[Methodological views on the SF-36 summary scores based on the adult German population]

The SF-36 is a widely used instrument to measure health-related quality of life that provides a profile of eight scales [Physical Functioning (PF), Role-Physical (RP), Bodily Pain (BP), General Health (GH), Vitality (VT), Social Functioning (SF), Role-Emotional (RE) and Mental Health (MH)]. To reduce the dimension of the profile, a principal component analysis with the data of the US SF-36 Survey was undertaken. As a result, the Physical Component Summary (PCS) and the Mental Component Summary (MCS) were gained by weighting and summing up of the original scales of the SF-36. The weights were assumed to be country specific. Comparing the weights from the representative samples of nine European countries with those from the USA it has been suggested that the weights gained from the US sample could be applied to all datasets for purposes of comparability and simplicity. In this paper, for a new representative German population sample completed in 1998, the calculation results are compared with those for the older German population sample from 1994. PCS and MCS are calculated using the weights from the new German population sample as well as using the American weights. In this paper, the calculations for the 1998 representative population sample were repeated based on the US weights developed in 1994. The German weights were also calculated based on the 1998 sample. The results were compared and the pro and cons of using the American weights are discussed. In essence, the American weights continue to be acceptable for international comparisons. The separate evaluation of physical and mental health for men and women allows additional insight into gender-specific changes. In this paper, the summary scales for the 1998 German normative population are calculated gender specific. The weights to be used in future studies are also published. With respect to gender mainstreaming, this approach should be taken into consideration in quality of life research. Regarding the methodological and factorial difficulties, the questions arise if the summary scales should be applied instead of the original scales of the SF-36 questionnaire.