Predicting inpatient costs with admitting clinical data

William M. Tierney, John F. Fitzgerald, Michael E. Miller, Margaret K. James, Clement J. McDonald

Research output: Contribution to journalArticle

38 Citations (Scopus)

Abstract

Hospital cost-containment programs should themselves be cost-effective, targeting high-cost physicians (which requires adjusting for case mix) and patients (which requires early identification). In this study, clinical data available within 24 hours of admission from an electronic medical record system were used to develop statistical models to predict hospital costs. In this retrospective analysis of clinical data and diagnosis-related groups (DRGs), study subjects were 2,355 patients admitted for at least 1 day to the medicine service at an urban teaching hospital with sophisticated electronic medical records. Of these 2,355 patients, 1,663 (71%) had one of the 41 most common DRGs. Predictive models were derived on a random subset of two thirds of the patients and were validated on the remaining third. The following patient data were obtained: admission and prior diagnostic test results, diagnoses, vital signs; demographic data; prior inpatient and outpatient visits; tests and treatments ordered within 24 hours of admission (discretionary data); DRGs; and total inpatient costs (estimated from charges). Diagnosis-related groups explained 24% of the variance in total costs in the derivation patient set and 16% in the validation set. When only nondiscretionary data were used, the models retained only clinical laboratory results and prior diagnoses, explaining 20% of the derivation set variance in total costs and 16% in the validation set. Adding DRGs increased the variance explained in the derivation set to 34%, but decreased to 24% in the validation set. Adding discretionary data substantially increased the explained variance in the derivation and validation patient sets. The models’ median predicted costs underestimated true costs by 10% to 13%, with the lowest error in the models using all types of variables. Clinical data gathered during routine clinical care can be used to adjust for case mix and identify high-cost patients early in their hospital stays, when they could be targeted by cost-containment interventions.

Original languageEnglish (US)
Pages (from-to)1-14
Number of pages14
JournalMedical care
Volume33
Issue number1
DOIs
StatePublished - Jan 1995

Fingerprint

Diagnosis-Related Groups
Inpatients
Costs and Cost Analysis
costs
Cost Control
Hospital Costs
Electronic Health Records
cost containment
Group
electronics
Vital Signs
Urban Hospitals
Statistical Models
predictive model
Routine Diagnostic Tests
Teaching Hospitals
study group
Length of Stay
Outpatients
diagnostic

Keywords

  • Case mix
  • Cost containment
  • Database research

ASJC Scopus subject areas

  • Public Health, Environmental and Occupational Health
  • Nursing(all)
  • Health(social science)
  • Health Professions(all)

Cite this

Tierney, W. M., Fitzgerald, J. F., Miller, M. E., James, M. K., & McDonald, C. J. (1995). Predicting inpatient costs with admitting clinical data. Medical care, 33(1), 1-14. https://doi.org/10.1097/00005650-199501000-00001

Predicting inpatient costs with admitting clinical data. / Tierney, William M.; Fitzgerald, John F.; Miller, Michael E.; James, Margaret K.; McDonald, Clement J.

In: Medical care, Vol. 33, No. 1, 01.1995, p. 1-14.

Research output: Contribution to journalArticle

Tierney, WM, Fitzgerald, JF, Miller, ME, James, MK & McDonald, CJ 1995, 'Predicting inpatient costs with admitting clinical data', Medical care, vol. 33, no. 1, pp. 1-14. https://doi.org/10.1097/00005650-199501000-00001
Tierney WM, Fitzgerald JF, Miller ME, James MK, McDonald CJ. Predicting inpatient costs with admitting clinical data. Medical care. 1995 Jan;33(1):1-14. https://doi.org/10.1097/00005650-199501000-00001
Tierney, William M. ; Fitzgerald, John F. ; Miller, Michael E. ; James, Margaret K. ; McDonald, Clement J. / Predicting inpatient costs with admitting clinical data. In: Medical care. 1995 ; Vol. 33, No. 1. pp. 1-14.
@article{d9c83ff7ec194423a2d38fcddd627b77,
title = "Predicting inpatient costs with admitting clinical data",
abstract = "Hospital cost-containment programs should themselves be cost-effective, targeting high-cost physicians (which requires adjusting for case mix) and patients (which requires early identification). In this study, clinical data available within 24 hours of admission from an electronic medical record system were used to develop statistical models to predict hospital costs. In this retrospective analysis of clinical data and diagnosis-related groups (DRGs), study subjects were 2,355 patients admitted for at least 1 day to the medicine service at an urban teaching hospital with sophisticated electronic medical records. Of these 2,355 patients, 1,663 (71{\%}) had one of the 41 most common DRGs. Predictive models were derived on a random subset of two thirds of the patients and were validated on the remaining third. The following patient data were obtained: admission and prior diagnostic test results, diagnoses, vital signs; demographic data; prior inpatient and outpatient visits; tests and treatments ordered within 24 hours of admission (discretionary data); DRGs; and total inpatient costs (estimated from charges). Diagnosis-related groups explained 24{\%} of the variance in total costs in the derivation patient set and 16{\%} in the validation set. When only nondiscretionary data were used, the models retained only clinical laboratory results and prior diagnoses, explaining 20{\%} of the derivation set variance in total costs and 16{\%} in the validation set. Adding DRGs increased the variance explained in the derivation set to 34{\%}, but decreased to 24{\%} in the validation set. Adding discretionary data substantially increased the explained variance in the derivation and validation patient sets. The models’ median predicted costs underestimated true costs by 10{\%} to 13{\%}, with the lowest error in the models using all types of variables. Clinical data gathered during routine clinical care can be used to adjust for case mix and identify high-cost patients early in their hospital stays, when they could be targeted by cost-containment interventions.",
keywords = "Case mix, Cost containment, Database research",
author = "Tierney, {William M.} and Fitzgerald, {John F.} and Miller, {Michael E.} and James, {Margaret K.} and McDonald, {Clement J.}",
year = "1995",
month = "1",
doi = "10.1097/00005650-199501000-00001",
language = "English (US)",
volume = "33",
pages = "1--14",
journal = "Medical Care",
issn = "0025-7079",
publisher = "Lippincott Williams and Wilkins",
number = "1",

