C35-18 Evaluating COPD in an Urban Safety Net Hospital: Concordance Between Clinical Diagnosis and Spirometry Results

Abstract Rationale Chronic obstructive pulmonary disease (COPD) is a leading cause of morbidity and mortality worldwide. Accurate diagnosis is essential for appropriate treatment and effective disease management. However, in low-resource settings, factors such as anchoring bias may contribute to misclassification. This study aimed to assess the concordance between ICD-based COPD diagnoses and spirometry-confirmed obstruction in an urban safety-net hospital. We hypothesized that many patients with an ICD-10 COPD diagnosis will not have spirometric evidence of obstruction, indicating low concordance between clinical diagnosis and objective testing. Methods We performed a retrospective chart review of patients referred for consideration to a pulmonary rehabilitation study who had an ICD-10 COPD diagnosis seen at Zuckerberg San Francisco General Hospital from January 1, 2017 to October 1, 2025. Demographics, spirometry results, insurance type, smoking status, and prescribed respiratory medications were extracted from the electronic health record. Patients clinically diagnosed with COPD were identified using ICD-10 codes (J40-J44) and spirometric confirmation of COPD was defined as a post-bronchodilator FEV1/FVC ratio 0.70 when available. We calculated the proportion of patients meeting spirometry criteria, described characteristics of those without confirmed obstruction, and summarized medication prescribing patterns. Results To date, we reviewed 298 patients with active ICD-10 COPD codes. This cohort was predominantly male (53%) and Black (49%). 152 (51%) patients did not have spirometry within the past two years, and there was a significant difference in race/ethnic composition between those who did and did not have spirometry. Among those who had spirometry available (n = 146), 76 (52%) did not meet criteria for obstruction. Smoking data was available for 293 patients, of whom 45% were current smokers. Among non-current smokers (n = 160), 79% reported prior smoking. Patients with pack-year data (n = 78) had a median smoking history of 25 pack-years. 30% of Medicare patients (65+) did not have spirometric confirmation for COPD compared to 25% of Medicaid patients. Among those who did not meet spirometric criteria for obstruction, 51 (67%) patients were prescribed LAMA, LABA, or ICS therapy. Conclusion Over half of our reviewed patients with an active COPD diagnosis in our urban safety-net hospital lacked spirometry within the preceding two years. Among those who did have available spirometry, many received medications despite not meeting diagnostic criteria for COPD. Our findings highlight significant challenges in accurately diagnosing COPD among patients from low-resourced communities. Targeted interventions are needed to improve adherence to guideline-based diagnostic practices to ensure accurate COPD diagnosis and optimal treatment. This abstract is funded by: NHLBI