Data envelopment analysis models for identifying and benchmarking the best healthcare processes. (9th December 2008)

• Record Type: Journal Article
• Authors: Benneyan, James C.; Sunnetci, Aysun; Ceyhan, Mehmet Erkan
• Is Part Of: International journal of six sigma and competitive advantage. Volume 4:Number 3(2008)
• Page Start: 305
• Page End: 331
• Publication Date: 2008-12-09
• Subjects: benchmarking -- healthcare processes -- data envelopment analysis -- DEA -- weight restrictions -- proportional data -- hyper-efficiency -- healthcare comparisons -- quality improvement -- six sigma; Six sigma (Quality control standard) -- Periodicals; Quality control -- Periodicals; 658.4013
• Languages: English
• ISSNs: 1479-2494

Abstract

We illustrate the use of Data Envelopment Analysis (DEA) models within process improvement work for identifying and benchmarking the best healthcare systems, in terms of most efficiently producing desirable outcomes from consumed resources. This approach is useful when comparing several systems that use multiple types of inputs (e.g., operating costs, clinicians, staff) to produce multiple outputs (e.g., outcomes, satisfaction, access), such as those commonly found in balanced scorecards and dashboard datasets, and provides the analyst with relative scores and rankings for each system, targets for each measure that would move inefficient systems to the best performance frontier, and a list of other systems to benchmark and emulate in order to improve. Modified DEA models are proposed to address four common issues that frequently arise in such contexts, including rationally constraining the weights given to each measure and handling missing, estimated or proportional data (such as adverse event or mortality rates). These models can be used to compare hospitals, departments, national healthcare systems, and regional or state systems and are useful to help understand how to improve sub-optimal processes and set feasible targets. This approach is illustrated at department, hospital, state, and country levels, with overall results showing very little correlation with less quantitative benchmarking studies.