Return on investments in the Health Extension Program in Ethiopia.

BACKGROUND: Since 2003, the government of Ethiopia has trained and deployed more than 42,000 Health Extension Workers across the country to provide primary healthcare services. However, no research has assessed the return on investments into human resources for health in this setting. This study aims to fill this gap by analyzing the return on investment within the context of the Ethiopian Health Extension Program. METHODS: We collected data on associated costs and benefits attributed to the Health Extension Program from primary and secondary sources. Primary sources included patient exit interviews, surveys with Health Extension Workers and other health professionals, key informant interviews, and focus groups conducted in the following regions: Amhara, Oromia, Tigray, and the Southern Nations Nationalities and Peoples' Region. Secondary sources consisted of financial and administrative reports gathered from the Ministry of Health and its subsidiaries, as well as data accessed through the Lives Saved Tool. A long-run return on investment analysis was conducted considering program costs (personnel, recurrent, and capital investments) in comparison to benefits gained through improved productivity, equity, empowerment, and employment. FINDINGS: Between 2008-2017, Health Extension Workers saved 50,700 maternal and child lives. Much of the benefits were accrued by low income, less educated, and rural women who had limited access to services at higher level health centers and hospitals. Regional return ranged from $1.27 to $6.64, with an overall return on investment in the range of $1.59 to $3.71. CONCLUSION: While evidence of return on investments are limited, results from the Health Extension Program in Ethiopia show promise for similar large, sustainable system redesigns. However, this evidence needs to be contextualized and adapted in different settings to inform policy and practice. The Ethiopian Health Extension Program can serve as a model for other nations of a large-scale human resources for health program containing strong economic benefits and long-term sustainability through successful government integration.

Significant Improvement in Blood Pressure Levels Among Older Adults With Hypertension in Rural South Africa.

BACKGROUND: Sub-Saharan Africa is undergoing an epidemiologic transition from infectious diseases to cardiovascular diseases. From 2014 to 2019, sociodemographic surveillance was performed in a large cohort in rural South Africa. METHODS: Disease prevalence and incidence were calculated using inverse probability weights. Poisson regression was used to identify disease predictors. The percentage of individuals with controlled (<140/90 mm Hg) versus uncontrolled hypertension was compared between 2014 and 2019. RESULTS: Compared with 2014 (n=5059), study participants in 2019 (n=4176) had similar rates of obesity (mean body mass index, 27.5±10.0 versus 27.0±6.5) but higher smoking (9.1% versus 11.5%) and diabetes (11.1% versus 13.9%). There was no significant increase in hypertension prevalence (58.4% versus 59.8%; age adjusted, 64.3% versus 63.3%), and there was a significant reduction in mean systolic blood pressure (138.0 versus 128.5 mm Hg; P<0.001). Among hypertensive individuals who reported medication use in 2014 and 2019 (n=796), the proportion with controlled hypertension on medication increased from 44.5% to 62.3%. Hypertension incidence was 6.2 per 100 person-years, and age was the only independent predictor. Among normotensive individuals in 2014 (n=2257), 15.2% developed hypertension by 2019, with the majority already controlled on medications by 2019. CONCLUSIONS: The hypertension prevalence and incidence are plateauing in this aging cohort. There was a statistically and clinically significant decline in mean blood pressure and a substantial increase in individuals with controlled hypertension on medication. The prevalence of cardiometabolic risk factors did not decrease over time, suggesting that the blood pressure decrease is likely due to increased medication access and adherence, promoted by local health systems.

Estimating population level 24-h sodium excretion using spot urine samples in older adults in rural South Africa.

BACKGROUND: South Africa has introduced regulations to reduce sodium in processed foods. Assessing salt consumption with 24-h urine collection is logistically challenging and expensive. We assess the accuracy of using spot urine samples to estimate 24-h urine sodium (24hrUNa) excretion at the population level in a cohort of older adults in rural South Africa. METHODS: 24hrUNa excretion was measured and compared to that estimated from matched spot urine samples in 399 individuals, aged 40-75 years, from rural Mpumalanga, South Africa. We used the Tanaka, Kawasaki, International Study of Sodium, Potassium, and Blood Pressure (INTERSALT), and Population Mean Volume (PMV) method to predict 24hrUNa at the individual and population level. RESULTS: The population median 24hrUNa excretion from our samples collected in 2017 was 2.6 g (interquartile range: 1.53-4.21) equal to an average daily salt intake of 6.6 g, whereas 65.4% of participants had a salt excretion above the WHO recommended 5 g/day. Estimated population median 24hrUNa derived from the INTERSALT, both with and without potassium, showed a nonsignificant difference of 0.25 g (P = 0.59) and 0.21 g (P = 0.67), respectively. In contrast, the Tanaka, Kawasaki, and PMV formulas were markedly higher than the measured 24hrUNa, with a median difference of 0.51 g (P = 0.004), 0.99 g (P = 0.00), and 1.05 g (P = 0.00) respectively. All formulas however performed poorly when predicting an individual's 24hrUNa. CONCLUSION: In this population, the INTERSALT formulas are a well suited and cost-effective alternative to 24-h urine collection for the evaluation of population median 24hrUNa excretion. This could play an important role for governments and public health agencies in evaluating local salt regulations and identifying at-risk populations.

Associations Between Practices and Behaviors at the Health Facility Level and Supply Chain Management for Antiretrovirals: Evidence from Cameroon, Namibia, and Swaziland.

BACKGROUND: While measuring, monitoring, and improving supply chain management (SCM) for antiretrovirals (ARVs) is understood at many levels of health systems, a gap remains in the identification and measurement of facility-level practices and behaviors that affect SCM. This study identifies practices and behaviors that are associated with SCM of ARVs at the hospital level and proposes new indicators for measurement. METHODS: We performed an in-depth literature review to identify facility-level practices and behaviors and existing indicators that are associated with SCM. We used the United States Agency for International Development's 2013 National Supply Chain Assessment Toolkit to define 7 supply chain function areas to frame the study. Qualitative, semistructured key informant and focus group interviews were conducted in hospitals with health professionals from Cameroon, Namibia, and Swaziland to understand facility-level practices and behaviors. RESULTS: Using the results from 54 key informant and focus group interviews from 12 hospitals, we identified 30 practices and behaviors that may affect ARV SCM at the facility level. The following practice areas were particularly associated with SCM: order verification, actions taken when ARV stock is received, changes in prescription and dispensing due to ARV stock-out, actions to ensure patient adherence, and communication with other affiliated facilities and higher-level SCM. We subsequently developed measurable indicators for future research. CONCLUSION: This study characterizes facility-level practices and behaviors that can affect ARV SCM. It also identifies gaps in their measurement. While this study uses ARVs as a tracer medicine to understand gaps in practices at the facility level, many of the findings are more broadly applicable to other medicines in an integrated setting. This study provides real-world evidence and the groundwork for further research to characterize the link between 30 facility-level practices and behaviors and ARV SCM at the facility and central levels.