A Clinical Phenotyping Algorithm to Identify Cases of Chronic Obstructive Pulmonary Disease in Electronic Health Records

Martucci, Victoria; Liu, Naian; Kerchberger, V.E.; Osterman, Travis; Torstenson, Eric S.; Richmond, Bradley W.; Aldrich, Melinda C.

doi:10.1101/716779

Cited by 3 publications

(4 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…5; Supplementary Fig . S3) [41]. We first examined the top 25 codes under the COPD topic’s distribution ϕ COPD ( Fig .…”

Section: Resultsmentioning

confidence: 99%

“…To investigate the concordance and discrepancy between MixEHR-G’s inferred topics and the existing phenotyping rules, we focused within the PopHR dataset on chronic obstructive pulmonary disease (COPD), a notoriously difficult phenotype to define with rules ( Fig.5 ; Supplementary Fig. S5) [58]. We first examined the top 25 codes under the COPD topic’s distribution ϕ COPD ( Fig.5a ; Supplementary Fig.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

MixEHR-Guided: A guided multi-modal topic modeling approach for large-scale automatic phenotyping using the electronic health record

Ahuja

Zou

Verma

et al. 2021

Preprint

View full text Add to dashboard Cite

Electronic Health Records (EHRs) contain rich clinical data collected at the point of the care, and their increasing adoption offers exciting opportunities for clinical informatics, disease risk prediction, and personalized treatment recommendation. However, effective use of EHR data for research and clinical decision support is often hampered by a lack of reliable disease labels. To compile gold-standard labels, researchers often rely on clinical experts to develop rule-based phenotyping algorithms from billing codes and other surrogate features. This process is tedious and error-prone due to recall and observer biases in how codes and measures are selected, and some phenotypes are incompletely captured by a handful of surrogate features. To address this challenge, we present a novel automatic phenotyping model called MixEHR-Guided (MixEHR-G), a multimodal hierarchical Bayesian topic model that efficiently models the EHR generative process by identifying latent phenotype structure in the data. Unlike existing topic modeling algorithms wherein, the inferred topics are not identifiable, MixEHR-G uses prior information from informative surrogate features to align topics with known phenotypes. We applied MixEHR-G to an openly available EHR dataset of 38,597 intensive care patients (MIMIC-III) in Boston, USA and to administrative claims data for a population-based cohort (PopHR) of 1.3 million people in Quebec, Canada. Qualitatively, we demonstrate that MixEHR-G learns interpretable phenotypes and yields meaningful insights about phenotype similarities, comorbidities, and epidemiological associations. Quantitatively, MixEHR-G outperforms existing unsupervised phenotyping methods on a phenotype label annotation task, and it can accurately estimate relative phenotype prevalence functions without gold-standard phenotype information. Altogether, MixEHR-G is an important step towards building an interpretable and automated phenotyping system using EHR data.

show abstract

“…5; Supplementary Fig . S3) [41]. We first examined the top 25 codes under the COPD topic’s distribution ϕ COPD ( Fig .…”

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

MixEHR-Guided: A guided multi-modal topic modeling approach for large-scale automatic phenotyping using the electronic health record

Ahuja

Zou

Verma

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…As previously described, a unique algorithm using regular expressions and pattern matching was created to extract structured, quantitative data from all RHC reports 20 . Individuals with COPD were identified using an internally validated algorithmic approach based on the presence/absence of international classification of diseases (ICD) codes relevant to COPD, as previously described 21 . An age cutoff of 45 was used to increase specificity for COPD disease, which is rare among younger adults.…”

Section: Methodsmentioning

confidence: 99%

“…20 Individuals with COPD were identified using an internally validated algorithmic approach based on the presence/absence of international classification of diseases (ICD) codes relevant to COPD, as previously described. 21 An age cutoff of 45 was used to increase specificity for COPD disease, which is rare among younger adults. Exclusion criteria included indeterminate status (i.e., inadequate coding to assign a COPD status), ICD codes for other restive and obstructive lung diseases (Table S1).…”

Section: Study Populationmentioning

confidence: 99%

Clinical insights into pulmonary hypertension in chronic obstructive pulmonary disease

Cook

Martucci

et al. 2022

Pulm. circ.

