Dataset Information

Predicting Discharge Dates From the NICU Using Progress Note Data.

ABSTRACT: Discharging patients from the NICU may be delayed for nonmedical reasons including the need for medical equipment, parental education, and children's services. We describe a method to predict which patients will be medically ready for discharge in the next 2 to 10 days, providing lead time to address nonmedical reasons for delayed discharge.A retrospective study examined 26 features (17 extracted, 9 engineered) from daily progress notes of 4693 patients (103,206 patient-days) from the NICU of a large, academic children's hospital. These data were used to develop a supervised machine learning problem to predict days to discharge (DTD). Random forest classifiers were trained by using examined features and International Classification of Diseases, Ninth Revision-based subpopulations to determine the most important features.Three of the 4 subpopulations (premature, cardiac, gastrointestinal surgery) and all patients combined performed similarly at 2, 4, 7, and 10 DTD with area under the curve (AUC) ranging from 0.854 to 0.865 at 2 DTD and 0.723 to 0.729 at 10 DTD. Patients undergoing neurosurgery performed worse at every DTD measure, scoring 0.749 at 2 DTD and 0.614 at 10 DTD. This model was also able to identify important features and provide "rule-of-thumb" criteria for patients close to discharge. By using DTD equal to 4 and 2 features (oral percentage of feedings and weight), we constructed a model with an AUC of 0.843.Using clinical features from daily progress notes provides an accurate method to predict when patients in the NICU are nearing discharge.

SUBMITTER: Temple MW

PROVIDER: S-EPMC5524203 | biostudies-literature | 2015 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting Discharge Dates From the NICU Using Progress Note Data.

Temple Michael W MW Lehmann Christoph U CU Fabbri Daniel D

Pediatrics 20150801 2

<h4>Background and objectives</h4>Discharging patients from the NICU may be delayed for nonmedical reasons including the need for medical equipment, parental education, and children's services. We describe a method to predict which patients will be medically ready for discharge in the next 2 to 10 days, providing lead time to address nonmedical reasons for delayed discharge.<h4>Methods</h4>A retrospective study examined 26 features (17 extracted, 9 engineered) from daily progress notes of 4693 p ...[more]

PMID: 26216319

Similar Datasets

Project description:Background: Acute heart failure (AHF) represents a leading cause of unscheduled hospital stays, frequent rehospitalisations, and mortality worldwide. The aim of our study was to develop a bedside prognostic tool, a multivariable predictive risk score, that is useful in daily practice, thus providing an early prognostic evaluation at admission and an accurate risk stratification after discharge in patients with AHF. Methods: This study is a subanalysis of the STADE HF study, which is a single-centre, prospective, randomised controlled trial enrolling 123 patients admitted to hospital for AHF. Here, 117 patients were included in the analysis, due to data exhaustivity. Regression analysis was performed to determine predictive variables for one-year mortality and/or rehospitalisation after discharge. Results: During the first year after discharge, 23 patients died. After modellisation, the variables considered to be of prognostic relevance in terms of mortality were (1) non-ischaemic aetiology of HF, (2) elevated creatinine levels at admission, (3) moderate/severe mitral regurgitation, and (4) prior HF hospitalisation. We designed a linear model based on these four independent predictive variables, and it showed a good ability to score and predict patient mortality with an AUC of 0.84 (95%CI: 0.76-0.92), thus denoting a high discriminative ability. A risk score equation was developed. During the first year after discharge, we observed as well that 41 patients died or were rehospitalised; hence, while searching for a model that could predict worsening health conditions (i.e., death and/or rehospitalisation), only two predictive variables were identified: non-ischaemic HF aetiology and previous HF hospitalisation (also included in the one-year mortality model). This second modellisation showed a more discrete discriminative ability with an AUC of 0.67 (95% C.I. 0.59-0.77). Conclusions: The proposed risk score and model, based on readily available predictive variables, are promising and useful tools to assess, respectively, the one-year mortality risk and the one-year mortality and/or rehospitalisations in patients hospitalised for AHF and to assist clinicians in the management of patients with HF aiming at improving their prognosis.

Dataset Information

Predicting Discharge Dates From the NICU Using Progress Note Data.

Publications

Predicting Discharge Dates From the NICU Using Progress Note Data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets