A detailed review of Anders Sørensen's guide to coming off psychiatric drugs — what it gets right, where the science falls short, and what's dangerously missing.
If you're reading this, chances are you're either taking psychiatric medication and thinking about stopping, you're in the middle of a taper and struggling, or you're someone who's been through withdrawal hell and you're looking for answers.
Maybe your doctor told you to just cut your dose in half. Maybe you tried to stop cold turkey and ended up in the worst experience of your life. Maybe you've been told your symptoms are "just anxiety returning" when you know — you KNOW — something else is happening in your body.
So when a book comes along called "Crossing Zero: The Art and Science of Coming Off — and Staying Off — Psychiatric Drugs," written by someone with a PhD who runs a dedicated tapering clinic, you pay attention. This book by Anders Sørensen has been getting a lot of buzz in the withdrawal community. It's got endorsements from David Taylor, who co-wrote the Maudsley Guidelines. Now if you've seen my review of the Maudsley deprescribing guidelines, you know that's not as important to me as it might be for many others, given my criticism of that book. And funnily enough, Crossing Zero actually contradicts Maudsley in some ways — for example, the tapering tables in Maudsley are much more simplistic and linear at the tail of the taper, sometimes with quite big steps like 1mg to 0.5mg of diazepam in one go. That's something Anders keeps emphasising should not be done.
Here's what I've done: I've gone through this book in detail. I've traced the references. I've compared what it says to what the research actually shows. And I've compared it to what people in the trenches — the people actually going through withdrawal — have learned the hard way.
I also want to note that while I read most of the book in detail, I did use AI partially to research some claims, had conversations with my mum about my findings, and verified things myself. I'm working on a separate piece about using AI for research — I believe that if you use multiple tools, let them cross-check each other, and always use the best models available, you can get very high accuracy. But I always try to verify claims independently. Everything I say here is my opinion.
Anders Sørensen is a Danish clinical psychologist with a PhD in psychiatry from the University of Copenhagen. His doctoral research, completed in 2022, focused on the relationship between antidepressant doses and serotonin transporter occupancy. He runs a tapering clinic in Copenhagen and claims to have worked with many people coming off psychiatric medications — I believe over 10 years.
The book is divided into three parts. Part One covers how psychiatric drugs work and critiques the chemical imbalance theory — it's quite long and very detailed. Part Two is the practical tapering section, which I expected would be the meat of the book. Part Three focuses on psychological strategies for managing life without medication, heavily emphasising something called "detached mindfulness."
The book is 413 pages with over 200 references. It covers antidepressants, antipsychotics, benzodiazepines, mood stabilisers, and provides tapering examples for over 20 specific medications in an appendix. That's not a lot compared to the Maudsley Deprescribing Guidelines, where each drug gets its own section.
On the surface, this looks like exactly what the withdrawal community has been waiting for — a scientifically grounded, comprehensive guide to coming off psychiatric drugs safely. And in some ways, it is. But in other ways, it falls significantly short.
The central argument of this book is correct and important. Sørensen explains that the relationship between dose and drug effect in the brain is not linear — it's what he and Mark Horowitz call hyperbolic.
What this means: when you cut your dose in half, you're not reducing the effect in your brain by half. Depending on where you are on the dose-response curve, cutting from 50mg to 40mg of diazepam might barely change anything, but cutting from 2mg to 1mg might be enormous.
This is based on receptor occupancy — the percentage of receptors in your brain blocked by the drug at any given dose. Sørensen's own PhD research, published in Molecular Psychiatry in 2021, examined PET scan data across 17 studies and 294 participants. It showed that for antidepressants, around 80% of serotonin transporters are already occupied at the minimum recommended doses.
What this means practically: the final milligrams of your taper are potentially the hardest. This is why many people can cut from high doses without problems, then hit a wall at lower doses. It's not psychological. It's pharmacology.
This principle — that tapers need to slow down as doses get lower — is scientifically sound (with caveats I'll discuss later) and clinically important.
