Similarly, one might infer, based on Table 5,8 that diabetes was “protective” for rehospitalizations and that “complicated diabetes” provided even more benefit. This inference might be correct specific to “COPD-specific” hospitalizations but probably not to all-cause hospitalizations or mortality. Furthermore, it is also possible that a higher proportion of patients with “diabetes” or complicated diabetes died during the index hospitalization, providing a population biased toward survivors. Overall, is the presence of diabetes a risk factor for COPD hospitalizations? Based on the prevalence of comorbid diabetes (17.8%) and complicated diabetes (1.6%) in the studied population, which seem higher than the level one might expect in a similar, unhospitalized population, the answer is probably yes. Of course, the best way to evaluate this is to look at a population-based cohort and look for COPD hospitalizations in follow-up among those with and without diabetes.