Do SF-36 summary component scores accurately summarize subscale scores ?

Standard scoring algorithms were recently made available for aggregating sc ores from the eight SF-36 subscales in two distinct, higher-order summary s cores: Physical Component Summary (PCS) and Mental Component Summary (MCS). Recent studies have suggested, however, that PCS and MCS scores are not in dependent and may in part be measuring the same constructs. The aims of thi s paper were to examine and illustrate (1) relationships between SF-36 subs cale and PCS/MCS scores, (2) relationships between PCS and MCS scores, and (3) their implications for interpreting research findings. Simulation analy ses were conducted to illustrate the contributions of various aspects of th e scoring algorithm to potential discrepancies between subscale profile and summary component scores. Using the Swedish SF-36 normative database, corr elation and regression analyses were performed to estimate the relationship between the two components, as well as the relative contributions of the s ubscales to the components. Discrepancies between subscale profile and comp onent scores were identified and explained. Significant correlations (r = - 0.74, -0.67) were found between PCS and MCS scores at their respective uppe r scoring intervals, indicating that the components are not independent. Re gression analyses revealed that in these ranges PCS primarily measures aspe cts of mental health (57% of variance) and MCS measures physical health (65 % of variance). Implications of the findings were discussed. It was conclud ed that the current PCS/MCS scoring procedure inaccurately summarizes subsc ale profile scores and should therefore be revised. Until then, component s cores should be interpreted with caution and only in combination with profi le scores.