Background

Readmissions after hospitalization for pneumonia are common, but the few risk-prediction models have poor to modest predictive ability. Data routinely collected in the electronic health record (EHR) may improve prediction.

Objective

To develop pneumonia-specific readmission risk-prediction models using EHR data from the first day and from the entire hospital stay ("full stay").

Design

Observational cohort study using stepwise-backward selection and cross-validation.

Subjects

Consecutive pneumonia hospitalizations from 6 diverse hospitals in north Texas from 2009-2010.

Measures

All-cause nonelective 30-day readmissions, ascertained from 75 regional hospitals.

Results

Of 1463 patients, 13.6% were readmitted. The first-day pneumonia-specific model included sociodemographic factors, prior hospitalizations, thrombocytosis, and a modified pneumonia severity index; the full-stay model included disposition status, vital sign instabilities on discharge, and an updated pneumonia severity index calculated using values from the day of discharge as additional predictors. The full-stay pneumonia-specific model outperformed the first-day model (C statistic 0.731 vs 0.695; P = 0.02; net reclassification index = 0.08). Compared to a validated multi-condition readmission model, the Centers for Medicare and Medicaid Services pneumonia model, and 2 commonly used pneumonia severity of illness scores, the full-stay pneumonia-specific model had better discrimination (C statistic range 0.604-0.681; P < 0.01 for all comparisons), predicted a broader range of risk, and better reclassified individuals by their true risk (net reclassification index range, 0.09-0.18).

Conclusions

EHR data collected from the entire hospitalization can accurately predict readmission risk among patients hospitalized for pneumonia. This approach outperforms a first-day pneumonia-specific model, the Centers for Medicare and Medicaid Services pneumonia model, and 2 commonly used pneumonia severity of illness scores. Journal of Hospital Medicine 2017;12:209-216.

Rationale

Predicting which patients are at highest risk for readmission after hospitalization for pneumonia could enable hospitals to proactively reallocate scarce resources to reduce 30-day readmissions.

Objectives

To synthesize the available literature on readmission risk prediction models for adults who are hospitalized because of pneumonia and describe their performance.

Methods

We systematically searched Ovid MEDLINE, Embase, The Cochrane Library, and Cumulative Index to Nursing and Allied Health Literature databases from inception through July 2015. We included studies of adults discharged with pneumonia that developed or validated a model that predicted hospital readmission. Two independent reviewers abstracted data and assessed the risk of bias.

Measurements and main results

Of 992 citations reviewed, 7 studies met inclusion criteria, which included 11 unique risk prediction models. All-cause 30-day readmission rates ranged from 11.8 to 20.8% (median, 17.3%). Model discrimination (C statistic) ranged from 0.59 to 0.77 (median, 0.63) with the highest-quality, best-validated model, the Centers for Medicare and Medicaid Services Pneumonia Administrative Model performing modestly (C Statistic of 0.63 in 4 separate multicenter cohorts). The best performing model (C statistic of 0.77) was a single-site study that lacked internal validation. The models had adequate calibration, with patients predicted as high risk for readmission having a higher average observed readmission rate than those predicted to be low risk. None of the studies included pneumonia illness severity scores, and only one included measures of in-hospital clinical trajectory and stability on discharge, robust predictors of readmission.

Conclusions

We found a limited number of validated pneumonia-specific readmission models, and their predictive ability was modest. To improve predictive accuracy, future models should include measures of pneumonia illness severity, hospital complications, and stability on discharge.