Loss to follow-up and associated factors among adult people living with HIV at public health facilities in Wakiso district, Uganda: a retrospective cohort study

Abstract

Background: The Human Immunodeficiency Virus/ Acquired Immune deficiency Syndrome (HIV/AIDS) epidemic is still a major global public health challenge, particularly to resource constrained countries like Uganda. The availability of Anti-retroviral Therapy (ART) for HIV treatment and prevention has helped to suppress the virus. However, there is still a problem of loss to follow- up (LTFU) since patients have to stay on treatment for a lifetime. With the current policy of ‗test and treat‘ for HIV/AIDS, we anticipate the problem of LTFU to be more pronounced given that more people need to be immediately initiated on ART. Several studies have been conducted in Centres of excellence‘ where patients are provided with extra support to ensure that they are retained in care. However, we need more evidence from the public health care delivery system. This study sought to determine the incidence of and the factors that are associated with LTFU amongst adult HIV positive patients in Wakiso district. Methods: The study employed a concurrent- nested mixed methods (quantitative and qualitative) design of data collection and analysis. For the quantitative study, a retrospective cohort design was used to review 646 records of HIV positive individuals who were registered for care in 13 health facilities of Wakiso district between January 2015 and December 2017. The study used systematic sampling technique to randomly select the patients to study from the list of patients receiving ART in the selected health facility. The data was extracted from the patient ART cards in the health facilities using a data extraction form. The study described the properties of patients lost to care. The cox proportional hazards regression model was used to determine the factors that are associated with LTFU. Subgroup analysis was done using the log rank test to compare the patients that were enrolled on a ‗Test and Treat‘ strategy to those enrolled on the basis of CD4 count or World Health Organization (WHO) Clinical staging. For the qualitative study, we conducted in-depth interviews with patients and key informant interviews with health service providers to explore perceptions about LTFU. The information was coded and synthesized to conduct a thematic analysis. Results: Out of 646 patients, 216 were LTFU, 359 were still in care, 55 were transferred out, and 16 died. The overall incidence rate of LTFU was 21 per 1000 person-months (95%Confidence Interval (CI): 18- 25 per 1000 person- months). Normal weight patients were 36 percent less likely to be LTFU (Adjusted Hazard Ratio (aHR) =0.64, 95%CI: 0.453- 0.903) compared to underweight patients; and hospital level patients were 78 percent less likely to be LTFU (aHR=0.22, 95%CI: 0.121- 0.408) compared to health centre III level patients. The patients with no telephone contact were twice as likely to be LTFU (aHR=2.16, 95%CI: 1.330- 3.511). The patients initiated on the basis of the traditional CD4 count had an incidence rate of LTFU of 30 per 1000 person-months (95%CI: 24- 39 per 1000 personmonths) compared to ‗Test and Treat‘ strategy with an incidence rate of LTFU of 33 per 1000 person months (95%CI: 23- 49 per 1000 person-months). The p-value of the difference was 0.231. Perceptions about stigmatization, competing life activities, and long waiting time were prominent among the lost patients compared to the active patients. Conclusions: The incidence rate of LTFU is higher in the public health care setting, mainly due to high patient loads, compared to the ‗Centers of excellence‘. The factors such as normal weight, hospital level and not having telephone contact are important while addressing the problem of LTFU in the public health care setting.