Most drug interactions are not catastrophic. They are quieter: a medication that suddenly does not work, or works too well, or works in a way it was not supposed to. Sometimes the change is obvious — sedation, nausea, a tremor that wasn’t there last week. Sometimes it is invisible, showing up only on a lab result months later. The interactions worth knowing are the ones that could meaningfully change how a psychiatric medication treats you, and most of them are preventable with one habit: telling every prescriber, dentist, urgent care, and pharmacist every prescription, over-the-counter medication, supplement, and recreational substance you use. Not most of them. All of them.
This essay walks through the categories of interactions that come up most often in psychiatric care — what they are, why they happen, and what to do about them. It is not a substitute for a medication review with your prescriber and your pharmacist. Use it as a map.
How drug interactions actually work
Most interactions in psychiatry fall into one of three buckets.
Pharmacokinetic interactions change how much of a medication is in your bloodstream — by speeding up or slowing down how the liver breaks the drug down, by changing how the kidneys clear it, or by changing how the gut absorbs it. The liver does most of this work through a family of enzymes called the cytochrome P450 system. Names like CYP2D6, CYP2C19, and CYP3A4 are the specific enzymes most relevant to psychiatry. A medication or substance that inhibits one of these enzymes will raise the blood level of any other medication that depends on that enzyme to be cleared. A medication or substance that induces the enzyme will lower the blood level of medications that depend on it.
Pharmacodynamic interactions happen when two medications act on the same brain or body system. Two serotonin-active medications taken together can drive serotonin signaling too high. Two sedating medications taken together cause more sedation than either alone. Two QT-prolonging medications taken together raise the risk of a dangerous heart rhythm.
Physical/absorption interactions happen in the gut. Taking certain medications close in time to antacids, calcium, iron, or large amounts of food can change how much of the medication actually gets into your body. The lurasidone-and-meal rule is one example: without enough calories in the stomach, lurasidone is not absorbed well enough to do its job.
You do not have to remember which enzyme does what. You do have to know that interactions exist, that they can be silent, and that they are why your prescriber asks the same questions about supplements and substances at every visit.
Antidepressants and serotonin syndrome
Serotonin syndrome is the interaction most people have heard of and most people misunderstand. It is not a common side effect of starting an SSRI. It is what happens when two or more medications drive serotonin activity past a tolerable threshold at the same time. The presentation ranges from mild — agitation, sweating, tremor, diarrhea, dilated pupils, fast heart rate — to severe, with high fever, muscle rigidity, confusion, and in rare cases, death.
The combinations that most often produce serotonin syndrome:
SSRI or SNRI plus an MAOI — the classic, dangerous combination. The two classes are never used together. Your prescriber will require a washout period of weeks between stopping one and starting the other, and the washout for fluoxetine is especially long because the drug stays in the system for weeks after the last dose.
SSRI or SNRI plus tramadol — tramadol is a pain medication that also acts on serotonin. The combination is one of the most common real-world triggers for serotonin syndrome, partly because people do not think of tramadol as a serotonergic drug. If you are on an SSRI or SNRI and given tramadol for a back injury, dental pain, or after surgery, tell the prescriber. Many will choose a different pain medication.
SSRI or SNRI plus linezolid — linezolid is an antibiotic with MAOI-like activity. It is given for certain serious infections, often in hospital. The interaction is well known to hospitalists but can be missed in outpatient transitions.
SSRI or SNRI plus triptans (sumatriptan, rizatriptan, others) — the official labeling warns about this combination for migraine treatment. In practice, the risk at typical doses appears smaller than the warning implies, but it is not zero. Your prescriber should know.
SSRI or SNRI plus St. John’s wort — St. John’s wort is an over-the-counter herbal supplement marketed for depression. It increases serotonin activity. Combined with a prescription antidepressant, it can push serotonin past safe levels. Many people do not think of an herbal supplement as a medication and do not mention it. St. John’s wort is also a potent CYP3A4 inducer (see below), which makes it one of the most interaction-heavy supplements on the shelf.
