Introduction and Objectives Oral corticosteroids (OCS) are often used in patients with chronic obstructive pulmonary disease (COPD) experiencing exacerbations, but can result in short- and long-term adverse outcomes. Here, we evaluated associations between COPD-related OCS exposure and adverse outcome incidence.

Methods This observational, individually matched historical cohort study used electronic medical records (1987−2019) from the UK Clinical Practice Research Datalink with Hospital Episode Statistics (HES) linkage to compare patients with COPD who used OCS (OCS cohort) or never used OCS (non-OCS cohort) during the observational period. Patients had a COPD diagnosis or monitoring/review on or after 1 April 2003 (per introduction of the Quality and Outcomes Framework), and continuous practice records for ≥1 year prior to index date (OCS cohort: first COPD-related OCS prescription; non-OCS cohort: nearest primary care practice visit to matched-case index date). Cohorts were individually matched by index date, age, sex, HES linkage availability and smoking status. OCS exposure was measured from index date to incidence of each adverse outcome or end of observation. Multivariable Cox proportional hazard regressions, adjusting for confounders, assessed adverse outcome risk in the OCS versus non-OCS cohorts and within the OCS cohort by cumulative dose (<0.5 g, 0.5−<1.0 g, 1.0−<2.5 g, 2.5−<5.0 g, 5.0−<10.0 g, ≥10.0 g), treating cumulative dose as a time-varying measure.

Results In total, 106,775 patients received COPD-related OCS prescriptions; 53,299 pairs were matched in the OCS and non-OCS cohorts (mean ± SD age, 64.6±12.5 years; 59.8% male). The OCS cohort had significantly increased risk versus the non-OCS cohort for all adverse outcomes (figure 1), with the highest risk observed for pneumonia, osteoporosis with/without fractures, sleep disorders and anxiety/depression. Within the OCS cohort, cumulative OCS exposure ≥0.5 g was associated with increased hazard ratios (95% CIs) for multiple adverse outcomes, including osteoporosis with/without fractures (0.5−<1.0 g, 1.45 [1.30, 1.62]; 1.0−<2.5 g, 1.89 [1.70, 2.11]; 2.5−<5.0 g, 2.92 [2.58, 3.30]; 5.0−<10.0 g, 4.27 [3.74, 4.86]; ≥10.0 g, 6.39 [5.56, 7.35]) and type 2 diabetes mellitus (0.5−<1.0 g, 1.20 [1.10, 1.30]; 1.0−<2.5 g, 1.40 [1.29, 1.52]; 2.5−<5.0 g, 1.47 [1.32, 1.63]; 5.0−<10.0 g, 1.75 [1.54, 1.98]; ≥10.0 g, 2.23 [1.95, 2.54]).

• Download figure
• Open in new tab
• Download powerpoint

Abstract S88 Figure 1

HRs comparing adverse outcome risk for the OCS vs non-OCS cohorts (multivariable Cox regression)

Conclusions OCS use was associated with increased risk of adverse outcomes, with risk increasing as cumulative exposure increased.

Please refer to page A211 for declarations of interest related to this abstract.