Will payment based on diagnosis-related groups control hospital costs?

N Engl J Med. 1984 Aug 2;311(5):295-300. doi: 10.1056/NEJM198408023110505.


Previous studies have shown that the admission rates for a few surgical procedures, such as hysterectomy, vary extensively among hospital market areas, apparently because of differences in physicians' practice styles. To see whether such variations occur for most causes of admission, we classified all nonobstetrical medical and surgical hospitalizations in Maine for the years 1980 through 1982 into diagnosis-related groups (DRGs) and measured the variations in admission rates among 30 hospital market areas. Hysterectomy rates varied 3.5-fold, but 90 per cent of medical and surgical admissions fell into DRGs for which admission rates were even more variable, suggesting that professional discretion plays an important part in determining hospitalization for most DRGs. Losses in hospital revenues resulting from the DRG payment system could be offset if physicians modified their admission policies to produce more profit, well within the current limits of medical appropriateness. If this occurred, the net effect of a DRG program would be to exacerbate hospital cost inflation. We conclude that, to be successful, cost-containment programs based on fixed, per-admission hospital prices will need to ensure effective control of hospitalization rates.

Publication types

  • Research Support, Non-U.S. Gov't
  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Cost Control
  • Costs and Cost Analysis*
  • Data Collection
  • Diagnosis-Related Groups*
  • Hospitalization / economics*
  • Humans
  • Length of Stay
  • Maine
  • Patient Admission / economics
  • Physicians
  • Reimbursement Mechanisms
  • Statistics as Topic
  • Surgical Procedures, Operative