Let’s all agree first of all, that we need more patient ‘voice’ in the NHS. It’s actually quite painful to review research on how bad the NHS has often been at dealing with complaints and attempts to get patients (and the public more generally) involved in health organisation. We need to do better.
However, I’m going to argue that one of the main forms we use to achieve this, anonymous patient surveys, aren’t that helpful in achieving improvements.
It’s a cornerstone of research, not just in the NHS but more generally, that anonymity is hugely important in responding to surveys. I understand why this is – especially in a setting such as healthcare where there are obvious power and information imbalances between patients and clinicians. However, the assumption has serious down-sides that are also not often acknowledged.
If we survey patients to ask them to think about how good (or otherwise) their experience has been of healthcare, we want them to be as honest as possible. We don’t want them to fear repercussions later on for saying when they’ve received bad service or clinicians who seem more concerned with having an easy life than than in being challenged in what they are doing. I think that this only applies to a minority of clinicians, but they are out there.
The trick is in getting patients to respond honestly to patient surveys. Is anonymity the best way? Intuitively it is, and many research bodies insist upon it. However, I’m concerned that research in areas such as behavioural economics suggests otherwise. Research reported by Dan Ariely, for example, suggests that anonymity can lead to cheating and exaggeration. Are patients doing this? We don’t know.
What I do suspect is that, in an environment where patients are increasingly encouraged to consider themselves as customers, and law firms are advertising medical negligence systems hard, the danger of vexatious complaints increases. Most patients will treat feedback systems responsibly, but a few won’t, accusing clinicians of the most appalling things. The asymmetry is that these accusations will be investigated and clinicians face potentially career-ending claims against them, even when there are no grounds underpinning those complaints, but have little or no comeback where a complaint or accusation is found to be groundless.
This situation is amplified when we consider websites that gather patient feedback. In these circumstances both lavish praise and out-of-proportion criticism can appear with little or no moderation, and no repercussions. It only takes one patient who feels hard done-by, but perhaps has significant problems of their own, to potentially tarnish the reputation of a clinician on-line. Once posted a comment is potentially there for years, with a clinician able to do little about it in response.
Anonymity also raises a problem in that it becomes just about impossible to see if the comments left are representative, or the result of organised action on behalf of an aggrieved patient who has posted several times, as well as perhaps getting friends and family to do so. This can result in a clinician or service being vilified on-line and appearing to provide poor outcomes when it may have comparatively few problems, but a vocal group who want to campaign against it. No clinical service wants to leave any patients feeling they have been poorly serviced, but it’s inevitable that some will feel that way.
Anonymity is great in encouraging the powerless to find a voice, but it is also a means by which people can make complaints and claims that are unsupportable. I suspect we are going to see a great deal more of this problem in years to come as feedback websites multiply and are incorporated into systems for measuring the quality of care. In such situations patient feedback can be brought out or presented with little regard as to whether it is representative or without having checked whether the complaints are justified.
It is crucial that we incorporate patient voice into measures of outcome, but I’m not sure anonymous comments are the best way of achieving this. Possibly unrepresentative messages about poor practice are too important not to be specifically investigated, or to be made in an unaccountable way.