A study published on bmj.com
says that patient deaths at three London hospitals have been cut by almost 15% after introducing treatment checklists.
The researchers say their methods could be used to reduce mortality in many other hospitals.
The North West London Hospitals NHS Trust serves a population of about 500,000 at three sites: Northwick Park Hospital, Central Middlesex Hospital and St Mark's Hospital. In 2005 and 2006, the trust was subject to a series of adverse media stories which impacted on staff morale and also on patients' perceptions of care.
So a group of senior clinicians set out to reduce hospital inpatient mortality and thus increase public confidence in the quality of patient care at the trust.
They developed eight care bundles for 13 diagnostic areas with the highest number of deaths at the trust in 2006-7. These included treatments for stroke, heart failure and chronic obstructive pulmonary disease (COPD).
Care bundles are a collection of treatment checklists based on clinical guidelines that, when combined, improve the effectiveness and safety of patient care.
Adjusted hospital mortality for 2007-8 (the year the care bundles were introduced) was then compared with the previous year using the hospital standardised mortality ratio (HSMR).
The HMSR is a comparative measure of a hospital's overall mortality. It focuses on a group of diagnoses that account for 80% of all hospital deaths nationally and provides a tool for analysing hospital outcomes over time.
The results show that the overall HSMR of the trust fell from 89.6 in 2006-7 to 71.1 in 2007-8, to become the lowest among acute trusts in England.
In 2007-8, 174 fewer deaths occurred in the trust in the targeted diagnoses, and 255 fewer deaths occurred in the HMSR diagnoses compared to 2006-7. This represents a 14.5% decrease in actual deaths from 2006-7 to 2007-8.
This study demonstrates that it is possible to target care bundles across a wide range of diagnoses in a busy acute hospital trust and that this can be associated with a significant reduction in mortality in the targeted diagnostic areas, conclude the authors. These methods could also be generally applicable, they add.