Objective: To develop a health status measure in older African-American women with type 2 diabetes.
Research design and methods: African-American women, age > or =40 years with type 2 diabetes, were recruited from central North Carolina to participate in three sequential phases: 1) Seven focus groups were convened and transcripts evaluated to generate questions and identify plausible domains; 2) Ten one-on-one cognitive response interviews were performed to ensure clarity and cultural appropriateness of the questions; and 3) 217 women participated in psychometric evaluation to establish the internal consistency and validity of the instrument.
Results: Three broad categories--mental, physical, and social well-being--captured important issues generated during the focus groups. "My diabetes" was added during the cognitive response interviews as a way of separating the impact of diabetes from coexisting issues that affect health status. The response option was changed from a six- to a four-point Likert scale to accommodate subject preference. Using principal components and subsequent promax rotation, we identified two hierarchical domains (mental and social well-being) and a physical symptom index. The internal consistency (Cronbach's alpha) of the mental and social well-being subscales are 0.83 and 0.93, respectively. A priori hypothesized correlations between subscales along with each subscale and glycated hemoglobin, diabetes duration, physical activity, and a perceived health competence scale helped establish the construct validity of the instrument.
Conclusions: A culturally appropriate disease-specific health status measure for older African-American women with type 2 diabetes has been developed. We have established the internal consistency, construct validity, and factor analytic properties of the measure. This measure should prove useful for investigators who seek a health status instrument that addresses issues germane to African-American women with type 2 diabetes.