SSRI or SNRI plus dextromethorphan (DXM, the cough suppressant in many cold remedies) — usually fine in normal cold-remedy doses, dangerous at the high doses some people use recreationally.
SSRI or SNRI plus MDMA, cocaine, or methamphetamine — recreational stimulants that act on serotonin can produce serotonin syndrome, sometimes severe. The risk is highest in the days after starting an antidepressant.
If you develop unexplained agitation, sweating, tremor, muscle twitching, fever, or diarrhea after starting or adjusting any of these combinations, treat it as a medical concern, not a side effect to ride out.
The MAOI tyramine story
Monoamine oxidase inhibitors (MAOIs) — phenelzine, tranylcypromine, isocarboxazid, and selegiline — are older antidepressants that are highly effective for certain treatment-resistant depressions but rarely used as first-line treatment today, in part because of their dietary and drug-interaction profile.
The famous interaction is with tyramine, an amino acid present in certain aged, fermented, cured, and air-dried foods. Aged cheeses, cured meats, draft beer, fava beans, and several fermented soy products carry meaningful tyramine. When tyramine is absorbed normally, monoamine oxidase in the gut breaks it down before it can raise blood pressure. When MAO is blocked by an MAOI, tyramine reaches the bloodstream intact, releases norepinephrine, and can drive blood pressure to dangerous levels — sometimes high enough to cause stroke or heart attack. This is the “cheese reaction.”
If you are on an oral MAOI, your prescriber will give you a specific dietary list. The selegiline patch (Emsam) at the lowest dose does not require dietary restriction, but higher patch doses do. The interaction is not theoretical and is not a relic. People still have hypertensive crises from MAOI-tyramine reactions every year.
MAOIs also interact dangerously with many cold remedies (pseudoephedrine, phenylephrine), with stimulants, and with most other antidepressants. If you are on an MAOI, do not start a new medication of any kind — over-the-counter included — without checking with your prescribing clinician or a pharmacist who knows you are on one.
Alcohol
Alcohol interacts with almost everything in psychiatry, but the interactions vary in shape.
With SSRIs and SNRIs: moderate alcohol use is generally compatible with these medications. It is not what most patients expect to hear, and it is not a license to drink heavily, but the older blanket warning that any alcohol with any SSRI is dangerous has not held up. The bigger concerns are that alcohol is itself a depressant — meaning heavy use undercuts the treatment of depression and anxiety it is supposed to be supporting — and that some people experience worse sleep and worse mood the day after even a moderate drink. People in active recovery from alcohol use disorder are a separate case; for them, the safer rule is zero.
With benzodiazepines, sleep medications (the “Z-drugs”: zolpidem, eszopiclone, zaleplon), and opioids: combinations to avoid. All of these medications depress breathing, and alcohol compounds that effect. Deaths from accidental overdose are most often combinations, not single substances. If you take any of these medications, alcohol is not a casual decision.
With mood stabilizers: alcohol can both worsen mood instability and complicate lithium monitoring (dehydration). With valproate, the combination raises the risk of liver injury and pancreatitis. With carbamazepine, alcohol increases sedation. None of these are absolute prohibitions, but all are reasons to be honest with your prescriber about your drinking.
With antipsychotics: alcohol adds to sedation and to the risk of low blood pressure on standing, which can cause falls — especially in older adults.
With disulfiram (Antabuse) or naltrexone or acamprosate: disulfiram is designed to make alcohol unpleasant; even small amounts can produce severe reactions. Naltrexone and acamprosate do not cause that reaction but are typically prescribed to people abstaining from alcohol.
Cannabis and CBD
The cannabis section has changed more in the last two years than almost any other interaction topic in psychiatry. The relevant pharmacology splits cleanly into two pieces: tetrahydrocannabinol (THC), which is psychoactive and produces the impairment most people associate with cannabis, and cannabidiol (CBD), which is not directly psychoactive but has a wide and clinically meaningful set of liver-enzyme effects.
What CBD actually does to your medications
CBD inhibits two of the major liver enzymes that break down psychiatric medications: CYP2C19 and CYP3A4. The clinical consequence is that CBD can raise the blood levels of medications that depend on those enzymes to be cleared.
