Modelling adverse treatment outcomes of HIV-infected adolescents attending public-sector HIV clinics in Lusaka

• Published in: Public health (London) 2017-06, Vol.147, p.8-14
• Main Authors: Fwemba, I, Musonda, P
• Format: Article
• Language: English
• Subjects: Adequacy; Adolescent; Adolescents; Adult; Adults; Adverse; Age Factors; Ambulatory Care Facilities; Anti-HIV Agents - therapeutic use; Antiretroviral agents; Antiretroviral therapy; CD4 antigen; CD4 Lymphocyte Count - statistics & numerical data; Child; Clinical outcomes; Clinics; Confidence intervals; Data processing; Female; Hazards; HIV; HIV clinics; HIV Infections - drug therapy; HIV Infections - mortality; HIV-infected; Human immunodeficiency virus; Humans; Immune response; Infectious Disease; Internal Medicine; Long term health care; Long-term care; Male; Modelling; Mortality; Outcomes; Patient Compliance - statistics & numerical data; Public health; Public Sector; Recovery; Regression analysis; Retrospective Studies; Risk; Risk reduction; Side effects; Statistical analysis; Statistical models; Teenagers; Treatment; Treatment Outcome; Young Adult; Young adults; Zambia - epidemiology
• ISSN: 0033-3506; 1476-5616
• DOI: 10.1016/j.puhe.2017.01.022

Abstract Background In resource-limited setting, there is scarce evidence comparing antiretroviral therapy (ART) outcomes among HIV-infected adolescents to that of other age groups. Methods and study design We analysed data from 25 ART facilities in Lusaka District, comparing treatment-naïve ART-eligible young adolescents (10–14 years), older adolescents (15–19) and young adults (20–24 years) initiating first-line ART to those aged 24 years or older. The adjusted relative risk (RR) of failure to achieve an adequate CD4 response (defined as failure to increase CD4 count by ≥ 50 cells/mm3 at 6 months or by ≥ 100 cells/mm3 ) at 6 or 12 months after ART initiation was modelled using log-binomial regression. The effect of age group on mortality and loss to follow-up (LTFUP; ≥60 days since scheduled visit date) was estimated using adjusted Cox proportional hazards models, respectively. This was a routine retrospective design using program data. Results Of the 94,023 patients initiating ART from May 2004 to February 2011, 1303 (1.4%) were young adolescents, 1440 (1.5%) were older adolescents and 5825 (6.2%) were young adults. 85,455 (90.9%) were 24 years or older at the time of ART initiation. Compared with adults, both young adolescents (RR: 0.88, 95% confidence interval [CI]: 0.76–1.01 at 6 months and RR: 0.80, 95% CI: 0.69–0.93 at 12 months) and older adolescents (RR: 0.82, 95% CI: 0.71–0.95 at 6 months) were less likely to achieve adequate CD4 response. No evidence of a difference in mortality risk was observed among older adolescents (hazard ratio [HR] 1.20, 95% CI: 0.93–1.56) compared with adults; however, there was a reduced risk of mortality in young adolescents compared with adults (HR: 0.61, 95% CI: 0.40–0.92). Young adolescents were less likely to be LTFUP following ART initiation (HR: 0.74, 95% CI: 0.59–0.92), while older adolescents and young adults were reported to be more likely to drop out of care (HR: 1.54 95% CI: 1.33–1.78; HR: 1.51 95% CI: 1.40–1.63 respectively). Conclusion Older adolescents and young adults had poorer ART treatment outcomes, including failure to achieve adequate CD4 recovery and failure to remain in long-term care, when compared with adults. Interventions are necessary to help increase outcomes and retention in care.