After a long holiday it must surely be time to write here again. I had occasion to read Gloucestershire’s current needs assessment for the mental health and wellbeing of children and young people, although it’s nearly a year old. Here’s a brief review.
In fact my break from writing has not entirely been a holiday. It’s partly that I’ve been doing things that can’t easily be blogged about in the public domain. I hope to push some of those boundaries and get back to writing over the coming months.
The needs assessment arises out of the Government and NHS England’s Future in Mind report. In Gloucestershire this led to Gloucestershire’s Future in Mind, a transformation plan for 2015 – 2020. The needs assessment is Appendix 3 to that document, but it’s published as a separate document: Needs Assessment
The needs assessment was produced by Gloucestershire County Council, which has a responsibility for public health. This means it should be independent of healthcare commissioners and providers—theoretically.
I looked at the needs assessment in two ways. First I looked at the overall logic of the assessment, and then I looked at some specific examples of mental health issues to see how well the logic worked.
The model that forms the logical basis of the needs assessment is represented by a diagram like this (p. 8):
Clearly this has some peculiarities. Supply looks bigger than demand, and the areas overlap in a strange way. It’s not obvious that the diagram means very much, except that there is a thing we can call Need, another thing we can call Demand, and another we can call Supply. Also, it would be good if need, demand and supply can be made to align perfectly. I’ll call this the 3-way model.
A more informative way to draw the diagram, I think, would be to show the overlaps between need, demand and supply explicitly. The sizes of the three areas have no meaning, so I show them as equal circles. The result is seven areas, which I have invented my own labels for:
OK, the grey area in the middle, represents the perfect alignment of need, demand and supply. For example you feel ill, you go to the doctor, and the doctor gives you treatment that cures the illness. Note that the Coercion area, described below, is also well-aligned.
Hidden need, the blue area at the top, represents need that no one (or no one in the healthcare system) knows about. For example, you are a teenager who feels intensely distressed and you regularly cut yourself, but you keep the scars hidden and no one knows. There is no demand for treatment in the system and you don’t get treatment from the system.
Coercion, the purple area upper right, represents where the system meets a need without any demand. For example, a teenage boy is picked up by the police after being found wandering around town, apparently not knowing where he is. He tries to run away and doesn’t want help. They arrest him and take him to a place of safety for a mental health assessment. There was no demand, but the system coerced him and responded to the need anyway.
Unmet need, the blue-green area upper-left, represents the need that results in demand, but there’s no corresponding supply. For example you feel ill, you go to your doctor, and the doctor tells you there’s no treatment available for that illness in Gloucestershire, so your need can’t be met. (If you think that’s far-fetched, read on.)
Failure, the red area at the right, represents where the system supplies something that doesn’t meet either a need or a demand. For example, your teenage son has become extremely anxious about going to school, so you go to the doctor and she arranges a referral to a service that sends a nurse along once a week for a cup of tea and a chat with your son. He remains anxious about going to school. You keep mentioning this to the doctor. What the system supplied failed to meet either the need or the demand.
False demand, the green area bottom left, is demand that doesn’t reflect any need, so the system ignores it. For example, you take your teenage daughter to the doctor because you think she’s obese. The doctor measures and weighs her and says there’s nothing wrong. Don’t worry. Go away. The demand was false. (I know, I know, the doctor should wonder what the real reason for the visit is, but let’s keep it simple.)
Collusion, the brown area bottom centre, is where the there’s supply to meet false demand. For example, you’re feeling resentful and agitated after you weren’t picked for a team you felt you deserved to be in, so you go to the doctor. The doctor knows you aren’t ill and only need to give yourself time to adjust, but she colludes by giving you a prescription for fluoxetine anyway.
Even this 7-way model isn’t always easy to apply. For example, work on prevention is hard to classify.
The body of the assessment simply ignores the 3-way model presented in the introduction. Instead of need, demand and supply there are chapters on incidence, prevalence and risk. Then there’s a chapter on supply and a chapter on prevention.
In places it’s possible to determine need, demand and supply, but never on the same page because supply is in a chapter of its own. This makes the needs assessment almost impossible to make sense of.
