Adolescents living with HIV are at higher risk of death and loss to follow up from care: Analysis of cohort data from eight health facilities in Ethiopia.

Background
There are limited data on the treatment outcomes of adolescents living with HIV. Our objective was to compare mortality and loss to follow up (LTFU) rates between adolescent and younger age groups at enrollment in care.
Methods
This was a retrospective cohort study carried out in eight health facilities in two regions of Ethiopia. Adolescents (age 10–14 and 15–19 year) and children (age 0–9 year) enrolled in chronic HIV care between 2005 and 2013 constituted the study population. We reviewed the individual patient charts between March and June 2014 and updated the data on the status of each patient through December 2015. We used death and loss-to-follow up as primary endpoints and used the Cox-regression analysis where age, categorized as adolescent versus child, was the main predictor variable.
Results
Of 2058 participants studied, 52.1% were adolescents. The cohort contributed 2422 person-years of observation (PYO) during the pre-ART follow-up, whereas 1531 patients put on ART contributed 5984 PYO. Of those put on ART, 209 (13.7%) LTFU and 92 (6%) deaths were reported. Adolescents in age group 15–19 yr had the highest risk of LTFU [adjusted hazard ratio, aHR (95% CI) = 3.1 2.1, 5.0 ] followed by those in age group 10–14 yr (aHR = 1.5 [0.9, 2.3]) compared with children aged 0–9 yr. Mortality hazard was significantly higher among younger adolescents (aHR = 2.8 [1.4, 5.4]) and older adolescents (aHR = 2.3 [1.1, 4.9]) compared with children.
Conclusions
Adolescents are at higher risk of mortality and LTFU as compared to children ages 0–9. Younger adolescents and children had similar LTFU rates. Narrow age band disaggregated analysis can serve as useful guide for tailoring interventions to the specific needs of different age groups.