Improving photoprotection in adults with xeroderma pigmentosum: personalisation and tailoring in the ‘XPAND’ intervention

Abstract

ABSTRACT Background Individualised behaviour change interventions can result in greater effects than one-size-fits-all approaches. Factors linked to success include dynamic (vs. static) tailoring, and tailoring on behaviour, multiple theoretical variables, and participant characteristics. XP is a very rare (∼100 UK patients) genetic disease, involving an inability to repair ultraviolet radiation (UVR)-induced damage, resulting in skin cancers and eye damage from an early age, and mean life expectancy of 32-years. Management involves rigorous UVR photoprotection, which is often inadequate, and no interventions have been published. UK-based care is personalised and delivered by a multidisciplinary team at the National XP Service in London. Following an intensive, mixed-methods formative phase with patients diagnosed with XP (n-of-1, qualitative interviews, objective UVR measurement, cross-sectional survey) and relevant stakeholder consultation (clinical and patient/public teams), the ‘XPAND’ intervention was developed. This paper describes the comprehensive and novel tailoring and personalisation processes used to deliver the intervention. Methods XPAND consists of core and personalised modules targeting cue-based (time of day, weather, symptoms), belief-based (motivation, priority), self-regulatory (effort, barriers, planning), and emotional (stress, self-consciousness, mental exhaustion) factors, social support, disclosure, habit, and willingness, using appropriately-matched BCTs. A-priori, phase I data and a baseline profiling questionnaire (data sources) were used to allocate modules to participants (‘personalisation’) and to adapt module content (‘tailoring’). Iterative decisions about delivery were based on patient response to feedback, identification of additional barriers (e.g. reasons for varying protection across contexts), and emergence of new barriers as improvements in protection were attempted or achieved (e.g. appearance concerns). Conclusions Dynamic multi-level personalisation and tailoring based on mixed-methods in XPAND allowed for insights and decision-making not possible with cross-sectional quantitative or qualitative methods alone. Data collection and allocation/adaptation methods may be of use in other rare conditions where small patient numbers mean that within-participant, individual-level delivery is well-suited and feasible.