Shame and Working Alliance in Healthcare Relationships

Abstract

Working alliance – the collaborative, purpose-driven elements of the relationship between healthcare provider and consumer – is a well-evidenced predictor of successful and effective healthcare treatments in a variety of settings, both medical and allied health. It is therefore important to identify and understand which characteristics of consumers, clinicians and environments contribute to the development and maintenance of a robust alliance. Healthcare interactions have long been seen as environments in which the experience of shame is commonplace, requiring the revelation of private information and the body, and discussion of one’s own behaviour and its impacts on health and wellbeing. Research also suggests that shame is an aversive and unpleasant experience which frequently motivates avoidance, withdrawal, defensiveness, and anger. It is reasonable to suppose that a tendency to experience intense shame may be related to a weaker working alliance, with flow-on effects on the efficacy and tolerability of treatment. Limited research exists in this area. The aim of this research was to develop an improved understanding of shame’s role in working alliance in a variety of healthcare settings. Study one was a scoping review which examined literature on shame and alliance. Studies were included if they pertained to healthcare relationships, were quantitative in nature, written in English, and included empirical measures of shame and working alliance (or proxies). The scoping review identified 13 studies which met inclusion criteria. The synthesis suggested a small, negative relationship between shame and alliance. Methodological limitations were also identified in past research. It was determined that additional research was needed to understand the hypothesised shame-alliance relationship. Study two was a pilot study conducted with community members (N=127) who had regularly seen a GP over the past year. Participants were asked to reflect on their relationship with their GP, and completed instruments assessing shame, as well as working alliance and satisfaction. Small, negative correlations were found between shame measures and working alliance, as well as patient satisfaction. One specific subtype of shame (external shame) had an indirect effect on patient satisfaction through working alliance. The findings from this study suggested that shame was an important factor to consider in the provision of medical care to maximise the quality of patient experience and working alliance. Study three replicated and expanded upon the methodology of the pilot study, including additional measures to explore potential mechanisms to explain the shame-alliance relationship, such as items assessing generalised and specific tendencies to disclose distressing information. A similar set of measures to the pilot was included for this sample (N=177 community members), with the addition of a secret-specific shame measure. Participants with past psychotherapy experience (n=104) were also asked to retrospectively rate their working alliance with their therapist. Contrary to the hypotheses and to the findings of the pilot study, there were no statistically significant correlations between shame measures and working alliance. Those correlations that were evident were typically small and negative. It was determined that the sample was underpowered to detect an effect of the predicted size. Study four was concurrent with study three and aimed to investigate the shame/alliance relationship in a clinical sample. Participants were current psychotherapy clients (N=18) as well as their treating psychologists (N=6). This study included both bidirectional ratings of working alliance (client and therapist) as well as ratings over time, with participants providing data across 1-10 treatment sessions. Hierarchical linear modelling was used to assess for a shame/alliance relationship. The hypothesised relationship was not detected in the sample. As with study three, the lack of statistically significant findings was attributed to the low power of the study. This series of studies represents a preliminary investigation of the hypothesised shame/alliance relationship in healthcare. It identified a small, negative relationship; however, the evidence was modest and sometimes did not reach statistical significance. Overall, the results suggest that those with a greater tendency to experience shame may report somewhat weaker alliances – as such, clinicians may benefit from remaining alert for signs of high dispositional shame or including brief assessments of shame as part of an intake questionnaire, to better serve highly shame-prone individuals.