The term "hyperbolic" is technically correct for describing the dose-response curve — the Michaelis-Menten equation produces a hyperbolic shape. But when people say "hyperbolic tapering," the taper schedule itself isn't hyperbolic. It's exponential decay. You reduce by a percentage of your current dose each time. These are two different mathematical functions.
I think we should call it what it is: exponential decay tapering, or simply percentage-based tapering. That would help more people understand and trust the method. Furthermore, to do TRUE receptor-occupancy-guided tapering, you'd need to know the EC50 value for your specific drug — and as I'll discuss, that data comes from studies on healthy volunteers taking single doses, not long-term patients. So what we're all actually doing is using the percentage method as a practical approximation. And that's okay.
To his credit, unlike Maudsley which doesn't even mention a jeweller's scale — the number one tool for proper dose precision used in the lived experience community — Anders does. If you don't use this tool, you have to round to human-friendly numbers and available tablet sizes, which is exactly what the Maudsley tables do. Anders acknowledges how critically important it is to go in tiny steps as you descend.
Also to his credit, Anders mentions quite often that our bodies are complex and what works for one person could look very different for another.
The book doesn't gaslight when it comes to withdrawal symptoms. It advocates for you, the person who suffers. It clearly states that withdrawal symptoms affect approximately 50% of antidepressant users, 50% of antipsychotic users, and 40–90% of benzodiazepine users. It describes symptoms that can last weeks, months, or even years. It explicitly says these symptoms are physiological, not psychological.
This matters. Many people have been told by their doctors that withdrawal doesn't exist, that what they're experiencing is their "illness returning." This book validates what the patient community has been saying for decades.
Sørensen provides a clear explanation of why withdrawal happens using the concept of homeostasis. When you take a drug that increases or blocks a neurotransmitter, your brain adapts by downregulating receptors. When you reduce the dose, the drug leaves your body faster than your brain can readjust. This mismatch is the withdrawal experience.
He uses a helpful analogy of an accelerator and brake pedal. The drug is the accelerator. Your brain's adaptation is the brake. When you lift off the accelerator, the brake is still pressed. That lag time is what you experience as withdrawal.
The book provides four clues to distinguish withdrawal from relapse: the appearance of new symptoms unrelated to your original condition, symptoms that feel more "chemical" than psychological, symptoms that resolve with a slower taper at the same dose, and symptoms that pass with time. This is genuinely helpful guidance that many people need.
The writer mentions a technique he claims helped his patients called "detached mindfulness." It reminded me a lot of ACT, the DARE approach, and other techniques. The first chapters provide detailed psychological information — perhaps too much at the expense of practical guidelines, but still valuable content from someone with genuine clinical experience.
Here's something critical: Sørensen's PhD research is a systematic review of existing PET scan studies. It's not a clinical trial where he tapered long-term patients and measured outcomes.
His systematic review gathered data on 10 different antidepressants across 17 studies and 294 participants. That's reasonably comprehensive for receptor occupancy research. However, the actual study characteristics reveal major limitations.
Most participants were healthy volunteers, not long-term patients. Only a handful of studies used people with actual depression, and even those were relatively new to the medication.
Six of the seventeen studies only gave participants a single dose. Just one pill. The longest anyone was on medication was about two months. Not a single study looked at people who had been on antidepressants for years — exactly the population who struggles most with withdrawal.
Occupancy increases with longer treatment. Duloxetine showed 65% occupancy after a single dose but 78% after just four days. What happens after five years? We simply don't know.
The paper itself acknowledges this: "As many patients take antidepressants for years, studies should also study patients after long treatment duration." So even Sørensen knows this data is missing.
When the book presents specific numbers for where the curve "bends" or what percentage to reduce by, these are extrapolations — educated guesses, not proven protocols for long-term users. This is especially concerning given that he suggests the first cuts can be quite big. No one actually knows when the curve bends for any individual, so I would never tell someone to cut 20–30% just because they take a high dose and it "probably" won't affect them much.
