Estimating the distribution of external causes in hospital data from injury diagnosis

Kavi Bhalla, Saeid Shahraz, Mohsen Naghavi, Rafael Lozano, Christopher Murray

Research output: Contribution to journalArticle

Abstract

Hospital discharge datasets are a key source for estimating the incidence of non-fatal injuries. While hospital records usually document injury diagnosis (e.g. traumatic brain injury, femur fracture, etc.) accurately, they often contain poor quality information on external causes (e.g. road traffic crashes, falls, fires, etc.), if such data is recorded at all. However, estimating incidence by external causes is essential for designing effective prevention strategies. Thus, we developed a method for estimating the number of hospital admissions due to each external cause based on injury diagnosis. We start with a prior probability distribution of external causes for each case (based on victim age and sex) and use Bayesian inference to update the probabilities based on the victim's injury diagnoses. We validate the method on a trial dataset in which both external causes and injury diagnoses are known and demonstrate application to two problems: redistribution of cases classified to ill-defined external causes in one hospital data system; and, estimation of external causes in another hospital data system that only records nature of injuries. In comparison with age-sex proportional distribution (the method usually employed), we found the Bayesian method to be a significant improvement for generating estimates of incidence for many external causes (e.g. fires, drownings, poisonings). But the method, performed poorly in distinguishing between falls and road traffic injuries, both of which are characterized by similar injury codes in our datasets. While such stop gap methods can help derive additional information, hospitals need to incorporate accurate external cause coding in routine record keeping.

Original languageEnglish (US)
Pages (from-to)1822-1829
Number of pages8
JournalAccident Analysis and Prevention
Volume40
Issue number6
DOIs
StatePublished - Nov 2008
Externally publishedYes

Fingerprint

cause
Wounds and Injuries
Fires
incidence
Information Systems
road traffic
Incidence
Probability distributions
Sex Distribution
Bayes Theorem
Hospital Records
Brain
Femur
Poisoning
redistribution
coding
brain
Datasets

Keywords

  • Bayesian inference
  • Hospital discharge
  • ICD
  • Injury surveillance
  • Missing data
  • Non-fatal injuries

ASJC Scopus subject areas

  • Public Health, Environmental and Occupational Health
  • Safety, Risk, Reliability and Quality
  • Human Factors and Ergonomics
  • Law
  • Medicine(all)

Cite this

Estimating the distribution of external causes in hospital data from injury diagnosis. / Bhalla, Kavi; Shahraz, Saeid; Naghavi, Mohsen; Lozano, Rafael; Murray, Christopher.

In: Accident Analysis and Prevention, Vol. 40, No. 6, 11.2008, p. 1822-1829.

Research output: Contribution to journalArticle

Bhalla, Kavi ; Shahraz, Saeid ; Naghavi, Mohsen ; Lozano, Rafael ; Murray, Christopher. / Estimating the distribution of external causes in hospital data from injury diagnosis. In: Accident Analysis and Prevention. 2008 ; Vol. 40, No. 6. pp. 1822-1829.
@article{88ef31884d4d4999a406897031baa38a,
title = "Estimating the distribution of external causes in hospital data from injury diagnosis",
abstract = "Hospital discharge datasets are a key source for estimating the incidence of non-fatal injuries. While hospital records usually document injury diagnosis (e.g. traumatic brain injury, femur fracture, etc.) accurately, they often contain poor quality information on external causes (e.g. road traffic crashes, falls, fires, etc.), if such data is recorded at all. However, estimating incidence by external causes is essential for designing effective prevention strategies. Thus, we developed a method for estimating the number of hospital admissions due to each external cause based on injury diagnosis. We start with a prior probability distribution of external causes for each case (based on victim age and sex) and use Bayesian inference to update the probabilities based on the victim's injury diagnoses. We validate the method on a trial dataset in which both external causes and injury diagnoses are known and demonstrate application to two problems: redistribution of cases classified to ill-defined external causes in one hospital data system; and, estimation of external causes in another hospital data system that only records nature of injuries. In comparison with age-sex proportional distribution (the method usually employed), we found the Bayesian method to be a significant improvement for generating estimates of incidence for many external causes (e.g. fires, drownings, poisonings). But the method, performed poorly in distinguishing between falls and road traffic injuries, both of which are characterized by similar injury codes in our datasets. While such stop gap methods can help derive additional information, hospitals need to incorporate accurate external cause coding in routine record keeping.",
keywords = "Bayesian inference, Hospital discharge, ICD, Injury surveillance, Missing data, Non-fatal injuries",
author = "Kavi Bhalla and Saeid Shahraz and Mohsen Naghavi and Rafael Lozano and Christopher Murray",
year = "2008",
month = "11",
doi = "10.1016/j.aap.2008.07.002",
language = "English (US)",
volume = "40",
pages = "1822--1829",
journal = "Accident Analysis and Prevention",
issn = "0001-4575",
publisher = "Elsevier Limited",
number = "6",

