Attrition from antiretroviral treatment and the factors associated with attrition from antiretroviral treatment among adult clients : a case of clients at TASO in Uganda
The main objective of the study was to establish the factors associated with attrition of clients receiving ART among adult clients. The study utilized secondary data from TASO, in which data for a cohort of clients initiated from January 2010 to December 2013 were reviewed. A total of 10,061 adult clients were included. At the first level of analysis, a non-parametric technique was employed in which the survival function was used to describe the distribution of time to attrition from ART. At this level, proportions, frequency tables and percentages were used to summarize the data. At the second level of analysis, the log rank test was used to investigate differentials in time to attrition; this was done to test the equality of time to attrition of the various categorical variables. At the multivariate level of analysis, the Cox proportional hazards model was used to regress time to attrition from ART on the predictor variables. The median time to attrition was 2.52 years (IQR: 1.77-3.42). The following factors were significantly associated with attrition from ART: Unmarried clients had increased risk of attrition as compared to the married (HR=1.69, p-value=0.000), the study showed wide variability of attrition rates between different sites of care, for instance whereas at Entebbe site, clients had increased risk of attrition from ART (HR: 2.36, p-value=0.000) as compared to their counterparts at Jinja site, clients at Mulago site had reduced risk of attrition from ART as compared to their counterparts at Jinja site. Low CD4 cell count (< 50 cells/µl) at initiation increased clients attrition from ART as compared to higher CD4 cell count,>=200 cells/ µl (HR:1.87,p-value=0.001), the initial ARV first line regimen combinations, efavirenz/tenoforvir/lamivudine and nevirapine/lamivudine/tenoforvir, increased the risk of attrition from ART as compared to the ARV first line combination, Nevirapine/Zidovudine/Lamivudine (HR: 3.62, p-value=0.000) and (HR: 2.85, p-value=0.000) respectively. Clients attending care from community drug distribution points had reduced risk of attrition as compared to those attending care at TASO facilities(HR:0.75,p-value=0.000). Basing on the findings of this study the researcher recommends the following: Ministry of Health in collaboration with implementing partners should hold learning sessions to have experience sharing among the different health facilities. This will facilitate sharing of good practices which led to reduced risk of loss to follow or death at Mbarara/Masindi, Mulago and Soroti should be implemented as a strategy for improving performance at the sites with high attrition rates and the country at large. Ministry of Health should design policies which will encourage the unmarried clients to remain in care; implementing partners together with health workers should employ quality improvement approaches in identifying what might be leading to high drop-out rates among the unmarried and implement innovative methods to address this. The researcher further recommends that heath workers should initiate ART at a higher CD4 cell count since this might reduce the mortality and loss to follow up of clients after ART initiation, but this should be accompanied with interventions to achieve early diagnosis of HIV and with improvement in linkage between HIV-diagnosis and ART care centers. HIV care providers should implement ART care delivery models which make clients access services easily.