User login
Access to Germline Genetic Testing through Clinical Pathways in Veterans With Prostate Cancer
Background
Germline genetic testing (GGT) is essential in prostate cancer care, informing clinical decisions. The Veterans Affairs National Oncology Program (VA NOP) recommends GGT for patients with specific risk factors in non-metastatic prostate cancer and all patients with metastatic disease. Understanding GGT access helps evaluate care quality and guide improvements. Since 2021, VA NOP has implemented pathway health factor (HF) templates to standardize cancer care documentation, including GGT status, enabling data extraction from the Corporate Data Warehouse (CDW) rather than requiring manual review of clinical notes. This work aims to evaluate Veterans’ access to GGT in prostate cancer care by leveraging pathway HF templates, and to assess the feasibility of using structured electronic health record (EHR) data to monitor adherence to GGT recommendations.
Methods
Process delivery diagrams (PDDs) were used to map data flow from prostate cancer clinical pathways to the VA CDW. We identified and categorized HFs related to prostate cancer GGT through the computerized patient record system (CPRS). Descriptive statistics were used to summarize access, ordering, and consent rates.
Results
We identified 5,744 Veterans with at least one prostate cancer GGT-relevant HF entered between 02/01/2021 and 12/31/2024. Of these, 5,125 (89.2%) had access to GGT, with 4,569 (89.2%) consenting to or having GGT ordered, while 556 (10.8%) declined testing. Among the 619 (10.8%) Veterans without GGT access, providers reported plans to discuss GGT in the future for 528 (85.3%) patients, while 91 (14.7%) were off pathway.
Conclusions
NOP-developed HF templates enabled extraction of GGT information from structured EHR data, eliminating manual extraction from clinical notes. We observed high GGT utilization among Veterans with pathway-entered HFs. However, low overall HF utilization may introduce selection bias. Future work includes developing a Natural Language Processing pipeline using large language models to automatically extract GGT information from clinical notes, with HF data serving as ground truth.
Background
Germline genetic testing (GGT) is essential in prostate cancer care, informing clinical decisions. The Veterans Affairs National Oncology Program (VA NOP) recommends GGT for patients with specific risk factors in non-metastatic prostate cancer and all patients with metastatic disease. Understanding GGT access helps evaluate care quality and guide improvements. Since 2021, VA NOP has implemented pathway health factor (HF) templates to standardize cancer care documentation, including GGT status, enabling data extraction from the Corporate Data Warehouse (CDW) rather than requiring manual review of clinical notes. This work aims to evaluate Veterans’ access to GGT in prostate cancer care by leveraging pathway HF templates, and to assess the feasibility of using structured electronic health record (EHR) data to monitor adherence to GGT recommendations.
Methods
Process delivery diagrams (PDDs) were used to map data flow from prostate cancer clinical pathways to the VA CDW. We identified and categorized HFs related to prostate cancer GGT through the computerized patient record system (CPRS). Descriptive statistics were used to summarize access, ordering, and consent rates.
Results
We identified 5,744 Veterans with at least one prostate cancer GGT-relevant HF entered between 02/01/2021 and 12/31/2024. Of these, 5,125 (89.2%) had access to GGT, with 4,569 (89.2%) consenting to or having GGT ordered, while 556 (10.8%) declined testing. Among the 619 (10.8%) Veterans without GGT access, providers reported plans to discuss GGT in the future for 528 (85.3%) patients, while 91 (14.7%) were off pathway.
Conclusions
NOP-developed HF templates enabled extraction of GGT information from structured EHR data, eliminating manual extraction from clinical notes. We observed high GGT utilization among Veterans with pathway-entered HFs. However, low overall HF utilization may introduce selection bias. Future work includes developing a Natural Language Processing pipeline using large language models to automatically extract GGT information from clinical notes, with HF data serving as ground truth.
Background
Germline genetic testing (GGT) is essential in prostate cancer care, informing clinical decisions. The Veterans Affairs National Oncology Program (VA NOP) recommends GGT for patients with specific risk factors in non-metastatic prostate cancer and all patients with metastatic disease. Understanding GGT access helps evaluate care quality and guide improvements. Since 2021, VA NOP has implemented pathway health factor (HF) templates to standardize cancer care documentation, including GGT status, enabling data extraction from the Corporate Data Warehouse (CDW) rather than requiring manual review of clinical notes. This work aims to evaluate Veterans’ access to GGT in prostate cancer care by leveraging pathway HF templates, and to assess the feasibility of using structured electronic health record (EHR) data to monitor adherence to GGT recommendations.
Methods
Process delivery diagrams (PDDs) were used to map data flow from prostate cancer clinical pathways to the VA CDW. We identified and categorized HFs related to prostate cancer GGT through the computerized patient record system (CPRS). Descriptive statistics were used to summarize access, ordering, and consent rates.
Results
We identified 5,744 Veterans with at least one prostate cancer GGT-relevant HF entered between 02/01/2021 and 12/31/2024. Of these, 5,125 (89.2%) had access to GGT, with 4,569 (89.2%) consenting to or having GGT ordered, while 556 (10.8%) declined testing. Among the 619 (10.8%) Veterans without GGT access, providers reported plans to discuss GGT in the future for 528 (85.3%) patients, while 91 (14.7%) were off pathway.
Conclusions
NOP-developed HF templates enabled extraction of GGT information from structured EHR data, eliminating manual extraction from clinical notes. We observed high GGT utilization among Veterans with pathway-entered HFs. However, low overall HF utilization may introduce selection bias. Future work includes developing a Natural Language Processing pipeline using large language models to automatically extract GGT information from clinical notes, with HF data serving as ground truth.