Evaluation of the effects of the COPDnet comprehensive care model; an interim analysis
Nieuwe ontwikkelingen in behandeling en zorg (Nederlands)Introduction
We developed an evidence-based, comprehensive care model for patients with COPD referred to secondary care, i.e. the COPDnet model (Koolen et al., Int J COPD 2018; 13: 2225-35). The COPDnet model has been implemented in two hospitals and affiliated primary care region in the Netherlands. The value of the COPDnet model is currently investigated with a mixed-model design (Koolen et al., Int J COPD 2018; 2237-44). Primary outcome measure is the pre- to postintervention change in health status of patients. An interim analysis on the primary outcome was performed and results presented here.
The Clinical COPD Questionnaire (CCQ) was used to compare health status at baseline (T0), that is at the time of referral to secondary care, and 8 months later (T1) assuming that 8 months would be a sufficient period of time to obtain any effect of non-pharmacological interventions. An a priori power analysis revealed that 199 patients would be required to detect a small effect size (0.2) on the CCQ.
For the interim analysis data from 73 COPD patients were available: 54% male, age 63±8, FEV1 1.69±0.72 L. (56±20% pred.), CCQ total score baseline 1.88±1.05 points. At T1 the CCQ total score was 0.40± 0.73 points lower (i.e. better) compared to T0 (95% confidence interval 0.56 till 0.22; P<0.001). Cohens' d effect size was 0.44. In addition, 45% of the patients had a ≥ 0.4 (i.e. the minimal clinically important difference) reduction in CCQ score.
The interim analysis revealed a statistically significant change in health status at T1 compared to T0 and an effect size more than twice as large as assumed in the a pariori sample size calculation approaching a medium effect.