Session IV Scientific Basis for Fracture Care


Saturday, September 28, 1996 Session IV, 10:16 a.m.

The Musculoskeletal Functional Assessment Instrument: Validity Comparisons with the SF-36

Marc F. Swiontkowski, MD, Julie Agel, MA, ATC, Ruth Engelberg, PhD, Diane Martin, PhD

Harborview Medical Center, Seattle, WA

Purpose: To assess direct comparisons of the Musculoskeletal Functional Assessment instrument (MFA) and the SF-36, in terms of construct validity in trauma patients.

Conclusions: In a population of injured patients, the MFA was shown to be more responsive than the SF-36.

Materials and Methods: The MFA instrument has been developed over the past 3 years. In the first field trial, the instrument's reliability and validity (content criterion and construct) were supported and published. In the second field trial, the MFA was compared with 2 generic health status instruments: the SF-36 and the Sickness Impact Profile (SIP). The MFA and the SF-36 were administered twice to a sample of 327 injured patients, 188 of whom had bone and soft tissue injuries to the lower extremity, 121 with bone or soft tissue injuries to the upper extremity, and 15 with both. These patients were selected from a Level One trauma center and 2 community orthopaedic clinics.

Results: Median scores, maximum scores, and minimum scores were similar for the MFA and the SF-36. The SF-36 scores are inverted for easier comparison, where the higher the score is, the greater the patient's functional disability. The score means and effect sizes are reported in Table 1 and 2.

An assessment of the effect size and relative efficiency analyses comparing the SF-36 with the MFA, showed that the MFA was more responsive than the SF-36. Particularly noteworthy was the performance in the upper extremity. When a subset of patients self-reported that they were "much better" at the time of the second survey than at the time of the first survey (n=57), the MFA effect size demonstrated more responsiveness (.21 to 1.09) than the SF-36 effect sizes (.10 to .198). For subscales that were similar to MFA domains (Table 3) the MFA was more sensitive for 8 of the 13 comparisons, while the SF-36 for only 1 of 13 comparisons.

Discussion: The MFA was demonstrated to be more sensitive and efficient than the SF-36 in this comparative study.

Table 1

 MFA Category  Time 1 Mean (sd)  Time 2 Mean (sd)  p   Time 1-Time 2 Mean (sd)  Effect Size  SRM
 Lower (n=188)  30.23 (16.58)  20.40 (16.74) .000  9.83 (12.60)  .56  .84
 Upper (n=121)  23.12 (16.95)  14.45 (15.59) .000  8.67 (10.91)  .50  .74
 Both (n=18)  43.73 (16.94)  29.81 (20.77)   .000  13.92 (9.90)  .80  1.19

Table 2

SF-36   Category  Time 1 Mean (sd)  Time 2 Mean (sd)  p   Time 1-Time 2 Mean (sd)  Effect Size  SRM
 Lower (n=188)  42.56 (28.16)  29.61 (25.54) .000  12.95 (24.15)  .44  .60
 Upper (n=121)  19.17 (22.87)  14.63 (22.59) .000  4.55 (16.06)  .16  .21
 Both (n=18)  64.17 (24.93)  35.83 (29.96)   .000  28.33 (24.37)  .97  1.31

Table 3

 SF-36 Categories  MFA Categories
 Physical Functioning Leisure/Recreation
Self-Care
Housework
Mobility
 Physical Role Function Employment/Work
Leisure/Recreation
Housework
Life Change Item
 Bodily Pain  Worst Pain Item Average Pain Item
 General Health  Emotional Adjustment, Coping, Adaptation
 Vitality  Emotional Adjustment, Coping, Adaptation Sleep/Rest
 Social Function Family Relationships Emotional Adjustment, Coping, Adaptation
 Emotional Role Function Family Relationships
Employment/Work
Emotional Adjustment, Coping, Adaptation
 Mental Health  Emotional Adjustment, Coping, Adaptation Family Relationships