Dataset Information

Predicting Patients at Risk for Prolonged Hospital Stays.

ABSTRACT:

Background

Patients with prolonged hospitalizations account for 14% of all hospital days in US hospitals. Predicting which medical patients are at risk for prolonged hospitalizations would allow early proactive management to reduce their length of stay.

Methods

Using the National Inpatient Sample, we examined risk factors for prolonged hospitalizations among adults hospitalized on the medicine service in 2014. We defined prolonged hospitalizations as those lasting 21 days or longer. We divided the sample into derivation and validation sets, and used logistic regression to identify significant risk factors in the derivation set, which were validated in the validation set. We used the estimates from the model to derive a risk score for prolonged hospitalizations.

Results

Our sample included 2,997,249 hospitalizations (median age of 66 y, 53.5% female). 1.2% of hospitalizations were 21 days or longer. Patients with prolonged hospitalizations were younger, and had a greater number of chronic diseases. A prolonged hospitalization risk score, derived from the many significant predictors in our model, performed well in discriminating between prolonged and nonprolonged hospitalizations, with c-statistics of 0.80 in both the derivation and validation sets.

Conclusions

Our predictive model using readily available administrative data was able to discriminate between prolonged and nonprolonged hospitalizations in a national sample of medical patients, and performed well on internal validation. If prospectively validated, such a tool could be of use to hospitals and researchers interested in targeting development, testing, and/or deployment of programs to reduce length of stay.

SUBMITTER: Doctoroff L

PROVIDER: S-EPMC7444462 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundThe increasing number of patients treated with peritoneal dialysis (PD) and their consistently high rate of hospital admissions have placed a large burden on the health care system. Early clinical interventions and optimal management of patients at a high risk of prolonged length of stay (pLOS) may help improve the medical efficiency and prognosis of PD-treated patients. If timely clinical interventions are not provided, patients at a high risk of pLOS may face a poor prognosis and high medical expenses, which will also be a burden on hospitals. Therefore, physicians need an effective pLOS prediction model for PD-treated patients.ObjectiveThis study aimed to develop an optimal data-driven model for predicting the pLOS risk of PD-treated patients using basic admission data.MethodsPatient data collected using the Hospital Quality Monitoring System (HQMS) in China were used to develop pLOS prediction models. A stacking model was constructed with support vector machine, random forest (RF), and K-nearest neighbor algorithms as its base models and traditional logistic regression (LR) as its meta-model. The meta-model used the outputs of all 3 base models as input and generated the output of the stacking model. Another LR-based pLOS prediction model was built as the benchmark model. The prediction performance of the stacking model was compared with that of its base models and the benchmark model. Five-fold cross-validation was employed to develop and validate the models. Performance measures included the Brier score, area under the receiver operating characteristic curve (AUROC), estimated calibration index (ECI), accuracy, sensitivity, specificity, and geometric mean (Gm). In addition, a calibration plot was employed to visually demonstrate the calibration power of each model.ResultsThe final cohort extracted from the HQMS database consisted of 23,992 eligible PD-treated patients, among whom 30.3% had a pLOS (ie, longer than the average LOS, which was 16 days in our study). Among the models, the stacking model achieved the best calibration (ECI 8.691), balanced accuracy (Gm 0.690), accuracy (0.695), and specificity (0.701). Meanwhile, the stacking and RF models had the best overall performance (Brier score 0.174 for both) and discrimination (AUROC 0.757 for the stacking model and 0.756 for the RF model). Compared with the benchmark LR model, the stacking model was superior in all performance measures except sensitivity, but there was no significant difference in sensitivity between the 2 models. The 2-sided t tests revealed significant performance differences between the stacking and LR models in overall performance, discrimination, calibration, balanced accuracy, and accuracy.ConclusionsThis study is the first to develop data-driven pLOS prediction models for PD-treated patients using basic admission data from a national database. The results indicate the feasibility of utilizing a stacking-based pLOS prediction model for PD-treated patients. The pLOS prediction tools developed in this study have the potential to assist clinicians in identifying patients at a high risk of pLOS and to allocate resources optimally for PD-treated patients.