A clinical pharmacokinetic study published in 2026 confirmed what earlier in vitro studies suggested: CBD at the doses found in typical consumer products raised citalopram exposure by approximately 43 percent in healthy volunteers.1 That is a clinically meaningful increase. It means that a patient stable on a particular dose of citalopram, who starts a CBD product for anxiety or sleep, may experience higher serotonin-related side effects — sometimes including dose-dependent increases in QT interval — without changing their prescription dose at all. The same mechanism applies, with varying strength, to escitalopram, sertraline (less so), and many other psychotropic medications metabolized through CYP2C19 or CYP3A4.
This is the opposite of what many patients have been told. The older idea — that cannabis or CBD makes antidepressants less effective — does not reflect the clinical pharmacology. The well-documented risk is the other direction: CBD raises levels.
If you are taking any psychiatric medication and using a CBD product — for sleep, anxiety, pain, or any other reason — tell your prescriber. The interaction does not mean you have to stop the CBD, but the medication may need adjustment, monitoring may need to be more frequent, and the combination should be a known thing rather than a hidden one.
A note on what CBD is and is not: the only FDA-approved cannabidiol product is Epidiolex, approved for certain rare seizure conditions. Over-the-counter CBD products are not FDA-approved for any psychiatric indication, and independent product testing has repeatedly shown that the actual CBD content of many consumer products differs from what is on the label — sometimes substantially. The interactions described here are with the active compound, regardless of the label.
THC and cannabis itself
THC’s interactions with psychiatric medications are more about pharmacodynamics than enzymes. It adds to the sedation of benzodiazepines, sleep medications, and antipsychotics. It can worsen the cognitive flatness some patients describe on certain antidepressants. It can destabilize bipolar disorder — particularly in younger patients — and can trigger or worsen psychotic symptoms, especially in people with a personal or family history of psychosis. For patients with bipolar disorder, schizophrenia, schizoaffective disorder, or a history of cannabis-induced psychotic episodes, the clinical recommendation in current guidelines is to avoid cannabis. This is not a moral judgment; it is a risk calculation based on what the medication is trying to do and what cannabis can undo.
Cannabis and PTSD deserves a specific note because the question comes up so often. Many patients with PTSD use cannabis for nightmares, sleep disturbance, and hyperarousal. The 2023 VA/DoD Clinical Practice Guideline for PTSD strongly recommends against cannabis use for PTSD treatment, based on a systematic review finding insufficient evidence of benefit and emerging signals of worsened avoidance, emotional numbing, and cannabis use disorder.2 The only completed randomized trial in this population (Bonn-Miller and colleagues, 2021) did not find a statistically significant advantage over placebo on the primary symptom outcome. A second trial is ongoing. The clinical bottom line is that, for now, cannabis is not part of evidence-based PTSD treatment, even when patient experience says otherwise.
Lithium — the interactions that matter most
Lithium is filtered out of the body by the kidneys, and anything that changes how the kidneys handle salt and water can change how much lithium is in your blood. The interactions worth knowing are not exotic. Most of them are common medications.
NSAIDs (ibuprofen, naproxen, indomethacin, others) — taken regularly, these raise lithium levels significantly. Acetaminophen (Tylenol) does not have this effect and is the preferred over-the-counter pain reliever for people on lithium. If you need an NSAID for a few days, tell your prescriber.
ACE inhibitors and angiotensin receptor blockers (lisinopril, losartan, valsartan, others) — common blood pressure medications that can raise lithium levels.
Thiazide diuretics (hydrochlorothiazide and combinations) — raise lithium levels by reducing sodium clearance. Loop diuretics (furosemide) have a less predictable effect.
Salt restriction — sodium and lithium share renal transporters. A sudden change in dietary salt can shift lithium levels.
