In the first study of its kind, Dr Richard Pinder and colleagues at Imperial found that hospitals in England with lower mortality rates were more likely to have members of staff satisfied with the quality of care they provide.
Despite the researchers' initial assumptions, satisfaction levels among non-clinical staff were found to be as closely tied to a hospital's performance as those of doctors. A stronger correlation was found among nursing staff.
The paper appears in BMJ Quality and Safety
days after the long-awaited publication of the Francis report into failures at the Mid-Staffordshire Foundation Trust between 2005 and 2009. In his report, Robert Francis QC recommended that NHS trusts across the UK renew their focus on compassion and the creation of a caring environment for patients. The report also highlighted the central role of staff in raising concerns about poor quality care.
In the new research, the team determined levels of satisfaction by examining data from the NHS's 2009 staff survey. In particular, they focused on whether or not staff would recommend their NHS trust to a friend or colleague, whether they felt that care was their trust's priority, and if they were themselves happy with the standard of care they provided to patients. Over 60,000 responses were collected across the 147 acute general NHS hospitals in England, from doctors and nurses as well as administrative and support staff.
These results were then compared with the individual Hospital Standardised Mortality Ratios (HSMRs), a figure obtained by comparing the expected rate of death in a hospital with the actual rate of death. Although this has been called a crude approach to assessing the quality of healthcare, the Department of Health continue to use overall mortality figures to assess hospital performance. Dr Pinder says: "HSMR isn't perfect, but it's a useful indicator that gives you a steer on performance and has a role in identifying the best- and worst-performing hospitals. In this paper, what we are suggesting is that staff willingness to recommend their hospital may actually be a more sensitive indicator of the quality of care than HSMR."
Dr Pinder and his colleagues suggest that further research would be needed to establish the mechanism behind the correlation. "What this work does is demonstrate that staff satisfaction is correlated with organisational performance. The logical next question is about establishing whether happier staff provide better care, or if better care creates happier staff. We don't know yet, but it's probably the case that both processes are at work. Better organisations attract better staff, who work harder. It's a cycle of improvement or a cycle of degeneration for many of these hospitals," he said.
The findings suggest that staff satisfaction could be used as an early warning system to help spot more serious institutional failings. Regular surveys asking questions such as 'would you recommend this hospital to friends and family?' might have been able to prevent the deterioration of hospital standards that occurred at the Mid-Staffordshire trust.
The researchers say their study might also help patients to make informed choices. "It's difficult for patients to make decisions based on the intricacies of adjusted mortality rates. If you want to choose between two hospitals, knowing that 98 per cent of doctors and nurses working there would recommend their hospital, compared with 60 per cent elsewhere is a useful thing to know," said Dr Pinder.