[Factorial variance of the SCL-90-R in a Spanish out-patient psychiatric sample].

OBJECTIVE: To test the consistency of the proposed structure for the SCL-90R.
METHOD: 598 first appointments of a Mental Health Centre were evaluated. 352 of these patients completed the SCL-90-R during their first appointment and also 6 and 12 months later. The structure of this questionnaire from the first appointment as well as from follow-up observations is analysed separately in men and women by exploratory factor analysis. Moreover, confirmatory factor analysis have been applied in order to compare the relative adjustment with the data observed during the first appointment in the original model of Derogatis et al., as well as in 3 other factor models.
RESULTS: Exploratory factor analysis rendered a different factor structure with all other contrasted models, that were rejected by confirmatory factor analysis as well, in men and women. Only a relative temporal stability in factor structure, different for men and women, was found.
CONCLUSION: It seems that this tool may be rather more useful as a unitary measure for global distress. In addition, our results suggest that the factor structure of the SCL-90-R may vary in the same sample depending on the gender and also possibly varying throughout the time of observation. According to this, the benefit of the SCL-90-R for descriptive and monitoring studies throughout the time is questionable for this type of samples.