Development and Validation of the Krio Version of the WHOQOL-BREF for Use in Sierra Leone

Formal measures of subjective well-being are needed to assure that health and psychosocial support services in Sierra Leone are objectively evaluated and evidence-based. The purpose of this study was to evaluate a Krio version of the brief form of the World Health Organization’s Quality of Life (WHOQOL-BREF) questionnaire. A convenience sample of 425 adult Sierra Leoneans was verbally interviewed in the northern province of Bombali, using a previously piloted version of the scale, as well as a short demographic survey. Descriptive, correlational, and inferential statistics were used to assess evidence of reliability and validity in a manner similar to the WHOQOL-BREF international field trial. Cronbach’s alpha for the four domains ranged from 0.55 to 0.72 and improved when the three negatively phrased items requiring reversed scoring were removed, as well as when all items were treated as a unidimensional scale (alpha = 0.83). Patients scored significantly lower scores on all four subscales and the question assessing overall quality of health question. Persons with no formal education had lower scores on Physical, Psychological, and Environmental domains, as well as the two overall quality of life and health questions. Age was negatively correlated with Physical, Psychological, and Social domains, as well as the questions for overall quality of life and quality of health. Multiple regression analyses found Physical, Psychological, and Environmental domains to be significantly associated with questions assessing overall quality of life and quality of health. Twenty-three out of 24 items correlated highest to their expected domain, but 10 also correlated above 0.40 on another domain. Confirmatory factor analysis allowing for two pairs of error variances to co-vary showed good fit with the original scale’s four domain model (χ2/df = 3.02; CFI = .833; RMSEA = .069, 90% CI = 0.063–0.076). The results indicate that the new scale shares many of the same psychometric properties as the original WHOQOL-BREF and is appropriate for health-related research. Future studies with the WHOQOL and other quality of life instruments should use caution when developing negatively phrased items in scale development, particularly when planned for use across multiple settings, as they may yield unwanted method effects and adversely impact test reliability and internal structure.