Friday, April 23, 2004

ADHD: We don't know how to treat it

National Institute of Mental Health Multimodal Treatment Study of ADHD follow-up: 24-month outcomes of treatment strategies for attention-deficit/hyperactivity disorder.

I've not read the study, but I've read commentaries that in some ways may be more illuminating. The bottom line -- we don't know how to manage severe ADHD/ODD. It looked like stimulant meds were a cost-effective approach, but their efficacy seems to decline over time. The best results seemed to come from very intensive interventions, but there's no money for them and even their efficacy is not clear over extended periods.

The other result of the study is that growth limitation from stimulants is real -- 1/2" at least. Since many of these kids are very short to begin 1/2" to 1" is significant -- it's not the difference between 5'11" and 6'.

Overall, humility is indicated all around. Maybe everyone will stop beating up on the "bumbling" primary care docs who get stuck managing so much of this problem (not likely).

For physicians and patients I think the take home will be that the stimulants appear to work and they also have significant side-effects. Parents and teachers like behavioral interventions, even though they don't seem to change school outcomes. Stimulant efficacy decreases over time and doses rise.

I suspect we'll now see the "consensus" move towards more intermittent use of stimulants, with higher doses during critical times and lower or no doses on weekends, summers, etc. This is partly a return to old habits, but I think the prn dosing is something we need to learn to accept -- or at least study. We do need to train and incent psychologists to learn how to work with these kids and their parents and teachers. Maybe more school based psychologists? I have a hunch much of what we've learned with autism therapy can carry over to these kids.

And it's very clear we need more basic research and clinical research. I'm intrigued by recent studies of short-term and very-short-term working memory.

No comments: