Will a single rating for hospitals limit or support patient choice?

This content relates to the following topics:

I've been planning a weekend city break in Europe and wondering which hotel to stay in. I looked at the 3 and 4 star hotels on the tourist information site. They varied hugely in price and I couldn't tell which ones had good customer service. I asked some friends and family who had been there, but no-one had been recently. I read some of the reviews on TripAdvisor. In the end I booked one that was close to the city centre with a swimming pool and availability for the dates I needed. No-one had written a bad review so, while it’s only 3 star, I hope it will be fine.

Now read this again but substituting hospital for hotel. I need major surgery and am wondering which hospital to attend. I looked at the 3 and 4 star hospitals on the official rating site. The clinical outcomes varied hugely and I couldn't tell which ones had good patient care. I asked some friends and family who had been there before but it was for a different procedure. I read some of the reviews on Patient Opinion. In the end I booked one that was close to home and that had the earliest availability. No-one had written a bad review so, while it is only 3 star, I hope it will be fine.

The Nuffield Trust has been leading a health and social care ratings review for the government on the feasibility and desirability of a summary rating for hospitals and other care providers. They are due to report later this month. In our submission to the review we highlighted the challenges of producing a single summative measure. The above scenario illustrates the limitations of such ratings to support patient choice.

The reality is that patient choice is limited. It applies only to those patients who are referred to a specialist. Many patients are not aware of their right to choose and are not routinely offered choice by their GPs. If these patients then require elective treatment they are not usually offered a further choice of where to have the procedure. Most patients never have the time to consider their options as they need urgent treatment.

Even if they do have the opportunity to choose, different patients value different things. While many put safety and effectiveness of clinical care before location/ proximity and patient experience, in reality most choose to go local. A summative rating gives weightings that reflect the values of those constructing it, not necessarily the information valuable to patients.

Patients also want information specific to them and their condition. If you are going in for a heart operation you want to know the outcomes for patients having the same procedure, not the average outcomes for all patients undergoing surgery in the hospital. The NHS Commissioning Board has committed to publishing more of this detailed clinical information on individual surgeons. However, this information is often complex for patients to interpret as it is usually presented as probabilities and is adjusted for case mix. Hospitals are complex organisations providing a wide range of services, reducing these to a single measure makes the information less useful and is likely to mask a multitude of sins.

In reality, many patients ask their GP or friends and family for advice. Personal testimonies such as those now on NHS Choices and Patient Opinion have a powerful effect on decisions but they skew perceptions, even when presented alongside objective data on quality. It is likely that these will play an increasing role in patient decision-making in future but this does not necessarily mean they will be well-informed decisions.

From April 2013, every NHS hospital will be required to ask patients in A&E and on the wards whether they would recommend the hospital to a friend or relative if they needed treatment. The so called family and friends test aims to get real-time patient feedback at ward level, but again its value is limited by its reporting as a summative score. It is the qualitative statements behind the scores that are important; they can be acted on by clinical teams and the organisation can analyse themes which might suggest systemic problems or issues which need addressing.

The government is right to want to make more information about quality of care available to patients and the public. But the information must be tailored to individual preferences, at a level of detail that is relevant to the patient (ie, specialty level), presented in a way that is understandable and not skewed by qualitative perception. Ideally patients should also have the opportunity to discuss their decision with someone.

A more powerful reason for collecting and publishing data on quality is to drive improvements by providers. There is good evidence that this is enhanced when benchmarked data is published. The reputational effect on providers, particularly those who are poor performing, is a powerful driver of change. While no one would advocate a return to naming and shaming under the old star ratings, there is a role for public ratings if only to make boards sit up and pay attention to quality.


Paul Hodgkin

Patient Opinion
Comment date
14 March 2013
Thank you for a thoughtful blog. As one of the team behind Patient Opinion I would agree with many of your points. In particular the compelling nature of stories and the difficulty of using them to make rational choices. This is compounded because, as you say, people want very granular level information if it is to be useful. Given that we can handle feedback about the 100,000 clinical services or organisations across health and social care listed on Patient Opinion there will always be a problem about getting realistic volumes of feedback at the level patients want it.
A useful way round this is to measure how organisations respond to patient feedback rather than the feedback itself: how often does a hospital respond? How often do they make a service improvement? How are responses rated by authors and by the public? How many staff at what grade have actually read any given story? This both produces useful real improvements and generates a response which is what motivates most authors in the first place.

Mark Rickenbach

Professor Healthcare education and quality,
Wessex Deanery for Postgraduate Medical Education
Comment date
31 March 2013
The best source of advise about a hospital are those who refer patients to the hospital and those who work in the hospital. How we move from anecdote to an effective summary of opinion is the question?
An open easy to access system of feedback works if we look at the pattern of feedback rather than every detail. This will guide organisations on where to focus resources and to ask more about the issues identified and then ask in more detail locally
When it comes to grading hospital and community trusts it helps to consider the unit size that is reported. A department (specialty in a location) is the most useful unit of reporting that then can be aggregated in different views and give the overview needed. The pattern of grading in departments across one trust is more useful to a user than an overall grade for a trust. The user (doctor or patinet) will want to look at the grade for the department they are referring to. This is already done for healthcare education in the Wessex Deanery for postgraduate medical education

Add your comment