Improving the quality of written communication at patient discharge: triangulation of qualitative analyses and intervention co-design

BACKGROUND:Poor handovers between hospital and primary care threaten safe discharges, with elderly and frail patients most at risk of harm. Using Behavioural Science we explored influences and identified relevant behaviour change techniques (BCTs) to improve written handovers and safety during discharge. METHODS:We conducted two qualitative studies: (1) ethnographic observations (>80 h) collected by five researchers in five purposively sampled clinical areas of a London teaching hospital, investigating routine work and interactions of hospital staff involved in discharges; and (2) 12 semi-structured interviews with hospital staff involved in discharge exploring influences on preparations of written handovers. Written consent was sought from clinical leads for ethnographic observations and from interview participants. Ethnographic fieldnotes and interview transcripts were thematically analysed using inductive and deductive approaches, respectively. Study findings were triangulated to identify key influences, mapped onto the Theoretical Domains Framework (TDF). We identified appropriate BCTs to address observed influences within each TDF domain using the Theory and Techniques Tool. Health-care workers (n=15), patients (n=2) and carers (n=2) selected and designed an intervention to improve written handovers in two workshops. Hospital workshop participants were involved with preparing written discharge handovers. Public participants had either recently been discharged from hospital or cared for someone recently discharged, including patients from groups especially vulnerable during discharge. FINDINGS:Triangulation of study findings generated 11 key influences on preparations of written handovers within five TDF domains: knowledge (eg, lack of awareness of guidelines), skills (staff experience), social or professional role and identity (effective communication), environmental context and resources (working patterns), and social influences (lack of feedback). 14 BCTs were identified to address these influences, including behavioural rehearsal or practice, instruction on how to perform a behaviour, and social support (practical). Workshop participants selected and designed a multifaceted educational intervention to improve written handovers. INTERPRETATION:The quality of handover documentation prepared by hospital staff for primary care teams is affected by influences from multiple domains, requiring a multifaceted approach to improve handovers. Although only based on findings from one hospital, the designed intervention should be tested in clinical settings with key stakeholders, including primary care staff, to evaluate impact on quality of written handovers and patient safety. FUNDING:National Institute for Health and Care Research (NIHR) Imperial Patient Safety Translational Research Centre.