July 17, 2026 · 8 min read

SSRIs Quietly Thin the Blood — and One Over-the-Counter Painkiller Turns That Into a GI Bleed — Here’s What Every PMHNP Student Needs to Know

An amber prescription bottle with white tablets spilled on a marble counter beside a glass of water

Here is a scenario that will feel almost too ordinary to be dangerous. A 74-year-old woman comes to you low, tearful, not sleeping, and you do exactly what your training says: you start an SSRI. She improves. A few months later her knee flares, she picks up ibuprofen at the pharmacy because it is over the counter and therefore feels harmless, and nobody — not her, not the pharmacist, not you — thinks twice. Three weeks after that she is in the emergency department with an upper GI bleed. Nothing about your prescribing was reckless. And yet a completely predictable, preventable event happened on your watch, because of an interaction that hides in plain sight inside two of the most common medications in America.

This is the kind of thing that separates a clinician who memorized SSRIs as “safe” from one who actually understands them. SSRIs are remarkably safe compared with the drugs they replaced — that is why they became first-line. But “safe” is not “consequence-free,” and the bleeding risk that comes with serotonin reuptake inhibition is exactly the sort of quiet, mechanism-based hazard that shows up on your boards and, more importantly, on your patients. Let’s make sure you see it coming.

Why this is back in the conversation

Bleeding risk with antidepressants is not new information, but it just got a serious update. In March 2026, the Association of Medicine and Psychiatry published an expert consensus in the Journal of Psychiatric Research — a group of physicians who work at the seam between medical and psychiatric illness sat down, combed the literature, and built practical recommendations for prescribing antidepressants in patients who already carry bleeding risk. They even produced an “AMP Bleeding Risk Score” to help identify the patients you should worry about, along with a consensus algorithm for what to do about it. When a specialty organization goes to the trouble of building a scoring tool around a problem, that is a signal the problem is common enough and consequential enough to deserve your attention.

It also echoes a theme The Carlat Psychiatry Report has hammered for years. In its September 2025 “Drug Interactions That Matter,” Dr. Jason Mallo made the case that clinicians waste anxiety on interactions that rarely bite while under-appreciating the handful that genuinely do — and the serotonergic contribution to bleeding is one of the ones that genuinely does, especially when a second drug is layered on top. Carlat has been flagging serotonergic antidepressants and abnormal bleeding since long before it was fashionable. For a PMHNP student, that consistency is the tell: this is not a fringe worry. It is a bread-and-butter safety issue you will manage over and over.

The mechanism you can actually reason from

Here is the piece that makes everything else make sense, and it is worth truly understanding rather than memorizing, because once it clicks you will never lose it. Platelets need serotonin to do their job. When a vessel is injured, platelets release serotonin from their dense granules, and that serotonin drives vasoconstriction and helps platelets clump together into a plug. But platelets cannot manufacture serotonin themselves. They have to scavenge it from the bloodstream using — and this is the beautiful, terrible part — the exact same serotonin transporter that an SSRI is designed to block.

So when you inhibit that transporter systemically, you are not only raising serotonin in the synapses of the brain, where you want the effect. You are also quietly starving the platelets of the serotonin they normally store. Over weeks of therapy, platelet serotonin content falls, and with it falls one arm of the body’s clotting response. The platelets are still there in normal numbers — this will not show up on a routine platelet count — but they are less effective at forming a plug. There is a proposed second mechanism too: SSRIs may increase gastric acid secretion, which raises the odds of an ulcer in the first place. Put those together and you have a drug that both makes a bleed slightly more likely to start and slightly harder to stop.

Notice what this mechanism predicts. If the effect depends on how strongly a drug blocks the serotonin transporter, then the antidepressants with the highest affinity for that transporter should carry the most bleeding risk — and that is exactly what the data show. Agents like fluoxetine, paroxetine, and sertraline sit at the higher-risk end. Antidepressants that barely touch serotonin reuptake, like mirtazapine and bupropion, sit at the lower-risk end. You did not have to memorize that list. You can derive it from the mechanism, which is precisely the kind of reasoning boards reward and rushed clinicians forget.

A single deep-red droplet blooming and spreading across a pristine white ceramic surface

How big is the risk, really?

Let’s keep this honest, because overstating it is its own kind of error. On its own, an SSRI raises the risk of upper GI bleeding modestly — meta-analyses generally land somewhere in the range of a 1.5- to roughly 3-fold increase compared with non-users. In absolute terms, for a healthy younger adult with no other risk factors, that is a small bump on a small baseline. You should not be frightening low-risk patients away from effective treatment over it.

