Dataset Information

Using electronic health record data to develop inpatient mortality predictive model: Acute Laboratory Risk of Mortality Score (ALaRMS).

ABSTRACT: OBJECTIVE: Using numeric laboratory data and administrative data from hospital electronic health record (EHR) systems, to develop an inpatient mortality predictive model. METHODS: Using EHR data of 1,428,824 adult discharges from 70 hospitals in 2006-2007, we developed the Acute Laboratory Risk of Mortality Score (ALaRMS) using age, gender, and initial laboratory values on admission as candidate variables. We then added administrative variables using the Agency for Healthcare Research and Quality (AHRQ)'s clinical classification software (CCS) and comorbidity software (CS) as disease classification tools. We validated the model using 770,523 discharges in 2008. RESULTS: Mortality predictors with ORs >2.00 included age, deranged albumin, arterial pH, bands, blood urea nitrogen, oxygen partial pressure, platelets, pro-brain natriuretic peptide, troponin I, and white blood cell counts. The ALaRMS model c-statistic was 0.87. Adding the CCS and CS variables increased the c-statistic to 0.91. The relative contributions were 69% (ALaRMS), 25% (CCS), and 6% (CS). Furthermore, the integrated discrimination improvement statistic demonstrated a 127% (95% CI 122% to 133%) overall improvement when ALaRMS was added to CCS and CS variables. In contrast, only a 22% (CI 19% to 25%) improvement was seen when CCS and CS variables were added to ALaRMS. CONCLUSIONS: EHR data can generate clinically plausible mortality predictive models with excellent discrimination. ALaRMS uses automated laboratory data widely available on admission, providing opportunities to aid real-time decision support. Models that incorporate laboratory and AHRQ's CCS and CS variables have utility for risk adjustment in retrospective outcome studies.

SUBMITTER: Tabak YP

PROVIDER: S-EPMC3994855 | biostudies-other | 2014 May-Jun

REPOSITORIES: biostudies-other

ACCESS DATA

Publications

Using electronic health record data to develop inpatient mortality predictive model: Acute Laboratory Risk of Mortality Score (ALaRMS).

Tabak Ying P YP Sun Xiaowu X Nunez Carlos M CM Johannes Richard S RS

Journal of the American Medical Informatics Association : JAMIA 20131004 3

<h4>Objective</h4>Using numeric laboratory data and administrative data from hospital electronic health record (EHR) systems, to develop an inpatient mortality predictive model.<h4>Methods</h4>Using EHR data of 1,428,824 adult discharges from 70 hospitals in 2006-2007, we developed the Acute Laboratory Risk of Mortality Score (ALaRMS) using age, gender, and initial laboratory values on admission as candidate variables. We then added administrative variables using the Agency for Healthcare Resear ...[more]

PMID: 24097807

Similar Datasets

Project description:ImportanceMany health systems are considering increasing price transparency at the time of order entry. However, evidence of its impact on clinician ordering behavior is inconsistent and limited to single-site evaluations of shorter duration.ObjectiveTo test the effect of displaying Medicare allowable fees for inpatient laboratory tests on clinician ordering behavior over 1 year.Design, setting, and participantsThe Pragmatic Randomized Introduction of Cost data through the electronic health record (PRICE) trial was a randomized clinical trial comparing a 1-year intervention to a 1-year preintervention period, and adjusting for time trends and patient characteristics. The trial took place at 3 hospitals in Philadelphia between April 2014 and April 2016 and included 98 529 patients comprising 142 921 hospital admissions.InterventionsInpatient laboratory test groups were randomly assigned to display Medicare allowable fees (30 in intervention) or not (30 in control) in the electronic health record.Main outcomes and measuresPrimary outcome was the number of tests ordered per patient-day. Secondary outcomes were tests performed per patient-day and Medicare associated fees.ResultsThe sample included 142 921 hospital admissions representing patients who were 51.9% white (74 165), 38.9% black (55 526), and 56.9% female (81 291) with a mean (SD) age of 54.7 (19.0) years. Preintervention trends of order rates among the intervention and control groups were similar. In adjusted analyses of the intervention group compared with the control group over time, there were no significant changes in overall test ordering behavior (0.05 tests ordered per patient-day; 95% CI, -0.002 to 0.09; P = .06) or associated fees ($0.24 per patient-day; 95% CI, -$0.42 to $0.91; P = .47). Exploratory subset analyses found small but significant differences in tests ordered per patient-day based on patient intensive care unit (ICU) stay (patients with ICU stay: -0.16; 95% CI, -0.31 to -0.01; P = .04; patients without ICU stay: 0.13; 95% CI, 0.08-0.17; P < .001) and the magnitude of associated fees (top quartile of tests based on fee value: -0.01; 95% CI, -0.02 to -0.01; P = .04; bottom quartile: 0.03; 95% CI, 0.002-0.06; P = .04). Adjusted analyses of tests that were performed found a small but significant overall increase in the intervention group relative to the control group over time (0.08 tests performed per patient day, 95% CI, 0.03-0.12; P < .001).Conclusions and relevanceDisplaying Medicare allowable fees for inpatient laboratory tests did not lead to a significant change in overall clinician ordering behavior or associated fees.Trial registrationclinicaltrials.gov Identifier: NCT02355496.

