More red flags to watch for when prescribing controls, and how to manage them in practice.
CHRIS AIKEN: We continue with the second part of our 7th Psychopharm Commandment: Controlled Substances Shall Be Controlled.
Welcome to the Carlat Psychiatry Podcast, keeping psychiatry honest since 2003. I’m Chris Aiken, the editor in chief and the author of the Depression and Bipolar Workbook.
KELLIE NEWSOME: And I’m Kellie Newsome, a psychiatric NP and a dedicated reader of every issue.
Last week we entered the hidden vaults of the DEA website where they keep a list of red flags that pharmacists are supposed to resolve or report when filling controlled substances.
Let’s recap them, but first a preview of the CME quiz for this episode:
1. Which sleep medication is considered relatively safe in the elderly by the Beers criteria
A. Trazodone (Desyrel)
B. Lemborexant (Dayvigo)
C. Amitriptyline (Elavil)
D. Low-dose zolpidem (Ambien)
CHRIS AIKEN: Here’s a recap of last week’s red flags:
- Patients who avoid insurance when filling controls
- Patients on high doses or multiple controls that tend to be misused together, particularly the “Holy Trinity”: a combination of benzos, opioids, and a muscle relaxant like soma that produces rapid euphoria.
- Early refills
- Pharmacists also look for clinicians who prescribe high volumes of controls or prescribe a lot of controls to patients who fill them out of state
KELLIE NEWSOME: And here are some more on the DEA’s list.
Getting controls from multiple providers, or filling them at multiple pharmacies. Driving long distances to fill their prescriptions, and picking up controls filled in other patient’s names. Also patients who ask for specific brands, or refuse XR formulations – although with the exception of Vyvanse and the new tamper resistant methylphenidate Azstarys, most XRs can be easily converted into instant release by crushing them.
CHRIS AIKEN: Here’s another one I would take seriously: Patients who refuse to let you get their old records or get input from their relatives. You can’t be flying blind when prescribing controls, and if they do end up misusing their medication, it’s usually those relatives who are going to know first, and who are going to turn you in to the boards or the DEA.
When pharmacists see suspicious behavior they have to “resolve it,” which usually means contacting the prescriber to find out what is going on. Ultimately, controlled substances can only be filled if they satisfy this definition, supplied by the DEA. The medication “must be issued for a legitimate medical purpose by an individual practitioner acting in the usual course of his professional practice.”
That’s from the Code of Federal Regulation 1306 section 4, but what is a legitimate medical purpose? It means the drug is considered safe and effective for a specific condition. In theory, that should apply to all prescriptions, but in reality it does not. For example, once I inherited a patient with panic disorder who had responded very well to mirtazapine and olanzapine after trying lots of more reasonable options. Neither of those meds have good evidence in panic disorder, but I kept her on them empirically – based on her own response.
Suppose instead she presented on Dexedrine and Hydrocodone and told me those were the only medications that got rid of her panic. That would be more difficult to justify. The ideal justification is FDA approval, like clonazepam or alprazolam in panic disorder; lorazepam or temazepam for insomnia. Beyond that, practice guidelines or a positive controlled trial may suffice. Outside of that, it’s the wild west, and you best have some cover.
KELLIE NEWSOME: Back to the red flags. The DEA has advised pharmacists to look out for high doses of controlled meds, but what is a high dose? The FDA is not always clear on this either. Take Adderall. It is FDA approved up to 60 mg/day in narcolepsy, but the labeling is vague on whether the max dose is 40 or 60mg in ADHD. Here’s what it says: “Only in rare cases will it be necessary to exceed a total of 40 mg per day.” And while that’s not a hard and fast line, it’s actually consistent with the evidence.
The pivotal trial that got Adderall XR FDA approved in adult ADHD tested 3 doses: 20, 40, and 60mg/day. Surprisingly, all 3 brought about the same benefit. It was only on a secondary analysis that the 60mg worked better, and only in a subgroup with the most severe levels of ADHD.
For methylphenidates like Ritalin and Concerta the upper limits are more clear, but they vary by formulation, with higher doses allowed for those with longer durations. As a general rule, the max dose equates to 20 mg methylphenidate every 3-4 hours.
