Improving palliative care in selected settings in England using quality indicators: a realist evaluation

Background: There is a gap between readily available evidence of best practice and its use in everyday palliative care. The IMPACT study evaluated the potential of facilitated use of Quality Indicators as tools to improve palliative care in different settings in England.

Methods: 1) Modelling palliative care services and selecting a set of Quality Indicators to form the core of an intervention, 2) Case studies of intervention using the Quality Indicator set supported by an expert in service change in selected settings (general practice, community palliative care teams, care homes, hospital wards, in-patient hospices) with a before-and-after evaluation, and 3) realist evaluation of processes and outcomes across settings. Participants in each setting were supported to identify no more than three Quality Indicators to work on over an eight-month period in 2013/2014.

Results: General practices could not be recruited to the study. Care homes were recruited but not retained. Hospital wards were recruited and retained, and using the Quality Indicator (QI) set achieved some of their desired changes. Hospices and community palliative care teams were able to use the QI set to achieve almost all their desired changes, and develop plans for quality improvements. Improvements included: increasing the utility of electronic medical records, writing a manual for end of life care, establishing working relationships with a hospice; standardising information transfer between settings, holding regular multi-disciplinary team meetings, exploration of family carers' views and experiences; developing referral criteria, and improvement of information transfer at patient discharge to home or to hospital. Realist evaluation suggested that: 1) uptake and use of QIs are determined by organisational orientation towards continuous improvement; 2) the perceived value of a QI package was not powerful enough for GPs and care homes to commit to or sustain involvement; 3) the QI set may have been to narrow in focus, or more specialist than generalist; and 4) the greater the settings' 'top-down' engagement with this change project, the more problematic was its implementation.

Conclusions: Whilst use of QIs may facilitate improvements in specialist palliative care services, different QI sets may be needed for generalist care settings.