Abstract
Much has been written on the kind of culture which supports safe patient care. One of the founding members of Clinical Risk, Professor Charles Vincent, has been a prolific contributor to this debate. His recent book with Rene Almaberti, 1 is essential reading for anyone interested in patient safety, not least for its cogent mapping of the development of thinking in this field, and for the analysis of where we are now in the journey towards maturity. I have been unable to resist the temptation of applying (very loosely!) the three models of safety identified in the book to recent news stories concerning the well-being of children, each of which has provoked debate for a wide range of reasons.
The first model is the ‘ultra-adaptive’ model, where autonomy and expertise trump organisational rules or hierarchies. When things go wrong, this may be assigned to a lack of the ‘right stuff’ – if you are lucky, you learn by your mistakes. Consider, (but please do not emulate …) the recent experience of the seven-year-old Japanese boy deserted in bear-infested woods by his parents for bad behaviour. He was, after several days, found safe in a shelter of his own making, having found water, and survived without lasting physical harm. We should not, perhaps, seek to link his naughty behaviour with his impressive adaptive skills.
The second model is the ‘high-reliability’ model, more associated with industries such as nuclear and aviation, where individual expertise is valued but within an organised, team-based approach. This model is characterised by a systematic approach to learning from experience in order to improve both management and prevention of harm. The shooting of a gorilla in the Cincinnati zoo, after a four-year-old boy tumbled into its enclosure, split opinion as to the appropriateness of the response. In terms of my analogy, the decision to shoot the animal was implemented using professional expertise but doubtlessly taken within the parameters of procedure.
The ‘ultra-safe’ model is where prevention is key: exposure to hazards is minimised and there is little need for improvisation or adaptive behaviour. The wisdom of protecting children from exposure to any possible risk is controversial, but examples abound: the introduction of safe play surfaces to prevent grazes; Health and Safety Executive research showing half of children aged between 7 and 12 are not allowed to climb a tree without an adult present, and one in five children of the same age have been stopped from playing conkers. 2
Vincent and Almaberti point out that safety results can be produced by all of these models – that rules and constraints can be effective in some situations, just as the flexibility and ingenuity of individuals can in others.
Understanding where rules and behaviour may be implicated in the occurrence of harm depends on the quality of information we have about failure and potential failure. All types of safety models in healthcare will depend to varying degrees on a healthy reporting culture, in order to ensure that systems learn from things that go wrong.
Central to a strong reporting culture is the willingness of staff to raise concerns: the ‘openness, transparency and candour’ which Sir Robert Francis identified as necessary in his report into standards of care 3 and which his more recent review ‘Whistleblowing: freedom to speak up’, 4 put under the spotlight. The review concluded ‘there is a culture within many parts of the NHS which deters staff from raising serious and sensitive concerns and which not infrequently has negative consequences for those brave enough to raise them’.
The three models of safety described above are each context dependent, and change and improvement in healthcare is limited by the culture, values and demands of government, regulation and society. In terms of addressing the limitations of the culture identified by the Freedom to Speak Up review, this gives us a problem. Putting aside the recriminations that whistleblowers may, with good reason, fear, the review also found that anxiety about the possibility of ill-founded, or even malicious, allegations was cited as a barrier to dealing with concerns, and that the potential effects on an individual’s career or reputation were also deterrents. A review of social media or the popular press will evidence that the odds are stacked in favour of ‘where there’s smoke there’s fire’ rather than ‘innocent until proven guilty’. Experience of incident investigations shows that actions to protect patients may be delayed due to caution and reluctance to call clinical practice or behaviour into question. Unfortunately, the belief that, once the concern has been raised and an investigation started, the damage to reputation has been done, is not without justification. The need to blame is still embedded in our society and also in our medical culture, where individual autonomy and responsibility are naturally highly valued.
While we cannot alter society’s values overnight, there are still organisational actions to be taken to protect the patient, the whistleblower and the practitioner. Once the NHS can share an anthology of examples of a positive response to concerns, all approaches to improving safety will be more effective.
Safe healthcare relies on both rules and adaptive behaviour; on professional expertise and innovation as well as the identification and implementation of safe standards of care. Recognising the right moment and setting to apply these different approaches is the skill of the healthcare practitioner: it may yet help us to steer clear of those bears.