Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 21, 2022AnesthesiologyOpen Access

Risk Stratification Index 3.0, a Broad Set of Models for Predicting Adverse Events during and after Hospital Admission

View Full Paper
Ask AI
Bookmark
Share

Key result

Predictive analytical modeling based on administrative claims history achieved an average area under the curve of 0.76 (range, 0.70 to 0.82) for predicting adverse events and care utilization.

Why the study?

Can predictive analytical modeling based on administrative claims history accurately predict adverse events and care utilization outcomes for hospitalized patients?

Population

9,085,968 development and 5,336,265 validation Medicare patients with inpatient admissions

Comparison

Logistic regression vs five commonly used machine learning methods

Design

Retrospective model development and prospective validation study

Follow-up

Through 90 days after admission

Authors

SGScott GreenwaldGCGeorge F. ChamounNCNassib G. Chamoun

Discussion

Loading...

Member takes

Overview

May aid perioperative risk stratification using claims data; hypothesis-generating pending prospective validation.

Study Design

Type

Observational (n=14,422,233)

Structured PICO

Can predictive analytical modeling based on administrative claims history accurately predict adverse events and care utilization outcomes for hospitalized patients?

P
Population
14,422,233 patients with Medicare inpatient admissions between 2017 and 2019 used to develop and validate predictive models for adverse events up to 90 days post-admission.
E
Exposure
Risk Stratification Index 3.0 (predictive models based on ICD-10 diagnostic and procedural codes and patient demographic/coding history in the year before admission)
C
Comparator
Comparison between logistic regression and five commonly used machine learning methods
O
Outcome
Prediction of unplanned hospital admissions, discharge status, excess length of stay, in-hospital and 90-day mortality, acute kidney injury, sepsis, pneumonia, respiratory failure, and a composite of major cardiac complications

Main Result

Effect estimate: average AUC 0.76

Predictive analytical modeling based on administrative claims history can provide individualized risk profiles at hospital admission with good accuracy (average AUC 0.76).

Cite This Study

Greenwald et al. (2022) conducted an observational in Hospitalized patients (n=14,422,233). Risk Stratification Index 3.0 (predictive analytical modeling based on administrative claims) vs. Machine learning methods was evaluated on Adverse events and care utilization outcomes (including mortality, readmissions, and complications) (average AUC 0.76). Predictive analytical modeling based on administrative claims history achieved an average area under the curve of 0.76 (range, 0.70 to 0.82) for predicting adverse events and care utilization.

synapsesocial.com/papers/6a7c71dd0449e558d7c2f441https://doi.org/10.1097/aln.0000000000004380
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Broadly Applicable Risk Stratification System for Predicting Duration of Hospitalization and Mortality2010 · 109 citations
  2. 2Comparison of an Updated Risk Stratification Index to Hierarchical Condition Categories2017 · 15 citations
  3. 3Deficit Accumulation and Phenotype Assessments of Frailty Both Poorly Predict Duration of Hospitalization and Serious Complications after Noncardiac Surgery2019 · 29 citations
  4. 4Machine learning methods for perioperative anesthetic management in cardiac surgery patients: a scoping review2021 · 16 citations
  5. 5Outcomes Measures and Risk Adjustment2013 · 52 citations