Aims: An empirical comparison of the performance characteristics of 3 generic health-related quality of life (HRQL) measures in pain patients.
Methods: The Nottingham Health Profile (NHP), the Medical Outcomes Study 36-item Short-Form Health Survey (SF-36), and the German Life Satisfaction Scale (FLZ-M) by Henrich et al. were simultaneously employed in a multicenter survey measuring the impact of pain on quality of life. The HRQL- instruments were incorporated into the German Pain Questionnaire (pain variables, CES-D, Pain Disability Index).
Results: Characteristics of 3294 pain patients of 13 pain facilities are detailed in tables 1-3. Six of the 8 SF-36- and 4 of the 6 NHP-scales show satisfactory item-total correlations, bottom- and ceiling-effects, and internal consistency. FLZ-M reliabilities are satisfactory. The item weighting procedure of the FLZ-M proves to be unnecessary. Principle component analyses result in 7 factors for the SF-36 and the NHP. Six of the SF-36-factors are fairly homogeneous. The heterogeneity of the NHP- factors is marked. Correlations of the HRQL scales with depression (CES-D), anxiety (STAI) and physical functioning (FFbH-R-18) are high in all related contents. All instruments discriminate well between headache and back pain patients, between several pain grades (v. Korff) and the 3 Mainz pain chronicity stages.
Conclusions: The SF-36 has satisfactory to good psychometric properties in pain patients, the NHP item selection has to be improved. The FLZ-M weighting can be eliminated. The shortcomings of the SF-36 can be overcome by adding short scales on role functioning and pain (modular approach).