Author: Dr. David Healy
20 February 2015
Other guides are available in Guides & Papers.
RxISK & MISS
The most common query to RxISK has been from people dependent on or trying to withdraw from antidepressants, anticonvulsants, or other drugs. This paper provides some of the historical and theoretical background for two other papers on general withdrawal strategies and antidepressant withdrawal.
Confusions arise whenever anyone starts talking about dependence and withdrawal, because of a close-to-complete denial of the issues on the part of both pharmaceutical companies and the medical establishment. This paper picks a path through the confusions and their implications and calls on doctors, pharmacists, patients and others to identify problems and their possible solutions — to put this MISSing area of medical practice on a proper footing.
For decades, perhaps centuries, we drew distinctions between drugs and medicines. Medicines were used in medical care. Drugs were available over the counter or on the street and were often abused.
Alcohol, opiates, cocaine and later nicotine were the classic drugs. These were linked to social problems; addicts used them. Addicts and later their drugs were demonized — and in the case of opiates and cocaine made available on prescription only.
Addicts were initially viewed as sinful creatures, and later as personality disordered. The demon lay in them rather than in the drugs.
It was only in the 1950s that research made it clear that alcohol withdrawal could cause alcoholism and opiate withdrawal could lead to opiate dependence even in normal people. This led to a distinction between the physical dependence on opiates and addiction. Some very famous and highly functioning surgeons and others were for instance dependent on opiates but were clearly not junkies.
Physical dependence in this sense refers to changes induced in the body by a drug or a medicine that can lead to problems on withdrawal from the drug.
Addiction was different — it referred to drug seeking and often criminal behaviors. A dependent individual need not be a junkie.
This led to a hunt to detect what drugs might cause addiction — problem drugs. In the mid-1960s, this produced the concept of abuse liability. Drugs that had a high abuse liability induced pleasure or caused craving, and were linked to the development of a tolerance so that the person needed escalating doses.
This cluster of features led to idea of substance abuse and the confusing idea of drug dependence (WHO 1965). Drug dependence predisposed a user to the risks of withdrawal but also threatened to make them a “junkie”.
This kind of drug dependence leaves academics and companies able to say our antidepressant, steroid, or antihypertensive does not cause either addition or dependence.
When they hear that a drug does not cause dependence, most people assume that it will not cause withdrawal. But physical dependence causes withdrawal not drug dependence. Many drugs with high abuse liability do not cause withdrawal.
By the early 1960s it was clearly outlined in articles and agreed at meetings that the antidepressants and the antipsychotics cause physical dependence and withdrawal.
Despite the high profile of this issue, the recognition of antidepressant and antipsychotic withdrawal syndromes vanished from professional and public awareness soon after.
At the same time, the dependence and withdrawal that other drugs can cause was stillborn. This happened because comparatively few drugs in the 1950s and 1960s were given chronically. Drugs like the antibiotics were given for a week only. The fact that antihypertensives and other drugs did not apparently act on behavior also blocked any sense that these “other” drugs could cause physical dependence.
Retrospectively, we can see that the steroids which were introduced in the 1950s cause dependence. Doctors learned pretty quickly that even after only a few weeks exposure these drugs had to be tapered carefully — not doing so could kill patients.
When the beta-blockers were introduced and given long term for hypertension, stopping them was linked to a dramatic increase in heart rate and blood pressure. But instead of calling this dependence and withdrawal, a new term was introduced — rebound.
Despite later denial that the antidepressants could cause withdrawal, some doctors recognizing they too could cause problems on stopping described the problem as stemming from cholinergic rebound.
Rebound emerged just as discontinuation effects did 3 decades later as a term to cover symptoms on stopping drugs — seemingly something completely different to withdrawal effects.
Others may have better explanations for what happened, but here’s one. Amnesia can stem from a brain trauma or other injury, but it more commonly stems from social factors. Where there are inhibitory demands subjects will not be able to remember details linked to one set of demands. This is almost always the case when there is a mass amnesia.
In the 1960s, psychiatry was faced with the enemy at the gate. The world was in upheaval. After the Second World War, rising affluence and a tense stalemate between superpowers gave rise to democratic movements that questioned the legitimacy of states, East and West. At the same time there was rising drug intake by college students and other members of the middle-class. This was a group never previously linked to drug abuse who could not be portrayed as addicts. The crisis occurred in 1968.
The focus was on the hallucinogen group of drugs — LSD, psilocybin and phencyclidine.
