161 6 Cochrane review on integrated disease management for COPD 2006; Zwar 2016). All but two trials randomly assigned participants to either IDM or usual care. The other two trials had two di erent intervention groups and one usual care group (Vasilopoulou 2017; Wijkstra 1994). We included both intervention groups as separate comparisons and split the usual care group in half. A description of the included studies is provided in Table 1, Table 2 , and Characteristics of included studies. Participants A total of 21,086 COPD patients were randomised in the 52 studies, with a range of 29 to 8171 patients per study. Of these, 16,390 (84%) patients completed the studies (range 23% to 100%). At the moment of inclusion, the mean age of the intervention population was 67.1 years (SD 9.27), with 65% male (range 25% to 99%). In the usual care group, mean age was 67.2 years (SD 9.26) and 67% (range 30 to 100%) were male. Interventions Patients were treated in all types of healthcare settings: primary care (15 studies), secondary care (22 studies), tertiary care (5 studies), and a combination of primary and secondary health care (10 studies). The numbers of healthcare professionals involved ranged from 2 to 7, with a mean number of 3. The number of components per programme ranged from 2 to 8, with a mean number of 4. Interventions also varied in terms of duration - between 3 and 48 months- with varying intensity of separate intervention components. Some interventions consisted of a clearly de ned intensive intervention period and a subsequent maintenance or structural follow-up period (Bourbeau 2003; Fan 2012; Gottlieb 2011; Güell 2000; Jimenez-Reguera 2020; Ko 2016; Sridhar 2008; van Wetering 2010; Vasilopoulou 2017). One study had an intervention with a variable duration of 2 years minimum and 5 years maximum (Kalter-Leibovici 2018). Following the subgroup analysis performed in the previous version of this review, we determined the dominant component of the IDM programme from all newly included studies. The dominant component could be determined directly from the objective or title of the study for eight studies (Aboumatar 2019; Bernocchi 2017; Fan 2012; Haesum 2012; Kruis 2014; Öztürk 2020; Vasilopoulou 2017; Zwar 2016). For the remaining 18 studies, we contacted study authors to ask what they considered the dominant intervention component. Eleven study authors did not provide a response. Of the seven who responded, three indicated that the intervention did not have a dominant component. To perform a subgroup analysis on types of interventions, we chose the dominant component as the component with the greatest intensity in terms of duration. Given the increased use of telemonitoring and its distinguished features to monitor patients from a distance, we decided to include telemonitoring as a separate dominant component. In Vasilopoulou 2017, usual care was compared to two types of interventions: home-based and hospital-based pulmonary rehabilitation. As interventions were characterised by di erent dominant
RkJQdWJsaXNoZXIy MTk4NDMw