Stigma and statistics: why we need to look beyond ‘one in four’
By Professor Anthony S David, Director, UCL Institute of Mental Health
After years of silence, it seems that suddenly everyone is talking about mental health. With the COVID-19 outbreak causing increasing numbers of people to stay at home, it appears to be a pressing concern for all. Recently, we’ve seen public figures open up about their mental health, including sports men and women, artists, writers, and even members of the royal family. Some politicians, too, have been heard talking about their mental health issues.
Or have they? The phrase ‘mental health issues’ is a case in point; why do we not use the term mental illness? Or even psychiatric illness or disorder? Somehow such terms sound a bit too — serious? stigmatising? And what issues or illnesses are we most likely to hear described? Anxiety and depression must surely top the list. This is only reasonable, as they are what even the psychiatry profession will tend to call ‘common mental disorders’.
Yet very few people talk openly about living with conditions like psychosis, let alone schizophrenia, the most severe disorder in the psychosis spectrum. When was the last time you saw someone rattling a tin outside the supermarket collecting for ‘schizophrenia research’ or heard of a philanthropist donating several millions to build a new wing of a hospital for psychiatric treatments?
Most would argue that stigma against mental illness is at the heart of this. This welcome new but qualified openness reveals how deep this stigma runs and suggests that counteracting it is not going to be as easy.
For those in the field of mental health, stigma usually narrows down to prejudice and discrimination against people with mental illness. Combatting stigma has been the subject of many concerted campaigns around the world; providing education and information is at the heart of these. This entails both the sharing of “facts” as well as a certain amount of rhetoric, with the aim of challenging stereotypes and myths around mental illness, particularly that people with such problems aren’t necessarily violent, unpredictable, incurable or immoral.
Part of the rhetoric is to emphasise the ubiquity of mental health problems in the population, closely allied to the idea that mental illness lies on a continuum with normality, the implication being that “we’re all a bit mad”. An example of this is the UK’s ‘one-in-four campaign’. The ‘one in four’ statistic greets you from the NHS England official webpage and has been taken up as a useful slogan by many mental health charities, including the respected and redoubtable Rethink — but it has not been without controversy.
While this estimate is said to have been around since the late 1980s, a commonly cited source of the statistic is the 2007 British Adult Psychiatric Morbidity Survey (APMS), an NHS-funded household survey that interviews 7,500 people roughly every seven years, thought to be the best source of reliable data on the prevalence of mental disorder in the UK.
However, the January 2020 edition of the British Journal of Psychiatry carried an article trying to dissect the ‘one in four’ figure, written by Paul Bebbington from UCL and Sally McManus from NatCen, who are respected psychiatric epidemiologists and part of the APMS. The authors reasoned that loose talk around statistics based on the survey could damage its credibility and, given the level of interest and impact such discourse has, it is important to get the facts straight.
The problems with this survey begin with the question of which conditions it includes. As noted earlier, depression and anxiety are obvious, as are bipolar disorder and psychosis. But the authors wonder, what about others like attention deficit disorder, common in children but not adults — or PTSD, which is often self-ascribed? Then there’s addictions — but not everyone would think of alcohol and drug dependence as mental disorders. There’s also personality disorders — controversial even among mainstream psychiatrists in terms of whether they should be seen as ‘things that you get’ or ‘things that you are’. And what do you do about people who seem to have more than one condition? Do you just choose the most serious one?
Given the number and type of conditions included, this also posed methodical issues in terms of timescale. In a census, information about what’s happening on a particular day is collected. With mental health, however, the convention is to set a slightly longer timeframe — but surveys differ significantly; in the 2007 APMS survey, anxiety and depression were measured by identifying symptoms demonstrated over the past week, while alcohol dependence referred to the past six months, and drug dependence measured based on the past year. The longer the timeframe, the greater the number of people likely to be counted. Hence unless the timeframe is specified and clear, we do not really know what a bald prevalence figure means.
Why does this matter?
The purpose of using statistics in anti-stigma campaigns is seemingly polemical. The aim appears to be to impress on us that mental illness is not a rare and peculiar thing — even if, until recently, it was something hidden and rarely discussed. But grouping all these conditions together and using statistics like ‘one in four’ could have the opposite intended effect; prejudice may lead people to assume that mental problems are ‘just an excuse’ for laziness, failure or poverty, for example. Some may look at their own lives and question the validity of the statement — that one in four people just sounds like too many, and that some claiming to be mentally ill, must in fact, not be — or the threshold must be too low. This could then lead to mental illness not being taken seriously enough.
Even rigorous surveys like the APMS have to choose a cut-off for their measures and, even when calibrated against a notion of a level of symptoms that requires treatment, there is a certain element of arbitrariness about it. This can lead people to question the accuracy or rather the validity of the estimates.
Implications for policy
Although campaigns and statistics such as one in four aim to end the stigma around mental health, the crucial challenge to be addressed is that of funding and resources for those who suffer from mental illness; no one in the mental health field will deny the chronic underfunding of services, a result no doubt of the stigma we are all trying to counter.
The NHS Long Term Plan, published in January 2019, states that “NHS England has made a renewed commitment that expenditure on mental health services will grow faster than the overall NHS budget, creating a new ring-fenced local investment fund worth at least £2.3 billion a year by 2023/24.”
One commitment is to enable an additional 380,000 adults to access Increasing Access to Psychological Therapies (IAPT) services by 2023/24 — which are designed for those with less severe, common mental disorders. This will require increased recruitment and training of practitioners, mostly psychology graduates; however, it will take almost five years for them to join the NHS frontline, and does not address the current widespread reluctance of medical graduates to go into psychiatry, where at least a further six years of training is required.
The promise of ring-fenced investment is to be applauded — and may even be a positive consequence of ‘one in four’ — but such funding is required across the full spectrum of disorders, including those most severely affected. It is my concern that while emphasising the ubiquity of mental disorders, at the same time the one in four statistic risks trivialising them.
In the meantime, let’s continue to campaign for better facilities, greater investment, more training, and further resources being put into the treatment of people with mental illness in our hospitals and clinics — goodness knows there are plenty of them.
________________________________________________________________
Reference:
Bebbington, P., & McManus, S. (2020). Revisiting the one in four: The prevalence of psychiatric disorder in the population of England 2000–2014. British Journal of Psychiatry, 216(1), 55–57. doi:10.1192/bjp.2019.196
