Substance or style? Valuing aspects of personalisation of the support for self-management

Problem

Personalisation of healthcare delivery is one of the tenets of 21st century health services and is fundamental to the ethos of primary care. But personalisation is a complex and multifaceted construct; so which aspects of personalised care do patients value most, and how much is one aspect valued more than another?

Approach

We conducted the first discrete choice experiments (DCE) to assess the value that people with long term conditions place on different aspects of personalisation. The context was the support people receive for self-management, firstly for chronic pain, then in a replication study, for chronic lung disease.The DCE included 4 attributes of support, each with two levels of personalisation: information (“relevant to you” vs. “the same for everyone”); communication style (“friendly and personal” vs “neutral professional”); situation relevance (“makes suggestions which fit your situation” vs. “takes little account of your situation”); and what matters in life (“works with you on what you want to get from life” vs. “seems to think everyone wants to get the same from life”). A cost attribute was included so that willingness to pay (WTP), a monetary measure of value, could be estimated for all attributes. The DCE questionnaire was developed through focus groups and think-aloud studies. A pilot DCE with 126 people was used to check feasibility, define markers of data quality, and to design the main DCE with optimal statistical efficiency. The final DCE comprised 15 sets of choices between 3 options for services. We recruited participants from an online panel: 517 with self-reported chronic pain for the main DCE and 200 with breathlessness due to chronic lung disease for the replication study.

Findings

500 participants (96.4%) met our criteria for data quality. Participants revealed positive valuations for increased personalisation of all four attributes. Greater personalisation on the attributes relating to situation relevance (£15.51) and what matters in life (£14.10) were valued most highly; tailored information was valued less (WTP £10.86) and more friendly communication valued least (WTP £3.46).The replication study in 200 people with breathlessness due to chronic lung disease showed very similar results. We also examined heterogeneity in valuation between individuals and found this was present, particularly around the relative value of tailored information.

Consequences

The direction of effects here (with situation-relevance and what matters in life being most highly valued) may not surprise researchers of person-centred healthcare, but this demonstration of the magnitude of effect is new, striking and consistent across two conditions. It should lead us to rethink how we make support for self-management (which, indirectly, includes much of what we do in primary care outside of diagnosis) properly person-centred. Person-centred healthcare must emphasise the substance of personalisation and not just the style.