Performance of the Okwawuman Mutual Health Insurance Scheme and Impact on the Utilization and Outcome of Maternal Care Services in Ghana

Abstract

Background: The Okwawuman Mutual Health Insurance Scheme (OMHIS) had been operating since 2002. However, performance reviews have reported mainly on deficits/surpluses, coverage rates and effects on health services utilization. Objective: The study seeks to assess the performance of the scheme with respect to the performance ratios and its impact on the utilization of differential components of health services and also on maternal care outcomes including caesarean section, vacuum extraction, stillbirth and maternal mortality. Method: A cross-sectional and time series study design was used to collect both primary and secondary data through self administered questionnaires, structured interview guides and checklists. Secondary data were collected from the District Health Directorates and the District Hospitals spanning a period of eight years: Four years before and after the introduction of the OMHIS. The financial and membership records of the OMHIS were also studied. Primary data were gathered through a cross-sectional community survey, hospital exit interviews and key informant interviews. Logistic regression was employed for the test of associations. Results: The overall combined ratio of the OMHIS for the period under review was <1.0 making it a viable scheme. Health services utilization was generally increased against a declining length of stay which was influenced by complications of caesarean section and vacuum extraction (p= 0.000) rather than mode of payment. Drug availability was 95.29% and the prescription of generic (rINN) or proprietary drugs was not influenced by insurance status (P rINN = 0.777; P Proprietary = 0.213). The mean annual caesarean section rate was high, 21-22% and that for vacuum extraction was low, 0.95- 1.35%. There was a mean annual excess caesarean section of 151 cases. Caesarean section and vacuum extraction were not significantly affected by health insurance [OR C/S = 0.337, 95% CI= 0.025-4.493 (Vs. Uninsured); OR V/E = 2.970, 95% CI= 0.223- 39.627 (Vs. Uninsured)]. Insurance impacted significantly on stillbirth (OR= 3.223, 95% CI= 1.527-6.800). Among the medical determinants of stillbirth, low birth weight (<2.5kg) had the strongest impact (OR= 89.979, 95% CI= 12.002-674.564). Stillbirth was more affected by per capita ANC attendance than supervised delivery. Insurance did not impact significantly on maternal mortality [OR= 1.587, 95% CI= 0.325-7.761 (Vs. Insured)] but rather medical determinants; with complications of labour having the strongest impact (OR= 30.320, 95% CI= 9.748-94.309). Supervised delivery played a more pivotal role in influencing maternal mortality than per capita ANC attendance. Conclusion: The Okwawuman Mutual Health Insurance Scheme was viable with government subvention. The scheme caused an increased utilization of the health services but did not influence length of stay in hospital and prescription pattern. The scheme also significantly impacted on stillbirth but was not significantly associated with maternal mortality, caesarean section and vacuum extraction.