Dropout from physical activity interventions in people living with HIV: a systematic review and meta-analysis

ABSTRACT

Physical activity (PA) interventions have been shown to improve the health of people living with HIV (PLWH), yet treatment dropout poses an important challenge. We conducted a meta-analysis to investigate the prevalence and predictors of treatment dropout in PA interventions in PLWH. Electronic databases were searched for records up to September 2016. Randomized control trials of PA interventions in PLWH reporting dropout rates were included. Random effects meta-analysis and meta-regression analyses were employed. In 36 studies involving 49 PA intervention arms, 1128 PLWH were included (mean age = 41.6 years; 79.3% male; 39% White). The trim and fill adjusted treatment dropout rate was 29.3% (95% CI = 24.5–34.7%). There was a significant lower dropout rate in resistance training interventions compared with aerobic (p = 0.003) PA interventions, in studies utilizing supervised interventions throughout the study period (p < 0.001), and in studies using adequately qualified professionals (p < 0.001). Exerciser/participant variables that moderated higher dropout rates were a lower percentage of male participants (β = 1.15, standard error (SE) = 0.49, z = 2.0, p = 0.048), a lower body mass index(BMI) (β = 0.14, SE = 0.06, z = 2.16, p = 0.03), and a lower cardiorespiratory fitness (β = 0.10, SE = 0.04, z = 2.7, p = 0.006). The dropout from PA interventions is much higher in PLWH than in many other populations with chronic morbidities. Qualified professionals (i.e., exercise physiologists, physical educators, or physical therapists) should be incorporated as key care providers in the multidisciplinary care of HIV/AIDS and should prescribe supervised PA for PLWH in order to enhance adherence and reduce the burden of HIV/AIDS. Special attention should be given men, those with a higher BMI, and those with a lower cardiorespiratory fitness.

KEYWORDS:

• HIV
• AIDS
• physical activity
• exercise

Disclosure statement

No potential conflict of interest was reported by the authors.