The health care system in Kenya comprises of public and private facilities

Public facilities include those run by the MoH and are categorized into six health levels. Private facilities can be divided into private for-profit and private not-for-profit and are owned and run by non-governmental organizations, faith based organizations or private commercial (for-profit) entities.

Regardless of facility type, the MoH often has a role in guiding on care and reporting standards and protocols [5]. In Kenya, the MoH mandates that all facilities that take care of HIV-positive patients report on various indicators YICHANG. Indicator reporting has been done through the DHIS2 system since 2011. This study will therefore focus on the public and private facilities that have reported HIV indicator data to the DHIS2 from the year 2011 to 2018, hence providing insights to MoH as to how facility type affects the performance of reporting by health facilities.

The study involved all health facilities reporting to DHIS2 across the 47 counties in the country. A retrospective observational study was conducted in order to identify the relationship between facility type and performance on HIV indicator reporting in Kenya. Reporting data used was from years 2011–2018.

From DHIS2, we extracted indicators from the following HIV programmatic areas based on Kenya’s MoH7311 summary form: (i) HIV Counseling and Testing (CT), (ii) Prevention of Mother-to-child Transmission (PMTCT) of HIV, (iii) Care and Treatment (CRT), (iv) Voluntary Medical Male Circumcision (VMMC), (v) Post-Exposure Prophylaxis (PEP), and (vi) Blood Safety (BS). Health facilities only report on the indictors for which they provide services on as mandated by the MoH. Hence, each programmatic area constitutes to a report. Analysis was thus conducted for each programmatic area verses 1 Comprehensive HIV/AIDS Facility Reporting Form. A Retrospective Observational Study 39 aggregating all the areas.

This explains the difference in facility number (N), in each of the programmatic areas. Prior to analysis, a systematic procedure was used to clean the reported data within the data cleaning procedure consisting of three phases, namely: data screening, diagnosis, and treatment.

Health facilities that had never reported HIV indicator data into DHIS2 were excluded from the study TPE gloves. As such, only facilities offering any of the HIV services and reporting data to DHIS2 were included. Duplicates and other erroneous data were also excluded. The DHIS2 HIV indicator dataset was merged with the Kenya Master Facility List (MFL) in order to obtain information of facility ownership type.

The MFL comprises of a list of all registered health facilities in Kenya, and has detailed health facility attributes such as facility ownership PE Gloves, facility type, and facility level. For this study, we excluded facilities with missing facility ownership information. In order to determine the relationship of facility ownership type to performance on HIV indicator reporting, Mann-Whitney U tests were conducted given that the data was not normally distributed.

This was based on normality tests conducted using Shapiro-Wilk tests and test of Homogeneity of Variance. The dependent variables included completeness and timeliness in reporting, which were measured as continuous variables, while the independent variables included private and public facilities, which are categorical variables.