Abstract
Evidence suggests that HIV-infected adolescents experience elevated levels of fatigue that impacts their functioning at school and other important aspects of their lives. Yet, fatigue has not been measured amongst this population group. In this cross-sectional, mixed-methods study we explored the psychometric properties of the 11-item Chalder Fatigue Questionnaire (CFQ) amongst 134 South African adolescents (11-18 years old) receiving antiretroviral therapy (ART) in the Western Cape. Using the Likert scoring method for the CFQ, the mean total score was 14.89 (SD=3.83) and about a quarter (n=33, 24.63%) of adolescents scored ≥ 18, indicating problematic levels of fatigue. The CFQ demonstrated good internal consistency with a Cronbach's alpha of 0.83 (0.84, with item 11 deleted), which is comparable to other validation studies. On the first iteration of the principal component analysis (PCA) with a Varimax rotation, three factors accounted for 59.15% of the variance. However, the third factor consisted of one item only which we chose to remove from the analysis. On the second iteration of the PCA, a two factor solution was retained that accounted for 54.24% of the variance. The first factor, 'Physical fatigue', represented the first eight items on the scale. The second factor 'Mental fatigue' represented items nine and ten. The CFQ also demonstrated modest content and construct validity. The CFQ is a short, easy-to-use and cost-effective measure of fatigue, and demonstrates reliability and validity amongst a South African sample of adolescents. Given the high rate of fatigue amongst our participants, we recommend that future interventions be developed to reduce fatigue among HIV-infected adolescents, which may ultimately lead to improvements in school performance and social functioning. Future research may include test-retest reliability of the CFQ in order to show stability over time.