A comparison of natural language processing to ICD-10 codes for identification and characterization of pulmonary embolism

Johnson, Stacy A.; Signor, Emily; Lappé, Katie; Shi, Jianlin; Jenkins, S.L.; Wikstrom, Sara W.; Kroencke, Rachel D.; Hallowell, David M.; Jones, Aubrey; Witt, Daniel M.

doi:10.1016/j.thromres.2021.04.020

Cited by 17 publications

(9 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With respect to ICD-10 code validation, PE-EHR+ has several strengths compared with the existing investigations and will complement their findings. [33][34][35][36][37][38][39][40] Unlike several other studies, PE-EHR+ has a pre-specified power calculation. In addition, discharge records will be reviewed from both community hospitals and large referral hospitals with a diverse patient population.…”

Section: Discussionmentioning

confidence: 99%

“…However, there are important knowledge gaps related to the optimal approach for case selection of patients with PE. The existing studies using ICD-10 (►Table 1 with codes, ►Table 2 with studies) [33][34][35][36][37][38][39][40] or NLP (►Table 3) [27][28][29][30][31]38,39 have had limitations including small number or being from a single center, lack of sharing sufficient details including about the location of the codes (in the principal discharge diagnosis position vs. secondary discharge diagnosis position), or limited cross-validation. The PE-EHRþ study has been designed to address these gaps in knowledge and to validate efficient tools for identification of patients with PE in electronic databases.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Developing Validated Tools to Identify Pulmonary Embolism in Electronic Databases: Rationale and Design of the PE-EHR+ Study

Bikdeli

Khairani

et al. 2023

Thromb Haemost

View full text Add to dashboard Cite

Background: Contemporary pulmonary embolism (PE) research relies on data from electronic health records (EHRs) and administrative databases that use International Classification of Diseases (ICD) codes. Natural language processing (NLP) tools can be used for automated chart review and patient identification. However, there remains uncertainty with the validity of ICD-10 codes or NLP algorithms for patient identification. Methods: The PE-EHR+ study has been designed to validate ICD-10 codes as Principal Discharge Diagnosis, or Secondary Discharge Diagnoses, as well as NLP tools set out in prior studies to identify patients with PE within EHRs. Manual chart review by two independent abstractors by predefined criteria will be the reference standard. Sensitivity, specificity, and positive and negative predictive values will be determined. We will assess the discriminatory function of code subgroups for intermediate– and high-risk PE. In addition, accuracy of NLP algorithms to identify PE from radiology reports will be assessed. Results: A total of 1,734 patients from the Mass General Brigham health system have been identified. These include 578 with ICD-10 Principal Discharge Diagnosis codes for PE, 578 with codes in the secondary position, and 578 without PE codes during the index hospitalization. A smaller subset of patients will also be identified from the Yale-New Haven Health System. Data validation and analyses will be forthcoming. Conclusions: The PE-EHR+ study will help validate efficient tools for identification of patients with PE in EHRs, improving the reliability of efficient observational studies or randomized trials of patients with PE using electronic databases.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Developing Validated Tools to Identify Pulmonary Embolism in Electronic Databases: Rationale and Design of the PE-EHR+ Study

Bikdeli

Khairani

et al. 2023

Thromb Haemost

View full text Add to dashboard Cite

show abstract

“…At our institute, the diagnosis of PE was based on the presence of an intraluminal filling defect in the pulmonary artery tree on CTPA. In accordance with a previous study ( 26 ), the International Classification of Diseases-10 (ICD-10) codes were used to identify patients with PE (I26.0 and I26.9). ICD-10 I26.0 represents PE patients with acute cor pulmonale, while I26.9 represents PE patients without acute cor pulmonale.…”

Section: Methodsmentioning

confidence: 99%

Anatomic distribution of lower extremity deep venous thrombosis is associated with an increased risk of pulmonary embolism: A 10-year retrospective analysis

Zhang

Chen

Wang

et al. 2023

Front. Cardiovasc. Med.

