Perrone, Lucy AGloyd, StephenHe, Yao2023-09-272023-09-272023-09-272023He_washington_0250E_26119.pdfhttp://hdl.handle.net/1773/50639Thesis (Ph.D.)--University of Washington, 2023The open-source electronic laboratory information system (OpenELIS) in Côte d’Ivoire has been in routine implementation at clinical laboratories since 2009. The data dashboard that displays aggregate data on HIV viral load (VL) testing and early infant diagnosis (EID) are becoming tools that support people to use OpenELIS data in HIV service delivery. Although various theories, models, and frameworks exist to evaluate or explain implementation effectiveness of health information systems and public health data dashboards, evidence is limited in routine clinical laboratory settings in low- and middle-income countries (LMICs). This dissertation consists of three aims that examine the OpenELIS implementation, context of data-driven decision-making, and user perceptions of the VL and EID data dashboard in Côte d’Ivoire. The first aim quantifies the effects of implementing OpenELIS on data quality at 21 clinical laboratories through an interrupted time series analysis from 2014-2020. The second aim uses a mixed-methods design to assess the sustainment of OpenELIS use at 27 laboratories, identify implementation facilitators and barriers, and explain how high-sustainment laboratories differ from low-sustainment ones. The third aim is a qualitative study of the values, attitudes, and experiences regarding data-driven decision-making and the VL and EID dashboard among existing and potential dashboard users. The implementation of OpenELIS led to an immediate five-fold increase in data timeliness (OR=5.27; 95% confidence interval [CI]: 4.33, 6.41; p<0.001) and a 3.6-fold increase in completeness (OR=3.59; 95% CI: 2.40, 5.37; p<0.001), sustained until 72 weeks after adoption. Weekly post-implementation improvements in completeness were significant (OR=1.03; 95% CI: 1.02, 1.05; p<0.001). However, validity did not show a significant immediate change (OR=1.34; 95% CI 0.69, 2.60; p=0.38). Sustainment of routine OpenELIS use at the laboratories ranged from 9.0% to 99.3% of the implementation period when test results were recorded in OpenELIS. Reported common implementation facilitators included relative advantages and access to knowledge, while common barriers included adaptability issues and power outages. Barriers specific to low-sustainment sites included lower motivation and capability and less reflection and feedback about implementation. Participants valued data-driven decision-making and found the VL and EID data dashboard easy to use and useful for monitoring and decision-making. This dissertation identifies lessons about implementing a laboratory information system and data dashboards that may be useful for other routine settings in LMICs.application/pdfen-USnonedata dashboarddata-driven decision-makingelectronic laboratory information systemroutine implementationPublic healthGlobal HealthThesis