When I dug into the references, I found something interesting. The receptor occupancy curves that form the foundation of this entire tapering methodology — they primarily cite Sørensen's own systematic review and Horowitz and Taylor's papers. When you look at Horowitz's 2019 SSRI paper, it also uses Meyer 2004 as its primary PET data source — the same study in Sørensen's review.
So we have overlapping citations using the same underlying data, presented as if they're independent sources of evidence. This isn't fraud — it's just how academic citation works. But readers should understand these aren't three separate bodies of evidence. They're largely drawing from the same well.
Both Horowitz papers explicitly state that their proposals are hypotheses that need to be tested in clinical trials.
"This paper offers a pharmacologically-informed guide to withdraw from SSRIs. Its validity should be confirmed by randomised controlled trials." — Horowitz 2019, SSRI Paper
"The hypothesis put forward in this paper should be tested in further tapering trials of antipsychotics." — Horowitz 2021, Antipsychotic Paper
Yet Anders presents these untested hypotheses as established guidance. The papers themselves are more cautious than the book that cites them.
In Horowitz's 2019 paper, the diazepam binding curve that illustrates the hyperbolic principle comes from a 1991 study by Brouillet et al. — "In vivo bidirectional modulatory effect of benzodiazepine receptor ligands on GABAergic transmission evaluated by positron emission tomography in non-human primates."
Non-human primates. Monkeys.
Anders notably does NOT cite the Maudsley Deprescribing Guidelines, which rely on these same primate studies for their benzodiazepine curves. Instead, he simply claims the "law of mass action" makes the curve universal — a theoretical argument, not an empirical one for benzos.
For antipsychotics, the underlying theory for dopaminergic hypersensitivity comes partly from rat studies — rats given antipsychotics for 2 weeks (Samaha 2007) and rats on haloperidol for 9 months (Joyce 2001). Horowitz's paper provides curves for only 6 antipsychotics where the data fit reasonably well. Many commonly prescribed antipsychotics aren't covered.
Crucially, Horowitz warns that for drugs with short half-lives like clozapine or quetiapine, "more caution may be required." This caution is NOT clearly conveyed in Anders' book.
On page 196, the book claims that the shape of the occupancy curve is "the same across all psychiatric drugs" due to the "law of mass action." This is a massive overreach. Different drugs have different receptor binding profiles, different regional brain distributions, and different active metabolites.
The book even admits in a footnote on page 246 that for benzodiazepines and mood stabilisers, "we just know less in terms of research about the dose at which tapering typically needs to slow down." Yet despite admitting they don't have the research, they still provide specific tapering ranges in Table 5.
In Horowitz's 2019 SSRI paper, there's this acknowledgement: "We would like to acknowledge the contribution made to ideas in this manuscript from posts on survivingantidepressants.org."
The patient community actually contributed to this "scientific" paper. The lived experience community's knowledge fed INTO the academic work, not just the other way around. Yet Anders' book dismisses some of that same community knowledge as potentially harmful.
On pages 241–242, the book states that reinstatement becomes "increasingly risky after about three to four weeks in severe withdrawal." To be fair, the text hedges this — he writes that "how long this window lasts no one knows, and is obviously individual." But putting a specific number like "three to four weeks" in print can take on the weight of fact for anxious readers looking for concrete guidance.
The layperson community has documented successful reinstatements months later. It has also documented cases where even a week after was unsuccessful. When you're in withdrawal and you read a specific number, that number sticks, regardless of how many qualifiers surround it.
The book mentions that people differ in genetics and metabolism, but provides no practical guidance. The only CYP enzyme mentioned is CYP1A2 — once, on page 240, regarding smoking and clozapine metabolism.
