Abstract Rationale Health-related social needs (HRSN) are modifiable social factors associated with worse outcomes in asthma and chronic obstructive pulmonary disease (COPD). Among prominent barriers to quantifying these effects is the concern that people screened will differ systematically from those not screened (i.e., selection effects). Characterizing HRSN screening penetration is critical to health systems’ decisions to maintain or expand screening programs and for making HRSN causal effect inferences. Thus, we evaluated differences in patient characteristics between screened and non-screened asthma and COPD patients from a national US health system. Methods We performed a retrospective cohort study of asthma and COPD patients with an HRSN screening-eligible encounter within Trinity Health from January 2020-November 2024 (i.e., outpatient primary care; family, geriatric, and internal medicine; and obstetrics and gynecology encounters; and inpatient hospital admissions). Asthma and COPD were defined by active problem list ICD-10 codes J45 and J44, respectively, for screen-eligible encounters. To evaluate for selection bias, we compared unadjusted rates of screened vs non-screened patients by age, sex, race, ethnicity, preferred language, insurance type, rurality, social deprivation index quartile, and Elixhauser comorbidity index. We summarized between-group differences using standardized mean differences (SMD; |SMD|0.25 was considered imbalanced). Due to different screening workflows, we stratified comparisons by care setting. We analyzed asthma and COPD cohorts separately. Results Among 146,630 asthma patients, 96,500 (66.1%) completed HRSN screening, with higher outpatient uptake (76.5% vs 20.9%). Outpatient and inpatient screening rates were similar between marginalized and referent groups. For example, Black vs White (outpatient 70.4% vs 77%; inpatient 21.0% vs 20.9%) and Medicaid vs commercial insurance (outpatient 70.1% vs 78.8%; inpatient 20.4% vs 20.5%), with no subgroup comparisons differing significantly in either setting (all SMD≤0.25). Among 79,723 COPD patients, 39,505 (50.1%) completed HRSN screening with higher outpatient uptake (70.1% vs 23.1%). Again, outpatient and inpatient screening rates were similar between marginalized and referent groups. For example, Black vs White (outpatient 66.6% vs 71.3%; inpatient 24.8% vs 22.8%) and Medicaid vs commercial insurance (outpatient 72.5% vs 75.8%; inpatient 23.9% vs 22.1%), with no subgroup comparisons differing significantly in either setting (all SMD≤0.25) (Table). Conclusions Among asthma and COPD patients eligible for HRSN screening in a national health system, screened and non-screened patients were similar across measured characteristics within outpatient and inpatient care settings, supporting the use of these data for system-level inferences. This abstract is funded by: T32 HL098054
Patel et al. (Fri,) studied this question.