These drugs associated with the counter-culture of the 1960s were perceived as subversive of the social order, causing problems for instance with the Vietnam War effort. They were banned and portrayed as drugs of abuse, even though they do not cause dependence or addiction. In short order, the bad drugs became drugs to which subjects became dependent and correspondingly the good drugs, therapeutic drugs, which restore individuals to their place in the social order by curing diseases, were drugs to which people could not get hooked.
The political reality of the time over-rode the emerging science. While physical dependence is a pharmacological issue, addiction is a social one with political implications and as a result antipsychotic and antidepressant dependence was written out of the picture and only began to re-emerge in the mid-1990s. In a delicious historical irony the re-emergence was partly fuelled by a marketing war between Lilly’s Prozac and SmithKline’s Seroxat/Paxil.
The first indication of where the corpses were buried in the 1960s came in the 1980s when a dispute blew up between the public and politicians on the one side with medicine on the opposite side over the risks of dependence on the benzodiazepines.
The benzodiazepines produce a clear physical dependence. This develops on low doses of the drugs — in contrast to the doses of alcohol that alcoholics used and of opiates that opiate addicts used. The dependence happens in patients taking the drugs for therapy rather than for recreation. It happens in normal people, not “addicts”.
Confused doctors refused to accept that there could be a serious dependence problem with the therapeutic use of a drug. But the “victims” of “benzos” received public support and sympathy in a way that “addicts” never do. This was the first time that medicine had been accused of creating addicts. The profession reacted as if stung.
The American Psychiatric Association tried to separate addiction and dependence. “Historically, long-term, high dose, physiological dependence was called addiction, a term that applies to recreational use. In recent years, however, it has been apparent that physiological adaptation develops and discontinuance syndromes can appear after regular therapeutic dose administration…. in some cases after a few days or weeks of administration. Since therapeutic prescribing is clearly not recreational use, the term dependence is preferred to addiction, and the abstinence syndrome is called a discontinuance syndrome” (APA 1990).
In plain language this says if you get hooked to a drug being used for therapy, this is okay — please don’t use the term dependence or withdrawal to describe what’s going on.
This is not unreasonable but the public don’t buy it. For many people being unable to stop a drug is what they mean by being hooked or addicted and if there is a risk of not being able to stop they want to be told this before they start treatment. The public will never allow doctors or pharmaceutical companies to decide this one. It is a classic consent issue.
The benzodiazepines are relatively pleasurable — opiate light. Not a surprise that people might get hooked to them. But the antipsychotics and antidepressants are not opiate light. They are not pleasurable and they blow a hole in both medical and public views of addiction.
Here’s the problem. If you stop a mood-stabilizer antipsychotic, you may be hit out of the blue with tardive dyskinesia — a severe and disfiguring neurological condition. This has all the classic features of physical dependence. It appears on withdrawal but it can also appear in the course of treatment. It shows tolerance as alcohol does — you can treat the problem by increasing the dose of the drug or by restarting treatment.
For most of us, withdrawal is something that goes on for a few weeks at most. After that if a person is complaining about problems, doctors don’t believe them. But in the case of tardive dyskinesia, this very visible problem continues for months or years and in some cases indefinitely. Nothing about it is in the mind.
This is not the kind of problem doctors want to think they could cause. It’s only one of several enduring problems that antidepressants, steroids and other drugs can cause that don’t look much like opiate or alcohol addiction.
Symptoms on stopping (SOS)
If the train hadn’t come off the tracks in the 1960s, what might dependence look like now?
We would not be so focused on the brain.
When patients stop antipsychotics, a quarter have neurological problems. The other three-quarters have increased sensitivity to heat or stress, gut problems, or rapid alterations in mood.
These are not features of the original illness. They appear almost instantly on withdrawal where an illness relapse would only be expected weeks or months later. Re-starting treatment with a low dose of drug relatively instantly suppresses the problem, whereas treating a new illness episode usually requires hefty drug doses and can take weeks or months to restore control.
When we think of withdrawal from psychotropic drugs, we think the problems must be primarily mental or psychological but in this case the problems are intensely physical.
Some doctors or patients might agree that the symptoms are physical but still say that the problems stem from the brain. But the problems that come with steroid withdrawal do not stem from the brain, so do antidepressant withdrawal problems have to come from the brain?
In the case of the antidepressants, only 5% of our serotonin is in our brains. The gut has 50% of our serotonin and the cardiovascular system 40% with 5% elsewhere. It’s difficult to avoid linking withdrawal to the brain but many of the symptoms and antidepressant withdrawal come from the inner ear or the blood vessels to the scalp and eyes. The increased risk of blood clots certainly doesn’t come from a brain effect.
We would recognize that many other drugs cause dependence and withdrawal.
