Abstract
Background
Associations between health services utilization (HSU) and objectively measured airway inflammation in severe asthma or chronic obstructive pulmonary disease (COPD) patients remain unclear.Research question
Is there an association between quantitative sputum cell counts and HSU (hospitalizations, physician visits, emergency department visits) patterns in patients with asthma and COPD?Study design and methods
In this retrospective cohort study, clinical data, spirometry, and sputum cell counts (% of eosinophils, neutrophils) obtained during routine clinical assessment of patients attending a university tertiary airway disease clinic were linked to Ontario health administrative databases (2006-2019). Sputum inflammatory cell counts were treated as fixed baseline exposure variables and negative binomial regression models using generalized estimating equations estimated RRs and 95% CIs for subsequent HSU adjusting for demographic, socioeconomic, and clinical covariates.Results
Of the 4821 patients included, 3714(77%) had asthma only and 1107(23%) had COPD only. Among patients with asthma, sputum eosinophilia (>3%) was associated with higher asthma-specific HSU. Compared with patients with sputum eosinophils 15% eosinophil groups had similarly increased rates of asthma-specific emergency department visits (RR=1.92, 95%CI:1.49-2.47 and RR=1.88, 95%CI:1.17-3.03). Conversely, sputum eosinophil counts were not significantly associated with HSU in COPD, whereas high sputum neutrophil counts were associated with increased all-cause hospitalizations (RR=1.71, 95%CI:1.11-2.64). Overall, those with higher eosinophil (>15%) and lower neutrophil counts (Interpretation
Sputum eosinophilia (>3%) is associated with increased asthma-related HSU, whereas sputum eosinophils were not associated with increased HSU in COPD. These findings support the clinical relevance of airway inflammometry in asthma and highlight important differences in inflammatory phenotypes between asthma and COPD. Further studies are needed to better understand the mechanisms underlying these associations.