Development and Validation of an Age-Risk Score for Mortality Predication after Thermal Injury

Abstract

Background: In burn patients, the risk of mortality typically decreases as children mature, reaches a nadir at age 21, rises linearly thereafter, and levels off in old age. We hypothesized that a single age-risk score (AGESCORE), incorporating a cubic functional form, can be used in predictive models for mortality after burns. Methods: Data from 6,395 thermally injured patients admitted to a single burn center between January 1, 1950, and December 31, 1999, were used. Variables included age, total burn size, year of discharge, and survival. AGESCORE was defined as follows: -5(age) + 14(age2 /100) -7(age3 /10,000). Logistic regression verified the cubic functional form of the age- mortality relationship. Models using a general cubic functional form of age, and AGESCORE, were compared for lack of fit. The stability of AGESCORE was assessed over six distinct treatment eras within the 50-year period. AGESCORE was also validated using data from a different burn center. Results: AGESCORE provided an accurate method for modeling mortality in burn patients across different age groups, burn sizes, eras, and burn centers. Conclusion: The benefits of a stan- dardized index of age risk include ease of comparison, reduction of bias, and in- creased efficiency attributable to statistical parsimony. The applicability of this approach to nonthermal trauma petients remains to be seen.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 2005
Accession Number
ADA628085

Entities

People

  • Aimee R. Moreau
  • Arthur D. Mason Jr.
  • Leopoldo C. Cancio
  • Peter H. Westfall

Organizations

  • United States Army Institute of Surgical Research

Tags

DTIC Thesaurus Topics

  • Age Groups
  • Burns
  • Combat Casualty Care
  • Data Sets
  • Dermatologic Agents
  • Health Services
  • Infection
  • Infection Control
  • Intensive Care Units
  • Patient Care
  • Regression Analysis
  • Risk Factors
  • Statistical Analysis
  • United States
  • Validation
  • Wound Infections
  • Wounds And Injuries

Fields of Study

  • Medicine

Readers

  • Computational Modeling and Simulation
  • Mathematics or Statistics
  • Trauma or Military Medicine