Validity and reliability of the Setswana translation of the Short Form-8 health-related quality of life health survey in adults

Abstract

BACKGROUND: The absence of culturally relevant measures in indigenous languages could pose a challenge to epidemiological studies on health-related quality of life (HRQoL) in developing nations AIM: To explore the feasibility and determine the validity and reliability of the Setswana translation of the HRQoL Short Form-8 (SF-8) among Setswana-speaking adults SETTING: Potchefstroom in the North West province. METHODS: Sixty healthy men (n = 26) and women (n = 34), aged 45.5 ± 9.3 years, completed a Setswana translation of the SF-8 questionnaire and the original English version twice, with a 4-week interval between completions. RESULTS: The Setswana SF-8 presented good concurrent validity with the Spearman's correlation coefficients (ρ) of 0.72 for role physical to 0.91 for social functioning. The Cronbach's alpha coefficients for the first and second measurements were 0.87 and 0.87, respectively, for the Setswana-translated SF-8 and 0.86 and 0.89 for the original English SF-8. The reliability coefficients were moderate for the mental health (ρ = 0.60), social functioning (ρ = 0.56) and role emotional (ρ = 0.50) domains, as well as the mental component summary (ρ = 0.50) and physical component summary (ρ = 0.45), but fair for the role physical (ρ = 0.43), body pain (ρ = 0.43), general health (ρ = 0.42), physical functioning (ρ = 0.41) and vitality (ρ = 0.38) domains on the translated Setswana version of the SF-8. CONCLUSION: The Setswana SF-8 version was feasible, acceptable and had acceptable concurrent validity and fair to moderate evidence of test-retest reliability for assessing HRQoL among adult Setswana-speaking community dwellers.