Upper limits for benzodiazepines are less clear. The FDA gives usual dose ranges, like
- 2-5 mg/day for alprazolam, Xanax
- 2-6 mg/day for lorazepam, Ativan, and
- 1-2 mg/day for clonazepam, Klonopin
But they then go on to say that higher doses may be needed in rare cases, up to 10 mg/day for lorazepam and alprazolam and 4 mg/day clonazepam.
CHRIS AIKEN: The guidance is rough, but the consequences are harsh. What is a clinician to do? Keep in mind that the FDA authorizes those high doses in rare cases, so as long as you’re not going that high for most patients you should be OK. If you do, document the rational in the chart and add a note in the prescription so the pharmacist is not alarmed.
One thing I’d avoid is justifying high doses by saying the patient must be a rapid metabolizer. Rapid metabolizer status is very rare – much more rare than poor metabolism – and with the stimulants we don’t have any evidence that it’s relevant. Stimulant metabolism is complex and poorly understood, the studies are few and it likely occurs through through multiple pathways. In one of the few studies out there, blocking the CYP 2D6 enzyme made no difference in methylphenidate levels. With the benzodiazepines, the CYP interactions are better understood, but if you expect that the patient’s benzo levels are going to be significantly altered by a drug interaction or a genetic difference in CYP enzymes, it’s better to switch them to a benzo that won’t be affected, as correcting the dose is too much of an inexact science. Lorazepam – Ativan – is often the safer bet in these cases.
KELLIE NEWSOME: Patients will often say they metabolize meds quickly, but we shouldn’t accept that as science. That’s what makes controlled substance prescriptions so difficult – we can’t just accept the patient’s report at face value like we do when dosing antidepressants. These drugs are inherently rewarding – for all of us – not just for people with substance use disorders. Most patients will say they feel better at a higher dose of Xanax or Adderall, but that doesn’t mean it’s treating their panic or ADHD any better.
CHRIS AIKEN: Here’s a tip. I often inherit patients who come in for their first visit on high doses of stimulants – Adderall 120 mg/day, that sort of thing. I’ll let them know that I can’t prescribe that much and why – besides the cardiac effects, high doses are also neurotoxic – and I’ll show them a picture of what that neurotoxicity looks like in the brain. We then lower it slowly – perhaps by 10 mg/month – toward the therapeutic range, and I’ll have them rate their symptoms on a self-rated ADHD scale like the ASRS every visit. Patients are afraid to lower the med – they tell me they can’t possibly manage their work without it – and indeed at every visit they say they are doing worse on the lower dose. But – and here’s the surprise – their self rated scale usually improves as the dose goes down. More is not better with stimulants. High doses make people more perseverative. They over think things, doubt their decisions, and have trouble shifting gears. They are less spontaneous. But high doses do make people more confident, even euphoric. As the dose is lowered, they don’t feel as good, but they function better.
KELLIE NEWSOME: Higher doses also carry greater risks with benzos. Tolerance and withdrawal are worse when patients take higher doses, as is the risk of traffic accidents, falls, and cognitive problems, and those risks are greater in the elderly.
CHRIS AIKEN: All this talk about controlling controlled substances is not just about catching people who misuse substances and protecting your license. These are risky drugs; that is why they are controlled, and the risks are greater for all controls in the elderly. Biologically, old age begins at age 50, but we'll use 55-65 as a cut off, with the lower number for patients in poor physical health. That’s when the risks go up, things like falls (from low blood pressure or imbalance), cognitive problems (including delirium), anticholinergic effects (from tricyclics or antipsychotics), and cardiac problems from stimulants or meds that prolong QTc.
The Beers criteria highlights meds that are problematic in the elderly, and they are pretty harsh on sleep meds. Most sleep meds are dinged on the Beers list – benzos and z-hypnotics like zolpidem/Ambien cause falls, memory problems, and traffic accidents. Even trazodone is not safe – in a cohort study of over 300,000 patients with insomnia from 2022, trazodone had a greater risk of falls then the benzos and the z-hypnotics.
Here are the sleep meds considered reasonably safe in the Beers criteria:
Sleep meds that are safer in the elderly:
• Ramelteon (Rozerem)
• The orexin antagonist (daridorexant, lemborexant, suvorexant)
• Melatonin, which is prescription in many countries but over the counter in the US, dosed 1-3 mg at night, and there is a study where melatonin worked better in the elderly when taken with magnesium 225mg and zinc 11.25mg – which enhance release of endogenous melatonin.
