We read with great interest the article by Fernández-García et al.1 recently published in Archivos de Bronconeumología on the assessment of dependence as a predictor of mortality following hospitalization for COPD exacerbations. The authors developed 3 scores for predicting mortality that included the variables age > 60 years, FEV1 < 50%, and Charlson Index (CI) ≥ 3. They were surprised to find that dependence, measured by the Barthel and Lawton and Brody indices are independent predictors of mortality that carry a greater weight than other conventional variables.

Despite the existence of different dependency scales2, few publications in the scientific literature allow us to correctly evaluate their weight in the morbidity and mortality of our patients3,4. Our group recently reported data from a cohort of 305 octogenarians with multiple diseases and a prevalence of COPD of 21.3%, in which different scales, including the OARS (Older Americans Resources and Services) questionnaire, were used to evaluate dependence in both basic activities of daily living (BADL) and instrumental activities of daily living (IADL)5. Using the age-adjusted Charlson index, 83.3% of our patients had a predicted 1-year mortality rate of 85%, compared with a real mortality rate of 57.7% at 18 months. We too were surprised by the power of dependence in terms of mortality prediction, and that the degree of dependence in the subgroup with a predicted 1-year mortality rate of 85% was higher for both BADL (6.6 ± 5 vs. 4.1 ± 4; p = 0.001) and IADL (9.7 ± 4.6 vs. 6.7 ± 4.8; p < 0.001).

Multiple factors influence the overall mortality of our patients. Dependence is a factor that can predict mortality more reliably than other conventional indices such as the Charlson Index. As we mentioned above, this index overestimates the risk of 1-year mortality and proves ineffective, while the assessment of dependence adds more weight to the prediction of short-term mortality. In a healthcare setting in which demand is progressively increasing, we need better tools to predict mortality and to help us use resources rationally based on the benefit we can provide to our patients.