Factors Associated With Receipt of Molecular Testing and its Impact on Time to Initial Systemic Therapy in Metastatic Non-Small Cell Lung Cancer.

BACKGROUND: Despite recommendations for molecular testing irrespective of patient characteristics, differences exist in receipt of molecular testing for oncogenic drivers amongst metastatic non-small cell lung cancer (mNSCLC) patients. Exploration into these differences and their effects on treatment is needed to identify opportunities for improvement. PATIENTS AND METHODS: We conducted a retrospective cohort study of adult patients diagnosed with mNSCLC between 2011 and 2018 using PCORnet's Rapid Cycle Research Project dataset (n = 3600). Log-binomial, Cox proportional hazards (PH), and time-varying Cox regression models were used to ascertain whether molecular testing was received, and time from diagnosis to molecular testing and/or initial systemic treatment in the context of patient age, sex, race/ethnicity, and multiple comorbidities status. RESULTS: The majority of patients in this cohort were ≤ 65 years of age (median [25th, 75th]: 64 [57, 71]), male (54.3%), non-Hispanic white individuals (81.6%), with > 2 comorbidities in addition to mNSCLC (54.1%). About half the cohort received molecular testing (49.9%). Patients who received molecular testing had a 59% higher probability of initial systemic treatment than patients who were yet to receive testing. Multiple comorbidity status was positively associated with receipt of molecular testing (RR, 1.27; 95% CI 1.08, 1.49). CONCLUSION: Receipt of molecular testing in academic centers was associated with earlier initiation of systemic treatment. This finding underscores the need to increase molecular testing rates amongst mNSCLC patients during a clinically relevant period. Further studies to validate these findings in community centers are warranted.

Mapping Clinical Documents to the Logical Observation Identifiers, Names and Codes (LOINC) Document Ontology using Electronic Health Record Systems Structured Metadata.

As Electronic Health Record (EHR) systems increase in usage, organizations struggle to maintain and categorize clinical documentation so it can be used for clinical care and research. While prior research has often employed natural language processing techniques to categorize free text documents, there are shortcomings relative to computational scalability and the lack of key metadata within notes' text. This study presents a framework that can allow institutions to map their notes to the LOINC document ontology using a Bag of Words approach. After preliminary manual value- set mapping, an automated pipeline that leverages key dimensions of metadata from structured EHR fields aligns the notes with the dimensions of the document ontology. This framework resulted in 73.4% coverage of EHR documents, while also mapping 132 million notes in less than 2 hours; an order of magnitude more efficient than NLP based methods.

Investigation of the Utility of Features in a Clinical De-identification Model: A Demonstration Using EHR Pathology Reports for Advanced NSCLC Patients.

BACKGROUND: Electronic health record (EHR) systems contain a large volume of texts, including visit notes, discharge summaries, and various reports. To protect the confidentiality of patients, these records often need to be fully de-identified before circulating for secondary use. Machine learning (ML) based named entity recognition (NER) model has emerged as a popular technique of automatic de-identification. OBJECTIVE: The performance of a machine learning model highly depends on the selection of appropriate features. The objective of this study was to investigate the usability of multiple features in building a conditional random field (CRF) based clinical de-identification NER model. METHODS: Using open-source natural language processing (NLP) toolkits, we annotated protected health information (PHI) in 1,500 pathology reports and built supervised NER models using multiple features and their combinations. We further investigated the dependency of a model's performance on the size of training data. RESULTS: Among the 10 feature extractors explored in this study, n-gram, prefix-suffix, word embedding, and word shape performed the best. A model using combination of these four feature sets yielded precision, recall, and F1-score for each PHI as follows: NAME (0.80; 0.79; 0.80), LOCATION (0.85; 0.83; 0.84), DATE (0.86; 0.79; 0.82), HOSPITAL (0.96; 0.93; 0.95), ID (0.99; 0.82; 0.90), and INITIALS (0.97; 0.49; 0.65). We also found that the model's performance becomes saturated when the training data size is beyond 200. CONCLUSION: Manual de-identification of large-scale data is an impractical procedure since it is time-consuming and subject to human errors. Analysis of the NER model's performance in this study sheds light on a semi-automatic clinical de-identification pipeline for enterprise-wide data warehousing.