Management Practices Associated with Viral Load Sample Collection among Healthcare Practitioners in Selected Health Facilities in Machakos County

Abstract

Effective viral load (VL) sample management is critical for reliable laboratory monitoring, timely clinical decision-making, and improved patient outcomes. Viral load testing is essential for effective HIV management; however, in Machakos County, Kenya, persistent gaps in sample handling, processing, and transportation have led to rejection rates exceeding 2% annually, peaking at 6.28% in 2023. Hemolysis is the leading cause of rejection, delaying patient care and compromising result reliability. Limited local research has examined the operational factors contributing to these challenges. This study aimed to establish management practices associated with viral load sample collection among healthcare practitioners in selected health facilities in Machakos County. A mixed-methods convergent parallel design was employed, integrating quantitative and qualitative approaches. The study was conducted in public and private health facilities across Machakos County, Kenya, including four viral load hub sites Machakos Level 5, Matuu Level 4, Athi River Level 4, and Kangundo Level 4 hospitals and their 71 satellite facilities. A total of 205 healthcare practitioners involved in VL sample collection, storage, and transportation participated, representing a 94.04% response rate. Quantitative data were collected through questionnaires while qualitative data were gathered using semi-structured interviews. Quantitative data were analyzed using descriptive statistics, Fisher’s Exact Test, odds ratios, and multivariate logistic regression, with model assumption checks performed. Predictive modeling compared logistic regression, random forest, gradient boosting, and support vector machine (SVM) algorithms to identify factors influencing sample management effectiveness. Qualitative data were analyzed thematically using NVIVO version 12, following a six-step framework, and integrated with quantitative findings through triangulation. Ethical approval was obtained from the Kenya Medical Research Institute Scientific and Ethics Review Unit (KEMRI-SERU), and informed consent was obtained from all participants. Participation was voluntary, and confidentiality was maintained throughout the study. Quantitative results revealed a significant association between documentation quality and sample management effectiveness (p = 0.01). Training on sample collection showed a weaker, non-significant association with compliance (p > 0.05), indicating that training alone may be insufficient without robust compliance monitoring. Logistic regression achieved the highest predictive accuracy, identifying significant predictors such as training on sample collection (β = -0.35, p = 0.02), barriers to effective management (β = 0.25, p = 0.07), and quality assurance practices (β = 0.15, p = 0.25). Qualitative findings indicated that the majority of participants reported persistent challenges, including inadequate training, insufficient equipment maintenance, inconsistent documentation, reliance on informal skills transfer, and poor calibration of temperature monitoring devices. Several participants emphasized that without consistent quality assurance protocols, training initiatives have limited long-term impact. The study concludes that effective VL sample management requires a combination of structured training, regular equipment calibration, and rigorous quality assurance monitoring. It recommends that the county government prioritize standardized training programs, enforce equipment maintenance schedules, and strengthen quality assurance systems to enhance VL sample management outcomes