RESUMEN
BackgroundThe impact of COVID-19 in Africa remains poorly defined. We sought to describe trends in hospitalisation due to all medical causes, pneumonia-specific admissions, and inpatient mortality in Kenya before and during the first five waves of the COVID-19 pandemic in Kenya. MethodsWe conducted a hospital-based, multi-site, longitudinal observational study of patients admitted to 13 public referral facilities in Kenya from January 2018 to December 2021. The pre-COVID population included patients admitted before 1 March 2020. We fitted time series models to compare observed and predicted trends for each outcome. To estimate the impact of the COVID-19 pandemic, we calculated incidence rate ratios (IRR) and corresponding 95% confidence intervals (CI) from negative binomial mixed-effects models. ResultsOut of 302,703 patients hospitalised across the 13 surveillance sites (range 11547 to 57011), 117642 (39%) were admitted to adult wards. Compared with the pre-COVID period, hospitalisations declined markedly among adult (IRR 0.68, 95% CI 0.63 to 0.73) and paediatric (IRR 0.67, 95% CI 0.62 to 0.73) patients. Adjusted in-hospital mortality also declined among both adult (IRR 0.83, 95% CI 0.77 to 0.89) and paediatric (IRR 0.85, 95% CI 0.77 to 0.94) admissions. Pneumonia-specific admissions among adults increased during the pandemic (IRR 1.75, 95% CI 1.18 to 2.59). Paediatric pneumonia cases were lower than pre-pandemic levels in the first year of the pandemic and elevated in late 2021 (IRR 0.78, 95% CI 0.51 to 1.20). ConclusionsContrary to initial predictions, the COVID-19 pandemic was associated with lower hospitalisation rates and in-hospital mortality, despite increased pneumonia admissions among adults. These trends were sustained after the withdrawal of containment measures that disrupted essential health services, suggesting a role for additional factors that warrant further investigation.
RESUMEN
BACKGROUND: The concept of benefit sharing to enhance the social value of global health research in resource poor settings is now a key strategy for addressing moral issues of relevance to individuals, communities and host countries in resource poor settings when they participate in international collaborative health research.The influence of benefit sharing framework on the conduct of collaborative health research is for instance evidenced by the number of publications and research ethics guidelines that require prior engagement between stakeholders to determine the social value of research to the host communities. While such efforts as the production of international guidance on how to promote the social value of research through such strategies as benefit sharing have been made, the extent to which these ideas and guidelines have been absorbed by those engaged in global health research especially in resource poor settings remains unclear. We examine this awareness among stakeholders involved in health related research in Kenya. METHODS: We conducted in-depth interviews with key informants drawn from within the broader health research system in Kenya including researchers from the mainstream health research institutions, networks and universities, teaching hospitals, policy makers, institutional review boards, civil society organisations and community representative groups. RESULTS: Our study suggests that although people have a sense of justice and the moral aspects of research, this was not articulated in terms used in the literature and the guidelines on the ethics of global health research. CONCLUSION: This study demonstrates that while in theory several efforts can be made to address the moral issues of concern to research participants and their communities in resource poor settings, quick fixes such as benefit sharing are not going to be straightforward. We suggest a need to pay closer attention to the processes through which ethical principles are enacted in practice and distil lessons on how best to involve individuals and communities in promoting ethical conduct of global health research in resource poor settings.