GLP-1 medications (semaglutide, tirzepatide, liraglutide and others) — a newer story. Several case series since 2025 have documented lithium toxicity in patients who started or escalated a GLP-1 medication for diabetes or weight management.3 The mechanism is the GI side effects — nausea, vomiting, reduced fluid intake — pulling lithium levels up the way any other dehydration event would. If you are on lithium and starting a GLP-1, your prescriber should plan more frequent lithium monitoring during the titration weeks. This interaction is also relevant in the antipsychotic space: one published case report documented supratherapeutic ziprasidone levels in a patient who started semaglutide.
Heat, illness, dehydration, hard exercise — all of these are lithium-toxicity risk factors. If you live in the Inland Empire, summer is a season you plan around if you are on lithium.
Lamotrigine and hormonal contraception
Estrogen-containing oral contraceptives — the combined birth control pill, the patch, and the vaginal ring — lower lamotrigine blood levels by approximately half through induction of the liver enzymes that clear it. The practical consequence is that a woman stabilized on lamotrigine for bipolar disorder or epilepsy who starts a combined contraceptive may experience return of symptoms within weeks. Conversely, stopping the contraceptive without lowering the lamotrigine dose can lead to dose-related side effects (dizziness, unsteadiness, double vision). Progesterone-only methods — the hormonal IUD, the implant, the progesterone-only pill, and the contraceptive injection — do not have this interaction and are generally the simpler choice for people on lamotrigine.4 This interaction is covered in more detail in the essay on mood stabilizers.
Grapefruit
Grapefruit juice and whole grapefruit are potent inhibitors of CYP3A4 in the gut wall. The effect lasts roughly twenty-four hours after a single serving, longer with repeated use. The clinically relevant interactions in psychiatry include certain anti-anxiety medications (most notably buspirone, where grapefruit can raise blood levels several-fold), several benzodiazepines (especially midazolam and triazolam — less so the more common ones like lorazepam), and a few of the second-generation antipsychotics. Outside psychiatry, the interaction with several statins, calcium channel blockers, and immunosuppressants is more clinically significant. The simplest rule, if you take any of these medications, is to avoid grapefruit and grapefruit juice routinely.
Other citrus — orange, lemon, lime, tangerine — does not produce the grapefruit effect. Pomelo and Seville orange do.
St. John’s wort — the supplement that acts like a medication
St. John’s wort (Hypericum perforatum) is the most interaction-heavy over-the-counter supplement in psychiatry, and the one most often missed in medication reconciliation because patients categorize it as “herbal,” not as a drug. It does two things that matter.
First, it is mildly to moderately serotonergic. Combined with an SSRI, SNRI, MAOI, or several other serotonergic medications, it can contribute to serotonin syndrome.
Second, it is a potent inducer of CYP3A4 — meaning it speeds up the clearance of medications that depend on CYP3A4. The list of medications affected is long and clinically important: oral contraceptives (failure of contraception has been documented), several antidepressants, certain anticonvulsants, warfarin, many antivirals (including HIV medications), some chemotherapy agents, and others. The interaction direction is opposite to CBD: where CBD raises levels by inhibiting CYP3A4, St. John’s wort lowers them by inducing it.
If you are taking any psychiatric medication, do not start St. John’s wort without checking. If you are already taking St. John’s wort, tell your prescriber.
Pharmacogenomic testing — what it can and cannot tell you
Some people break down certain medications faster or slower than average because of genetic variation in the liver enzymes that do the work. This is most relevant for CYP2D6 (which metabolizes many antidepressants, antipsychotics, and pain medications) and CYP2C19 (which metabolizes citalopram, escitalopram, sertraline, and several tricyclic antidepressants).
The Clinical Pharmacogenetics Implementation Consortium (CPIC) — the international body that develops gene-based prescribing guidance — provides specific dose-adjustment recommendations for certain medication-genotype combinations when results are available. CPIC explicitly does not recommend routine preemptive genotyping for everyone before prescribing antidepressants; the guidance applies if results are in hand, not as a justification for testing.
The largest clinical trial of pharmacogenomic-guided antidepressant prescribing (the GUIDED trial, published in 2019) did not meet its primary endpoint of significantly improved symptom reduction at eight weeks, though it did show modestly better response and remission rates as secondary outcomes. The picture is more nuanced than the marketing of consumer-facing pharmacogenomic tests sometimes implies.
