CORRECTION to: “Global mortality and readmission rates following COPD exacerbation-related hospitalisation: a meta-analysis of 65 945 individual patients” (ERJ Open Res 2024; 10: 00838-2023)
Waeijen-Smit K., Crutsen M., Keene S., Miravitlles M., Crisafulli E., Torres A., Mueller C., Schuetz P., Ringbæk TJ., Fabbian F., Mekov E., Harries TH., Lun CT., Ergan B., Esteban C., Quintana Lopez JM., López-Campos JL., Chang CL., Hancox RJ., Shafuddin E., Ellis H., Janson C., Ulrik CS., Gudmundsson G., Epstein D., Dominguez J., Lacoma A., Osadnik C., Alia I., Spannella F., Karakurt Z., Mehravaran H., Utens C., De Kruif MD., San Ko FW., Trethewey SP., Turner AM., Bumbacea D., Murphy PB., Vermeersch K., Zilberman-Itskovich S., Steer J., Echevarria C., Bourke SC., Lane N., Batlle JD., Sprooten RTM., Russell R., Faverio P., Cross JL., Prins HJ., Spruit MA., Simons SO., Houben-Wilke S., Franssen FME.
In the originally published version of this article there was an error in the follow-up time and total number of readmissions in two datasets of the hospital readmission data subset. This concerned the datasets of QUINTANA et al. [1] and LOPEZ-CAMPOS et al. [2]. Instead of a follow-up time of 365 days, the correct follow-up times were 60 and 90 days, respectively. In addition, instead of total readmission rates of 19.5% and 26.6%, the correct readmission rates were 26.0% and 35.1%, respectively. As a result, the median follow-up time in the hospital readmission data subset changed from 365 days to 90 days, whereas the overall readmission rate changed from 15 195 (32.8%, 95% CI 32.4–33.3%) to 16 646 (36.0%, 95% CI 35.5–36.4%). Corrections have been made accordingly to the Results, including table S7 and figure S8, and to the Discussion. In addition, there was an error in the coding for the 30-, 90- and 365-day categories related to post-discharge mortality and hospital readmission. The initial coding failed to capture all possible conditional statements needed to accurately capture outcome statuses based on follow-up times. This was particularly relevant in studies where the exact time until the event was not known, and was instead represented by the predetermined, or set, study follow-up period. These corrections altered the pooled 30-, 90- and 365-day post-discharge mortality and hospital readmission rates from 1.8% to 2.0%, from 5.5% to 6.4%, from 10.9% to 12.2%, and from 7.1% to 11.8%, from 12.6% to 26.5% and from 32.1% to 38.2%, respectively. Corrections have been made accordingly to the Abstract, Results (including figures 5 and 7), and to the Discussion. Figures S3 and S7 have been aligned to display only the percentages of patients with a known time of event, categorised by time intervals during follow-up after hospital discharge from the index event. Importantly, the corrections do not change the scientific conclusions drawn in the article.