A nursing information model process for interoperability

Abstract The ability to share nursing data across organizations and electronic health records is a key component of improving care coordination and quality outcomes. Currently, substantial organizational and technical barriers limit the ability to share and compare essential patient data that inform nursing care. Nursing leaders at Kaiser Permanente and the U.S. Department of Veterans Affairs collaborated on the development of an evidence-based information model driven by nursing practice to enable data capture, re-use, and sharing between organizations and disparate electronic health records. This article describes a framework with repeatable steps and processes to enable the semantic interoperability of relevant and contextual nursing data. Hospital-acquired pressure ulcer prevention was selected as the prototype nurse-sensitive quality measure to develop and test the model. In a Health 2.0 Developer Challenge program from the Office of the National Coordinator for Health, mobile applications implemented the model to help nurses assess the risk of hospital-acquired pressure ulcers and reduce their severity. The common information model can be applied to other nurse-sensitive measures to enable data standardization supporting patient transitions between care settings, quality reporting, and research.

Download Full-text

Exchanging personal health data with electronic health records: A standardized information model for patient generated health data and observations of daily living

International Journal of Medical Informatics ◽

10.1016/j.ijmedinf.2018.10.006 ◽

2018 ◽

Vol 120 ◽

pp. 116-125 ◽

Cited By ~ 3

Author(s):

Panagiotis Plastiras ◽

Dympna O’Sullivan

Keyword(s):

Electronic Health Records ◽

Daily Living ◽

Information Model ◽

Health Data ◽

Personal Health ◽

Health Records ◽

Electronic Health

Download Full-text

Health Care Analytics With Time-Invariant and Time-Variant Feature Importance to Predict Hospital-Acquired Acute Kidney Injury: Observational Longitudinal Study (Preprint)

10.2196/preprints.30805 ◽

2021 ◽

Author(s):

Horng-Ruey Chua ◽

Kaiping Zheng ◽

Anantharaman Vathsala ◽

Kee-Yuan Ngiam ◽

Hui-Kim Yap ◽

...

Keyword(s):

Machine Learning ◽

Acute Kidney Injury ◽

Electronic Health Records ◽

Lead Time ◽

Kidney Injury ◽

False Positives ◽

Health Records ◽

Electronic Health ◽

Hospital Acquired ◽

Time Invariant

BACKGROUND Acute kidney injury (AKI) develops in 4% of hospitalized patients and is a marker of clinical deterioration and nephrotoxicity. AKI onset is highly variable in hospitals, which makes it difficult to time biomarker assessment in all patients for preemptive care. OBJECTIVE The study sought to apply machine learning techniques to electronic health records and predict hospital-acquired AKI by a 48-hour lead time, with the aim to create an AKI surveillance algorithm that is deployable in real time. METHODS The data were sourced from 20,732 case admissions in 16,288 patients over 1 year in our institution. We enhanced the bidirectional recurrent neural network model with a novel time-invariant and time-variant aggregated module to capture important clinical features temporal to AKI in every patient. Time-series features included laboratory parameters that preceded a 48-hour prediction window before AKI onset; the latter’s corresponding reference was the final in-hospital serum creatinine performed in case admissions without AKI episodes. RESULTS The cohort was of mean age 53 (SD 25) years, of whom 29%, 12%, 12%, and 53% had diabetes, ischemic heart disease, cancers, and baseline eGFR <90 mL/min/1.73 m<sup>2</sup>, respectively. There were 911 AKI episodes in 869 patients. We derived and validated an algorithm in the testing dataset with an AUROC of 0.81 (0.78-0.85) for predicting AKI. At a 15% prediction threshold, our model generated 699 AKI alerts with 2 false positives for every true AKI and predicted 26% of AKIs. A lowered 5% prediction threshold improved the recall to 60% but generated 3746 AKI alerts with 6 false positives for every true AKI. Representative interpretation results produced by our model alluded to the top-ranked features that predicted AKI that could be categorized in association with sepsis, acute coronary syndrome, nephrotoxicity, or multiorgan injury, specific to every case at risk. CONCLUSIONS We generated an accurate algorithm from electronic health records through machine learning that predicted AKI by a lead time of at least 48 hours. The prediction threshold could be adjusted during deployment to optimize recall and minimize alert fatigue, while its precision could potentially be augmented by targeted AKI biomarker assessment in the high-risk cohort identified.

Download Full-text

Health Care Analytics With Time-Invariant and Time-Variant Feature Importance to Predict Hospital-Acquired Acute Kidney Injury: Observational Longitudinal Study

Journal of Medical Internet Research ◽

10.2196/30805 ◽

2021 ◽

Vol 23 (12) ◽

pp. e30805

Author(s):

Horng-Ruey Chua ◽

Kaiping Zheng ◽

Anantharaman Vathsala ◽

Kee-Yuan Ngiam ◽

Hui-Kim Yap ◽

...

Keyword(s):

Machine Learning ◽

Acute Kidney Injury ◽

Electronic Health Records ◽

Lead Time ◽

Kidney Injury ◽

False Positives ◽

Health Records ◽

Electronic Health ◽

Hospital Acquired ◽

Time Invariant

Background Acute kidney injury (AKI) develops in 4% of hospitalized patients and is a marker of clinical deterioration and nephrotoxicity. AKI onset is highly variable in hospitals, which makes it difficult to time biomarker assessment in all patients for preemptive care. Objective The study sought to apply machine learning techniques to electronic health records and predict hospital-acquired AKI by a 48-hour lead time, with the aim to create an AKI surveillance algorithm that is deployable in real time. Methods The data were sourced from 20,732 case admissions in 16,288 patients over 1 year in our institution. We enhanced the bidirectional recurrent neural network model with a novel time-invariant and time-variant aggregated module to capture important clinical features temporal to AKI in every patient. Time-series features included laboratory parameters that preceded a 48-hour prediction window before AKI onset; the latter’s corresponding reference was the final in-hospital serum creatinine performed in case admissions without AKI episodes. Results The cohort was of mean age 53 (SD 25) years, of whom 29%, 12%, 12%, and 53% had diabetes, ischemic heart disease, cancers, and baseline eGFR <90 mL/min/1.73 m2, respectively. There were 911 AKI episodes in 869 patients. We derived and validated an algorithm in the testing dataset with an AUROC of 0.81 (0.78-0.85) for predicting AKI. At a 15% prediction threshold, our model generated 699 AKI alerts with 2 false positives for every true AKI and predicted 26% of AKIs. A lowered 5% prediction threshold improved the recall to 60% but generated 3746 AKI alerts with 6 false positives for every true AKI. Representative interpretation results produced by our model alluded to the top-ranked features that predicted AKI that could be categorized in association with sepsis, acute coronary syndrome, nephrotoxicity, or multiorgan injury, specific to every case at risk. Conclusions We generated an accurate algorithm from electronic health records through machine learning that predicted AKI by a lead time of at least 48 hours. The prediction threshold could be adjusted during deployment to optimize recall and minimize alert fatigue, while its precision could potentially be augmented by targeted AKI biomarker assessment in the high-risk cohort identified.

Download Full-text