There's no mention of CYP2D6 variations (affecting paroxetine, fluoxetine, venlafaxine, duloxetine, risperidone, aripiprazole), CYP2C19 variations (critical for diazepam metabolism — 15–20% of some populations cannot properly metabolise diazepam), or CYP3A4 variations. The book recommends switching to diazepam without any warning about CYP2C19 polymorphisms. From my own observations, many people in the forums don't tolerate this switch.
The book recommends switching to fluoxetine for SSRI and SNRI tapers because of its long half-life (listed as 4–6 days). What it fails to adequately address: fluoxetine's active metabolite, norfluoxetine, has a half-life of 7–15 days. If you switch to fluoxetine and have a bad reaction — akathisia, severe agitation — you cannot simply stop it. That metabolite will remain in your system for five to six weeks. You're trapped.
The book also claims serotonin syndrome "usually resolves within a few days of stopping the medication." This may be true for short-acting drugs, but it's dangerously misleading for fluoxetine specifically.
Fluoxetine is a very potent inhibitor of the CYP2D6 liver enzyme. If you're taking other medications metabolised by this enzyme — certain beta-blockers, painkillers, or antipsychotics — adding fluoxetine essentially "blocks the exit" for those drugs, causing their levels to skyrocket in your blood.
Interestingly, Horowitz's own 2019 paper acknowledges this. But Anders doesn't mention CYP2D6 at all — the letters don't appear once in his 413-page book.
And it cuts both ways: if you're stabilised on a cocktail of meds including fluoxetine and you stop it, the inhibition eventually lifts, and the levels of your other medications might plummet, causing withdrawal from THOSE drugs.
I should mention the concerned look on my mum's face — she was a psychiatrist for 30 years — when she heard he wants to ADD fluoxetine on top of another SSRI drug.
Clinical cross-tapering where two serotonergic drugs briefly overlap IS done by some prescribers. However, it significantly increases serotonin syndrome risk. Serotonin syndrome is not just uncomfortable — it can be life-threatening, causing seizures, loss of consciousness, or even death. And because of norfluoxetine, if something goes wrong with fluoxetine specifically, you could be dealing with it for weeks, not days.
The book also recommends switching SNRIs to fluoxetine despite acknowledging in a footnote that "the receptor overlap with fluoxetine is less significant compared to SSRIs." SNRIs affect both serotonin AND norepinephrine. Fluoxetine only covers serotonin. So you're essentially going cold turkey on the norepinephrine component.
The book presents diazepam as "by far the most taper-friendly option" for benzodiazepines, but there are several issues: the CYP2C19 problem already mentioned; diazepam is more sedating and potentially more depressogenic than many modern benzodiazepines; it has multiple active metabolites (nordiazepam, temazepam, oxazepam) each with different half-lives; and in some countries, diazepam tablets contain dyes including tartrazine, which can trigger mast cell reactions in susceptible individuals.
The book insists doses "must be converted very precisely" when switching to diazepam, but provides no equivalency tables. Someone reading this book has no way to know what dose of diazepam equals their current benzodiazepine dose.
The book presents long half-life drugs as having a "built-in taper." But this framing ignores a critical problem: long-acting drugs break the feedback loop.
With a short-acting drug, if you cut too much, you feel it within 24 hours. You can reinstate quickly. With diazepam or fluoxetine, you might feel fine for two to three weeks while the drug auto-tapers from its previous levels. You might even make another cut during this period, or decide to cut more because "it doesn't affect me." Then at week four, the cumulative withdrawal from multiple cuts hits simultaneously — and you don't know which cut caused it.
Short-acting drugs taken multiple times daily actually provide better real-time feedback about whether your taper is too fast. The book dismisses this by warning about "interdose withdrawal" without acknowledging its advantages.
For a book that's supposed to be a practical guide, the actual instructions for achieving small doses are remarkably brief and incomplete.
The book describes dissolving tablets in water but doesn't explain how long the solution remains stable, whether it needs refrigeration, light sensitivity issues, bacterial growth concerns, what to do with enteric coatings, or pH differences between water and stomach acid. It just says "stir well."
