Supplement prediction model 131 Discussion This supplement examined the potential causal predictors of exercise adherence in patients with chronic obstructive pulmonary disease following pulmonary rehabilitation in a primary physiotherapy practice. The results showed that the strongest predictors of exercise adherence were education, MRC-score, intention, depression, and alliance. Hierarchic regression analyses indicated that attitude, and PBC, explained 34.5% of variance in exercise intention during pulmonary rehabilitation, with attitude, and PBC each making significant unique contributions to intention. Subjective norm did not had a significant influence on exercise intention, which is consistent with general exercise literature [13], and previous research in phase II cardiac rehabilitation [14]. The explained variance of 34.5% is slightly less than the explained variance in exercise intention of 41% found in the general literature [15]. This may be due to the fact that this cohort of patients has an average of ten years of COPD, and had been following PR for an extended period of time (70% of patients followed PR for ³ 1 year). The TPB has demonstrated better predictions with shorter time intervals, indicating a possible reduced predictive power of the model with this long-term prediction [1]. Furthermore, exercise intention explained 16.9% of the variance in exercise adherence. This is less compared to the 22.5% explained variance in exercise adherence found in the general literature [15]. In health domain, intention should be the most important variable for explaining the health behavior; 66.2% of the explained variance should be attributed to intention (22.5% of the 34.0% explained variance in behavior) [15]. In this analysis, intention explained 44.9% (16.9% of 37.6%) of exercise adherence. Furthermore, PBC did not add any additional variance in exercise adherence, as this has been found in the general literature [15]. This is in accordance with results found in phase II cardiac rehabilitation [14]. The data suggest that PBC was related to the initial development of the intention, but not anymore to the actual execution of the behavior in these patients who already engage in the behavior for a longer time. Finally, alliance, depression, MRC-score, and education each made significant unique contributions to exercise adherence scores. These results may indicate that the TPB is not very adequate in explaining the maintenance of behavior over a longer time. On the other hand, in interpreting these data it is essential to acknowledge the importance of the operationalizations that were used. Although the typical TPB format of questions was generally applied in our measures, the scales were brief and possibly not assessing all relevant aspects of the TPB determinants in this particular sample of patients. Further research is needed that includes patients who are just diagnosed with COPD or just started pulmonary rehabilitation, and more elaborate measures might be developed to further explore these findings.
RkJQdWJsaXNoZXIy MjY0ODMw