Self Cite

View full text Add to dashboard Cite

Pulmonary hypertension (PH) is a common complication of chronic obstructive pulmonary disease (COPD). Little is known about the prevalence and clinical profiles of patients with COPD‐PH. We report the clinical characteristics, hemodynamic profiles, and prognosis in a large population of patients with COPD referred for right heart catheterization (RHC). We extracted data from all patients referred for RHC between 1997 and 2017 in Vanderbilt's deidentified medical record. PH was defined as mean pulmonary artery pressure >20 mmHg. Pre‐ and postcapillary PH were defined according to contemporary guidelines. COPD was identified using a validated rules‐based algorithm requiring international classification of diseases codes relevant to COPD. We identified 6065 patients referred for RHC, of whom 1509 (24.9%) had COPD and 1213 had COPD and PH. Patients with COPD‐PH had a higher prevalence of diabetes, atrial fibrillation, and heart failure compared with COPD without PH. Approximately 55% of patients with COPD‐PH had elevated left ventricle (LV) filling pressure. Pulmonary function testing data from individuals with COPD‐PH revealed subtype differences, with precapillary COPD‐PH having lower diffusion capacity of the lungs for carbon monoxide (DLCO) values than the other COPD‐PH subtypes. Patients with COPD‐PH had significantly increased mortality compared with COPD alone (hazard ratio [HR]: 1.70, 95% confidence interval [CI]: 1.28–2.26) with the highest mortality among the combined pre‐ and postcapillary COPD‐PH subgroup (HR: 2.39; 95% CI: 1.64–3.47). PH is common among patients with COPD referred for RHC. The etiology of PH in patients with COPD is often mixed due to multimorbidity and is associated with high mortality, which may have implications for risk factor management.

show abstract

Performance of a Computational Phenotyping Algorithm for Sarcoidosis Using Diagnostic Codes in Electronic Medical Records: Case Validation Study From 2 Veterans Affairs Medical Centers

Seedahmed¹,

Mogilnicka²,

Zeng³

et al. 2022

JMIR Form Res

View full text Add to dashboard Cite

Background Electronic medical records (EMRs) offer the promise of computationally identifying sarcoidosis cases. However, the accuracy of identifying these cases in the EMR is unknown. Objective The aim of this study is to determine the statistical performance of using the International Classification of Diseases (ICD) diagnostic codes to identify patients with sarcoidosis in the EMR. Methods We used the ICD diagnostic codes to identify sarcoidosis cases by searching the EMRs of the San Francisco and Palo Alto Veterans Affairs medical centers and randomly selecting 200 patients. To improve the diagnostic accuracy of the computational algorithm in cases where histopathological data are unavailable, we developed an index of suspicion to identify cases with a high index of suspicion for sarcoidosis (confirmed and probable) based on clinical and radiographic features alone using the American Thoracic Society practice guideline. Through medical record review, we determined the positive predictive value (PPV) of diagnosing sarcoidosis by two computational methods: using ICD codes alone and using ICD codes plus the high index of suspicion. Results Among the 200 patients, 158 (79%) had a high index of suspicion for sarcoidosis. Of these 158 patients, 142 (89.9%) had documentation of nonnecrotizing granuloma, confirming biopsy-proven sarcoidosis. The PPV of using ICD codes alone was 79% (95% CI 78.6%-80.5%) for identifying sarcoidosis cases and 71% (95% CI 64.7%-77.3%) for identifying histopathologically confirmed sarcoidosis in the EMRs. The inclusion of the generated high index of suspicion to identify confirmed sarcoidosis cases increased the PPV significantly to 100% (95% CI 96.5%-100%). Histopathology documentation alone was 90% sensitive compared with high index of suspicion. Conclusions ICD codes are reasonable classifiers for identifying sarcoidosis cases within EMRs with a PPV of 79%. Using a computational algorithm to capture index of suspicion data elements could significantly improve the case-identification accuracy.

show abstract

A Clinical Phenotyping Algorithm to Identify Cases of Chronic Obstructive Pulmonary Disease in Electronic Health Records

Cited by 3 publications

References 53 publications

MixEHR-Guided: A guided multi-modal topic modeling approach for large-scale automatic phenotyping using the electronic health record

MixEHR-Guided: A guided multi-modal topic modeling approach for large-scale automatic phenotyping using the electronic health record

Clinical insights into pulmonary hypertension in chronic obstructive pulmonary disease

Performance of a Computational Phenotyping Algorithm for Sarcoidosis Using Diagnostic Codes in Electronic Medical Records: Case Validation Study From 2 Veterans Affairs Medical Centers

Contact Info

Product

Resources

About