Baseline and Follow-Up Scores for General Health–Related Quality of Life and Adjusted and Unadjusted Analysis of Differences Between Patients From Experimental and Control GP Offices
| SF-36 Concept Scalea | Experimental | Control | Primary Analysise | Secondary Analysisf | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Baseline (n = 258)b | Follow-Up (n = 163)c | Mean Difference (n = 131)d | Baseline (n = 286)b | Follow-Up (n = 231)c | Mean Difference (n = 181)d | Unadjusted P Valueg | Adjusted P Valueh | Unadjusted P Valueg | Adjusted P Valueh | |
| Physical functioning | 77.2 (22.8) | 75.1 (27.3) | 0.1 (16.6) | 72.9 (23.9) | 74.5 (24.4) | –1.4 (17.8) | .425 | .541 | .633 | .779 |
| Social functioning | 81.2 (20.8) | 81.3 (24.4) | –0.2 (22.5) | 81.0 (22.6) | 82.0 (20.5) | –2.2 (21.3) | .496 | .518 | .844 | .865 |
| Role limitations due to physical health problems | 70.6 (39.9) | 70.1 (40.2) | 3.4 (34.5) | 66.3 (40.5) | 70.8 (39.4) | –0.8 (37.0) | .371 | .465 | .458 | .566 |
| Role limitations due to emotional problems | 78.9 (35.5) | 80.2 (35.6) | 3.6 (39.3) | 76.6 (37.3) | 79.4 (34.6) | –0.2 (40.4) | .376 | .376 | .34 | .34 |
| General mental health | 77.2 (17.8) | 79.7 (17.6) | 1.4 (12.9) | 75.0 (19.1) | 77.9 (17.8) | 1.4 (14.8) | .676 | .676 | .959 | .959 |
| Vitality | 65.2 (19.1) | 66.2 (19.1) | 1.3 (14.6) | 64.5 (20.4) | 65.3 (19.7) | 0.03 (16.1) | .509 | .508 | .715 | .887 |
| Bodily pain | 74.7 (23.9) | 76.4 (24.3) | 2.1 (23.3) | 74.1 (24.7) | 76.2 (24.7) | 0.02 (23.1) | .513 | .513 | .669 | .669 |
| General health perceptions | 59.6 (17.5) | 59.9 (18.5) | 0.8 (15.9) | 57.5 (20.3) | 56.9 (19.1) | –1.8 (18.7) | .084 | .084 | .272 | .272 |
| Changes in health over time | 51.8 (19.7) | 51.4 (19.5) | 1.0 (24.7) | 53.4 (20.1) | 49.0 (19.3) | –5.9 (23.3) | .023 | .033 | .035 | .067 |
ANCOVA = analysis of covariance; GP = general practitioner; QoL = quality of life; SF-36 = 36-item Short Form Health Survey.
Note: Data are presented as mean (SD) for general health–related QoL scores (range 0-100).
↵a Concept scales according to Ware Jr and Sherbourne26 and Ware Jr.27
↵b These mean values were calculated using all completed questionnaires at baseline (n = 258 for the experimental group, n = 286 for the control group).
↵c These mean values were calculated using all completed questionnaires at follow-up (n = 163 for the experimental group, n = 231 for the control group).
↵d These mean differences were calculated using both completed baseline and follow-up questionnaires (n = 131 for the experimental group, n = 181 for the control group).
↵e Analysis of differences in general health–related QoL between the experimental and control group, including all GP offices in the analysis.
↵f Analysis of differences in general health–related QoL between the experimental and control group, excluding patients from GP offices with follow-up <60%.
↵g The unadjusted P values were calculated with ANCOVA, using baseline general health–related QoL scores as covariate.
↵h The adjusted P values were calculated with linear mixed-model analysis using baseline general health–related QoL scores and group allocation (experimental/control) as fixed factor and office number as random factor.