Adherence to tablet and liquid formulations of antiretroviral medication for paediatric HIV treatment at an urban clinic in Uganda.

BACKGROUND: Major obstacles remain in scaling up paediatric HIV treatment, including limited paediatric anti-retroviral drug options for resource-limited settings, challenges with adherence to liquid formulations and treatment fatigue with lifelong therapy.

AIM: To determine levels of adherence to HAART in HIV-infected children at 12, 24, 36 and 48 weeks of follow-up and to compare adherence levels before and after switching from syrup to fixed-dose combination (FDC)-tablet anti-retroviral formulations.

METHODS: HIV-infected children aged between 6 months and 12 years were initiated on anti-retroviral therapy at Makerere University-Johns Hopkins University Care Clinic, Kampala. Good adherence to HAART was defined as taking ≥95% of prescribed medications. Adherence levels were measured using pharmacy refill data, quarterly unannounced home-visit pill counts and caregiver self-reports. Data were analysed using STATA(®) version 10.0.

RESULTS: A total of 129 HIV-infected children were initiated on HAART with 14.7% on syrups and 85.3% on tablet formulations at enrollment. According to caregiver self-reporting, 99.2%, 100%, 100% and 99.2% achieved ≥95% adherence at 12, 24, 36 and 48 weeks, respectively. Using pharmacy refill data, the proportions were 89.9%, 95.4%, 93.8% and 93.0% and for unannounced home visits were 89.8%, 92.4%, 88.9% and 86.2%, respectively. Median adherence to syrup formulations (97%, IQR 93-98) was significantly lower than for tablets (100%, IQR 97-100, p = 0.012, n = 28) using pharmacy refill data. Viral suppression correlated with home visit and pharmacy refill adherence data.

CONCLUSION: The majority of children initiating HAART had good adherence when estimated by caregiver self-report and pharmacy refill data but lower adherence when measured by home-visit pill counts. Adherence to tablet formulation of HAART was significantly better than syrup formulation. Medication formulation did not significantly affect viral suppression.