Identifying behaviour change techniques to support medicines management for older people at care transitions

Abstract

\n \n \n Older patients are at significant risk of medicines-related harm following a hospital to home transition [1]. Strategies, such as information transfer and medicines reconciliation, have been suggested to promote successful transitions through better medicines management [2]. These interventions, however, are not often underpinned by patient experience or grounded in theory, and therefore have variable effects in real-world contexts.\n \n \n \n To identify which behaviour change techniques (BCTs) could be combined within a complex intervention to support post-discharge medicines management.\n \n \n \n Barriers and facilitators to post-discharge medicines management behaviours were identified from an analysis of interviews with twenty-seven older people (aged 75 years plus) and mapped to the Theoretical Domains Framework (TDF). These participants were recruited during a stay in two large hospitals in the North of England. All potential BCTs targeting the identified TDF domains were found using validated mapping exercises and refined by panel consensus. The panel consisted of two pharmacists, two health services researchers, a social work academic and a patient representative. Next, a convenience sample of key stakeholders (n=40; patients and family carers, healthcare professionals and researchers) were invited to take part in an online survey to prioritise remaining techniques. A total score for each BCT was calculated based on its potential i) effectiveness, ii) acceptability and iii) ability to cause unintended consequences. The BCTs were ranked from lowest to highest scores. As validation, the final selection of BCTs was assessed using the APEASE criteria and mapped back to the behavioural determinants to check the components would enhance facilitators and overcome barriers (Figure 1).\n \n \n \n Forty-six behavioural determinants were identified within interviews and classified as barriers (n=19), facilitators (n=10), or both (n=17). These were deductively mapped onto 11 domains of the TDF. Following the mapping, 50 discrete BCTs were identified, which were reduced to 35 after subsequent consensus discussions between panel members to remove any deemed inappropriate. The survey resulted in 25 responses (62.5% response rate). Following visual inspection of the ranking, a natural cut-off was identified by panel consensus at 88% of the total score. This, and further assessment using the APEASE criteria, prioritised eight BCTs that were highly rated to be effective in promoting post-discharge medicines management when combined within a complex intervention. These were: practical social support, goal/ target specified, prompts, triggers or cues, social processes of encouragement, motivational interviewing, rehearsal, review goals, and comparative imagining of future outcomes.\n \n \n \n This study has identified eight BCTs that could be valuable when combined within a complex intervention to support post-discharge medicines management for older people. Consensus allowed prioritisation of BCTs that were likely to be effective, acceptable to older people, practical and cost-effective to deliver within current healthcare organisations. Limitations included challenges of coding complex behavioural determinants to the TDF, defining the BCTs within the medicines management context and modest survey sample size. To overcome these limitations, input was sought from a health psychologist with expertise and consensus involved all relevant stakeholders.\n \n \n \n 1. Parekh, N., Ali, K., Stevenson, J. M., et al. Incidence and cost of medication harm in older adults following hospital discharge: a multicentre prospective study in the UK. British journal of clinical pharmacology 2018;84(8):1789–1797.\n 2. Tomlinson, J., Cheong, V., Fylan, B., et al. Successful care transitions for older people: a systematic review and meta-analysis of the effects of interventions that support medication continuity. Age and Ageing 2020;00:1–12.\n