Just like the SSRIs, aspirin causes a rebound increase in clotting, as does the statin group of drugs.
We have been hypnotized into thinking that because drugs active on the serotonin system have been marketed as “antidepressants” that they may be at risk of producing withdrawal effects where drugs like the statins don’t. But just as there is more serotonin in our gut than our brain, so also there is more lipid in our brain than in our blood stream.
The marketing of these drugs obscures the withdrawal problems produced by respiratory, cardiac and other drugs — doctors and patients think these cannot produce withdrawal problems because they don’t act on the brain.
But drugs like metoclopramide and prochlorperazine marketed for nausea and gut problems are in fact identical to antipsychotics like olanzapine and cause identical tardive dyskinesia and other withdrawal problems.
We would stop treating dependence and withdrawal as something psychological and would think pharmacologically.
Anxiety is a common feature of SSRI withdrawal. The fact that SSRIs have often been given for anxiety confuses doctors and patients into thinking their original anxiety is re-emerging and they need a non-drug treatment for this. We are commonly asked by patients in withdrawal from their SSRI, why is it that cognitive therapy (CBT) doesn’t help my anxiety?
One answer to this is that no-one expects CBT to help the anxiety and tension linked to influenza or mitral valve disease or cardiac palpitations. The anxiety linked to withdrawal from SSRIs and Steroids is no less physical than the anxiety linked to influenza and is no more likely to respond to CBT than steroid withdrawal would.
Healthy volunteers taking SSRIs for only two weeks can have marked anxiety and depressive symptoms — these are physical not mental.
We would recognize that getting off steroids and mood-stabilizers may be harder than getting off heroin.
Some people, who want to come off treatment, may be unable to do so without lengthy and significant discomfort.
We have excellent treatments for opiate and alcohol dependence but our treatments for severe therapeutic drug dependence are much less effective.
There is every reason to believe patients who have been hooked to opiates, alcohol, and antidepressants, when they say that getting off an antidepressant can be more difficult than getting off anything else.
There is a huge overlap in the symptoms of benzodiazepine and SSRI withdrawal. Where once doctors denied that benzodiazepines caused problems, they now argue SSRIs don’t cause the problems that everyone recognizes the benzodiazepines did.
We would recognize that treating and stopping is not the same as not treating.
In the case of many treatments, what this means is that it may become impossible once the individual is on beta agonist for asthma, antidepressants, proton pump inhibitors for GERD or other treatments to know where the treatment ends and the disease begins.
It is likely that a part of the neurotic and dysthymic pictures that are counted as negative features of schizophrenia are treatment-induced phenomena rather than manifestations of the illness. There is a lot of evidence that SSRIs may in similar fashion all too often become the problem for which they are the treatment. This is a prospect that medicine cannot view with equanimity.
A few examples may help bring out the extraordinary anomalies to which the “political” settlements of the late 1960s have given rise.
1. From the 1990s, there was a huge increase in use of stimulants like methylphenidate (Ritalin) for children, with few parents having qualms about this, even though Ritalin differs little in its pharmacological profile from cocaine.
More recently, stimulants are being promoted for adult attention deficit hyperactivity disorder (ADHD) in clinics many of whose patients had formerly been attendees of drug abuse clinics in the same treatment facilities.
2. By the 1990s, many physicians viewed Valium as more addictive than Heroin. This stemmed in large part from the marketing of SSRIs which companies sold as the non-addictive alternate to the benzodiazepines.
A replay of this marketing in future may lead to the SSRIs being more “addictive” than Heroin, which pharmacologically may in fact be true.
3. There have been more reports to the regulators about dependence on SSRIs than for any other drugs, including the benzodiazepines and even opioid analgesics. But despite widespread recognition of problems at the individual patient level, there remains a failure at the social, professional and scientific levels to acknowledge any difficulties, with clinicians typically portraying any withdrawal problems as mild and transient.
Given that the SSRIs have been linked to a doubling of the rate of congenital malformations and of miscarriages, and likely mental handicap in children of women who have been on these drugs in pregnancy, the issue of dependence on these drugs is of far greater public health importance than dependence on the benzodiazepines was.
The difficulties in withdrawing from SSRIs can be brought out by the fact that women who have had one child borne with congenital malformations can find it difficult to stop SSRIs even though they may wish to have another child. Stopping SSRIs may be harder than stopping anything else.
4. There is abundant historical evidence that therapeutic communities like AA do more for a larger number of proper addicts than drug treatments do. But with the risk of a disease model of addiction, this evidence is written out of the script. All we see in the media is news of new agents, such as naloxone and acamprosate, which have been brought to market for alcohol or opiate dependence.
