187 6 Cochrane review on integrated disease management for COPD Future research should shed light on the bene cial e ects of telemonitoring and its use in practice. Also, the applicability of evidence depends on the healthcare context in which the IDM programme is implemented, which di ered greatly among studies included in this review. Studies were conducted in many di erent countries across ve di erent continents. Subgroup analysis pointed towards a context-speci c e ect. This is in line with recent ndings from the COMET study performed in Germany, France, Italy, and Spain (Kessler 2018), which reported signi cant country-speci c di erences between study settings. Kessler attributed these to di erences in routine care, such as countryspeci c di erences in baseline hospitalisation practices, admission criteria, and bed availability. Hence, e ectiveness varying between study regions is likely related to variations in usual care that occur over time and are driven by national changes in policy and healthcare nancing. Also, country-speci c di erences in terms of cultural and societal norms may play a role in terms of implementation delity and therefore outcomes (Marsiglia 2015). For example, the four-year study in which the IDM group received a monthly one-hour health lecture, performed by Lou 2015 in China, reported dropout (other than death) of only 7%, and noted that 87% of the study population attended all 48 COPD lectures. Furthermore, the period in which included studies were published spanned 30 years, with the earliest published in 1991 and the latest in 2020. The clinical applicability of more recent studies is larger, given the embedding of IDM programmes into the healthcare system and the evolution of healthcare systems nationally and internationally. Hence, it would be worthwhile to investigate the relationship between advancements in usual healthcare over time and additional bene cial e ects of IDM. Furthermore, it would be interesting to explore ways in which more weight could be given to more recent studies or older studies with limited applicability for current health care could be left out in a legitimate way. Quality of the evidence There was clinical and methodological heterogeneity among studies, which likely results (at least in part) from the complexity of IDM interventions. We have incorporated heterogeneity into estimated e ects by using random-e ects analyses. Using the GRADE approach, we speci ed levels of quality of the evidence (high, moderate, low, and very low) in our ‘Summary of ndings’ table. According to this approach, we checked whether included trials had limitations in terms of design, indirectness of evidence, unexplained heterogeneity or inconsistency of results, imprecision of results, or high probability of publication bias. Such limitations may impact the certainty of evidence for all outcomes that are relevant to guideline formation, health policy development, and clinical guidance. We deemed the quality of evidence for HRQoL (as measured by the SGRQ) as moderate, and we observed a consistent e ect in favour of the intervention group for all SGRQ domains at medium-term follow-up. We downgraded the quality of evidence due to
RkJQdWJsaXNoZXIy MTk4NDMw