Pressing on regardless, I looked at the assessment’s approach to half a dozen mental health issues. They just happen to be issues I’m interested in for various reasons. They are:
- Anorexia nervosa, an eating disorder
- Psychosis, for example schizophrenia
- Attention-Deficit/Hyperactivity Disorder
- Borderline personality disorder
- Self harm
- Post-traumatic stress disorder
Anorexia nervosa has its own section in the chapter on prevalence and incidence. We’re told (p. 16) that the incidence has been estimated at 8 new cases a year per 100,000 total population. Then we’re immediately told that this is inaccurate, because of something to do with population increase that I didn’t understand a word of. We’re not specifically told that we can expect very roughly 50 new cases a year in Gloucestershire, but that seems to be the implication.
Overall prevalence is not reported at all. Only the prevalence in two age bands is reported. The age bands don’t match the population age bands given previously (p. 12) so it’s impossible to do anything with the figures. The implication is that no one knows how many people in Gloucestershire have anorexia.
Prevalence of eating disorders in general is estimated as potentially 8,335, with the word “potentially” implying that it’s not really as many as that. But there’s no attempt to break this down into the separate disorders.
So that’s all we know about the need.
In the chapter on supply anorexia isn’t mentioned.
However, Gloucestershire does have an eating disorders service. It’s possible it uses a transdiagnostic approach that doesn’t distinguish between anorexia and other disorders, but we’re not told. A chart (p. 64) shows the service’s caseload doubling over two years, while the text below it tells us referrals have remained static at 134 a year.
The chart and the referrals numbers paint a picture of an ineffective service where people are being steadily referred but they just pile up in the service and don’t get better. This would place many of them in the Failure sector of my 7-way model.
The charity Beat says:
“Research has found that 20% of anorexia sufferers will die prematurely from their illness.”
However, in the needs assessment we don’t see any reference to deaths. From very roughly 50 new cases a year we would expect very roughly 10 deaths a year if these estimates are correct. Perhaps people who become very ill are discharged, or sent out of county to specialist inpatient units, or have their diagnosis changed so that their deaths are invisible.
Psychosis is omitted from the chapters on prevalence and incidence.
In the chapter on supply there’s a section on the early intervention in psychosis team (known locally as GRiP for obscure reasons). Like the eating disorders team, caseload is shown rising but not so dramatically. There’s too little information on referrals to make anything of it.
Taken at face value, the needs assessment seems to imply that the GRiP service is pointless, but I don’t believe it can be taken at face value.
Prevalence and incidence of ADHD is not reported separately from other similar disorders (hyperkinetic disorders). There’s an estimate of somewhat over 1,100 cases of these disorders as a whole, but the assessment criticises the estimate for being out of date.
Hyperkinetic disorders are not mentioned at all in the chapter on supply, but ADHD is mentioned in connection with parenting programmes “currently led by external facilitators” (whatever that means).
The caseload is something over 700 cases, but the assessment doesn’t compare this to the 1,100 mentioned previously. Around 350 assessments a year are carried out, suggesting that either most of the assessments are negative, or ADHD can be cured very quickly. As other sources make it clear that ADHD can’t be cured, the relationship between need, demand and supply for ADHD lacks clarity. No, lacks clarity is not putting it well, it’s downright murky.
Borderline personality disorder
Personality disorders are not covered by the needs assessment.
On the old definition, borderline personality disorder was specified as beginning by early adulthood, which meant it could often be seen developing in people under 18. On the new definition there’s no age specified, so the situation remains the same. If anything, borderline personality disorder can now be diagnosed in even younger teenagers. Its omission from the needs assessment is mysterious.
Well, maybe not so mysterious. In the 2009–14 NHS Gloucestershire strategic commissioning plan, Achieving Excellence, the commissioning of a personality disorder service was shelved. It remained shelved until March this year when some dust was blown off it at a workshop. Then it was shelved again. It’s perhaps unsurprising that personality disorders have been airbrushed out of the needs assessment.
On self harm the needs assessment has some information on incidence and prevalance, but stops short of actual numbers of people. One statistic is that around 16% of children and young people have self harmed (p.19). From the demographic data given earlier (p. 12), this means around 20,000 self-harming children and young people in Gloucestershire. So that’s the overall need. There’s no information about overall demand.
Unusually there is supply data in the same section, a chart showing cases seen at A&E and minor injuries units have risen to about 450 a year, just over 2% of the overall need. Of course, A&E and MIUs only cater for the specialized need of people who have harmed themselves seriously.
