Responsiveness of outcome measurements in rehabilitation of patients with posterior pelvic pain since pregnancy.

摘要

STUDY DESIGN: A cohort study was conducted. OBJECTIVE: To develop a test battery for evaluating the course of posterior pelvic pain since pregnancy. SUMMARY OF BACKGROUND DATA: Properly validated scales to evaluate the course of posterior pelvic pain since pregnancy are scarce. Moreover, the use of many tests would be too strenuous for the patient and has an unfavorable cost-benefit ratio. METHODS: The ability of 48 effect measures to detect clinically relevant changes over time (responsiveness) was tested in patients with posterior pelvic pain since pregnancy. In this test, 35 measures were evaluated in a group of 44 patients, and 16 measures in a group of 56 patients (three measures were evaluated in both groups). All the tests were performed at baseline and after 8 weeks treatment. A global impression of improvement (improved or not improved) scored by the patient was used as the standard for assessing the course of the disease. Responsiveness was examined by calculating the standardized response mean of the improved patients and by using a two-tailed Mann-Whitney nonparametric test to compare the patients who had improved and those who had not improved. RESULTS: Of the 48 effect measures, 26 measures of five categories (activities of daily living, pain, hip muscle strength, spine mobility, and spine muscle strength) showed good correlation with the patient's global impression of improvement. The measures in the "mobility of the pelvic joints" category were insufficient for assessing clinical change in posterior pelvic pain since pregnancy. The measures in the "fatigue" and "pain provocation tests" categories correlated only moderately with clinical change. CONCLUSIONS: It seems possible to gain appropriate information about the course of posterior pelvic pain since pregnancy with a small test battery. The usefulness of the Quebec Back Pain Disability Scale, the hip adduction strength assessment, and the active straight-leg-raise test was proved by the current study. The value of 23 other instruments was substantiated, but further studies are needed to confirm their usefulness. The correlation of 22 evaluated measures with the patient's global improvement was too weak for them to be recommended as measures of clinical changes over time in posterior pelvic pain since pregnancy. It is recommended that clinicians and investigators compile a small test battery by selecting the best representatives of the five measurement categories that have good correlation with the patient's global impression of improvement.