Average length of stay is a cost measure wearing a clinical costume
Hospital scorecards are full of numbers doctors do not believe. Removing one changed how a fourteen-hospital group measures itself.
Average length of stay appears on almost every hospital operating scorecard in India. It is easy to calculate, comparable across facilities, and correlates with cost per admission. It is also, according to a growing number of clinician-administrators, actively harmful as a managed target.
The objection is not that the number is wrong. It is that it rewards discharging patients who should have stayed, and that everyone on the clinical side knows this while the metric continues to be reported upward as a quality indicator.
It is a cost measure wearing a clinical costume. Every doctor in the building knew we were optimising for the wrong thing, and none of them were asked.
Arjun Bhatia, Chief Operating Officer, Vaidya Health Networks
The replacement
A readmission-adjusted measure — length of stay penalised by readmissions within a defined window — is harder to compute, slower to report and less comparable across facilities with different case mixes. It also cannot be improved by discharging someone early, which is the only property that matters.
The broader lesson is about who gets consulted when an operating scorecard is designed. In most hospital groups the scorecard is built by finance and operations, reviewed by the executive committee, and never formally shown to the clinicians whose behaviour it is intended to shape.