Government uses new method to judge performance of hospitals to try to prevent “gaming” of system

BMJ 2011; 343 doi: (Published 28 October 2011) Cite this as: BMJ 2011;343:d7015
  1. Nigel Hawkes
  1. 1London

A new official indicator of the number of patients who die after treatment in hospital has found 14 NHS trusts in England with a higher than expected mortality and 14 whose mortality is lower than expected.

The summary hospital level mortality indicator (SHMI) compares the number of patients who die within 30 days of treatment at a trust with the number expected to die if the trust performed at the national average rate. The actual deaths are corrected for factors outside the trust’s control, such as age, deprivation, and comorbidities.

The new measure, developed for the NHS by an expert committee, is similar to the hospital standardised mortality ratio (HMSR) published by the health analysis company Dr Foster.

Its creation was inspired by the occasional embarrassing gaps between the conclusions reached by Dr Foster and those of the health service regulator, at that time the Healthcare Commission. Mid Staffordshire NHS Foundation Trust, now recognised to have been one of the most dysfunctional in the history of the NHS, was rated by Dr Foster among its five most improved hospitals …

View Full Text

Log in

Log in through your institution


* For online subscription