Predictive efficiency analysis: A study of us hospitals

Andrew L. Johnson, Chia-Yen Lee

Research output: Chapter in Book/Report/Conference proceedingChapter

1 Citation (Scopus)

Abstract

Demand fluctuations lead to variations in output levels, affecting a hospital's efficiency measures. For a typical efficiency measure, data envelopment analysis (DEA) is a nonparametric linear programming technique to estimate the production function and then efficiency. However, using units sold as an output measure could bias the efficiency estimate. In particular, DEA implicitly assumes hospitals will be able to perfectly predict customer demands for hospital services or that hospitals can adjust input resources without any time delays. The present study accounts for the expected output and uses the concept of "effectiveness measure" calculated from the input-truncated production function. A hospital can achieve effective production if its input levels are equal to the effective input levels identified by the forecast output in the next period and the input-truncated production function. That is, the effectiveness measure provides a predictive efficiency analysis. A low effectiveness measure implies that the hospital used more inputs than necessary or more than the forecasted demand, and the implication is consistent with a medical arms race. An empirical study of US hospitals in 2009-2011 was conducted to validate the proposed model and provide managerial insights to drive productivity in the hospital industry.

Original languageEnglish
Title of host publicationAdvances in DEA Theory and Applications
Subtitle of host publicationWith Examples in Forecasting Models
PublisherWiley-Blackwell
Pages404-418
Number of pages15
ISBN (Electronic)9781118946688
ISBN (Print)9781118946701
DOIs
Publication statusPublished - 2016 Oct 21

Fingerprint

Production Function
Output
Data Envelopment Analysis
Measure Data
Estimate
Empirical Study
Productivity
Forecast
Linear programming
Time Delay
Customers
Industry
Fluctuations
Imply
Predict
Resources
Unit
Necessary
Demand
Model

All Science Journal Classification (ASJC) codes

  • Mathematics(all)

Cite this

Johnson, A. L., & Lee, C-Y. (2016). Predictive efficiency analysis: A study of us hospitals. In Advances in DEA Theory and Applications: With Examples in Forecasting Models (pp. 404-418). Wiley-Blackwell. https://doi.org/10.1002/9781118946688.ch26
Johnson, Andrew L. ; Lee, Chia-Yen. / Predictive efficiency analysis : A study of us hospitals. Advances in DEA Theory and Applications: With Examples in Forecasting Models. Wiley-Blackwell, 2016. pp. 404-418
@inbook{c4a40fdba0664da4914e64dc5f1b6a68,
title = "Predictive efficiency analysis: A study of us hospitals",
abstract = "Demand fluctuations lead to variations in output levels, affecting a hospital's efficiency measures. For a typical efficiency measure, data envelopment analysis (DEA) is a nonparametric linear programming technique to estimate the production function and then efficiency. However, using units sold as an output measure could bias the efficiency estimate. In particular, DEA implicitly assumes hospitals will be able to perfectly predict customer demands for hospital services or that hospitals can adjust input resources without any time delays. The present study accounts for the expected output and uses the concept of {"}effectiveness measure{"} calculated from the input-truncated production function. A hospital can achieve effective production if its input levels are equal to the effective input levels identified by the forecast output in the next period and the input-truncated production function. That is, the effectiveness measure provides a predictive efficiency analysis. A low effectiveness measure implies that the hospital used more inputs than necessary or more than the forecasted demand, and the implication is consistent with a medical arms race. An empirical study of US hospitals in 2009-2011 was conducted to validate the proposed model and provide managerial insights to drive productivity in the hospital industry.",
author = "Johnson, {Andrew L.} and Chia-Yen Lee",
year = "2016",
month = "10",
day = "21",
doi = "10.1002/9781118946688.ch26",
language = "English",
isbn = "9781118946701",
pages = "404--418",
booktitle = "Advances in DEA Theory and Applications",
publisher = "Wiley-Blackwell",
address = "United Kingdom",

}

Johnson, AL & Lee, C-Y 2016, Predictive efficiency analysis: A study of us hospitals. in Advances in DEA Theory and Applications: With Examples in Forecasting Models. Wiley-Blackwell, pp. 404-418. https://doi.org/10.1002/9781118946688.ch26

Predictive efficiency analysis : A study of us hospitals. / Johnson, Andrew L.; Lee, Chia-Yen.

Advances in DEA Theory and Applications: With Examples in Forecasting Models. Wiley-Blackwell, 2016. p. 404-418.

Research output: Chapter in Book/Report/Conference proceedingChapter

TY - CHAP

T1 - Predictive efficiency analysis

T2 - A study of us hospitals

AU - Johnson, Andrew L.

AU - Lee, Chia-Yen

PY - 2016/10/21

Y1 - 2016/10/21

N2 - Demand fluctuations lead to variations in output levels, affecting a hospital's efficiency measures. For a typical efficiency measure, data envelopment analysis (DEA) is a nonparametric linear programming technique to estimate the production function and then efficiency. However, using units sold as an output measure could bias the efficiency estimate. In particular, DEA implicitly assumes hospitals will be able to perfectly predict customer demands for hospital services or that hospitals can adjust input resources without any time delays. The present study accounts for the expected output and uses the concept of "effectiveness measure" calculated from the input-truncated production function. A hospital can achieve effective production if its input levels are equal to the effective input levels identified by the forecast output in the next period and the input-truncated production function. That is, the effectiveness measure provides a predictive efficiency analysis. A low effectiveness measure implies that the hospital used more inputs than necessary or more than the forecasted demand, and the implication is consistent with a medical arms race. An empirical study of US hospitals in 2009-2011 was conducted to validate the proposed model and provide managerial insights to drive productivity in the hospital industry.

AB - Demand fluctuations lead to variations in output levels, affecting a hospital's efficiency measures. For a typical efficiency measure, data envelopment analysis (DEA) is a nonparametric linear programming technique to estimate the production function and then efficiency. However, using units sold as an output measure could bias the efficiency estimate. In particular, DEA implicitly assumes hospitals will be able to perfectly predict customer demands for hospital services or that hospitals can adjust input resources without any time delays. The present study accounts for the expected output and uses the concept of "effectiveness measure" calculated from the input-truncated production function. A hospital can achieve effective production if its input levels are equal to the effective input levels identified by the forecast output in the next period and the input-truncated production function. That is, the effectiveness measure provides a predictive efficiency analysis. A low effectiveness measure implies that the hospital used more inputs than necessary or more than the forecasted demand, and the implication is consistent with a medical arms race. An empirical study of US hospitals in 2009-2011 was conducted to validate the proposed model and provide managerial insights to drive productivity in the hospital industry.

UR - http://www.scopus.com/inward/record.url?scp=85047755035&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047755035&partnerID=8YFLogxK

U2 - 10.1002/9781118946688.ch26

DO - 10.1002/9781118946688.ch26

M3 - Chapter

AN - SCOPUS:85047755035

SN - 9781118946701

SP - 404

EP - 418

BT - Advances in DEA Theory and Applications

PB - Wiley-Blackwell

ER -

Johnson AL, Lee C-Y. Predictive efficiency analysis: A study of us hospitals. In Advances in DEA Theory and Applications: With Examples in Forecasting Models. Wiley-Blackwell. 2016. p. 404-418 https://doi.org/10.1002/9781118946688.ch26