More supply is mentioned in the chapter on supply, where self harm is mentioned as one of the roles of the level 3.5 service. There, referrals have risen to around 300 a year, but not all of these are for self harm. Also, level 3.5 is another service for serious cases, so it’s likely that many of the people who attended A&E and MIUs were referred on to level 3.5. The assessment doesn’t tell us how to interpret the numbers.
A counselling service in schools is mentioned, and it seems likely that it does have some clients who self harm, but there are no numbers to support this conjecture.
Also mentioned in the chapter on supply, a self harm helpline receives about 150 contacts a year from children and young people (p. 73). We’re not told how many are the same person making contact again and again, and how many overlap with usage of other services, which may have signposted people to the helpline.
The overall impression is that the system’s response to self harm is incoherent, and that it chips away at the tip of an iceberg.
Post-traumatic stress disorder
PTSD is mentioned in the contexts of asylum seekers and of child sexual exploitation and abuse. No other potential causes are mentioned (such as bullying, road traffic accidents). There are no estimates of need or demand, and no evidence of supply.
It seems likely that some traumatized children and young people do receive treatment for anxiety and depression, but it’s anybody’s guess how many or how effective this is.
A penultimate chapter presents a collection of bright ideas that have been tried in Gloucestershire and elsewhere, without attempting to relate any of them to the specific needs of the population of Gloucestershire as a whole.
In spite of the chapter title, it’s not entirely clear that any of the ideas work. Trying to determine their effectiveness from the information provided in the assessment is next to impossible. The phrase “established evidence” is used in this chapter four times, not one of them associated with a direct link to any evidence, established or not. One of the other links is to a web page that explicitly says research is still in progress.
One of the bright ideas mentioned in many places throughout the report is the idea of emotional resilience. Faith in this bright idea, like the others, seems to be unsupported by specific evidence. It seems to be no more than a magic word, a superstition.
The only reference quoted in support of resilience is a report about health inequalities among children with learning disabilities. This cites another report about health inequalities, which describes resilience in purely circular terms—school pupils who have fewer problems will have more resilience; those who have more resilience will have fewer problems. This makes resilience a way to blame the victim when life is hard—it isn’t hard for you because that’s how life is, it’s hard for you because you didn’t have enough resilience.
As a magic dust that will make life better for the children it is sprinkled on, resilience is elusive. This is because it isn’t defined as a cause of good outcomes in life. It’s defined as an effect of good outcomes in life. There isn’t any way to manufacture the magic dust from scratch. Unfortunately it appears public money is being spent trying.
Beyond worrying about the money, however, another worry is that teaching children resilience creates in them an expectation that they will be resilient, that life will not be hard, that for life to be hard would be failure and shameful.
This kind of feeling pretty closely matches the feeling that children and young people have when they self-harm. Life is hard for them but they’ve been told to be resilient, so they don’t reveal to anyone how hard life is. Instead they develop their own secret coping mechanisms in isolation and pain.
The story of Margie is famous in a quite different context:
Margie was holding tightly to the string of her beautiful new balloon.
Suddenly, a gust of wind caught it.
The wind carried it into a tree.
The balloon hit a branch and burst.
Margie cried and cried.
Public health policy, I fear, may be rewriting the last line:
Margie was resilient and did not cry.
Afterwards, in secret, she cut her arms.
This extremely poor report is based on an inadequate model of need, demand and supply, and it doesn’t even adhere to its own model.
Its coverage of the mental health needs of children and young people in Gloucestershire is patchy at best.
In particular, for some conditions where there the supply of services is missing or suspect, the assessment gives little or no indication of the level of need. This gives the unfortunate impression that information has been airbrushed out to protect commissioners and service providers from criticism. Some of the wording also gives the impression that text was copied and pasted, in places, from submissions by commissioners or providers. All this suggests a regrettable lack of independence.
Superstitious belief in magic buzzwords sabotages the discussion of what works to improve mental health.
There’s a huge pile of conclusions and recommendations. They are mostly vague. Many things, it seems, should be carefully considered or reviewed. Further work should be undertaken and momentum should be maintained. It all reminds me of Sir Humphrey Appleby. The recommendations lack focus and appear designed to be immediately forgettable.
For anyone interested in improving the mental health of children and young people in Gloucestershire, this needs assessment is not a useful document.