• attitudes

Alongside concern about avoidable mortality, one of the key findings of the public enquiry into failings at Mid Staffordshire NHS Foundation Trust,1 which ran Stafford Hospital in England, was the lack of compassion in care delivery. Sir Robert Francis, who led the enquiry, laid the blame for the compassion deficit at the door nursing and support staff. He recommended, among other things, that people should work as care assistants prior to nurse training and that values-based recruitment should be used to ensure that the ‘right’ people are recruited to be nurses. However, there has been little evidence to support these propositions. For example Snowden et al 2 found that nursing students who had previous care jobs scored no higher for emotional intelligence than those without prior experience.

More recently opinion has shifted to the impact of compassionate leadership3 and compassionate environments on team behaviours. This move in part reflects the definition of culture as the shared assumptions that are normalised and are therefore absorbed by new staff as the correct way to think and behave.4 In this issue of BMJ Quality and Safety, Bridges and colleagues5 describe an intervention to develop the workplace at ward level through the use of a structured approach to creating learning environments for compassionate care (CLECC). The intervention uses reflection and group activities, including mid-shift cluster discussions to reinforce a team climate of compassion with a view to increasing the compassionate behaviours of individual members of the team.

Bridges et al used normalisation process theory6 to evaluate the implementation of CLECC. They conclude that staff found it coherent and valued the principles. However, for some ward staff it remained a series of activities that they could articulate but had not become part of the shared assumptions of the ward. New starters were not socialised in its principles. Ownership of the intervention was variable: in some wards it was located outside the team and resided with practice development nurses. The authors suggest that the extent to which the intervention was faithfully implemented and normalised within individual wards was mediated by its relationship to the wider organisational agenda.

The partial implementation of an intervention is a common problem. In this case, the researchers observed a mismatch with organisational imperatives, perhaps indicating a lack of alignment between the priorities of the executive, the middle managers and the front-line staff.7 For managers, the drive for increased patient flow and cost efficiencies may be paramount, meaning that taking time out for reflection can appear to be counter cultural. This was illustrated by the manager described in this study who asked why staff were standing and drinking in the corridor: they were, in fact, undertaking a mid-shift cluster discussion.

The difference in priorities and therefore lack of apparent organisational support may well be part of the reason the intervention was not fully imbedded, but is it the whole story? Compassion is sometimes thought of as a fundamental human trait. But Benner and Wrubel8 assert that caring for strangers is significantly different from caring for friends and family. Professional caring involves emotional labour, which Hochschild9 defines as the induction or suppression of a feeling in order to produce in other people a sense of being cared for in a safe place. At times, this kind of labour can be overwhelming; staff can become stressed and less able to induce or suppress their feelings, and may avoid situations that require this behaviour. Isabel Menzies Lyth’s10 classic work in 1960 showed how nurses organised their work to reduce emotional labour by splitting contact with patients into tasks in a way that is often seen as the antithesis of compassionate care.

In the 21st century, person-centred care is widely espoused. But when emotional labour becomes unsustainable, a task orientation may become dominant. Ball et al’s11 study of English hospital wards found that 86% of nurses reported missed care on their last shift. They found a striking difference in the nature of what was missed; most commonly it was comfort/talking with patients (66%) and educating patients (52%), whereas the least likely to be missed were tasks such as pain management (7%) and treatment and procedures (11%). The fact that the nurses reported the care as missed implies insight into the needs of their patients for compassionate care. Both Ball and the authors of this paper propose that the omission is due to lack of staff time to undertake relational care; certainly, compassionate behaviours are not factored into calculations of staffing requirement. But this does not explain why two members of staff facing the same workload, within the same workplace climate, may behave differently. The answers may lie in a positive working environment and support for reflective practice that can help staff manage their emotional labour.12

Bridges et al showed that where the CLECC intervention was successfully implemented, there was an increase in staff morale and staff well-being. However, participating in CLECC involved talking about missed care, and it may be the case that the wards in this study which did not fully adopt CLECC were those in which staff were aware that they were omitting activities related to compassion. Rather than a lack of time or an unsupportive culture, nurses on these wards may have been actively managing their own well-being and avoiding cognitive dissonance by opting out of the intervention.

While personal agency cannot be the only explanation for lapses in compassion, it would be wrong to discount it. CLECC addressed the environmental conditions that facilitate compassion, and the challenge for implementing it further will be to integrate it with other factors including the management of individual emotional labour.