These are treatments for the craving opiates and alcohol can cause. They are treatments for dependence.
5. Lilly obtained a license to claim their antipsychotic, Zyprexa, is a mood stabilizer, on the basis of a trial in which a group of patients stable on Zyprexa were randomized to continue Zyprexa or be withdrawn to placebo. There is unquestionably a withdrawal syndrome to Zyprexa. So is the fact that a greater number of those put on placebo deteriorated compared to those on Zyprexa in the first few weeks after the switch, but not thereafter, a demonstration that Zyprexa stabilizes moods or a convincing demonstration of physical dependence on Zyprexa?
These examples indicate that addiction and theories of addiction are absolutely context dependent. Society it seems is happy to countenance individuals staying on particular psychotropic drugs, as with the antidepressants and antipsychotics currently. When this is the case issues of dependence or addiction are not raised, even though the difficulties in stopping may be extreme and amount to an enforced compliance.
The pleasurable effects of the benzodiazepines made it easy to portray them as opiate-like. The antidepressants and antipsychotics don’t look like opiates or cocaine. They don’t have a street value. They usually don’t cause tolerance.
We need some new words to describe the problems the mood-stabilizers, steroids and other drugs can cause.
This has led us to the idea of a stress syndrome.
Stress syndromes can be distinguished from side-effects:
- They don’t appear immediately.
- They may first appear or be most obvious on stopping treatment.
- They may disappear on re-starting or increasing the dose of treatment — conventional side effects do just the opposite.
- They develop a life of their own and may persist for months or years after the original drug has been removed.
- They may be sufficiently severe to produce a situation of enforced compliance.
- Stress syndromes may be more uncomfortable and dangerous than opiate withdrawal.
Tardive dyskinesia is the archetypal stress syndrome. But there are also tardive dystonia, tardive akathisia and tardive dysthymia states following mood-stabilizer and antidepressant use.
Stress syndromes may come in the form of rebound, stress, or legacy effects.
Many drugs force the body to do the opposite to what the drug was doing.
- So beta-blockers that slow the heart, force the heart to try to overcome this and on stopping for a while pulse rates rebound.
- Shock treatment causes the brain to fight against seizures and for a while afterwards the seizure threshold is higher — it is harder to have a convulsion.
- Opiate withdrawal is largely a rebound effect.
- Beta agonist inhalers also certainly cause some degree of wheezing that the inhaler then alleviates.
- Thrombosis after stopping aspirin, SSRIs or statins, is likely a rebound effect.
- Rebound effects will tend to have a limited lifespan.
Push a bodily system too hard in one direction and just like a spring it may lose its bounce-backability.
In contrast to rebound effects, these may last a long time. Tardive dyskinesia, tardive akathisia and tardive dysthymia and are all stress effects
These are effects of a treatment that endure after its stops. What endures is not something new that develops in the course of treatment as with stress effects or after the treatment is finished as happens with rebound effects. It begins from the start and is linked to the main effect of the drug.
- After stopping bisphosphonates, bones remain denser than before for 5 years. As it is not a natural density, this is a mixed effect.
- After stopping an oral contraceptive, periods may remain infrequent for months.
- After stopping an antidepressant, sexual dysfunction may persist indefinitely.
It was once assumed that when a patient improved on treatment the improvement was caused by the treatment. But it is now recognized that improvement may stem from the natural history of the underlying disorder, or from a series of “hygienic” interventions that are part of good clinical care or from patient expectations rather than from the drug effect. As a result, new drugs are only thought to work if their effect is greater than that of placebo.
Another set of trials, drug withdrawal trials in controls and patients are now needed to distinguish between Medicine Induced Stress Syndromes and the supposed effects of treatment on an underlying condition.
We need these trials because doctors and regulators have a marked bias to seeing problems that emerge on stopping medicines as evidence that the medicines were really working all the time. But if a patient gets worse when treatment stops, to assume that this change is due to a re-emergence of the original condition without controlled trials is like assuming that all the benefits of treatment stem from specific drug effects.
Until we undertake — or rather publish — such trials, any worsening of a clinical condition on stopping or reducing the dose of a drug should be assumed to be linked to a stress syndrome unless otherwise proven.
In the earliest studies of the SSRIs and related antidepressants, 66% of healthy volunteers exposed to these drugs for 1–2 weeks had clear withdrawal problems, including depression, anxiety and completed suicide. But none of these data have ever been published.
At present we place the burden of proof on patients to prove the drug is causing the problem rather than on companies and researchers to show its not.