It also casually suggests switching brands if your tablet doesn't dissolve well — but switching brands can itself trigger withdrawal reactions due to differences in fillers, binders, and bioavailability.
The weight method gets about half a page. It says you need a "high-quality precision scale" but doesn't specify what precision. Some consumer jeweller's scales have error margins of ±3–5 milligrams. If you're making tiny few-mg reductions, that error margin makes the measurement essentially random.
For capsules containing beads, the book admits bead counts can vary between capsules — up to 15%. If you're trying to do a "precise" 3% reduction with 15% capsule-to-capsule variation, you're not achieving precision. You could be doing nothing or cutting 18% depending on which capsule you opened.
No milk titration method (which the layperson community has developed and Maudsley actually recommends). No substantive discussion of compounding pharmacies beyond a brief mention. No discussion of pharmacy-grade liquid formulations.
One of the most frustrating aspects of this book is how it treats all psychiatric drugs as essentially the same.
Drugs like lithium, lamotrigine, pregabalin, gabapentin, valproate, and carbamazepine are mentioned about a dozen times in the entire 413-page book, mostly in passing. There is no dedicated chapter, no specific guidance, no detailed discussion of their unique challenges.
Despite admitting they don't have the research, the book still provides specific "bend point" ranges — claiming diazepam bends at 3–5mg, lithium at 300–450mg, pregabalin at 200–500mg, lamotrigine at 150–400mg. Where did these numbers come from if the research doesn't exist? Common mood stabilisers like valproate and carbamazepine aren't mentioned at all.
Agomelatine (Valdoxan) — an antidepressant acting as a melatonin agonist — is mentioned exactly ONCE on page 41 in a list. But it has serious unique considerations: liver toxicity requiring regular blood tests, an extremely short half-life of 1–2 hours, and a completely different receptor profile (melatonin MT1/MT2, serotonin 5-HT2C). The book's entire framework is built around serotonin transporter occupancy, which is irrelevant for agomelatine.
The book treats all psychiatric drugs as following the same hyperbolic curve, governed by the same "law of mass action." But SSRIs and SNRIs work on different receptor systems. Benzodiazepines affect GABA. Antipsychotics block dopamine and can cause supersensitivity psychosis. Mood stabilisers work through mechanisms we don't fully understand. Z-drugs have ultra-short half-lives. Atypical drugs like mirtazapine, bupropion, and agomelatine each have unique profiles.
A truly comprehensive guide would acknowledge these differences with drug-class-specific chapters. Instead, we get one-size-fits-all advice.
The book does not mention supplements. Not magnesium. Not CBD oil. Not B vitamins. Not vitamin D. Not omega-3s. Not Chinese herbs, chamomile tea, green tea and theanine — nothing. Not to recommend them, not to warn about them, not to discuss how withdrawal affects nutrient needs. This is extraordinary for a 400-page book about psychiatric drug withdrawal.
There's no mention of dangerous addictive substances like kratom that people sometimes turn to in desperation. No mention of therapies beyond "detached mindfulness" — no acupuncture, EMDR, hypnosis, red light therapy, breathing techniques, CBT, ACT, yoga nidra, or vagus nerve activation. No mention of medications that can help in some cases, like prazosin or propranolol.
On page 189, the book warns readers to "beware of the sometimes overly restrictive diets and strict lifestyle rules recommended on some online withdrawal forums." It adds: "If you decide to exclude foods, beverages, or activities that you normally enjoy, make sure that doing so actually helps (which it may) – and don't just do it because you read it online."
He does acknowledge dietary restrictions MAY help. But the framing — with the warning first and the acknowledgement buried in a parenthetical — risks being dismissive. For someone who has developed histamine intolerance or mast cell activation during withdrawal, those "restrictive diets" might be exactly what they need.
