Are costs derived from diagnosis-related groups suitable for use in economic evaluations? A comparison across nine European countries in the European Healthcare and Social Cost Database

Eur J Health Econ. 2022 Dec;23(9):1563-1575. doi: 10.1007/s10198-022-01444-y. Epub 2022 Feb 26.

Abstract

Background: Economic evaluation of health technologies requires healthcare resources, procedures and services to be valued at their opportunity cost. In practice, many economic evaluation studies use official databases of hospital Diagnosis-Related Groups (DRGs) as inputs where unit costs are required. This study describes the available data on costs of hospital DRG from official, publicly available sources in nine European countries (England, France, Germany, Italy, Poland, Portugal, Slovenia, Spain and Sweden), critically examines and compares the methodologies used to construct these databases and comments on the appropriateness of such unit cost data for economic evaluation.

Methods: A standardized semi-structured questionnaire was developed in order to obtain both official publicly available sources of inpatient DRG databases and documents explaining the costing methodology used in calculation of unit costs available in those databases.

Results: England stands out as a benchmark in terms of good practice. Other countries face more challenges in one or more items, whether in documenting and auditing processes, guaranteeing methodological rigour, including all relevant economic items such as depreciation of buildings and equipment and capital costs, conducting the process annually and completely, publishing the costs as well as tariffs and recognising sampling uncertainty or variation.

Conclusion: Analysts should evaluate carefully whether DRG costs or tariffs published in each country are appropriate for use in economic evaluation.

Keywords: Costs; Diagnosis-related groups; Economic evaluation; Healthcare.

MeSH terms

  • Cost-Benefit Analysis
  • Databases, Factual
  • Diagnosis-Related Groups*
  • Europe
  • Hospital Costs*
  • Humans