Predictors of Treatment Failure among HIV Positive Clients in Webuye Sub-County, Bungoma County, Kenya

Abstract

In Kenya, KHDS survey observed that approximately 1.43 million people live with HIV with over 77 % receiving ART while 73% achieving viral suppression. The specific objectives were to determine the profiles of clinical predictors of treatment failure among HIV positive clients attending Webuye sub-county Hospital, Bungoma County, Kenya. To explore the co-morbidities associated with HIV infection among HIV positive clients attending Webuye sub-county Hospital, Bungoma County, Kenya. To assess patient-related factors influencing treatment failure among HIV positive clients attending Webuye sub-county Hospital, Bungoma County, Kenya. The gaps in achieving 90:90:90 indicates the dangers of non-suppression (treatment failure) and the consequent socio-economic effects of viral non-suppressions. Due to the contextual gaps in studies examining first – line treatment failure, the study examined the factors determining the treatment failure among HIV positive clients attending Webuye sub-county hospital, Bungoma County, Kenya. The study employed a retrospective cross-sectional design and targeted a total of 3,231 adults who have been active on ART for more than six months. The inclusion criteria were that the clients were adult who had two viral loads test over the last 6 months while excluding patient younger than 18 years those defaulted on ART and too ill to participate. The study had a sample size of 356 clients selected through a formula, The sample size was determined by Yamane (1967), recommended formula of calculating the desired sample size as stated below, when the target population is less than 10,000 and used Simple random sampling to select the participants through the use of random numbers. The study used a document review tool to capture the clients’ data which was entered into a excel sheet before being transferred to a statistical package software (SPSS 22.0). The data was analysed descriptively and information presented in tabular and pictographic formats. The study used logistic regression analysis to assess the strength of the relationship between the variables at 0.05 significant levels, and chi-square statistics were utilized to investigate the nature of the relationship between the variables. The study measured treatment failure through successive virological load measures ≥ 1,000 copies/mL in between 3 – 6 months of tests. The results indicated that 4.5 per cent of the participants had first – line treatment failure. Age and sex of participant were found to be significant client-related factors through the chi-square test. while first – line ART regimen, time on ART drug adherence and WHO stage 3 were the clinical – related factors. Presence of opportunistic infections were related to treatment outcomes with alcohol consumption. The logistic regression indicated that client – related factors such as being male increases the likelihood of a treatment failure by 0.1065 while being aged 35 years and below odds ratio increases by 0.6015. The most significant clinical factors were WHO stage, Time on ART and drug adherence. A client with WHO stage 3 or 4 increase the odds ratio by 2.04399, CD4 count < 350 cells increase the odds ratio by 0.4498 while being ART is less than 10 years increases the odds ratio by 0.3960. Lastly, poor drug adherence increases the odds ratio by 4.1919. The presence of comorbidities (opportunistic infections) increased the odds ratio by 5.48079 (presence of tuberculosis) and 3.0320 (presence of diabetes). The study conclusion indicates that client- related and clinical – related factors significantly determine treatment failure among PLHIV in Webuye Sub-county while comorbidities was also a significant issue in treatment failure.