The statistics behind hospital mortality rates have been questioned by MPs after 14 hospital trusts were named as being under investigation.
Bernard Jenkin MP, who chairs the Public Administration Select Committee (PASC), wrote to Andrew Dilnot, chair of the UK Statistics Authority, expressing concern over the index used to name five hospitals with the highest rates of mortality, saying it is ‘relatively unproven’.
Following the recent publication of the Francis report, the government asked the NHS medical director, Sir Bruce Keogh, to conduct a further investigation into five hospitals with the highest mortality rates. The Francis report looked into what went wrong at Stafford Hospital, where an estimated 400-1200 patients are believed to have died between 2005 and 2009 as a result of poor care.
In his letter, Bernard Jenkin pointed out that one index used to identify five hospitals with the highest mortality rates, the Summary Hospital-level Mortality Indicator (SHMI), is ‘a new and relatively unproven indicator’. He also noted that after these five hospitals were identified using SHMI, further hospitals were named using a more established indicator, the Hospital Standardised Mortality Ratio (HSMR).
Bernard Jenkin asked Andrew Dilnot if he could advise the government on how these measures of hospital patient mortality could be used ‘without causing undue alarm in the media, and amongst patients, staff and the wider public’.
In his response, Andrew Dilnot pointed out that while the SHMI was intended to be a successor to the HSMR, it should not be used as a sole indicator of the quality of care and should be used as part of a range of indicators. As to why both indicators had been used to name the 14 hospitals currently being investigated, Dilnot said that it was a ‘reasonable approach’ since the introduction of the SHMI is relatively recent and both indicators were readily available.
The UK Statistics Authority chair promised that the Statistics Authority would ‘do all it can to ensure that these official statistics are properly understood.’
Dilnot also indicated that in response to the Francis Report, the Statistics Authority was planning to undertake an independent review of patient outcome statistics, working with the Office for National Statistics, the Department of Health and the Health and Social Care Information Centre (HSCIC). The findings of this review will be published in summer 2013.