APSR2026-1087

 
Late-Breaking
Submitted
Oral Presentation
AO22-004
 
 
https://storage.unitedwebnetwork.com/files/1285/1245788-46903-APSR-1087_441008.pptx
Format: PPTX Only
 
Format: PDF Only
07. COPD
6

No more than 15 authors can be listed.
Please ensure the authors are listed in the right order.
If there are multiple affiliations, please use the "+" button next to "Select Country/Region" to add additional affiliation fields.

YIJUN CHEN * the First Affiliated hospital of Guangzhou Medical University Guangzhou Institute of Respiratory Health Guangzhou China Macau University of Science and Technology Faculty of Medicine Macao Macao
WENHUA JIAN - the First Affiliated hospital of Guangzhou Medical University Guangzhou Institute of Respiratory Health Guangzhou China
CHI HOU PANG - University of Macau Faculty of Law Macao Macao
MIN TANG - the First Affiliated hospital of Guangzhou Medical University Guangzhou Institute of Respiratory Health Guangzhou China
JINPING ZHENG - the First Affiliated hospital of Guangzhou Medical University Guangzhou Institute of Respiratory Health Guangzhou China
SOOKJA KIM CHUNG - Macau University of Science and Technology Faculty of Medicine Macao Macao
 
 
 
 
 
 
 
 
 
Presenting Author Details
YIJUN
CHEN
3240008971@student.must.edu.mo
+85363249746
31-40
 
Original Research
Main-diagnosis-only algorithms undercapture comorbidity in spirometry-confirmed COPD
Maximum 25 words
Abstract Content
Comorbidities are central to chronic obstructive pulmonary disease (COPD) assessment, but electronic-record algorithms often rely on main diagnoses. We evaluated whether main-diagnosis-only coding adequately captures baseline comorbidity in a spirometry-confirmed COPD cohort.
We conducted a retrospective study at the First Affiliated Hospital of Guangzhou Medical University. Adults with COPD diagnosis and post-bronchodilator FEV1/FVC <0.70 undergoing eligible spirometry from 2018 to 2021 were included. The index date was the first eligible spirometry date. Baseline diagnoses within 3 years before index were extracted from inpatient and outpatient electronic records and classified as main, secondary, or suspected diagnoses. For the ten most frequent non-COPD conditions, we calculated main-diagnosis capture rate and under-ascertainment proportion, compared diagnosis-position distributions using Pearson χ² tests, and fitted an entry-level logistic model for main-diagnosis capture, using lung cancer as reference.
Among 1,437 patients, the ten most frequent non-COPD conditions generated 20,704 diagnosis entries. Main-diagnosis-only coding captured 7,973 entries (38.5%) and would miss 12,731 (61.5%). Diagnosis-position distributions differed markedly across conditions (χ²=4,647.1; df=18; P<0.001). Main-diagnosis capture was high for lung cancer (81.2%) but low for hypertension (16.6%), atherosclerotic heart disease (24.5%), cor pulmonale (8.2%), benign prostatic hyperplasia (9.6%), and type 2 diabetes (19.1%). In the logistic model, odds of main-diagnosis capture were much lower for cor pulmonale (OR 0.02, 95% CI 0.02–0.03), benign prostatic hyperplasia (OR 0.02, 95% CI 0.02–0.03), hypertension (OR 0.05, 95% CI 0.04–0.05), and type 2 diabetes (OR 0.05, 95% CI 0.05–0.07) than for lung cancer.
In spirometry-confirmed COPD, secondary diagnoses were not peripheral coding noise but the main source for many extrapulmonary comorbidities. Main-diagnosis-only algorithms may therefore misclassify multimorbidity burden. Diagnosis-position-aware extraction should be considered when building COPD registries, comorbidity indices, and risk-adjustment datasets, and may help identify patients requiring integrated cardiopulmonary and multisystem assessment.
 
 
 
 
 
No
No
 
JPG or JPEG or PNG format ONLY
Chronic obstructive pulmonary disease, Comorbidity, Electronic medical records
maximum 3 keywords and no abbreviations, separated by commas
We acknowledge the First Affiliated Hospital of Guangzhou Medical University for support in de-identified data access, data management, and technical infrastructure.
323