Prospective payment for psychiatry--feasibility and impact.

Area of Science:

• Psychiatry
• Health Services Research
• Health Economics

Background:

• Psychiatric hospitals are exempt from the prospective payment system (PPS) based on diagnosis-related groups (DRGs).
• Existing DRGs for psychiatric patients have not been rigorously evaluated for their predictive accuracy regarding length of stay and costs.

Purpose of the Study:

• To assess the predictive ability of the 15 proposed psychiatric DRGs for hospital length of stay and costs.
• To develop and compare alternative patient grouping systems for improved prediction of length of stay.

Main Methods:

• Retrospective analysis of 8816 randomly selected psychiatric hospital patient charts from 32 US psychiatric hospitals.
• Evaluation of the 15 psychiatric DRGs and development of alternative grouping models based on diagnostic categories, transfer status, age, and comorbidities.

Main Results:

• The 15 psychiatric DRGs explained only 3.9% of the variance in patient length of stay.
• The best alternative grouping model, incorporating diagnostic categories, transfer status, age, and comorbidities, explained only 7.8% of the variance in length of stay.
• All tested models demonstrated significant financial disparities, indicating potential for inappropriate incentives in patient care.

Conclusions:

• Diagnosis-related groups (DRGs) are inadequate for predicting length of stay and costs in psychiatric hospitals.
• Factors beyond diagnosis are better predictors of length of stay, but current grouping models create financial risks and misaligned incentives.
• The findings suggest that DRG-based payment systems may not be suitable for psychiatric care without substantial modification.