated needs and the probable uptake for HIV vaccines will represent a major public health challenge for the future. The potential advantages and disadvantages of targeted versus universal vaccination will have to be considered.

Estimating HIV/AIDS prevalence in countries with low-level and concentrated epidemics: the example of Honduras.

Estimates of HIV/AIDS prevalence are important, because they are the primary measure of the current state of the epidemic in a country. How estimates of HIV/AIDS are made depends on the level of the epidemic. For estimates of HIV/AIDS prevalence in low-level and concentrated epidemics it is necessary to disaggregate the total adult population into sub-groups based on the relative risk of infection. For each group, the major issues and questions are: identifying risk groups, estimating the size of the populations, and estimating HIV prevalence in these groups. The greatest difficulty in making estimates of prevalence in low-level and concentrated epidemics is often establishing the size of various populations. Because of the uncertainty inherent in making an estimate of population size for these groups at high risk, low and high estimates are used. In order to demonstrate the method the case of Honduras was used. The most recent HIV prevalence data and the estimates of population sizes were applied. It was estimated that Honduras, which has a total population of 6,575,000 (United Nations Population Division sources), has approximately 55,000 adults living with HIV/AIDS, with a range of uncertainty between 30,000 and 80,000. Estimations of the burden of HIV is a continuous process and should be updated on a regular basis according to the most recent and relevant information available.