Comment

Making the Friends & Family test work for the NHS

Toby Knightley-Day responds to the new guidance on the Friends and Family test published by the Department of Health.

We should all welcome the publishing of the Department of Health’s (DH) Friends and Family test (FFT) guidelines for the endorsement and clarifi cation they provide. Initial test work undertaken by NHS Midlands and East has proved that there is both an appetite and ability to operate a test of this kind, and this guidance identifi es a standard question and approach that will eventually allow for some form of comparability between trusts.

Nonetheless, though the guidelines have confi rmed and standardised a number of important aspects of the programme, they have not mandated other important elements or elaborated on the all-important score reporting mechanism.

What follows is an analysis of the Department of Health’s guidelines and suggestions for ways in which trusts can make the most of the test.

Wording of the final question

The standardised question reads:

‘How likely are you to recommend our [ward / A&E department] to friends and family if they needed similar care or treatment?’

This is representative of the question proposed by Fr3dom Health during their involvement with NHS Midlands and East, and provides the following advantages:

1. The qualifi cation statement ‘if they needed similar care or treatment’ is an important inclusion; without it, the natural response might be to think ‘I suggest my family does not get itself in this situation at all’, or variants thereof;

2. The question allows for localisation without diluting the core of the question. Points of service delivery will be identifi able down to ward level, something especially signifi cant should the FFT be extended to other sectors of the NHS beyond the acute;

3. The use of ‘our ward’ rather than ‘this ward’ gives a sense of ownership that will encourage staff to take on the responsibility for collecting the data and working to improve scores.

Used in all participating trusts, answers to this question will be comparable. More importantly, if used correctly, it provides scope for staff to react to patient feedback locally and make changes to the way in which they deliver services.

The question of free text

The mandated 48-hour period in which data has to be collected is vital as it is the only way the information can ever be reported quickly enough for front line staff to react to it. Although there are still challenges in relaying the information to this group in time, there is little point capturing experiential feedback at or near the point of experience and reporting it several weeks later.

The FFT differs considerably from the traditional survey instruments in this respect as it emphasises the importance of collecting data close to the experience. However, despite this step in the right direction, the DH has not yet explained the reporting process for the scores and has only recommended that trusts implement a free text question.

Originally, the point of asking the Friends and Family question was to allow patients’ experience data to be used to effect real change in service delivery. The inclusion of free text acknowledges that dry, quantitative data alone is not enough to drive change at the front line. While this element of the test has been made optional, we would suggest that this is where much of the true value lies.

Trusts must now make a choice: to use the Friends and Family test to change behaviour and improve service, or to collect data to provide a reporting mechanism that has no obvious frontline purpose.

The danger is that trusts do not grasp the fact that the real power of the FFT comes from its ability to drive change and improve service delivery. As a result, the measure of its success will in part be determined by the methods used to report results back to staff, as fi gures alone do not highlight areas in need of improvement.

If the patient-staff feedback loop is not closed quickly enough with a free text question explaining a patient’s score, then the FFT may not be the proactive, quality and patientcentred initiative it should be. At worst, it could become a turgid process in which the only outcome is a predictable correlation between national survey results and FFT scores.

The 6 point scale

The trials conducted in NHS Midlands and East utilised a Net Promoter Score-inspired (NPS) 11 point scale: 10 for those extremely likely to recommend, zero for those extremely unlikely.

The guidelines, however, have reduced this to six options, and made the numerical scale into a series of text answers. One of the obvious concerns this raises is how trusts will generate and report a ‘score’ from a text scale; something emphasised by the lack of guidance in the published document.

Although this move is not ideal, it is understandable – after all, the six point scale is closer to more traditional survey techniques that have been proved widely in healthcare settings. However, the obvious downside is the dramatic reduction of ‘granularity’ and sensitivity compared to the 11-point system.

This denies us the chance to drive a higher degree of precision into patient experience data. The loss of this valuable granularity may of course be mitigated by careful survey design, but as a result of this decision it is not inherent in the chosen scale. Whereas a small shift in responses on the NPS-style score would be visible very quickly, the reduced scale is less sensitive and, therefore, less able to differentiate between small and medium sized issues.

This in turn means that trusts will need to seriously consider overall survey strategy and align operational practices with it; after all, understanding how staff behaviour affects the score is the only way trusts can begin to improve it.

Making it all happen

In spite of the uncertainties introduced by this document, there is also good news: with the question and the free text response we have a great opportunity to make patient feedback work harder than it has before.

All that remains are the specifics of implementation, such as: how do we encourage trusts to embrace the free text question? How can free text be written, gathered, interpreted and acted upon in a timely manner? How much will it cost? What are the cultural and practical implications if implementing the FFT? Are there governance issues to consider?

The list could go on. Trusts must start to think about these and other issues to ensure that the test is not excessively disruptive or expensive, and to make the results meaningful and credible.