}

TY - JOUR

T1 - Estimating the distribution of external causes in hospital data from injury diagnosis

AU - Bhalla, Kavi

AU - Shahraz, Saeid

AU - Naghavi, Mohsen

AU - Lozano, Rafael

AU - Murray, Christopher

PY - 2008/11

Y1 - 2008/11

N2 - Hospital discharge datasets are a key source for estimating the incidence of non-fatal injuries. While hospital records usually document injury diagnosis (e.g. traumatic brain injury, femur fracture, etc.) accurately, they often contain poor quality information on external causes (e.g. road traffic crashes, falls, fires, etc.), if such data is recorded at all. However, estimating incidence by external causes is essential for designing effective prevention strategies. Thus, we developed a method for estimating the number of hospital admissions due to each external cause based on injury diagnosis. We start with a prior probability distribution of external causes for each case (based on victim age and sex) and use Bayesian inference to update the probabilities based on the victim's injury diagnoses. We validate the method on a trial dataset in which both external causes and injury diagnoses are known and demonstrate application to two problems: redistribution of cases classified to ill-defined external causes in one hospital data system; and, estimation of external causes in another hospital data system that only records nature of injuries. In comparison with age-sex proportional distribution (the method usually employed), we found the Bayesian method to be a significant improvement for generating estimates of incidence for many external causes (e.g. fires, drownings, poisonings). But the method, performed poorly in distinguishing between falls and road traffic injuries, both of which are characterized by similar injury codes in our datasets. While such stop gap methods can help derive additional information, hospitals need to incorporate accurate external cause coding in routine record keeping.

AB - Hospital discharge datasets are a key source for estimating the incidence of non-fatal injuries. While hospital records usually document injury diagnosis (e.g. traumatic brain injury, femur fracture, etc.) accurately, they often contain poor quality information on external causes (e.g. road traffic crashes, falls, fires, etc.), if such data is recorded at all. However, estimating incidence by external causes is essential for designing effective prevention strategies. Thus, we developed a method for estimating the number of hospital admissions due to each external cause based on injury diagnosis. We start with a prior probability distribution of external causes for each case (based on victim age and sex) and use Bayesian inference to update the probabilities based on the victim's injury diagnoses. We validate the method on a trial dataset in which both external causes and injury diagnoses are known and demonstrate application to two problems: redistribution of cases classified to ill-defined external causes in one hospital data system; and, estimation of external causes in another hospital data system that only records nature of injuries. In comparison with age-sex proportional distribution (the method usually employed), we found the Bayesian method to be a significant improvement for generating estimates of incidence for many external causes (e.g. fires, drownings, poisonings). But the method, performed poorly in distinguishing between falls and road traffic injuries, both of which are characterized by similar injury codes in our datasets. While such stop gap methods can help derive additional information, hospitals need to incorporate accurate external cause coding in routine record keeping.

KW - Bayesian inference

KW - Hospital discharge

KW - ICD

KW - Injury surveillance

KW - Missing data

KW - Non-fatal injuries

UR - http://www.scopus.com/inward/record.url?scp=54949127860&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=54949127860&partnerID=8YFLogxK

U2 - 10.1016/j.aap.2008.07.002

DO - 10.1016/j.aap.2008.07.002

M3 - Article

C2 - 19068282

AN - SCOPUS:54949127860

VL - 40

SP - 1822

EP - 1829

JO - Accident Analysis and Prevention

JF - Accident Analysis and Prevention

SN - 0001-4575

IS - 6

ER -