(SLA-062) Adjusting for residual confounding using high-dimensional propensity scores in a study of inhaled corticosteroids and COVID-19 outcomes

Background: In pharmacoepidemiologic studies of COVID-19, there were concerns about bias from residual confounding. Data-driven approaches such as high-dimensional propensity scores (HDPS) aim to optimise confounding adjustment by considering recorded data as proxy variables.

Objectives: To apply HDPS to adjust for unmeasured confounding in a case study investigating the effect of inhaled corticosteroids (ICS) on COVID-19 hospitalisation and death.

Methods: We selected a cohort of patients with chronic obstructive pulmonary disease in the UK on 01 March 2020 from Clinical Practice Research Datalink (CPRD) Aurum, comparing ICS/long-acting β-agonist (LABA) (+/- long-acting muscarinic antagonist (LAMA)) and LABA/LAMA users. ICS effects on the outcomes COVID-19 hospitalisation and death were assessed through Cox proportional hazards models. HDPS were estimated from primary care clinical records, prescriptions and hospitalisations. SNOMED-CT codes and dictionary of medicines and devices codes from CPRD Aurum were mapped to International Classification of Disease 10th revision codes and British National Formulary paragraphs respectively. Using logistic regression models, we estimated propensity scores (PS) combining prespecified and HDPS covariates, selecting the top 100, 250, 500, 750 and 1000 covariates ranked by confounding potential. IPTWs were used to weight Cox regression models to estimate the impact of ICS on the outcomes.

Results: When excluding triple therapy users, the conventional PS-weighted estimates showed weak evidence of increased risk of COVID-19 hospitalisation among ICS users (HR 1.19 (95% CI 0.92-1.54)). Results varied slightly based on the number of covariates included in HDPS (HR using 100 HDPS covariates 1.01 (95% CI 0.76-1.33), HR using 250 HDPS covariates 1.24 (95% CI 0.83-1.87)).
For COVID-19 death, conventional PS-weighted models showed weak evidence of harm of ICS when excluding triple therapy users (HR 1.24 (95% CI 0.87-1.75)). HDPS-weighting moved estimates toward the null, suggesting no effect of ICS on COVID-19 outcomes (HR using 250 HDPS covariates excluding triple therapy 1.08 (95% CI 0.73- 1.59)).

Conclusions: This is the first study to implement HDPS in CPRD Aurum. HDPS may have provided better confounding control for COVID-19 deaths and may be able to partially compensate for suboptimal comparison groups. HDPS results can be sensitive to the number of covariates included, highlighting the importance of sensitivity analyses.