Predictors of Loss to Follow-Up among Adult Patients Initiated on Antiretroviral Therapy in Nakuru West Sub-County Health Facilities, Kenya

The loss to follow up (LTFU) has reversed the successful strengthening of antiretroviral therapy (ART) programmes globally. The factors associated with LTFU still remain a challenge to many countries including Kenya, a critical barrier to effective scale-up of human immunodeficiency virus (HIV) services. Kenya is one of the four HIV high burden countries in Africa (alongside Mozambique and Uganda). HIV prevalence in the country is at 5.9 % with 29 % adult HIV co-morbidities, overburdening the health care system and the countrys economy. UNAIDS targeted to achieve 90 % in three key areas of HIV: identification in HIV positive patients; prompt linkage to treatment; and viral suppression by the year 2020. The second and third has not been met in Nakuru West Sub-county. This study aimed at determining predictors of LTFU among adult patients initiated on ART in Nakuru West Sub-County health facilities. A retrospective cohort study design was employed and 1131 participants enrolled/initiated on ART within 1st January 2016 to 31st December 2018 in the County Referral hospital, Kapkures and Rhonda health centers were examined for three years to determine their outcomes. Baseline patient records were extracted from Electronic Master Facility and ART Cohort registers. First, data were entered in SPSS version 21.0. Survival data analysis was done using Cox regression analysis. Kaplan Meier (KM) curves were drawn to estimate the probability of LTFU and proportionality hazards assumption was checked for covariates intended to be included in the final Cox model. From the results, a total of 1131 patients, had an overall LTFU prevalence of 29.8 % (337/1131). Four predictive variables were statistically significantly associated with LTFU in the final Cox regression model: low BMI of less than 18.4, (AHR = 1.88; 95 %CI: 1.40 2.54; p less than 0.0001) ; attending rural facility (AHR: 0.58; 95 %CI: 0.36 0.76; p less than 0.0001) ; using IPT AHR: 1.34; 95 %CI: 0.84 1.57; p