Measuring quality of life among cervical cancer survivors: preliminary assessment of instrumentation validity in a cross-cultural study.

BACKGROUND: With growing interest in cross-cultural and multicultural cancer-related quality of life studies, the need to assess reliability and validity of quality of life measures for linguistically and culturally diverse cancer survivors is pressing.
METHODS: Reliability and validity of the English and Spanish versions of the Functional Assessment of Cancer Therapy (FACT)-G subscales were tested with a sample of English-speaking European American (n = 273) and ethnic minority American (n = 194), and Spanish-speaking Latina (n = 199) cervical cancer survivors in the U.S.
RESULTS: Reliability coefficients (Cronbach's alpha) were 0.76 or higher across ethnic/linguistic groups except for the emotional wellbeing subscale among Spanish-speaking Latinas (alpha = 0.64). Factor analyses demonstrated overall measurement equivalence across groups with some ethnic/linguistic variations: there were greater differences between linguistic groups than between ethnic groups. Additionally, the scale's factor structure was less satisfactory for Spanish-speaking Latinas. The subscales had good concurrent validity with appropriate subscales of the Short Form (SF)-12 and Rand/SF-36 General Health subscale (Pearson's r 0.53-0.66), suggesting each subscale was assessing its intended construct.
CONCLUSION: The overall psychometric properties of the FACT-G were cross-culturally equivalent. However, more validation studies are needed for non-English speaking populations particularly with emotional wellbeing. In addition, disaggregated analyses on linguistic groups are recommended unless cross-cultural equivalence is established.