Abstract
Little is known about using electronic medical records to identify patients with chronic obstructive pulmonary disease to improve quality of care. Our objective was to develop electronic medical record algorithms that can accurately identify patients with obstructive pulmonary disease. A retrospective chart abstraction study was conducted on data from the Electronic Medical Record Administrative data Linked Database (EMRALD®) housed at the Institute for Clinical Evaluative Sciences. ed charts provided the reference standard based on available physician-diagnoses, chronic obstructive pulmonary disease-specific medications, smoking history and pulmonary function testing. Chronic obstructive pulmonary disease electronic medical record algorithms using combinations of terminology in the cumulative patient profile (CPP; problem list/past medical history), physician billing codes (chronic bronchitis/emphysema/other chronic obstructive pulmonary disease), and prescriptions, were tested against the reference standard. Sensitivity, specificity, and positive/negative predictive values (PPV/NPV) were calculated. There were 364 patients with chronic obstructive pulmonary disease identified in a 5889 randomly sampled cohort aged ≥ 35 years (prevalence = 6.2%). The electronic medical record algorithm consisting of ≥ 3 physician billing codes for chronic obstructive pulmonary disease per year; documentation in the CPP; tiotropium prescription; or ipratropium (or its formulations) prescription and a chronic obstructive pulmonary disease billing code had sensitivity of 76.9% (95% CI:72.2–81.2), specificity of 99.7% (99.5–99.8), PPV of 93.6% (90.3–96.1), and NPV of 98.5% (98.1–98.8). Electronic medical record algorithms can accurately identify patients with chronic obstructive pulmonary disease in primary care records. They can be used to enable further studies in practice patterns and chronic obstructive pulmonary disease management in primary care.
Highlights
Chronic obstructive pulmonary disease (COPD) is characterized by persistent airflow limitation and an enhanced chronic inflammatory airway response to noxious particles or gases such as tobacco smoke.[1]
Compared to people in the reference cohort, those with COPD were older and had a higher proportion of males. They were more likely to have smoking history recorded in their charts (70% compared to 61% in those without COPD), and to have documented pulmonary function test (PFT) results (40% vs. 5% in patients without COPD; see Table 1)
The algorithms tested for identifying patients with COPD in the Electronic medical record (EMR) all had high specificity and negative predictive value (NPV), but varied in their sensitivity and positive predictive value (PPV)
Summary
Chronic obstructive pulmonary disease (COPD) is characterized by persistent airflow limitation and an enhanced chronic inflammatory airway response to noxious particles or gases such as tobacco smoke.[1]. There is still a limited availability of population-wide data that can be used to build strategies for improvement of care, research and healthcare planning. Previous work identifying people with and evaluating the burden of COPD have primarily been based on cross-sectional survey data and population cohorts.[5] Self-reported measures for COPD in surveys have been validated against clinical records and physician diagnosis with relatively high accuracy, but are limited in clinical information.[9,10,11,12] While population cohorts have been derived from health care claims from some administrative databases ( for populations with comprehensive health and drug coverage),[13] they are limited in the depth and details of patient clinical information because they are created to manage financial transactions rather than for research purposes or patient care.[14]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.