Are NHS complaints really ‘the tip of the iceberg’?
“It is difficult to know how many people want to complain, but maybe we should be asking whether complaints are the right measure of quality….”
It’s been widely reported that complaints about the NHS are on the rise – and, worse, that the complaints that are received may reflect only “the tip of the iceberg” (eg    ). This follows the publication of new data from the Health and Social Care Information Centre (HSCIC), which shows that there were 174,872 written complaints about the NHS reported in 2013/14.
There’s no getting away from it: that’s a lot of complaints. But the NHS is a big service; it deals with over one million people every 36 hours. So how can we put this in perspective?
It turns out that the most commonly complained about service area is inpatient hospital treatment, with 34,422 reported complaints. It’s difficult, if not impossible, to determine an accurate rate of complaints – but if we assume that there were around 12 million completed inpatient episodes in 2013/14 then we can estimate that somewhere in the order of 0.3% of inpatient stays led to a complaint. Put differently: that’s three complaints for every 1,000 patients treated.
Perhaps this is good news: should you be admitted to hospital as an inpatient today, you only have a one in 350 chance that you – or someone else acting on your behalf – will make a written complaint about it in the future. But, of course, not making a complaint doesn’t necessarily equate to being content with your treatment. It’s quite possible that some people don’t know how to complain; don’t get around to it, or just don’t feel able to speak up. How do we account for these instances of poor care?
There are a couple of ways of approaching this. One way is to ask people how often they report poor care. This is exactly what Healthwatch England have done, via a YouGov survey of 1,676 adults across England. This found that that 30% of people had experienced or witnessed poor care within the last two years – but that 61% of this group did not make a complaint about it. They therefore estimate that “roughly 250,000 people every year are not only failing to get the standard of treatment and care they expect but feel unable to make their voices heard”1, .
This calculation is problematic, though, for a number of reasons. Firstly, by asking people about both experiencing orwitnessing poor care the survey creates double counting. For example, suppose my partner and I visited a relative in hospital and witnessed problems with their care. If all three of us were surveyed, we would each report having ‘experienced or witnessed’ poor care – but in practice it’s not realistic that we would make three separate complaints. Secondly, calculating in this way makes the assumption that everything a person might consider as “poor care” is also something that they would judge worthy of a complaint. This won’t always be the case: more than anything it will depend on how respondents interpret “poor care”, which could range from anything that could’ve been better right through to nothing less than incidents causing serious risk of harm. If we assume that at least some of the instances of “poor care” reported are ones that the respondents did not feel would merit a complaint, for whatever reason, then the estimate of the number of people “unable” to complain is again overstated.
What other methods could we use to investigate this issue? One would be to ask people who had recently used care services if they had ever wanted to make a complaint. We did this on behalf of the Healthcare Commission in the national NHS adult inpatient survey from 2007 to 2010 and found a steady 7-8% of recently discharged inpatients said that they did “want to complain about the care [they] received in hospital”. Given that there are 12 million completed inpatient episodes per year, this suggests somewhere in the region of 900,000 people every year wanted to complain about their hospital care: a much higher estimate than described above, and more than 26 times higher than the actual number of written complaints about inpatient care reported by HSCIC. But we can’t infer that only 1 in 26 people who want to make a complaint go on to do so, because HSCIC only report written complaints. We have now way at all of estimating how many verbal complaints are made – or, for that matter, resolved.
So: are the complaints reported by the HSCIC “just the tip of the iceberg”? In a way, yes; there are undoubtedly many more complaints made verbally and never recorded. But it’s difficult to make any meaningful estimate of how many people feel ‘unable’ to complain for whatever reason: there are too many confounding factors and the data we have is too limited. It’s clear that a very sizable number of people have poor experiences of healthcare in the NHS and that many people want to complain about this: but we don’t know enough about how ‘lower level’ complaints, particularly verbal complaints, are being handled and resolved on a day-to-day basis. Because of that, it’s clear that the recorded numbers of written complaints are a poor basis for estimating quality of care. The statistic I’d like to know is far more complicated, possibly prohibitively so, but would really tell us something new about standards of person-centred care: how many complaints are received and resolved ‘there-and-then’ at the front-line?
 Healthwatch England. (28th August 2014). NHS complaints data is just the ‘tip of the iceberg’. http://www.healthwatch.co.uk/news/nhs-complaints-data-just-tip-iceberg
 Telegraph, the. (29th August 2014). Complaints to NHS about staff attitude rise. http://www.telegraph.co.uk/health/healthnews/11061573/Complaints-to-NHS-about-staff-attitude-rise.html
 Times, the. (29th August 2014). Most people unwilling to report poor care in NHS. http://www.thetimes.co.uk/tto/health/news/article4190183.ece?CMP=OTH-gnws-standard-2014_08_28
 NHS Confederation. (27th August 2014). Key statistics on the NHS. http://www.nhsconfed.org/resources/key-statistics-on-the-nhs
 This appears to be calculated by estimating the total number of incidents of poor care as the number of complaints reported per annum (174,872) divided by the proportion of complaints reported (39%). This suggest approximately 448,000 incidents of poor care per annum, of which 273,000 (61%) are not reported as complaints.