The current method of remunerating hospitals by an average per diem fee tends to over-reimburse hospitals that have a concentration of departments whose true costs are less than the average price received. Hospitals with a high concentration of expensive high-technology service departments whose true costs are more than the average price received will be under-reimbursed and are obliged to cover their running deficits by other means, e.g., donations. Reimbursements on a per diem basis provide a 'perverse incentive' for all hospitals to maximize the length of patient stays in order to maximize their income. This paper briefly examines alternative methods to the deficient per diem method of reimbursing hospitals, such as fee for service, historical budgeting, capitation, gatekeeper's fees and diagnosis-related groups (DRGs). Fee for service or historical budgeting shows little or no advantage over the present system. However, a combination of capitation and/or DRG linked with some form of payment via physician gatekeepers appears to provide a favorable option for correcting the distortions of the per diem system. Department-specific DRG weights for each hospital's department admission mix are used to estimate the magnitude of the current distortion in resources allocated to hospitals. The calculation is based on the changes in hospital income were a DRG mechanism introduced instead of a per diem method. Such changes would increase the hospitalization income of hospitals with low lengths of stay and high bed turnover rates up to 39%. Regional hospital centers with high lengths of stays and low bed turnover rates would receive as much as 17% lower income in some cases. Only if DRG weights were available for each individual hospital would it be possible to ascertain whether differences in lengths of stay reflect differing severities of case loads or differing hospital efficiency levels.