Intended for healthcare professionals

Rapid response to:

Analysis Too Much Medicine

Attention-deficit/hyperactivity disorder: are we helping or harming?

BMJ 2013; 347 doi: https://doi.org/10.1136/bmj.f6172 (Published 05 November 2013) Cite this as: BMJ 2013;347:f6172

Rapid Response:

Re: Attention-deficit/hyperactivity disorder: are we helping or harming?

The article is generally welcome as reinforcement of NICE guidelines on ADHD.

Some comments about its more specific claims are made:

1. Overdiagnosis is a generalised issue, even in the self-assured proponents of 'physical' medicine. Pathologists report overdiagnosis in general hospital medicine in 25.4% of cases, as high as 57.7% in MI, and 61.8% in pneumonia [1].

2. Clinical subjectivity of impairment. This is true of most medical conditions than we allow ourselves to think. How many times have we seen that severely pathological slides of plaques and tangles are often not associated with much clinical impairment in dementia, and vice versa? Ditto endometriosis. Pain, the basic beginning of much medicine, is also a psychological matter, but we readily prescribe medication for it, including very addictive opiates, when pathology could never predict its severity.

To pretend that practising medicine is some one-dimensional objectively arbitrated science, is to deny its basic humanity.

3. Increases in prescription rates is a misleading marker of overdiagnosis, and the authors must know this. Most adult ADHD clinics in the UK have only been set up over the last 5 years. In our clinic's catchment of 1 million, a conservative estimate of 2% prevalence, estimates 20,000 adults. We have seen less than one tenth of that, since we started. Would such a situation be tolerable if we were discussing a treatable cancer? Rises in prescription rates are related to rises in recognition in the first instance.

3. Effectiveness of treatment. Let's compare some data on general medicines, using the NNT as an arbiter. Amoxicillin has an NNT of 30, for treating LRTIs in primary care. Ramipril has an NNT of 67, for preventing stroke , over four and a half years of use [2] . ADHD medications have an NNT reported between 1.1, to 2.5 [3]. Yes, that's not a misprint. The treatment definitely works, in a manner that consistently surprises both clinical staff and patients. In my experience, so many adults benefit to such a night-and-day extent that their lives are transformed. Subjective? Maybe. Effective? Undoubtedly. And it isn't because the ADHD drugs are euphoriants or addictive- this is simply untrue for the dosages that we prescribe. [4]

4. More 'open-ended' definitions as a cause of oversiagnosis. DSM5 uses 'interfere with or reduce the quality of social, academic, or occupational functioning', instead of 'clinically significant' in DSM4. If anything, this is sympathetic to any disability. Consider dementia. I see nothing valid about diagnosis that is rated entirely on what we as clinicians call pragmatically defined impairment: the patients themselves experience it, and they are rightly the reference points of impairment, even given the (forced and false) dichotomy of choosing who is more right between us and them. In practice, both are involved. The DSM widens and grows, yes. But it is not a pathology instructional textbook so much as it is a glossary of definitions. Do people worry about the Oxford English dictionary getting bigger?

5. Commercial Influence. This is not unique to ADHD or even psychiatry. It is prevalent in all of medicine. Patient websites promoting 'causes' from erectile dysfunction to cancer care. Why pick on ADHD without a balanced reference to everything else? We are in the business of medicine whether we have commercial interests or not. If we were not, we would not be paid a salary or make a living based on what we do in healthcare. All medicine fields court the hand of Pharma in a perilous ballet. Picking on psychiatry is easy because it is invisible, but this does nothing to improve the stigma.

6. Adverse events of ADHD medications. Sneakily argued. No comparators are offered. Stimulants do not cause dependence, anaphylaxis, discontinuation syndromes, metabolic disorders, night-time coughs, or permanent extrapyramidal side effects. So, not like benzos, antibiotics, antidepressants, cancer drugs, diuretics, or antipsychotics. In the best study of long term stimulant use in children so far, the risk of sudden death was unquantifiable as it was not possible to separate it from other risky behaviours as a cause. It was, overall, very rare. [5]

7. Psychological harms. Again, an interesting and somewhat inverted argument is presented by the authors. Clearly, labels can be troublesome if they are taken to be used as excuses for bad behaviour. However, where a genuine impairment is concerned, the label is a massive relief. Many times in my clinic, young adults who never got the diagnosis as children describe the enormous relief they feel when they are given the diagnosis as adults, which is further vindicated when they report such massive, sustained gains from treatment.

Let’s use a visual analogy: If a child had short sightedness, but this was never diagnosed. He couldn’t do schoolwork, was clumsy, and kept losing things for reasons that were not obvious to him or others, and people didn't believe him, or worse still, believed that short sightedness was a made-up disorder, how good would it feel for him to finally have his eyes tested aged 22? The label of short-sightedness is welcome, relieving, and treatable, provided the right diagnosis is made, and the risks of overdiagnosing are not as perilous as those of underdiagnosing. Perhaps that’s why we overdiagnose pneumonia too.

We should realise we are in the 21st century, and disbelief in anything but the purely physical and easily observed, is akin to being still amazed that planes can fly without strings.

If people think that ADHD and its treatment is of a fashion, societal value-based, and unlikely to be sustained, let them assert this with the many ex-prisoners we see in clinic who describe being completely free of their impulses to be violent, to commit repeated crimes a result of simple treatment. Moreover, they report being dramatically transformed by treatment, going back to education, and finding work. In short, they are given their 'glasses'. Where an intervention is potentially effective as being able to reduce recidivism in prisoners by up to 42% in a study reviewing 25,000 cases [6], let's not have too much cartoonish villainy about the disorder or its treatment.

The headlines and over-argued cases set out by this type of article doesn’t help move things to a more rational debate, especially where even medical professionals are prejudiced against the matter.

[1] Gibson et al. Discrepancies between Clinical and Post Mortem Diagnoses in Jamaica. International Journal of Pathology. 2006, Volume 6, Number 2.
[2] Badrinath P. Preventing stroke with Ramipril. BMJ. 2002 August 24; 325(7361): 439
[3] Wigal S. Efficacy and Safety Limitations of ADHD pharmacotherapy in Children and Adults. CNS Drugs, 2009; 23 Suppl 1:21-31
[4] Kollins S. Comparing the abuse potential of methylphenidate versus other stimulants: A review of the available evidence and relevance to the ADHD patient. J Clin Psychiatry 2003: 64(supp-14-18)
[5] Gould M et al. Sudden Deatn and use of medication in youth., Am J Psych, 2009. A1A. (1-10)
[6] Larsson et al. N Medication for ADHD and Criminality. N Engl J Med 2012; 367:2006-2014November 22, 2012DOI: 10.1056/NEJMoa1203241

Competing interests: No competing interests

20 December 2013
Ovais F Badat
Consultant Psychiatrist
AWP NHS Mental health Trust
ADHD Clinic, BS14 9BP