145 6 Cochrane review on integrated disease management for COPD (mean di erence (MD) -3.89, 95% con dence interval (CI) -6.16 to -1.63; 18 RCTs, 4321 participants; moderate-certainty evidence). A comparable e ect was observed at shortterm follow-up (MD -3.78, 95% CI -6.29 to -1.28; 16 RCTs, 1788 participants). However, the common e ect did not exceed the minimum clinically important di erence (MCID) of 4 points. There was no signi cant di erence between IDM and control for long-term follow-up and for generic QoL. IDM probably also leads to a large improvement in maximum and functional exercise capacity, as measured by six-minute walking distance (6MWD), at medium-term followup (MD 44.69, 95% CI 24.01 to 65.37; 13 studies, 2071 participants; moderate-certainty evidence). The e ect exceeded the MCID of 35 metres and was even greater at short- term (MD 52.26, 95% CI 32.39 to 72.74; 17 RCTs, 1390 participants) and long-term (MD 48.83, 95% CI 16.37 to 80.49; 6 RCTs, 7288 participants) follow-up. The number of participants with respiratory-related admissions was reduced from 324 per 1000 participants in the control group to 235 per 1000 participants in the IDM group (odds ratio (OR) 0.64, 95% CI 0.50 to 0.81; 15 RCTs, median follow-up 12 months, 4207 participants; high-certainty evidence). Likewise, IDM probably results in a reduction in emergency department (ED) visits (OR 0.69, 95%CI 0.50 to 0.93; 9 RCTs, median follow-up 12 months, 8791 participants; moderate-certainty evidence), a slight reduction in all-cause hospital admissions (OR 0.75, 95%CI 0.57 to 0.98; 10 RCTs, median follow-up 12 months, 9030 participants; moderate-certainty evidence), and fewer hospital days per person admitted (MD -2.27, 95% CI -3.98 to -0.56; 14 RCTs, median follow-up 12 months, 3563 participants; moderate-certainty evidence). Statistically signi cant improvement was noted on the Medical Research Council (MRC) Dyspnoea Scale at short- and medium-term follow-up but not at long-term follow-up. No di erences between groups were reported for mortality, courses of antibiotics/ prednisolone, dyspnoea, and depression and anxiety scores. Subgroup analysis of dominant intervention components and regions of study suggested context- and intervention-speci c e ects. However, some subgroup analyses were marked by considerable heterogeneity or included few studies. These results should therefore be interpreted with caution. Authors’ conclusions This review shows that IDM probably results in improvement in disease-speci c QoL, exercise capacity, hospital admissions, and hospital days per person. Future research should evaluate which combination of IDM components and which intervention duration are most e ective for IDM programmes, and should consider contextual determinants of implementation and treatment e ect, including process-related outcomes, long-term follow-up, and cost-e ectiveness analyses.
RkJQdWJsaXNoZXIy MTk4NDMw