Research questions
The Africa Centre for Health and Population Studies supports the Department of Health HIV treatment and care program in 16 primary care clinics in the rural Hlabisa sub-district of KwaZulu-Natal, South Africa. The focus early in ART roll-out was on increasing access to ART, with time however, the need to ensure quality of care (QoC) has also emerged. We investigate site and health care worker (HCW) reported factors which may hinder or facilitate QOC and examine the association of patient and site factors with time from ART eligibility to initiation (TTI) and other critical measure of QoC.
Fit
The study thus fits with the Africa Centre research agenda on determinants treatment success and treatment outcomes.
Data sources and methods
Between February and April 2009 we administered two surveys at each of the 16 ART sites. A site survey collected information on number of patients, staffing level, scope of services and medication supply. Health care workers (HCW) at each site were given a self-administered survey asking about factors associated with QoC, their own perceptions of QoC and barriers to improvement. Patient data were extracted from the Africa Centre’s clinical and demographic database. Linear regression analysis was used to evaluate the relationship between ART site and TTI while controlling for demographic variables (sex, age at initiation), health status variables (TB co-treatment, ART history), and socioeconomic variables (education, employment, dependent children, receipt of social security grant).
Findings
Between 68 and 1654 patients were on ART at the sites, with patient to HCW ratios ranging from 23 to 150. Twelve of 16 clinics reported at least one stock-out of ARVs in the past three months; 11 reported at least one stock-out of Bactrim. 49 out of 75 nurses and counselors completed the HCW survey. An additional 10 HCW surveys were completed by 4 physicians treating patients in the ART program. The three areas most frequently rated as high quality were: distribution of ART, interactions with patients, and provision of counseling and supportive services. The most frequently cited barrier to QOC was staff shortages. Other barriers included space and infrastructure, medication supply, receipt of laboratory results, poverty and stigma. 5149 patients had data available for the TTI analysis. TTI ranged by up to 49 days between clinics. ART site was a highly significant predictor of TTI in the regressions (joint significance of ART site variables p < 0.001).
Policy implications
· A number of potential barriers to QOC were identified through the site and HCW surveys. These represented site-specific (space, staffing) and system barriers (stock outs, laboratory access) which require interventions by the clinics and program, and other areas (stigma, poverty) which will need broad, multidisciplinary interventions involving the community.
· The site where a patient receives care is significantly associated with TTI. Identifying the clinic-level factors that affect quality of care, is critical for programs to achieve optimal outcomes for all patients.
Next steps include regression analysis of factors contributing to other QoC indicators and inclusion in the models of additional site factors identified by the HCWs as potential site factors determining the QOC received.