How care bundling drives down mortality rates in pneumonia care

A community-acquired pneumonia care bundle implemented by North West London Hospitals Trust has reduced mortality rates by 6 per cent, equivalent to saving 49 lives last winter. This is despite higher admissions of acutely ill patients with pneumonia and patients with influenza conditions.

The initiative is one of three projects with the National Institute for Health Research’s Collaborations for Leadership in Applied Health Research and Care programme in north west London, which supports quality improvement initiatives that can be shared across the NHS. The projects began in April 2010 and lasted for 18 months.

North West London Hospitals Trust has been improving standards using this care bundle approach. In April 2010, it published results for the top five common conditions with the highest mortality rates. By changing practice and implementing this approach the trust achieved a 15 per cent reduction in the hospital’s standardised mortality rate and a 60 per cent reduction in C difficile and MRSA rates.

This latest project focused on implementing a sustainable model of care for patients with pneumonia, using evidence based interventions. The implementation phase began in November 2010. The trust, which has two separate sites, used quality improvement methodology to identify the scope of the project, to discover changes that needed to be made through evidence based practice, and to monitor sustainability. It also identified outcome measures that would provide evidence of sustained change and service improvement.

Process mapping

The project team conducted a service review, using national guidelines to benchmark best practice for pneumonia, then used this information to develop a tool that would be owned by all the teams using it.

The review discovered that barriers to implementing the admission bundle for pneumonia were lower on one of the sites, because it had previously implemented care bundles and was more familiar with the concept. Therefore, the team concentrated on implementing and sustaining the care bundle on the site with the greatest understanding.

Organisational barriers were:

• a 16 per cent increase in emergency department attendances;
• an 13 per cent increase in admissions over the winter;
• service reconfiguration at the second emergency department, which was establishing an urgent care centre.

Following the review, the clinical teams identified effective interventions that now form the basis of the care bundle:

• chest x-ray to confirm diagnosis;
• pneumonia severity assessment using CURB-65;
• oxygen levels assessed and therapy given;
• antibiotics administered within four hours;
• fluid balance assessed;
• patient acuity assessed using an early warning score for managing acutely ill patients, known as MEWS.

Care bundle design

The tripartite care bundle design is unique to the trust. It consists of:

• patient and ward details;
• a peel-off information sticker attached to notes;
• an audit checklist.

The project team reviewed the wording and layout of the bundle, using plan, do, study, act cycles to incorporate staff feedback, and it now incorporates a checklist that can easily be audited and entered into a database, while colour coding enables staff to recognise the bundle in patient notes.

The team mapped the processes throughout the project to identify any gaps, as well as to allow the bundle to be refined to the point at which it was as simple as possible to use. All members of the multidisciplinary team were responsible for delivering the training, which was carried out in the areas where the bundle was being implemented through clinical teaching days, at nursing handovers and lunch time studying sessions.

This was followed up by discussions about patients who had not been admitted on to the bundle, with the clinical team identifying the reasons the bundle was not used.

Significant barriers to the process were:

• knowing where to get the bundle;
• dentifying which patients were on the bundle;
• dentifying how staff would know which patients had been assessed.

The trust changed the process accordingly, and:

• identified to all staff an area where the bundles are stored;
• identified a central area for audit sheets to be kept post-implementation;
• agreed a communication process, with a meeting for handing over information about compliance and the number of patients admitted to the hospital with community acquired pneumonia.

These efforts have been successful in driving sustainable implementation.

Communication was a vital part of the project, and junior doctors gave valuable feedback about the need for readily available information about antibiotics and the care bundle itself. As a result, the trust developed a pocket-sized antibiotics guideline booklet which all doctors pick up at induction. The emergency department uploaded information on USB memory sticks given to new staff.

Data for each bundle was entered into a web reporting tool, as well as information from daily admission sheets for patients with pneumonia but not admitted to the bundle. This mapped implementation week by week and was used to provide progress feedback to staff.

Outcomes

The team reviewed mortality rates for pneumonia from April 2009 to March 2011 to review whether the bundle had improved patient outcomes. The first analysis demonstrated a significant reduction of mortality rates in winter from November 2010 to February 2011, coinciding with implementation of the bundle and an overall reduction in mortality rates of 6 per cent.

More patients were admitted with pneumonia during these months compared with the same period the previous year. Most had moderate or severe pneumonia, which has a mortality rate of 40 per cent.

The trust demonstrated an improvement in mortality rates of 49 from April 2010 to April 2011. Thirty-five of the lives saved coincided with the first three months of the care bundle’s implementation between November 2010 and March 2011. We continue to monitor the progress of the bundle and are planning a communication exercise to raise awareness before this winter begins to coincide with higher pneumonia admissions due to respiratory infections.

We are also working on a safer handover sticker to be used in the emergency department and have developed a digital story that shows how the bundle is used. Our aim is to continue to use the bundle and implement it on our second site.

Find out more

Collaborations for Leadership in Applied Health Research and Care