Adherence to antiretroviral therapy and factors associated with non-adherence: a cohort study at two referral services in Brazil

ABSTRACT

This cohort study evaluated non-adherence to antiretrovirals at referral services in Pernambuco, Brazil, 2016/2017, through self-report. A generalized mixed-effects model for binary outcomewas used. We assessed 542 participants with an adherence rate of 85.50%. A greater chance of non-adherence was associated with:a low/moderate level of nicotine dependence (OR = 2.79, p = 0.00, IC = 1.44-5.41); ≥7 tablets/day (OR = 6.14, p = 0.00, IC = 3.42-11.02); LPV/r (OR = 1.49, p = 0.6, IC = 0.98-2.26), ddI (OR = 3.34, p = 0.03, IC = 1.12-9.97), ABC (OR = 4.02, p = 0.05, IC = 1.01-16.03), RAL (OR = 2.49, p = 0.01, IC = 1.32-4.70) and DTG (OR = 4.65, p = 0.01, IC = 1.42-15.16); 6–10 year seropositive diagnosis (OR = 2.17, p = 0.01, IC = 1.20-3.92) and symptoms of depression (OR = 1.55, p = 0.03, IC = 1.03-2.33). Protective factors for non-adherence weres: ≥50 years (OR = 0.67, p = 0.06, IC = 0.45-1.01), secondary/higher education (OR = 0.48, p = 0.00, IC = 0.34-0.70), embarrassment at health service (OR = 0.49, p = 0.04, IC = 0.24-0.97), good understanding of antiretrovirals (OR = 0.62, p = 0.03, IC = 0.40-0.96), adverse event (OR = 0.74, p = 0,06, IC = 0.54-1.01), use of TDF (OR = 0.62, p = 0.01, IC = 0.43-0.90), NVP (OR = 0.41, p = 0.05, IC = 0.71-1.00) and EFZ (OR = 0.48, p = 0.01, IC = 0.29-0.80) and good knowledge of HIV/AIDS/ART. (OR = 0.67, p = 0.07, IC = 0.43-1.04). Variables with stronger association were those linked to ART. Systematic use of self-report adherence is recommended for priority groups.

RESUMEN

Esta cohorte evaluó la no adherencia a los antirretrovirales en los servicios de referencia en Pernambuco, Brasil, 2016/2017, a través del autoinforme. Se utilizó un modelo generalizado de efectos mixtos para la respuesta dicotómica. Se evaluaron 542 participantes y se observó una tasa de adherencia del 85.50%. Tenían más oportunidad de no adherirse los participantes con un nivel bajo/medio de dependencia a la nicotina (OR = 2.79, p = 0.00, IC = 1.44-5.41); que estuviera tomando  ≥  7 tabletas/día (OR = 6.14, p = 0.00, IC = 3.42-11.02); que estuviera usando LPV/r (OR = 1.49, p = 0.6, IC = 0.98-2.26), ddI (OR = 3.34, p = 0.03, IC = 1.12-9.97), ABC (OR = 4.02, p = 0.05, IC = 1.01-16.03), RAL (OR = 2.49, p = 0.01, IC = 1.32-4.70) y DTG (OR = 4.65, p = 0.01, IC = 1.42-15.16); con diagnóstico de seropositividad 6-10 años (OR = 2.17, p = 0.01, IC = 1.20-3.92) y con síntomas de depresión (OR = 1.55, p = 0.03, IC = 1.03-2.33). Identificado como factor de protección para: tener 50 años o más (OR = 0.67, p = 0.06, IC = 0.45-1.01), haber cursado escuela secundaria/universidad (OR = 0.48, p = 0.00, IC = 0.34-0.70), vergüenza en el centro de salud (OR = 0.49, p = 0.04, IC = 0.24-0.97), buena comprensión de los antirretrovirales (OR = 0.62, p = 0.03, IC = 0.40-0.96), evento adverso(OR = 0.74, p = 0,06, IC = 0.54-1.01), TDF (OR = 0.62, p = 0.01, IC = 0.43-0.90), NVP (OR = 0.41, p = 0.05, IC = 0.71-1.00) y EFZ (OR = 0.48, p = 0.01, IC = 0.29-0.80) y buen conocimiento del VIH/SIDA/ART. (OR = 0.67, p = 0.07, IC = 0.43-1.04).Las variables con la mayor fuerza de asociación fueron las vinculadas a ART. Se recomienda el uso sistemático del autoinforme de adhesión para grupos prioritarios.

KEYWORDS:

• HIV
• treatment adherence and compliance
• antiretroviral therapy
• high activity
• risk factors
• self report

DESCRIPTORES:

• VIH
• cumplimiento adherencia al tratamiento
• terapia antirretroviral altamenteactiva
• factores de riesgo
• autoinforme

Acknowledgments

The authors thank all participants in the cohort and the staff of the Hospital Universitário Oswaldo Cruz Infectious and Parasitic Diseases Outpatient Clinic, the records department and pharmacy, and the Hospital Correia Picanço.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Ethical considerations

This research was approved by the Ethics Committee on May 27, 2015, with C.A.A.E. no. 42923514.4.0000.5192.

Additional information

Funding

This research was funded by the Health Surveillance Secretariat of the Brazilian Ministry of Health - Public Notice No. 20/2013, Studies and Applied Research in Health Surveillance. It was funded, in part, by the Coordination for the Improvement of Higher Education Personnel - Brazil (CAPES) - Finance Code 001. As well as the National Council for Scientific and Technological Development (309722 / 2017–9 to RAAX, 308590 / 2013–9 to DBMF), and through the Scientific Initiation Program of the Science and Technology Support Foundation of the State of Pernambuco and the National Council for Scientific and Technological Development.