Linking health worker motivation with their stated job preferences: A hybrid choice analysis in Ethiopia.

Understanding health worker job preferences can help policymakers better align incentives to retain a motivated workforce in the public sector. However, in stated preference choice modelling, health worker motivation to do their jobs has not been incorporated, perhaps surprisingly, as an important antecedent to health worker job choices. This paper is the first application of a hybrid choice model to measure the extent to which variations in the job preferences of community health workers (CHWs) can be explained by multidimensional motivation. We interviewed 202 CHWs in Ethiopia in 2019. Motivation was assessed quantitatively using a series of thirty questions, on a five-point Likert scale. Stated preferences for hypothetical jobs were captured using an unlabelled discrete choice experiment. We estimated three models and explored which best fitted choice data. We found that the hybrid choice model fitted better than simpler choice models and provides additional behavioural insight into the preferences of CHWs. Intrinsically motivated CHWs had strong disutility towards a higher than average salary, but preferred good facility quality and good health outcomes. On the contrary, CHWs who were assessed to be extrinsically motivated had disutility attached to a heavy workload and preferred higher than average salaries. We show a link between heterogeneity in the job preferences of CHWs and their motivation, demonstrating that its important for policy makers and managers to understand this link in order to get health workers to exert more effort in return for the right incentives and to retain a motivated workforce in the long run.

Discrete choice analysis of health worker job preferences in Ethiopia: Separating attribute non-attendance from taste heterogeneity.

When measuring preferences, discrete choice experiments (DCEs) typically assume that respondents consider all available information before making decisions. However, many respondents often only consider a subset of the choice characteristics, a heuristic called attribute non-attendance (ANA). Failure to account for ANA can bias DCE results, potentially leading to flawed policy recommendations. While conventional latent class logit models have most commonly been used to assess ANA in choices, these models are often not flexible enough to separate non-attendance from respondents' low valuation of certain attributes, resulting in inflated rates of ANA. In this paper, we show that semi-parametric mixtures of latent class models can be used to disentangle successfully inferred non-attendance from respondent's "weaker" taste sensitivities for certain attributes. In a DCE on the job preferences of health workers in Ethiopia, we demonstrate that such models provide more reliable estimates of inferred non-attendance than the alternative methods currently used. Moreover, since we find statistically significant variation in the rates of ANA exhibited by different health worker cadres, we highlight the need for well-defined attributes in a DCE, to ensure that ANA does not result from a weak experimental design.

Determinants of pre-eclampsia/Eclampsia among women attending delivery Services in Selected Public Hospitals of Addis Ababa, Ethiopia: a case control study.

BACKGROUND: Pre-eclampsia is a pregnancy-specific hypertensive disorder usually occurs after 20 weeks of gestation. It is one of the leading causes of maternal and perinatal morbidity and mortality worldwide. In Ethiopia, the major direct obstetric complications including pre-eclampsia/eclampsia account for 85% of the maternal deaths. Unlike deaths due to other direct causes, pre-eclampsia/ eclampsia related deaths appear to be increasing and linked to multiple factors, making prevention of the disease a continuous challenge. The aim of this study is to assess determinants of pre-eclampsia/eclampsiaamong women attending delivery services in selected public hospitals in Addis Ababa, Ethiopia. METHODS: Hospital based unmatched case control study design was employed. The study wasconducted in Addis Ababa among women attending delivery services in two public hospitals from December, 2015 G.C. to February, 2016 G.C. with sample size of 291 (97 cases and 194 controls). Women with pre-eclampsia/eclampsia were cases and women who had not diagnosed for pre-eclampsia/eclampsia were controls. Case-control incidence density sampling followed by interviewer administered was conducted using pretested questionnaire. The data was entered in Epi Info 7 software and exported to STATA 14 for cleaning and analysis. Descriptive statistics were used todisplay the data using tables compared between cases and controls. To compare categorical variables between cases and controls Chi-squared testwas used. Both bivariable and multivariable logistic regression analyses were computed to identify the determinants of pre-eclampsia/eclampsia. RESULTS: Factors that were found to have statistically significant association with pre-eclampsia or eclampsia were primigravida (AOR: 2.68, 95% CI: 1.38, 5.22), history of preeclampsia on prior pregnancy (AOR: 4.28, 95% CI: 1.61, 11.43), multiple pregnancy (AOR: 8.22, 95% CI: 2.97, 22.78), receiving nutritional counseling during pregnancy (AOR: 0.22, 95% CI: 0.1, 0.48) and drinking alcohol during pregnancy (AOR: 3.97, 95% CI: 1.8, 8.75). CONCLUSIONS: The study identified protective and risk factors for pre-eclampsia/eclampsia. To promptly diagnose and treat pre-eclampsia, health workers should give special attention to women with primigravida and multiple pregnancy. Besides, health care providers should provide nutritional counseling during ANC, including avoiding drinking alcohol during their pregnancy.

Newborn care practices at home and in health facilities in 4 regions of Ethiopia.

BACKGROUND: Ethiopia is one of the ten countries with the highest number of neonatal deaths globally, and only 1 in 10 women deliver with a skilled attendant. Promotion of essential newborn care practices is one strategy for improving newborn health outcomes that can be delivered in communities as well as facilities. This article describes newborn care practices reported by recently-delivered women (RDWs) in four regions of Ethiopia. METHODS: We conducted a household survey with two-stage cluster sampling to assess newborn care practices among women who delivered a live baby in the period 1 to 7 months prior to data collection. RESULTS: The majority of women made one antenatal care (ANC) visit to a health facility, although less than half made four or more visits and women were most likely to deliver their babies at home. About one-fifth of RDWs in this survey had contact with Health Extension Workers (HEWS) during ANC, but nurse/midwives were the most common providers, and few women had postnatal contact with any health provider. Common beneficial newborn care practices included exclusive breastfeeding (87.6%), wrapping the baby before delivery of the placenta (82.3%), and dry cord care (65.2%). Practices contrary to WHO recommendations that were reported in this population of recent mothers include bathing during the first 24 hours of life (74.7%), application of butter and other substances to the cord (19.9%), and discarding of colostrum milk (44.5%). The results suggest that there are not large differences for most essential newborn care indicators between facility and home deliveries, with the exception of delayed bathing and skin-to-skin care. CONCLUSIONS: Improving newborn care and newborn health outcomes in Ethiopia will likely require a multifaceted approach. Given low facility delivery rates, community-based promotion of preventive newborn care practices, which has been effective in other settings, is an important strategy. For this strategy to be successful, the coverage of counseling delivered by HEWs and other community volunteers should be increased.