The friends and family test: a different view

The friends and family test has had a rough ride, however, the tide seems to be turning as big players such as the Picker Institute acknowledge it can generate useful data. Even as an advocate, I appreciate the test is far from perfect but there are many reasons why it is step in the right direction for patient engagement in the NHS.

‘Collecting patient experience is irrelevant unless it is inclusive; everyone’s view matters, all the time’

Many commentators have attempted to discredit the test without suggesting any viable alternatives. Indeed, most seem bogged down with sampling issues rather than participation and involvement, which is a shame as the NHS belongs to us all.

The Francis report and others highlighted the barriers preventing the patient voice being heard, so, while I greatly respect the work of established national survey architects, I was surprised to read the Picker Institute’s Chris Graham suggest small base sizes make the friends and family test data “effectively meaningless”.

• Neil Churchill: Patient feedback is essential to the NHS’s future

Not only does this devalue personal feedback, it also misses the point. How do you explain to a patient that her feedback about her dying husband’s hospital experience was “meaningless” because it was a lone opinion?

Collecting patient experience is irrelevant unless it is inclusive; everyone’s view matters, all the time. This is the weakness of the national survey tools − collected between specific dates; they are delayed and controllable snapshots rather than live and inclusive listening tools.

Enormous numbers

The significance of this fact is best illustrated with statistics. The NHS deals with 463 patients every minute − that is 666,667 a day, or 1 million patients every 36 hours. These are enormous numbers.

It is surprising, then, that the national inpatient survey 2012 gathered just 64,505 responses, sampling less than 1 per cent of the annual inpatient population. To achieve this, a sample was drawn from patients that met specific criteria: they were aged 16 or older; and had at least one overnight stay in June-August 2012.

This was done to create a “representative” sample, although, from a different perspective, it meant your voice was irrelevant if you did not meet these conditions.

‘Listening to the incessant moaning by critics, you would have thought the score was the only way this data can be used

A comparison between this and the friends and family test speaks for itself: in three months, the test collected 440,000 responses − almost seven times the number collected by the survey. By that yardstick, its 15 per cent target is huge and it has the potential to gather 2 million responses by the end of the year in acute and accident and emergency settings alone.

Patient and staff engagement

In his article, Chris made reference to two issues: one glaringly obvious, the other inconsequential. The first was to point out that survey results need staff engagement to stimulate improvement. Is this really a surprise? Any organisation with the slightest interest in using patient feedback already does this.

The other was to question the impact of “mode effects” on the comparability of data from organisations using different test collection methods.

‘We should be encouraging a move away from purely analysing the league table so we see this tool for one of its strengths: providing experience information to the frontline’

In this case, do we not have to accept and adapt to the fact that the public uses multiple communication channels? After all, which is more important: making sure patients feel comfortable about engaging, or denying individuals a voice because their preferred method of communication doesn’t fit with your survey model?

Recognising the good

Similarly, too much criticism continues to be directed at the reporting of results.

Listening to the incessant moaning by critics or the grudging support from traditionalists, who now know the test is here to stay, you would have thought the score was the only way this data can be used. This is why we should be encouraging a move away from purely analysing the league table so we see this tool for one of its strengths: providing experience information to the frontline.

‘For it to be a success, we must focus on the friend and family’s ability to drive behavioural change’

The nature of the friends and family test and its comparatively large base size, combined with ward level metadata, means there is a constantly evolving picture of service quality − with this data, staff can identify problems and find solutions. The fact that this is such a deviation from traditional mechanics seems to be central to the resistance among traditionalists.

Established players have applied established tools to the NHS for years, and it is questionable whether this has delivered a safer or more responsive service. The new test is not perfect, nor is it a statistical tool. But it has never claimed to be.

The test is a blanket approach that demands services provide an opportunity for every patient to feed back. It will always be more inclusive, more widespread and more relevant at ward level than any more detailed report containing exclusively sampled, aged data.

For it to be a success, we must focus on the friend and family’s ability to drive behavioural change. On that note, it is pleasing to see more commentators endorse free text. As for the score − well, it’s an indicator, and a useful one given the scale. My advice, however, would be to focus on the relationship between consistency, range and comments − this will be considerably more productive.

Toby Knightley-Day is managing director at Fr3dom Health