Predictive modeling with electronic health record (EHR) data is anticipated to drive personalized medicine and improve healthcare quality. Constructing predictive statistical models typically requires extraction of curated predictor variables from normalized EHR data, a labor-intensive process that discards the vast majority of information in each patient’s record. We propose a representation of patients’ entire raw EHR records based on the Fast Healthcare Interoperability Resources (FHIR) format. We demonstrate that deep learning methods using this representation are capable of accurately predicting multiple medical events from multiple centers without site-specific data harmonization. We validated our approach using de-identified EHR data from two US academic medical centers with 216,221 adult patients hospitalized for at least 24 h. In the sequential format we propose, this volume of EHR data unrolled into a total of 46,864,534,945 data points, including clinical notes. Deep learning models achieved high accuracy for tasks such as predicting: in-hospital mortality (area under the receiver operator curve [AUROC] across sites 0.93–0.94), 30-day unplanned readmission (AUROC 0.75–0.76), prolonged length of stay (AUROC 0.85–0.86), and all of a patient’s final discharge diagnoses (frequency-weighted AUROC 0.90). These models outperformed traditional, clinically-used predictive models in all cases. We believe that this approach can be used to create accurate and scalable predictions for a variety of clinical scenarios. In a case study of a particular prediction, we demonstrate that neural networks can be used to identify relevant information from the patient’s chart.

/pdf/scalable-and-accurate-deep-learning-with-electronic-health-eu4vc3mdb6.pdf

Scalable and accurate deep learning with electronic health records

https://www.thelancet.com/pdfs/journals/lancet/PIIS0140673600023230.pdf

Early use of non-invasive ventilation for acute exacerbations of chronic obstructive pulmonary disease on general respiratory wards: a multicentre randomised controlled trial.

Predictive modeling with electronic health record (EHR) data is anticipated to drive personalized medicine and improve healthcare quality. Constructing predictive statistical models typically requires extraction of curated predictor variables from normalized EHR data, a labor-intensive process that discards the vast majority of information in each patient's record. We propose a representation of patients' entire, raw EHR records based on the Fast Healthcare Interoperability Resources (FHIR) format. We demonstrate that deep learning methods using this representation are capable of accurately predicting multiple medical events from multiple centers without site-specific data harmonization. We validated our approach using de-identified EHR data from two U.S. academic medical centers with 216,221 adult patients hospitalized for at least 24 hours. In the sequential format we propose, this volume of EHR data unrolled into a total of 46,864,534,945 data points, including clinical notes. Deep learning models achieved high accuracy for tasks such as predicting in-hospital mortality (AUROC across sites 0.93-0.94), 30-day unplanned readmission (AUROC 0.75-0.76), prolonged length of stay (AUROC 0.85-0.86), and all of a patient's final discharge diagnoses (frequency-weighted AUROC 0.90). These models outperformed state-of-the-art traditional predictive models in all cases. We also present a case-study of a neural-network attribution system, which illustrates how clinicians can gain some transparency into the predictions. We believe that this approach can be used to create accurate and scalable predictions for a variety of clinical scenarios, complete with explanations that directly highlight evidence in the patient's chart.

/pdf/scalable-and-accurate-deep-learning-for-electronic-health-55ghjalidu.pdf

Scalable and accurate deep learning for electronic health records

The Braden Scale for Predicting Pressure Sore Risk was developed to foster early identification of patients at risk for forming pressure sores. The scale is composed of six subscales that reflect sensory perception, skin moisture, activity, mobility, friction and shear, and nutritional status. Content and construct validity were established by expert opinion and empirical testing. Three studies of reliability are reported here, using raters who varied in level of educational preparation and geographic region. Two prospective studies of predictive validity were completed to determine the scale's sensitivity and specificity. Reliability ranged from r = .83 to r = .94 for nurses' aides and licensed practical nurses; when used by registered nurses, the reliability increased to r = .99. Predictive validity was calculated for each cut-off point of the scale. Using a cutoff point of 16, sensitivity was 100% in both studies. Specificity ranged from 64% to 90%. This instrument has highly satisfactory reliability when used by RNs, and greater sensitivity and specificity than instruments previously reported.

The Braden Scale for predicting pressure sore risk. Nursing Research, 36 (4), 205-210

Lactobacillus GG in the prevention of antibiotic-associated diarrhea in children

The Braden Scale for Predicting Pressure Sore Risk was developed to foster early identification of patients at risk for forming pressure sores. The scale is composed of six subscales that reflect sensory perception, skin moisture, activity, mobility, friction and shear, and nutritional status. Content and construct validity were established by expert opinion and empirical testing. Three studies of reliability are reported here, using raters who varied in level of educational preparation and geographic region. Two prospective studies of predictive validity were completed to determine the scale's sensitivity and specificity. Reliability ranged from r = .83 to r = .94 for nurses' aides and licensed practical nurses; when used by registered nurses, the reliability increased to r = .99. Predictive validity was calculated for each cut-off point of the scale. Using a cut-off point of 16, sensitivity was 100% in both studies. Specificity ranged from 64% to 90%. This instrument has highly satisfactory reliability when used by RNs, and greater sensitivity and specificity than instruments previously reported.

Antoinette Laguzza

Papers

The Braden Scale for Predicting Pressure Sore Risk.