Internal Consistency of the SF-6D as a Health Status Index in the Brazilian Urban Population.

OBJECTIVES

To investigate the internal consistency of the SF-6D as a health status index in the Brazilian urban population.

STUDY DESIGN

Cross-sectional population based study.

METHODS

Five thousand individuals, older than the age of 15, were assessed in the five regions of the country. Two different methods of scoring the SF-6D where compared: "weighting the items" of the questionnaire through the Brazilian official weight coefficients, and "unweighting the items" through a parallel non preference scoring rule solely based on patients' answers to SF-6D health classification system (SF-6D). Principal component factor analysis was used for the development of the SF-6D Pearson's, Spearman's, and intraclass correlation coefficients were used to assess the psychometric properties.

RESULTS

The SF-6D scoring formula summarized the pattern of factor loadings and the item-internal consistency (Cronbach's α = 0.858). The scale showed good item-internal consistency, exceeding the 0.70 standard. The association between weighted and unweighted (SF-6D) scores was extremely high (Spearman's ρ = 0.971). The correlations of the SF-6D with the Physical Component of the 12-Item Short-Form Health Survey (SF-12) and the Health Assessment Questionnaire was moderate to strong. The intraclass correlation coefficient obtained (0.917) also suggested that the concordance between the weighted and unweighted score distributions was prominent.

CONCLUSIONS

A nonweighted approach to score the SF-6D provides a reliable global measure of health status. The SF-6D health classification system is useful for assessing quality of life in a large and representative sample of the Brazilian population.