View full text Add to dashboard Cite

AimsTo investigate the potential relationship between anatomic distribution of lower extremity deep venous thrombosis (LEDVT) and pulmonary embolism (PE).MethodsA retrospective case-control study was performed in patients diagnosed with LEDVT, which were confirmed by bilateral lower extremity compression ultrasonography (CUS) examination. According to the ultrasound reports, thrombus sidedness was categorized as unilateral and bilateral lower extremity, thrombus location was classified into distal and proximal LEDVT. Anatomic distributions of LEDVT were further subdivided depending on the combination of thrombus sidedness and location. Patients with PE were identified using the International Classification of Diseases-10 (ICD-10) codes (I26.0 and I26.9), and divided into PE group and Non-PE group. Univariate and multivariate logistic regression analyses were used to assess the association between anatomic distribution of LEDVT and PE. Sensitivity analyses were also conducted.ResultsA total of 2,363 consecutive patients with LEDVT were included, of whom 66.10% and 33.90% were unilateral and bilateral LEDVT, as well as 71.39% and 28.61% were isolated distal and proximal LEDVT, respectively. After the diagnosis of LEDVT, 185 patients (7.83%) developed PE. The proportions of PE ranged between the lowest (4.07%) in unilateral-distal LEDVT and highest (14.55%) in bilateral-proximal LEDVT. Multivariate logistic regression analysis showed that bilateral LEDVT (odds ratios [OR] = 2.455, 95% confidence interval [CI]: 1.803–3.344, P < 0.001) and proximal LEDVT (OR = 1.530, 95% CI: 1.105–2.118, P = 0.010) were risk factors for developing PE. Moreover, unilateral-proximal (OR = 2.129, 95% CI: 1.365–3.320, P = 0.00), bilateral-distal (OR = 3.193, 95% CI: 2.146–4.752, P < 0.001) and bilateral-proximal LEDVT(OR = 3.425, 95% CI: 2.093–5.603, P < 0.001) were significantly associated with an increased risk of PE. Sensitivity analyses also confirmed the robustness of these associations.ConclusionPatients with unilateral-proximal, bilateral-distal or bilateral-proximal are more likely to suffer from PE than those with unilateral-distal LEDVT.

show abstract

“…The overall estimate of MST in women and men increased from 8.1% to 13.1% when NLP was applied [40]. In another EHR system, NLP has recently been shown to achieve high specificity for identifying patients with pulmonary embolism events (PE) and greater specificity than ICD-10 coding in the EHR [41]. A challenge here is that bias in ML models can be exacerbated in NLP.…”

Section: Hearing the Patient And Providers Voices-leverage Unstructured Case Note Datamentioning

confidence: 99%

Artificial Intelligence, Heuristic Biases, and the Optimization of Health Outcomes: Cautionary Optimism

Feehan

Owen

McKinnon

et al. 2021

JCM

View full text Add to dashboard Cite

The use of artificial intelligence (AI) and machine learning (ML) in clinical care offers great promise to improve patient health outcomes and reduce health inequity across patient populations. However, inherent biases in these applications, and the subsequent potential risk of harm can limit current use. Multi-modal workflows designed to minimize these limitations in the development, implementation, and evaluation of ML systems in real-world settings are needed to improve efficacy while reducing bias and the risk of potential harms. Comprehensive consideration of rapidly evolving AI technologies and the inherent risks of bias, the expanding volume and nature of data sources, and the evolving regulatory landscapes, can contribute meaningfully to the development of AI-enhanced clinical decision making and the reduction in health inequity.

show abstract

A comparison of natural language processing to ICD-10 codes for identification and characterization of pulmonary embolism

Cited by 17 publications

References 11 publications

Developing Validated Tools to Identify Pulmonary Embolism in Electronic Databases: Rationale and Design of the PE-EHR+ Study

Developing Validated Tools to Identify Pulmonary Embolism in Electronic Databases: Rationale and Design of the PE-EHR+ Study

Anatomic distribution of lower extremity deep venous thrombosis is associated with an increased risk of pulmonary embolism: A 10-year retrospective analysis

Artificial Intelligence, Heuristic Biases, and the Optimization of Health Outcomes: Cautionary Optimism

Contact Info

Product

Resources

About