NHS England is under fresh pressure to rethink the presentation of results of its friends and family test amid warnings those results based on few responses are damaging staff morale and needlessly alarming patients.

The Picker Insitute, a respected patient feedback analytical body, called on NHS England to “suppress” any results from wards where fewer than 10 patients filled in the satisfaction survey.

Picker director of policy and research Chris Graham told HSJ: “We strongly recommend that data based on very small numbers, for example fewer than 10 responses, should be suppressed when reporting to avoid creating unreliable impressions, unnecessarily damaging staff morale, and alarming local communities.”

The body analysed the 36 wards which were widely criticised in the national media for failing the test - by getting a negative score - in the first tranche of results released last month. Just three of them recorded a negative score in the latest sets of results, covering July, which were released yesterday.

Picker’s analysis of the 36 wards shows:

  • 29 out of them received a positive score in July, including three which swung from a score of -100 to +100. These were all based on less six responses or less
  • Two had a score of exactly zero (both based on 3 responses)
  • One had no responses (and no eligible patients)
  • One is not included in the dataset for July, although it remains unclear why.

Mr Graham added: “For July, there are 30 trusts with a score of less than zero, and once again the majority of those have very low numbers of responses: 21 of the 30 have fewer than 10 responses, and only four have more than 20.”

Picker’s analysis comes as some sources have told HSJ a number of nursing staff working on wards accused of “failing” had “considered resigning” following the national media outcry in the wake of the previous set of results.

And David Dalton, chief executive at Salford Royal Foundation Trust where two wards were judged to have failed under the June data, told HSJ it was “grossly unfair” wards with low response rates were identified in this way.

He added: “I do support this information being in the public domain but we need to standardise data collection… For us the most important thing is to use this for learning rather than comparing ourselves with other trusts.”

As reported by HSJ earlier, Yeovil Foundation Trust – which was labelled by NHS Choices as being “among the worst” trusts, has accused the respected NHS website of presenting the test in a “misleading” way and of “confusing” patients.

Overall, NHS trusts have continued to struggle to gather responses to the friends and family test from accident and emergency patients.

Across inpatient and A&E settings, 16 per cent of patients undertook the new patient satisfaction survey in July - above NHS England’s 15 per cent minimum requirement. However, only 10 per cent of A&E patients completed the test.

The overall response rate was bolstered by a 28 per cent collection rate in inpatient settings.

wide variation of response rates remains evident between trusts.

There are some correlations with data for the first three months of the test. The top five trusts on the inpatient score based on the July data are all specialist hospitals - as they were in previous months.

The Royal National Hospital for Rheumatic Diseases Foundation Trust topped the pile, scoring 100, while Robert Jones and Agnes Hunt Orthopaedic Hospital Foundation Trust and the Royal Marsden Foundation Trust also scored highly (see attached table).

Barking, Havering and Redbridge University Hospitals Trust came bottom, as it did in the last set of data released, scoring 39, on a response rate of 49 per cent.

The Royal Free London Foundation Trust, Lewisham Healthcare Trust, Royal Liverpool and Broadgreen University Hospitals Trust and Wirral University Teaching Hospital Foundation Trust make up the rest of the bottom five on inpatient score.

Hinchingbrooke Health Care Trust, run by private firm Circle, recorded the highest combined score for the second month in succession for an NHS trust which has not achieved foundation trust status, scoring 82.   

Giles Wilmore, director for patient and public voice at NHS England, said: “Where only a few patients have completed the survey in an individual ward, it is just as important that their voice is captured and feedback is not lost in national publication, whilst respecting their individual anonymity.

“This is why, when the number of responses for an individual ward is fewer than five, we are not publishing how individual patients responded, but still publishing the overall score for the ward in order to ensure we are being transparent.

“The friends and family test’s primary purpose is to give trusts information on how well they are meeting the needs of the patients they care for and to ensure they can make improvements.

“We have been clear that scores should be viewed in the context of response rates, and we hope that Trusts will continue to work hard to increase their response rates in order to gain the richest possible information, so that they can analyse and act upon the views of their patients.”