It does go both ways. A certain person once took a bunch of us from the benzo forums to start a crazy diet and regimen. Most of us bought juicers, blood pressure monitors, and all sorts of supplements. It caused me a horrible setback when I followed her advice — drinking celery and cucumber juice to "detoxify" — something you should never do when trying to maintain a stable drug level. That person became abusive and abruptly closed the group. Many people there had suicidal ideation.
The truth is in the good middle — the floating, often elusive point of truth for your own biology and current mental state. I highly suggest doing blood work every 2 months at least when in this situation, and especially before starting a taper.
The book mentions alcohol about 11 times and caffeine about 7 times, but mostly in other contexts — comparing how drugs work or explaining receptor mechanics. There's no substantive warning to avoid alcohol during withdrawal, despite alcohol affecting GABA receptors — the same receptors involved in benzodiazepine withdrawal. No warning that caffeine can worsen anxiety, sleep problems, and akathisia.
I personally know a person who became much more ill with horrible symptoms of crushing head pressure because he drank alcohol once and lit a joint after a period of improvement. There are many people who avoid alcohol, taper, get better, then drink one drink after a year and suddenly have symptoms again. In the benzo survivors community, it is accepted that for at least 2 years after your last symptom, you don't drink alcohol.
I found it surprising that grapefruit juice is mentioned as something to avoid, but alcohol — which can be even more problematic — doesn't receive comparable attention.
This might be the most significant omission.
On page 242, the book describes "kindling" as a state where the nervous system becomes hypersensitive to almost anything — the drug itself, caffeine, sugar, alcohol, supplements, even detergents and cleaning products. The system says "No more!" and puts up guardrails.
Does this sound familiar? This closely resembles what many in the lived experience community recognise as mast cell activation syndrome (MCAS). But the book offers no explanation of the mechanism and no treatment options.
A 2025 cohort study (medRxiv) found that 53.8% of patients undergoing benzodiazepine tapering showed clear symptoms of mast cell activation.
The lived experience community in the benzo world defines kindling differently: starting and stopping a drug multiple times — taking benzos for a few months, coming off, starting again, and finding it way harder to taper each subsequent time. The author doesn't seem to mention this definition.
He also writes: "Protracted withdrawal and kindling reactions are virtually never seen when tapering is done slowly and gradually." But just paragraphs earlier, he acknowledges the research on kindling "hasn't been done yet." If the research doesn't exist, you can't claim slow tapering virtually eliminates it.
Mast cells are immune cells throughout your body that release histamine and other inflammatory chemicals. They have GABA-A receptors — the same receptors benzodiazepines act on. During long-term benzodiazepine use, mast cells may be stabilised by enhanced GABA signalling. During withdrawal, they can become destabilised and hyperreactive.
When mast cells degranulate inappropriately, you get symptoms across multiple systems: cardiovascular (palpitations, flushing), GI (nausea, diarrhoea), neurological (brain fog, anxiety), skin (hives, itching). Compare this to the book's withdrawal symptom tables — the overlap is remarkable.
If symptoms are driven by mast cell activation, "detached mindfulness" isn't going to fix it. What patients potentially need — and this is not medical advice, but worth discussing with your physician — includes H1 antihistamines (cetirizine), H2 antihistamines (famotidine), mast cell stabilisers (quercetin, cromolyn), and possibly a low-histamine diet. The book offers none of this.
Even worse: the book recommends diazepam for benzo tapers, but many diazepam tablets contain dyes including tartrazine — a known mast cell degranulator. For someone developing MCAS during withdrawal, switching TO diazepam could make everything worse.
Part Three includes a chapter suggesting psychosis can be managed primarily through psychological techniques, framing psychotic symptoms as understandable responses to life circumstances.
"Yet I have never seen a mental illness — never encountered a condition, state, or set of symptoms that didn't make sense in light of the person's circumstances and life story." — Anders Sørensen, p. 385
Even as a statement about his clinical experience, this is a significant overstatement that could be harmful for people with genuine biological psychosis. While trauma certainly plays a role in some cases, claiming ALL conditions can be fully explained by life circumstances ignores genetic factors, neurobiological dysfunction independent of experience, and cases where psychosis appears without identifiable trauma.
