The demonstration of a theory-based approach to the design of localized patient safety interventions

BACKGROUND: There is evidence of unsafe care in healthcare systems globally. Interventions to implement recommended practice often have modest and variable effects. Ideally, selecting and adapting interventions according to local contexts should enhance effects. However, the means by which this can happen is seldom systematic, based on theory, or made transparent. This work aimed to demonstrate the applicability, feasibility, and acceptability of a theoretical domains framework implementation (TDFI) approach for co-designing patient safety interventions. METHODS: We worked with three hospitals to support the implementation of evidence-based guidance to reduce the risk of feeding into misplaced nasogastric feeding tubes. Our stepped process, informed by the TDF and key principles from implementation literature, entailed: involving stakeholders; identifying target behaviors; identifying local factors (barriers and levers) affecting behavior change using a TDF-based questionnaire; working with stakeholders to generate specific local strategies to address key barriers; and supporting stakeholders to implement strategies. Exit interviews and audit data collection were undertaken to assess the feasibility and acceptability of this approach. RESULTS: Following audit and discussion, implementation teams for each Trust identified the process of checking the positioning of nasogastric tubes prior to feeding as the key behavior to target. Questionnaire results indicated differences in key barriers between organizations. Focus groups generated innovative, generalizable, and adaptable strategies for overcoming barriers, such as awareness events, screensavers, equipment modifications, and interactive learning resources. Exit interviews identified themes relating to the benefits, challenges, and sustainability of this approach. Time trend audit data were collected for 301 patients over an 18-month period for one Trust, suggesting clinically significant improved use of pH and documentation of practice following the intervention. CONCLUSIONS: The TDF is a feasible and acceptable framework to guide the implementation of patient safety interventions. The stepped TDFI approach engages healthcare professionals and facilitates contextualization in identifying the target behavior, eliciting local barriers, and selecting strategies to address those barriers. This approach may be of use to implementation teams and policy makers, although our promising findings confirm the need for a more rigorous evaluation; a balanced block evaluation is currently underway.