HC Economist Models

PPS Economic Model

In an attempt to reduce costs, Medicare enacted a Prospective Payment System (PPS) in 1983. Medicare aimed to pay hospitals a fixed rate based on the Diagnosis Related Group (DRG) plus/minus an adjustment for location and local wage. Although this system gives hospitals the incentive to misclassify patients into high profit DRG, I will assume for simplicity that the hospital diagnose the patient’s illness with perfect accuracy. I briefly outline a model in order to analyze how PPS effects hospital (or providers) incentives.

The Model

The hospital makes a profit on each patient of: P-C(s)-c(q(s))

  • P is the reimbursement rate from Medicare based on the DRG; C(s) is a cost function depending on sickness, c(q(s)) is the additional cost incured by the hospital for additional quality of care. C’,C”,c’,c”,D’,D” are all strictly positive.

Total profits for the hospital are: D(q)*[P-C(s)-c(q)]; where D(q) is the consumer’s demand function. Firms maximize profits by choosing the quality level for each sickness type. The first order condition for the firm is:

  • D'[P-C(s)-c(q)]=D*c'(q)

If we totally differentiate the above equation (remember q is a function of s), we have:

  • d(q(s))/d(s)=[D”(P-C-c)-2D’c’-Dc”]/[D’C’] <0

Discussion
What does all this math mean? Well since dq(s)/ds<0, this means that discretionary quality falls with severity for all profitable patients and is set to zero for unprofitable patients. Since the PPS payment system does not reimburse providers for additional quality of their work with patients, these providers have an incentive to decrease quality. On the other hand, if we construct a 'cost-plus' system where hospitals are reimbursed at (1+x%) of cost, hospitals have an incentive to treat the most severe illnesses since they are the most profitable.

Models as developed in: Meltzer and Chung (2002) “Effects of Competition Under Prospective Payment on Hospital Costs Among High- and Low-Cost Admissions: Evidence from California in 1983 and 1993” Forum for Health Economics and Policy, Vol 5(4).