The Forgotten Horrors of Tardive Dyskinesia
Medicine is about hard choices, and this adverse effect is one.
The Frontier Psychiatrists is a daily newsletter. It's about Health. One of the things that your Author, Owen Scott Muir, does, is work for a company called iRxReminder. We have a grant with the National Institute of Mental Health, where we're working, along with our colleagues at Videra Health, to build algorithms to identify tardive dyskinesia. This is part of a larger technological approach to medication adherence and adverse effect mitigation in schizophrenia. I know that sounds wonky—but here is a podcast to explain.
Today's article is about what I've seen in the course of that research. I didn't see it as clearly in the course of clinical practice, because we are bad at getting practice seeing rate and subtle things.1
Antipsychotic medications, other than clozapine, all have a risk of a late-developing movement disorder called tardive dyskinesia. This is amply highlighted if you go to the Scientology-sponsored—but theoretically, not Scientology at all, because it is just the Citizens’ Commission on Human Rights, an all-Scientologist advocacy group that is not an official part of the church!— museum on the horrors of the field of psychiatry, “Psychiatry, an Industry of Death.”
Rarely am I in the business of agreeing with Scientologists about their theological opposition to the medical discipline of psychiatry. Still, in this case, they highlight a real risk, albeit for self-serving reasons.
A movement disorder is what happens when your body moves not because you're telling it to, but because your brain is telling it to without any conscious input. There are complex structures in the brain that gate movement, which means that you don't move when you shouldn't move, and you do move when you should. Most of our movements are the result of intentional mental states that lead to behavior. Most behavior is movement.
So when you lift your arm, your arm moves up. When you stick your finger out, it extends. Some people have involuntary movements, movements that aren't because their brain is telling their finger or arm to move. There are of variety of movement disorders, and there's a whole medical specialty, housed in the discipline of neurology, that addresses movement disorders. Movement disorders can be very impairing.
One of the things that disorders highlight is that some problems are functional. It's not that there's a problem with your arm physically, you can't x-ray somebody's arm with a movement disorder to see what the problem is. It's a problem that you need a video camera to capture—or a trained eyeball. I would argue psychiatric disorders end up being movement disorders inside the brain, and that's what we're seeing with MRI, but this is an aside!
Here is what Tardive Dyskinesia (TD for short) looks like, along with disturbingly uptempo midi acoustic guitar and drums…
Thanks, healthgrades. You will notice blinking, grimacing, lip puckering, and other involuntary movements. Writhing of fingers or toes can also occur. The trunk can also be involved.
This is relatively common:
A 4-year prospective community-based study involving 352 psychiatric outpatients compared the incidence of TD with first-generation antipsychotics (FGAs) versus second-generation antipsychotics (SGAs). The authors found a greater risk of developing TD with FGAs, but the risk was still present with SGAs (rate-ratio = 0.68; 95% CI, 0.29-1.64), and the incidence of TD remained relatively unchanged since the 1980s.2
The prevalence vs incidence data is confusing at first:
The TD prevalence rate for SGAs was 13.1% compared with a prevalence rate of 32.4% for FGAs (P < .0001). The annualized incidence rate of TD across all studies was 3.9% for SGAs and 5.5% for FGAs. The annualized incidence rate of TD across the 6 direct comparison studies was 4.2% for SGAs and 5.5% for FGAs.
My first research mentor, Christoph Correll, M.D., points out the differences in the study designs leading to such different numbers:
methodological reasons may explain why the incidence rates appear drastically lower than the prevalence rates. These data were obtained from long-term studies whereas the prevalence data were obtained from cross-sectional studies, for which patients tend to have had longer medication exposure. Due to the lower requirements for inclusion in these studies, participants often are more severely ill, have more comorbidities, and have greater chronicity, duration, and treatment refractoriness.3
We see more patients who are more ill-treated and with these medicines for a longer time develop this adverse effect, which is potentially permanent and disabling.
Recall: prevalence is how many people in the population end up with the problem, and annualized incidence is how many new times the problem strikes in a given year.
It’s between 13-32% of patients treated with antipsychotics eventually. But for patients who don’t absolutely need to take these medicines, it’s a cruel bargain to have a lurking possible side effect, and you have a 4-5% chance of developing a permanent movement disorder every year.
This is still a low rate in any one person’s clinical practice, and if you start a patient on these medicines, you have to follow them over many years to see the adverse effects—which means new practitioners almost by definition can’t get a sense of the risk. Let’s say you are training as a nurse practitioner. You have about 800 hours of shadowing a doctor or other NP before you are in practice. This is less than a year, in total. Even psychiatrists, in training for 4-6 years of residency, only start a portion of their patients on these medicines, and most of them won’t develop the side effect right away. If you are lucky enough to work in a clinic that has long-term patients, you might see it—but they have been on old, bad meds for a long time! It feels rare.
It’s not that rare—as I wrote before. But it feels like it is when you start your career. So naturally, you format the heuristic that this problem doesn’t happen as much, and it’s not your fault. Except, as I review hundreds of videos to train our AI algorithms to identify the problems, it is common.
And disabling. It was easy to convince ourselves it was rare, and miscalculate the risk, and keep causing eventual harm.
Jain, R., & Correll, C. U. (2018). Tardive dyskinesia: recognition, patient assessment, and differential diagnosis. The Journal of Clinical Psychiatry, 79(2), 20745.
Woods SW, Morgenstern H, Saksa JR, et al. Incidence of tardive dyskinesia with atypical versus conventional antipsychotic medications: a prospective cohort study. J Clin Psychiatry. 2010;71(4):463-474.
Correll CU, Schenk EM. Tardive dyskinesia and new antipsychotics. Curr Opin Psychiatry. 2008;21(2):151-15