Intended for healthcare professionals

Head To Head

Will doctor rating sites improve the quality of care? No

BMJ 2009; 338 doi: (Published 17 March 2009) Cite this as: BMJ 2009;338:b1033
  1. Margaret McCartney, columnist for Financial Times weekend
  1. 1 Glasgow
  1. margaret{at}

    The UK government has signalled its support for rating sites with the decision to set up its own through the NHS Choices website. Neil Bacon (doi:10.1136/bmj.b1030) believes they provide essential feedback for doctors, but Margaret McCartney is concerned that the potential harms are still unclear

    The health minister Ben Bradshaw thinks doctor rating sites are a great idea. “I wouldn’t think of going on holiday without cross referencing at least two guide books and using Trip Advisor. We need to do something similar for the modern generation in health care.”1 If we can ask the eaters of pizzas and drinkers of coffee for their ratings of the staff, why should doctors escape the judgment of their clients? And so, with a nod to patient choice, and no apparent need to consider the evidence, NHS approval of websites to rate your doctor was stamped.

    There is a wealth of information about how patient opinion is useful, even essential, when considering how services and research should be shaped and delivered. But the belief that doctor rating sites will promote trusting doctor-patient relationships or help patients find medical practitioners with a particular style are untested and potentially dangerous suppositions.

    Evidence is missing

    Recent medical history teaches us the hazards of not obtaining and acting on quality evidence.2 3 A recent example is the much heralded Choose and Book system, which was introduced to enable general practitioners to book secondary care appointments for patients. The system was supposed to provide numerous supposedly common sense benefits,4 one of which was a reduction in the number of people not turning up for appointments. In fact, the rate of non-attendance with Choose and Book has been found to be higher, at 18%, compared with 12% for traditionally organised appointments.5

    Failure to consider the possibility that a new intervention could have unintended adverse effects remains a significant problem in health care. While there is evidence that good communication and trust between patients and doctors leads to better outcomes, it is uncertain that doctor rating sites will contribute to this positively. A negative review read by a patient may prevent a previously good doctor-patient relationship from continuing with the same ease. An inaccurate positive review may lead to more patients attending a doctor they would not have chosen had they used their own judgment, or that of people known to them. Good doctors who get bad reviews may change clinical practice needlessly or even harmfully; bad doctors may be reassured by good reviews. I imagine that the serial killer Harold Shipman, who was liked by patients, would score well. There is no evidence that feedback collected on websites such as is representative or accurate.

    Will it work?

    Can patient feedback improve doctors’ performance? The first question to ask is what kind of feedback is “good.” Some negative feedback may signify good practice: even the best doctor may experience minor disgruntlement from patients over hypnotics, sedatives, or antibiotics not being prescribed or sick notes not being dispensed. But while such incidents may be forgotten—and may even be remembered with retrospective gratitude by the patient—online they persist forever. Some medical work—for example, child protection or psychiatry—has the constant potential for conflict. What will be the effect of a damning assessment of a doctor on a rating site? Confidentiality and anonymity means no medical right of reply: only libel will be taken down.

    There have already been concerns raised about other factors, such as socioeconomic gradients, which may affect satisfaction with general practice services: researchers have cautioned against using satisfaction as a proxy for quality.6 Could such online ratings make doctors more cautious about working in certain specialties or deprived areas? We do not know. Nor will we know if comment is fair or unfair. What will the impact be on that doctor, their current patients, their family, and their children? There are currently no data about that. And as far as I know, there is no plan to search for such potential adverse effects either.

    Additionally, negative or positive feedback delivered anonymously online has limited ability to be related to specific incidents. This would make it difficult, if not impossible, for a doctor to try to learn from posted comments. A systematic review examining the effect of patient assessments on doctors’ interpersonal skills found only limited evidence of benefit.7 However, most of these assessments of doctors were solicited by serial patient surveys, as currently required as part of the general practice contract. These surveys may not be of proved benefit, but at least by asking sequential patients to participate they attempt to obtain properly reflective data. The collection of data on rating sites is from self referrers only. This creates an unscientific skew.

    How much money will the NHS and charities be paying for this non-evidence based intervention which comes with potentially damaging strings attached? For selling information is what sites like plan to do. And when will we learn that if we do not look for adverse effects, we will be dangerously unaware of them? Instead of wasting resources on doctor rating websites it would be far better to spend it on implementing strategies that are known to improve patient satisfaction, such as continuity of care and longer consultations.8 9 10


    Cite this as: BMJ 2009;338:b1033


    • Competing interests: MMcC is a general practitioner.