Arnulf Langhammer

Background and objective Chronic obstructive pulmonary disease (COPD) is a heterogeneous disease often viewed as part of a multimorbidity complex. There is a need for better phenotyping of the disease, characterization of its interplay with other comorbidities and its association with long‐term outcomes. This study aims to examine how clusters of comorbidities are associated with severe exacerbations and mortality in COPD. Methods Participants with potential COPD were recruited from the second (1995–1997) and third (2006–2008) survey of the HUNT Study and followed up until April 2020. Ten objectively identified comorbidities were clustered using self‐organizing maps. Severe COPD exacerbations requiring hospitalization were assessed using hospital data. All‐cause mortality was collected from national registries. Multivariable Cox regression was used to calculate hazard ratios (HRs) with 95% CIs for the association between comorbidity clusters and all‐cause mortality. Poisson regression was used to calculate incidence rate ratios (IRRs) with 95% CI for the cumulative number of severe exacerbations for each cluster. Results Five distinct clusters were identified, including ‘less comorbidity’, ‘psychological’, ‘cardiovascular’, ‘metabolic’ and ‘cachectic’ clusters. Using the less comorbidity cluster as reference, the psychological and cachectic clusters were associated with all‐cause mortality (HR 1.23 [1.04–1.45] and HR 1.83 [1.52–2.20], adjusted for age and sex). The same clusters also had increased risk of exacerbations (unadjusted IRR of 1.24 [95% CI 1.04–1.48] and 1.50 [95% CI 1.23–1.83], respectively). Conclusion During 25 years of follow‐up, individuals in the psychological and cachectic clusters had increased mortality. Furthermore, these clusters were associated with increased risk of severe COPD exacerbations.