THC can trigger psychosis in sensitive individuals. A close relative of mine saw vivid hallucinations every night — wide awake and moving — with zero traumatic experiences. For someone with genuine biological psychosis, bipolar disorder, or other neurodivergent conditions, this perspective could delay necessary treatment or cause inappropriate guilt.
Horowitz's 2021 antipsychotic paper is actually more cautious than Anders' book — noting that relapse rates following antipsychotic discontinuation are elevated for 1–3 years and that tapering periods "probably need to be similarly prolonged."
The book relies heavily on "detached mindfulness" as the primary tool for managing withdrawal symptoms.
For anxiety-type symptoms that are clearly withdrawal-related, learning not to catastrophise and to let symptoms pass can genuinely help. The attention training exercises are legitimate psychological techniques.
But applying mindfulness to physiological emergencies is dangerous. You cannot mindfully detach from severe akathisia, serotonin syndrome, mast cell reactions, or seizure risk. The framing throughout emphasises psychological management in a way that could lead people to delay seeking help or blame themselves for not being "mindful enough."
You can tell the author is a psychologist first. The psychological sections are extensive and detailed, while the practical medical guidance is thin. This imbalance reflects his training but may not serve readers who need more pharmacological and less psychological guidance.
The fundamental issue: this book is written from a primarily academic and clinical perspective rather than from personal experience with withdrawal. The writing in some places treats suffering withdrawal symptoms as acceptable and manageable with the right mindset — cut the initial dose, and if you suffer, just go back up.
From experience, that can take weeks to stabilise. Until then, you're living at your sister's house, crying day and night, with burning sensations, jelly legs, living nightmares, sleeping 3–4 hours at most — feeling shameful, losing weight, scared to leave the house. I know because that's what happened to me. Multiple times.
Crossing Zero is not a bad book. In fact, I would read it again. The core science about slowing down as you go lower is sound in principle. The acknowledgement that withdrawal is real and can be severe is very important. The basic framework of going slow, listening to your body, and reinstating if things go badly wrong is correct and should be taught to every prescriber.
But it's an incomplete book presented as comprehensive. A theoretical framework presented as practical guidance. An academic's perspective presented as clinical wisdom.
When you trace the references: the SSRI curves come from Sørensen's own systematic review, overlapping with Horowitz's sources. The antipsychotic curves come from a meta-analysis of small studies with underlying theory from rat experiments. The benzodiazepine curves are essentially theoretical — based on old primate data. Both Horowitz papers explicitly call their proposals "hypotheses that should be tested in randomised controlled trials." Those trials haven't been done.
All those EC50 values, "bend points," and specific percentages come from studies on healthy volunteers taking single doses or short courses. Not a single study looked at people on these drugs for years. This doesn't mean the principle is wrong — but we should be honest about what we're actually doing: using exponential decay tapering as a practical approximation.
If you read this book, read it for the principles, not the protocols. Understand WHY you need to taper slowly, but don't assume the specific methods are complete or will work for everyone.
Supplement this book with knowledge from the layperson community forums — but be careful, as many people repeat things uncritically. Get information about MCAS and histamine intolerance, especially if coming off a benzo or Z-drug. Get guidance on supplements from practitioners who understand sensitised patients. Find emergency protocols from other sources. Get pharmacogenomic testing if possible. Understand drug interactions, especially CYP2D6 if considering fluoxetine.
And most importantly, work with a clinician who has actual experience with complex withdrawal cases — not just someone who's read about it, but someone who has walked patients through the worst of it. Or better yet: a good tapering coach working hand in hand with a prescriber who is flexible and will allow you to taper properly.
If you're currently struggling with withdrawal, please reach out. You don't have to go through this alone.
And if you're a clinician reading this: please listen to your patients. They know their bodies. When they tell you something is wrong, believe them.