}

TY - JOUR

T1 - Predicting inpatient costs with admitting clinical data

AU - Tierney, William M.

AU - Fitzgerald, John F.

AU - Miller, Michael E.

AU - James, Margaret K.

AU - McDonald, Clement J.

PY - 1995/1

Y1 - 1995/1

N2 - Hospital cost-containment programs should themselves be cost-effective, targeting high-cost physicians (which requires adjusting for case mix) and patients (which requires early identification). In this study, clinical data available within 24 hours of admission from an electronic medical record system were used to develop statistical models to predict hospital costs. In this retrospective analysis of clinical data and diagnosis-related groups (DRGs), study subjects were 2,355 patients admitted for at least 1 day to the medicine service at an urban teaching hospital with sophisticated electronic medical records. Of these 2,355 patients, 1,663 (71%) had one of the 41 most common DRGs. Predictive models were derived on a random subset of two thirds of the patients and were validated on the remaining third. The following patient data were obtained: admission and prior diagnostic test results, diagnoses, vital signs; demographic data; prior inpatient and outpatient visits; tests and treatments ordered within 24 hours of admission (discretionary data); DRGs; and total inpatient costs (estimated from charges). Diagnosis-related groups explained 24% of the variance in total costs in the derivation patient set and 16% in the validation set. When only nondiscretionary data were used, the models retained only clinical laboratory results and prior diagnoses, explaining 20% of the derivation set variance in total costs and 16% in the validation set. Adding DRGs increased the variance explained in the derivation set to 34%, but decreased to 24% in the validation set. Adding discretionary data substantially increased the explained variance in the derivation and validation patient sets. The models’ median predicted costs underestimated true costs by 10% to 13%, with the lowest error in the models using all types of variables. Clinical data gathered during routine clinical care can be used to adjust for case mix and identify high-cost patients early in their hospital stays, when they could be targeted by cost-containment interventions.

AB - Hospital cost-containment programs should themselves be cost-effective, targeting high-cost physicians (which requires adjusting for case mix) and patients (which requires early identification). In this study, clinical data available within 24 hours of admission from an electronic medical record system were used to develop statistical models to predict hospital costs. In this retrospective analysis of clinical data and diagnosis-related groups (DRGs), study subjects were 2,355 patients admitted for at least 1 day to the medicine service at an urban teaching hospital with sophisticated electronic medical records. Of these 2,355 patients, 1,663 (71%) had one of the 41 most common DRGs. Predictive models were derived on a random subset of two thirds of the patients and were validated on the remaining third. The following patient data were obtained: admission and prior diagnostic test results, diagnoses, vital signs; demographic data; prior inpatient and outpatient visits; tests and treatments ordered within 24 hours of admission (discretionary data); DRGs; and total inpatient costs (estimated from charges). Diagnosis-related groups explained 24% of the variance in total costs in the derivation patient set and 16% in the validation set. When only nondiscretionary data were used, the models retained only clinical laboratory results and prior diagnoses, explaining 20% of the derivation set variance in total costs and 16% in the validation set. Adding DRGs increased the variance explained in the derivation set to 34%, but decreased to 24% in the validation set. Adding discretionary data substantially increased the explained variance in the derivation and validation patient sets. The models’ median predicted costs underestimated true costs by 10% to 13%, with the lowest error in the models using all types of variables. Clinical data gathered during routine clinical care can be used to adjust for case mix and identify high-cost patients early in their hospital stays, when they could be targeted by cost-containment interventions.

KW - Case mix

KW - Cost containment

KW - Database research

UR - http://www.scopus.com/inward/record.url?scp=0029195573&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029195573&partnerID=8YFLogxK

U2 - 10.1097/00005650-199501000-00001

DO - 10.1097/00005650-199501000-00001

M3 - Article

C2 - 7823640

AN - SCOPUS:0029195573

VL - 33

SP - 1

EP - 14

JO - Medical Care

JF - Medical Care

SN - 0025-7079

IS - 1

ER -