If you have already had pharmacogenomic testing, the results are worth sharing with your prescriber. If you are wondering whether to pursue it, the conversation worth having is about what the result might actually change in your treatment — not about treating the test as a diagnostic shortcut.
The interactions to flag immediately
Some interactions deserve a phone call, not a wait-and-see.
Any combination of two serotonergic medications you did not know was a combination. Especially SSRI plus tramadol, SSRI plus linezolid, SSRI plus St. John’s wort, SSRI plus MDMA or cocaine, or SSRI plus a new antidepressant added without a washout.
Lithium plus a new NSAID, ACE inhibitor, diuretic, or GLP-1 medication — especially if you are also dehydrated, ill, or in the middle of a heatwave.
Lamotrigine plus a new combined hormonal contraceptive (or stopping one).
Any antidepressant plus St. John’s wort.
Any psychiatric medication plus a new CBD product.
MAOI plus essentially any new medication. This is the only class where the default assumption should be that an interaction exists until proven otherwise.
Any combination that adds to sedation when you drive, work in a safety-sensitive job, or care for young children.
The one habit that prevents most of this
There is no app, no genetic test, and no clever combination of medications that replaces a complete and current medication list shared with every prescriber and pharmacist who treats you. That list includes everything you take — prescription, over-the-counter, supplement, vitamin, herbal, recreational. It includes the doses and how often you take them. It includes things you take only occasionally. It includes the supplement someone in your family recommended that you have been taking for two weeks.
The pharmacy at your usual location keeps your prescription history, but it does not know about the supplements you bought at the grocery store, the CBD gummies you tried last weekend, the antibiotic an urgent care prescribed, or the medication a friend gave you to try. Your prescriber depends on you to fill those gaps. Most missed interactions are not missed because the system failed. They are missed because no one was told.
If you remember nothing else from this essay, remember this: when a new clinician asks what you take, the answer is not “the usual.” It is the actual list, including the things you would rather not mention. The conversation is worth having for a reason — what gets shared is what gets safe.
If you or someone you know is in crisis
Call or text 988 — Suicide & Crisis Lifeline, available 24/7 En español: 988 y oprima 2 Veterans: 988 y oprima 1 — or text 838255 LGBTQ+ youth: The Trevor Project — call 1-866-488-7386 or text START to 678-678 Crisis Text Line: Text HOME to 741741
San Bernardino County residents: SBC DBH ACCESS Line: 888-743-1478 (24/7) SBC Mobile Crisis / CCRT: 800-398-0018
For life-threatening emergencies, call 911 or go to your nearest emergency room. Arrowhead Regional Medical Center (ARMC) has a dedicated adolescent psychiatric ER (ages 13–17).
This essay is general clinical education and does not replace individualized advice from your prescriber and pharmacist. If you take more than one medication, supplement, or substance regularly, a structured medication review with someone who has your full history is worth more than any general guide.
Clinical review: Fady Boules, PMHNP-BC
Footnotes
-
Clinical study evaluating drug interactions of cannabidiol with citalopram and morphine. PMC. 2026. https://pmc.ncbi.nlm.nih.gov/articles/PMC12997498/ ↩
-
U.S. Department of Veterans Affairs and Department of Defense. VA/DoD Clinical Practice Guideline for the Management of Posttraumatic Stress Disorder and Acute Stress Disorder. 2023. https://www.healthquality.va.gov/guidelines/MH/ptsd/ ↩
-
Al-Soleiti M, Leung JG, Mubaydeen T, et al. Lithium toxicity following semaglutide initiation: case series. Mayo Clinic Proceedings. 2025. https://pubmed.ncbi.nlm.nih.gov/40999647/ ↩
-
Center for Women’s Mental Health, Massachusetts General Hospital. Oral Contraceptives Reduce Lamotrigine (Lamictal) Blood Levels. https://womensmentalhealth.org/posts/contraceptives-lamictal-blood-levels/ ↩