Methods

A phenomenological approach as a ‘sensitising’framework

The analytical frameworks developed in this book were all useful informulating an array of empirical markers pertaining to the existence,nature and/or limits of trust. More particularly, we were especiallyattentive to various modes by which trust was linked across differentdimensions of healthcare services, partially building on an initialframework set out by Gilson and colleagues (2005). Trust has been portrayedin this study as dynamic and contingent. Accordingly, changes in trustlevels over time and features associated with this (those that were held tolead to changes in trust, as well as features associated with the resultanteffects of changes in trust) received particular attention within theanalysis.

As articulated in Chapter One, trust – as a construction ofexpectations – is very much dependent on how attitudes and behavioursare interpreted and then inferred to indicate intentions ofaction – regarding trustees’ interests and embeddedness withinnormative structures (Möllering, 2005). These features of trust as aprocess of building knowledge from an array of experiential sourcesindicated the utility of a phenomenological approach as a‘sensitising’ framework (Hammersley and Atkinson, 2007), whichinformed both the format of interviews and the coding of data (see more onphenomenology as an approach to trust in Chapter One).

In this way, the difficulty of identifying and coding the potentiallynebulous concept of trust was overcome. A focus on the interpretiveexperiences of actors and the way these were drawn upon when constructingfuture expectations was recognised as especially salient when codingunderstandings of trust – as an ongoing, embodied, conditional andnegotiated process of sense-making and expectation construction (Smith andOsborn, 2003). When analysing the data, the informants’ use ofideal-types (Schutz, 1972) was also apparent and noted – for example,generalised views of the quality of mental health services, the motives ofpsychiatrists or the likely concordance of certain service users. Thisideal-typical knowledge formed important building blocks when constructingfuture expectations (Brown, 2009a).

Design

The dynamics of different trust relations were explored through a single casestudy involving informal interviews with service users, professionals andmanagers (n = 23) working in three services designed tomeet the needs of people experiencing psychosis – all within one NHSTrust (local health authority) in Southern England.