The story changes the moment a second insult enters the picture. Add an NSAID — ibuprofen, naproxen, the drugstore staples — and the risk does not just add, it multiplies. Studies of the combination have reported the risk of upper GI bleeding climbing to roughly six times that of people on neither drug, with odds ratios in the range of 4 and above. A 2026 systematic review and meta-analysis in the British Journal of Clinical Pharmacology looking at bleeding by specific SSRIs and SNRIs reinforced that the danger is real and that it varies by agent. Aspirin, other antiplatelets, anticoagulants, and corticosteroids stack risk the same way. This is why the interaction is so treacherous: the SSRI alone is a whisper, but the SSRI plus a common painkiller your patient buys without telling you is a shout.

The surgical suite deserves its own mention, because you will rotate through consult and peri-procedure settings. Serotonergic antidepressants have been linked to more perioperative bleeding and a greater need for blood transfusion — one large multi-hospital analysis found roughly doubled bleeding-related odds in surgical patients on SSRIs. And yet, tellingly, there is still no firm consensus that you should routinely stop the SSRI before surgery, because abrupt discontinuation carries its own risks — discontinuation symptoms and a destabilized patient heading into a stressful event. That unresolved tension is a perfect example of why this is judgment, not a rule you can automate.

The patients who should make you pause

If you internalize one clinical habit from this post, make it this: before you write or refill an SSRI, run a quick bleeding-risk scan the same way you would screen for a drug allergy. The AMP consensus and the broader literature keep pointing at the same red flags. Advanced age is at the top — SSRIs should be used with real caution in patients over 80. A prior upper GI bleed is a loud warning. Concurrent aspirin, NSAIDs, anticoagulants, antiplatelet agents, or systemic corticosteroids each move a patient up the risk ladder. Liver disease, a bleeding disorder, and heavy alcohol use belong on the list too.

There is also a newer wrinkle worth carrying with you: signals from adverse-event reporting databases suggest that combining antidepressants with direct oral anticoagulants may increase hemorrhage risk beyond what the anticoagulant alone would predict. As DOACs continue to displace warfarin, you will see this pairing constantly, and “the patient is already on a blood thinner” is exactly the moment to slow down and think about which antidepressant you reach for.

A quiet empty hospital pharmacy dispensing area with rows of shelving at dusk

What you actually do about it

Awareness is only useful if it changes your hand. Here is how this translates into decisions you will make on a real panel of patients.

First, choose the molecule to fit the patient. For someone who is elderly, already on an NSAID or antiplatelet, or has a bleeding history, a lower-serotonergic antidepressant is often the smarter opening move. Mirtazapine and bupropion carry meaningfully less bleeding liability because they do not lean on serotonin reuptake the way the classic SSRIs do. That is not a reason to abandon SSRIs — it is a reason to match the drug to the risk profile in front of you.

Second, attack the modifiable risk. Often the SSRI is not the thing that should change — the ibuprofen is. Ask, at every visit, what over-the-counter and PRN medications your patient is taking, because they will not volunteer “ibuprofen” as a real drug unless you specifically dig for it. Where an NSAID cannot be avoided, a proton pump inhibitor meaningfully lowers the GI bleeding risk and is a well-supported protective move for higher-risk patients. Co-prescribing gastroprotection is one of those small decisions that quietly prevents emergencies.

Third, counsel the patient in plain language. A single sentence — “while you’re on this antidepressant, check with me before taking ibuprofen, naproxen, or aspirin, and tell me right away if you see black stools or vomit that looks like coffee grounds” — converts an invisible risk into something your patient can help you watch for. That is participatory safety, and patients remember it.

Why this matters for you specifically

You are training to be the person who owns the whole prescription, not just the diagnosis. SSRIs will be among the most common drugs you ever write, which means their risks are not exotic trivia — they are the daily texture of your practice. On the board exam, the SSRI-plus-NSAID bleed in an older adult is a classic stem, and the concept of PPI gastroprotection and choosing a less serotonergic agent are the kinds of answers that separate a confident test-taker from a guesser. In the clinic, this is the knowledge that keeps a routine antidepressant refill from becoming an ER visit.

But there is a bigger lesson hiding in this one, and it is the reason topics like this are worth your time now. Good psychiatric prescribing is not about fearing medications; it is about understanding them deeply enough to see around corners. When you can trace a bleed back through a starved platelet to a blocked transporter, you are no longer memorizing warnings — you are reasoning like a clinician. That is the muscle that makes everything else in psychopharmacology easier, from anticipating drug interactions to explaining a side effect to a nervous patient in words they can actually use.

So take this one and make it yours. Learn the mechanism until you could sketch it on a napkin. Build the reflex of scanning for aspirin, NSAIDs, anticoagulants, and age before you sign the prescription. Keep mirtazapine and bupropion in your back pocket for the high-bleeding-risk patient, and keep a PPI in mind when an NSAID is unavoidable. Small habits, practiced early, are what turn a good student into a safe prescriber.

You did not choose this field to prescribe on autopilot — you chose it to actually protect people. When you are ready to turn high-yield clinical connections like this one into board-day confidence and real-world competence, come study with us at pmhnpreview.org. The prescriber who sees the bleed coming is simply the one who learned to look a little earlier than everyone else.