Benz T1,2, Lehmann S3, Elfering A2, Aeschlimann A3, Angst F3
1RehaClinic, Research Department and Pain Center, Bad Zurzach, Switzerland, 2Institute of Psychology, University of Bern, Bern, Switzerland, 3RehaClinic, Research Department, Bad Zurzach, Switzerland
Background: Chronic musculoskeletal pain leads to reduced physical activity, functional impairment, loss of health-related quality of life and employment. It is unclear which measurement scale performs best in evaluating the state and change of function.
Purpose: The aim of this study was to examine cross-sectional and longitudinal validity of function related measurements in patients with chronic unspecific low back pain.
Methods: This prospective cohort study assessed 142 patients at baseline (before) and after a specific, 4 week interdisciplinary pain management program by the Short Form 36 (SF-36), the Multidimensional Pain Inventory (MPI) and Oswestry Disability Index (ODI). All scores were transformed into a scale from 0 (= worst function, maximal pain, worst health) to 100 (= best function, minimal pain, best health). Assessment was completed by 2 functional performance tests: Back Performance Scale (BPS) and 6-Minute Walking Distance (6MWD). Statistical analysis included bivariate correlations and factor analysis.
Results: Patients included in this study were on average of 47.6 years (+/-11.8) and 61.0% female. Correlations of baseline scores ranged from r=0.06 (BPS with MPI Social and away from home activities) to r=0.81 (SF-36 Mental health with MPI Negative mood). SF-36 Physical functioning correlated highest with functional performance tests: r=0.53 BPS and r=0.67 6MWD. Together with MPI Interference with pain, SF-36 Social functioning and mental health dimensions of SF-36 and MPI loaded most strongly on the first factor (37.4% explained variance). SF-36 Physical functioning, BPS and 6MWD loaded most strongly on the second factor (28.6% explained variance) at baseline. Correlations of change scores (difference of follow-up - baseline score) all were weaker, ranging from r=-0.02 (SF-36 Social functioning with BPS) to r=0.64 (SF-36 Mental Health with MPI Negative Mood). SF-36 Physical functioning correlated highest with MPI Interference with pain (r=0.41). Factor load analysis showed 3 factors with an explained variance of 25.3% (strongest load: MPI Pain severity and Interference with pain, ODI Function and SF-36 Bodily pain), 19.9% (SF-36 Social functioning and Mental health, MPI Negative mood) and 16.1% (BPS and 6MWD).
Conclusion(s): Overall, construct overlap of function related measurements was moderate to weak and the construct of physical function explained limited variance of state and change of health in chronic low back pain. State and change of function was strongest associated to social and psychological factors. Highest validity for the measurement of function was obtained by the BPS and the 6MWD followed by the SF-36 Physical functioning. The construct of the ODI and the MPI Interference with pain comprised pain and pain interference.
Implications: On the condition-specific instruments, namely the MPI interference with pain and the ODI (function), the simultaneous inclusion of function, pain and impairment blurs the construct of specific function content. This issue prevented identification of one favorite scale with the highest validity in chronic unspecific low back pain. In contrast, the generic SF-36 (physical functioning) showed highest functional specificity and construct overlap to the functional performance tests. Based on our data, we recommend the BPS and the 6MWD for longitudinal measurements of physical function in chronic low back pain.
Keywords: Chronic back pain, Functional Scales, Validity
Funding acknowledgements: The present study was further supported by the Zurzach Rehabilitation Foundation SPA, Bad Zurzach, Switzerland.
Purpose: The aim of this study was to examine cross-sectional and longitudinal validity of function related measurements in patients with chronic unspecific low back pain.
Methods: This prospective cohort study assessed 142 patients at baseline (before) and after a specific, 4 week interdisciplinary pain management program by the Short Form 36 (SF-36), the Multidimensional Pain Inventory (MPI) and Oswestry Disability Index (ODI). All scores were transformed into a scale from 0 (= worst function, maximal pain, worst health) to 100 (= best function, minimal pain, best health). Assessment was completed by 2 functional performance tests: Back Performance Scale (BPS) and 6-Minute Walking Distance (6MWD). Statistical analysis included bivariate correlations and factor analysis.
Results: Patients included in this study were on average of 47.6 years (+/-11.8) and 61.0% female. Correlations of baseline scores ranged from r=0.06 (BPS with MPI Social and away from home activities) to r=0.81 (SF-36 Mental health with MPI Negative mood). SF-36 Physical functioning correlated highest with functional performance tests: r=0.53 BPS and r=0.67 6MWD. Together with MPI Interference with pain, SF-36 Social functioning and mental health dimensions of SF-36 and MPI loaded most strongly on the first factor (37.4% explained variance). SF-36 Physical functioning, BPS and 6MWD loaded most strongly on the second factor (28.6% explained variance) at baseline. Correlations of change scores (difference of follow-up - baseline score) all were weaker, ranging from r=-0.02 (SF-36 Social functioning with BPS) to r=0.64 (SF-36 Mental Health with MPI Negative Mood). SF-36 Physical functioning correlated highest with MPI Interference with pain (r=0.41). Factor load analysis showed 3 factors with an explained variance of 25.3% (strongest load: MPI Pain severity and Interference with pain, ODI Function and SF-36 Bodily pain), 19.9% (SF-36 Social functioning and Mental health, MPI Negative mood) and 16.1% (BPS and 6MWD).
Conclusion(s): Overall, construct overlap of function related measurements was moderate to weak and the construct of physical function explained limited variance of state and change of health in chronic low back pain. State and change of function was strongest associated to social and psychological factors. Highest validity for the measurement of function was obtained by the BPS and the 6MWD followed by the SF-36 Physical functioning. The construct of the ODI and the MPI Interference with pain comprised pain and pain interference.
Implications: On the condition-specific instruments, namely the MPI interference with pain and the ODI (function), the simultaneous inclusion of function, pain and impairment blurs the construct of specific function content. This issue prevented identification of one favorite scale with the highest validity in chronic unspecific low back pain. In contrast, the generic SF-36 (physical functioning) showed highest functional specificity and construct overlap to the functional performance tests. Based on our data, we recommend the BPS and the 6MWD for longitudinal measurements of physical function in chronic low back pain.
Keywords: Chronic back pain, Functional Scales, Validity
Funding acknowledgements: The present study was further supported by the Zurzach Rehabilitation Foundation SPA, Bad Zurzach, Switzerland.
Topic: Outcome measurement; Musculoskeletal: spine; Pain & pain management
Ethics approval required: Yes
Institution: Health Department Aarau
Ethics committee: Local Ethic Commission
Ethics number: EK AG 2008/026
All authors, affiliations and abstracts have been published as submitted.