Project description:BackgroundAcute heart failure (AHF) is associated with significant morbidity and mortality. Effective patient risk stratification is essential to guiding hospitalization decisions and the clinical management of AHF. Clinical decision support systems can be used to improve predictions of mortality made in emergency care settings for the purpose of AHF risk stratification. In this study, several models for the prediction of seven-day mortality among AHF patients were developed by applying machine learning techniques to retrospective patient data from 236,275 total emergency department (ED) encounters, 1881 of which were considered positive for AHF and were used for model training and testing. The models used varying subsets of age, sex, vital signs, and laboratory values. Model performance was compared to the Emergency Heart Failure Mortality Risk Grade (EHMRG) model, a commonly used system for prediction of seven-day mortality in the ED with similar (or, in some cases, more extensive) inputs. Model performance was assessed in terms of area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity.ResultsWhen trained and tested on a large academic dataset, the best-performing model and EHMRG demonstrated test set AUROCs of 0.84 and 0.78, respectively, for prediction of seven-day mortality. Given only measurements of respiratory rate, temperature, mean arterial pressure, and FiO2, one model produced a test set AUROC of 0.83. Neither a logistic regression comparator nor a simple decision tree outperformed EHMRG.ConclusionsA model using only the measurements of four clinical variables outperforms EHMRG in the prediction of seven-day mortality in AHF. With these inputs, the model could not be replaced by logistic regression or reduced to a simple decision tree without significant performance loss. In ED settings, this minimal-input risk stratification tool may assist clinicians in making critical decisions about patient disposition by providing early and accurate insights into individual patient's risk profiles.

Project description: To assess the short term association of inpatient implementation of electronic health records (EHRs) with patient outcomes of mortality, readmissions, and adverse safety events. Observational study with difference-in-differences analysis. Medicare, 2011-12. Patients admitted to 17 study hospitals with a verifiable "go live" date for implementation of inpatient EHRs during 2011-12, and 399 control hospitals in the same hospital referral region. All cause readmission within 30 days of discharge, all cause mortality within 30 days of admission, and adverse safety events as defined by the patient safety for selected indicators (PSI)-90 composite measure among Medicare beneficiaries admitted to one of these hospitals 90 days before and 90 days after implementation of the EHRs (n=28?235 and 26?453 admissions), compared with the control group of all contemporaneous admissions to hospitals in the same hospital referral region (n=284?632 and 276?513 admissions). Analyses were adjusted for beneficiaries' sociodemographic and clinical characteristics. Before and after implementation, characteristics of admissions were similar in both study and control hospitals. Among study hospitals, unadjusted 30 day mortality (6.74% to 7.15%, P=0.06) and adverse safety event rates (10.5 to 11.4 events per 1000 admissions, P=0.34) did not significantly change after implementation of EHRs. There was an unadjusted decrease in 30 day readmission rates, from 19.9% to 19.0% post-implementation (P=0.02). In difference-in-differences analysis, however, there was no significant change in any outcome between pre-implementation and post-implementation periods (all P?0.13). Despite concerns that implementation of EHRs might adversely impact patient care during the acute transition period, we found no overall negative association of such implementation on short term inpatient mortality, adverse safety events, or readmissions in the Medicare population across 17 US hospitals.

Dataset Information

Using electronic health record data to develop inpatient mortality predictive model: Acute Laboratory Risk of Mortality Score (ALaRMS).

Publications

Using electronic health record data to develop inpatient mortality predictive model: Acute Laboratory Risk of Mortality Score (ALaRMS).

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure