Indicators of mortality for patients treated in a nurse-led COPD unit

Background: COPD accounts for around 30,000 deaths per year in the UK (Nacul LC et al. Popul Health Metr. 2007; 5:8), with the highest risk seen in the 12 months following an exacerbation. A dedicated nurse-led COPD service at our hospital works to assess, treat, facilitate discharge and follow up patients following acute exacerbation of COPD. It receives referrals from wards, community healthcare teams and patients themselves.Aims: To ascertain risk factors for mortality in patients managed by our COPD unit in the ensuing 12 months after presentation.Methods: Data on 100 consecutive patients seen from January 2014 was assessed for factors which may contribute to disease progression, such as BMI, smoking status, deprivation index, comorbidity, sputum culture, and blood gas analysis. We tested these variables as risk factors for mortality using a binomial logistic regression analysis.Results: The 12 month mortality was 20% in our cohort, with ward-referral patients 2.3 times more likely than self-referrals to die (p=0.179). Of the other factors, patients with hypercapnoea (odds ratio 2.407, p=0.019, 95% CI 1.157 to 5.008) and sputum non-producers (odds ratio 5.938, p=0.043, 95% CI 1.057 to 33.350) had significantly higher mortality rates.Conclusion: Patients requiring hospitalization had a higher mortality rate, perhaps indicating more severe disease. In keeping with the literature, patients with hypercapnoea and sputum non-producers (emphysema phenotypes) are at higher risk of mortality(Groenwegen et al. Chest 2003;124:459-67). These factors should not only alert clinicians to patients requiring more intensive management and monitoring but also for whom advance care planning may be appropriate.