Abstract
BACKGROUND: Although it is known that a history of acute exacerbation of chronic obstructive pulmonary disease (AECOPD) is the most important predictor of future risk of AECOPD and readmission to hospitals, there is no scientific evidence that an event of only one chronic obstructive pulmonary disease (COPD)-related admission is a high risk of future readmission. We retrospectively analyzed the association of an incident of one COPD-related admission with future readmission risk. METHODS: This is a retrospective study. The data of AECOPD-related admissions and readmissions during 5 years were obtained and analyzed for the admission frequencies of patients with AECOPD and an association of the admission history with a future readmission risk. RESULTS: The admission frequency of patients with frequent admission (3 or more admissions within 5 years) was 4.1 times that of those with non-frequent admission (0.95 vs. 0.23 times per person per year). In each calendar year during the 5-year study period, most patients (88.2%) were hospitalized only once, and 11.8% had two or more admissions. Nevertheless, their average number of admissions in each calendar year was 3.3 times that of those who only had one admission each year (3.33 vs. 1.00 times per person per year). More importantly, the positive predictive value for future readmission due to AECOPD was only 14.8% in those who had one admission in the previous year. The patients with the greater readmission risk were those with two or more admissions due to AECOPD in the previous year [crude odds ratio (OR): 4.10, 95% confidence interval (CI): 1.24-13.58 and 7.51, 95% CI: 3.81-16.68]. CONCLUSIONS: There is a subtype of frequent admission due to AECOPD, and it can be distinguished by having three or more admissions in the past 5 years or two or more admissions in the previous year. Nevertheless, an incident of admission once a year is not a good predictor of future readmission.