Modelling Mortality of patients with Acute Exacerbation of Chronic Obstructive Pulmonary disease

Background/aim: Dyspnea, Eosinopenia, Consolidation, Acidemia and Atrial Fibrillation (DECAF) score is considered better predictor of mortality than other predictive tools. Mortality models are fitted to determine the better prediction of mortality in patients with acute exacerbation of chronic obstructive pulmonary disease (COPD) having DECAF score of 1-6 and within 7 days of hospital admission keeping actual mortality as gold standard. The data is collected from Chest Department, Fauji Foundation Hospital, Rawalpindi, Pakistan; 15 may 2017 to 15 Nov. 2019. Materials and methods: 146 Patients with COPD were selected from emergency and OPD of Fauji Foundation Hospital Rawalpindi, Pakistan. The procedure began after taking informed consent of the patients. Clinical features documented and investigations were carried out. The expense of all the tests was on hospital administration and not on the patient. A specially designed questionnaire was used for data collection. Results: Total 146 patients were included according to the inclusion criteria of study. The mean age (years) of patients is 64.90+0.93. Patients with DECAF score of 1 to 3 are 64%, the cases with DECAF score of 4 are 26.7% and the cases with DECAF score of 5 and 6 are 8.9%. The two fitted binary logistic regression models indicate that among the independent variables DECAF score and duration of smoking significantly contribute to mortality. The accuracy of the fitted model is 96.6% and 90.2% respectively and the values of sensitivity, specificity, positive and negative predictive values (i.e PPV and NPV) are found to be appropriate for fitted model. Conclusion: The study concludes that DECAF score and duration of smoking affect the mortality in patients admitted with acute exacerbation of COPD. Patients admitted in hospital with high DECAF score of and with more years of smoking should be admitted in intensive care unit because they may require invasive ventilation due to respiratory failure and high mortality. Keywords: Mortality, AECOPD, Risk ratios, Logistic regression, Model accuracy