The Beers also give the green-light to two antidepressants often used for sleep
Doxepin (3-6 mg) and mirtazapine (7.5-15mg), although mirtazapine comes with a risk of hyponatremia and falls.
Note that none of the tricyclics are approved by the beers, and Doxepin is only approved in the very low doses used for sleep. If insurance doesn’t cover those low doses, you can get them low-cost in the liquid form of doxepin.
KELLIE NEWSOME: With stimulants we are really in unchartered territory with the elderly. None of the randomized controlled trials of stimulants in ADHD enrolled patients over age 50. The data we have is limited to a few retrospective chart reviews, and those generally used lower doses – about half of what we tend to see in adults. The elderly are more prone to cardiac side effects on stimulants, and they may get a therapeutic effect at lower doses. The blood brain barrier is more permeable in older age, allowing more stimulants to pass through to the CNS, at least that is what we see in primate studies. Human data is lacking.
ADHD can continue into old age, and just because those patients are retired doesn’t mean the symptoms aren’t affecting their personal life, not to mention their driving. But we want to make sure we’re not overmedicating them, or using stimulants to treat other problems like age related cognitive decline; they don’t work for that.
Dr. Aiken, we’ve talked about red flags but what do you do when you see them. Do you just stop the medication?
CHRIS AIKEN: First, talk to the patient. Don’t accuse them of misuse, but let them know that the facts you have before you make it hard to continue the medication. You may stop the medication, or you may do something else to address it, like a random urine drug screen, or bringing in family to get collateral information. Unless the patient already has red flags for misuse – like a history of substance use disorder or overdoses – I usually give one warning before stopping. And I don’t end care with the patient. I just tell them that I can’t continue to prescribe the controlled substance, but will continue to help them in other ways, including tapering off the drug if needed.
KELLIE NEWSOME: The bottom line. We’re entering a new era of oversight with controlled substances, and you need to let patients know about these guardrails up front. They might be used to a collaborative approach with antidepressants and antipsychotics, but they shouldn’t expect that same degree of collaboration with controls. The DEA has notified pharmacists of red flags to watch for – like the Holy Trinity combination of benzos, opioids, and muscle relaxants – and while we would have liked for the DEA to notify us directly, we’re glad that information is on their website, even if it’s buried behind a dozen links. We need to know what the DEA expects us to look out for when prescribing controls.
CHRIS AIKEN: And now for the study of the day.
A few months ago a group of psychiatrists at the Mount Sinai School of Medicine sent a 60 page report to the World Health Organization. Their goal was to get methylphenidate on the list of “essential” medications, a list that recognizes the necessity of Prozac, benzos, antipsychotics, but no treatments for ADHD. They failed. The WHO rejected the application on the grounds that the long-term risks and benefits of methylphenidate are not clear. The studies are biased by industry sponsorship and a blind that is easy to break. And nearly all of those studies lasted 3 months, while stimulants are increasingly taken for a lifetime.
I sympathize with the WHO – and you’ll notice we often report on the expansion of ADHD diagnoses and overuse of stimulants, but today we have something different to report. It’s not a human study, but it’s close. It’s a landmark long-term study of methylphenidate on the primate brain. Researchers at the FDA’s toxicology center gave high and low doses of methylphenidate to primates for 12 years starting in adolescence. The equivalent of a lifetime. At the end, they abruptly stopped the medication and looked for every kind of change in the primate’s brain – blood flow, regional metabolism, dopamine transporter, etc. Reassuringly, they found no evidence of toxicity even at 5-times the clinical dose. In fact, there were no differences at all between the placebo and they methylphenidate groups.
That’s reassuring for people who take long-term methylphenidate for ADHD, but I’d be careful about translating those findings to the amphetamines like Adderall and Vyvanse. Most studies of the amphetamines find evidence of neurotoxicity – usually in high doses but a few found that in normal doses. To learn more, check out our Carlat webinar “Amphetamine vs Methylphenidate. It’s free at the Carlat Report.com.
KELLIE NEWSOME: Get daily research updates like this through Dr. Aiken’s linkedin or twitter feed - @ChrisAikenMD. Start collecting CME credits for this episode through the link in the show notes, or subscribe to the journal online. The Carlat Report is one of the few CME publications that depends entirely on subscribers. Thank you for helping us stay free of commercial support.