We all hope that Medicines will be Magic Bullets, but they are more like a shotgun than a rifle. Even when we hit the target, close to 90% of the shot probably hits other things and if so much shot misses MISS is almost inevitable unless a course of treatment is as short as the 5 days of an antibiotic.
A stress syndrome should be suspected when what was perceived as an acute and self-limiting illness requiring a time-limited course of treatment gradually becomes perceived as a chronic disorder requiring long-term treatment. This has been a pattern observed for many conditions from asthma to ulcers and more recently gastro-esophageal reflux disease (GERD).
It is found with the use of beta-agonists, as well as steroids, proton-pump inhibitors, dopamine agonists for Parkinson’s disease and other drugs in addition to antidepressants or antipsychotics.
MISS is a win-win for companies. It increases the demand for drugs by converting acute disorders into chronic conditions, and by creating new disease categories to be treated with the drug that has caused the problem in the first case.
Studies to investigate Stress Syndromes will have to be more creative than usual. If we don’t know the nature of the withdrawal effects in advance, trials cannot rely upon highly-focused and specified questionnaires — they will have explore all bodily systems in an open-ended way and follow up any hints of changes.
Our hope is that reporting to RxISK.org will provide leads for these studies, and strategies for managing the symptoms on stopping generated by a wide range of medicines.
Aldhous P (2010). Prescription: sobriety. New Scientist, 9th Jan, pp 40–43.
American Psychiatric Association (1990) Task force on Benzodiazepine in dependence, Benzodiazepine Dependence, Toxicity and Abuse. APA Washington DC.
Battegay R (1966). Drug dependence as a criterion for differentiation of psychotropic drugs. Comprehensive Psychiatry 7, 501–509.
Bury M, Gabe J (1991). Tranquillisers and health care in crisis. Social Science and Medicine 32, 449–454.
Coupland NJ, Bell CJ, Potokar JP (1996). Serotonin reuptake inhibitor withdrawal. Journal of Clinical Psychopharmacology 16, 356–362.
Fava G (1995). Holding on: depression, sensitization by antidepressant drugs, and the prodigal experts. Psychotherapy & Psychosomatics 64, 57–61.
Fitzgerald K, Healy D (1995). Dystonias and Dyskinesias of the jaw following the use of SSRIs. Human Psychopharmacology.
Gaskell D (1998). Drugs against drugs. Chemistry in Britain 34; 12, 27–32.
Gilbert PL, Harris J, McAdams LA, Jeste DV (1995). Neuroleptic withdrawal in schizophrenic patients”. Archives of General Psychiatry 52: 173–188. With commentaries by R Baldessarini and A Viguera, W Carpenter and C Taminga, J Greden and R Tandon, H Meltzer, K Nuechterlein et al, R Wyatt.
Healy D (2002). The Creation of Psychopharmacology. Harvard U Press, Cambridge Ma.
Healy D, Tranter R (1999). Pharmacologic Stress Diathesis Syndromes. J Psychopharmacology 13, 287–299.
Healy D, Mangin D, Mintzes B (2010). The ethics of randomized placebo controlled trials of antidepressants with pregnant women. Internat J of Risk and Safety in Medicine.
Hollister L (1998). From Hypertension to Psychopharmacology: A serendipitous career. In Healy D. The Psychopharmacologists Vol 2, Chapman & Hall, London pp 215–235.
Kramer J C, Kline D F, Fink M (1961). Withdrawal Symptoms Following Discontinuation of Imipramine Therapy. Am J Psychiatry 118, 549–550.
Mann AM, MacPherson AS (1959). Clinical experience with imipramine (G22355) in the treatment of depression. Can Psychiatric Assoc J 4, 38–47.
Nutt DJ (1996). Addiction, brain mechanisms and their treatment implications. Lancet 346, 31–36.
Temple R (2004). Comments at Feb. 2, 2004 Psychopharmacological drugs Advisory Committee meeting on pediatric antidepressant suicidality:
Tohen M, Calabrese JR, Sachs G et al (2006). Randomized, placebo-controlled trial of olanzapine as maintenance therapy in patients with bipolar I disorder responding to acute treatment with olanzapine. Am J Psychiatry 163, 247–256.
Tranter R, Healy D (1998). Neuroleptic Discontinuation Syndromes. Journal of Psychopharmacology.
Whitaker R (2010). Anatomy of an Epidemic. Crown Publishers. New York.
World Health Organisation (1965). Expert Committee on Dependence Producing Drugs, 14th Report, WHO Techn. Rep. cir. 312.
WHO Expert Committee on Drug Dependence, 28th Report, WHO Technical Report Series No 836, World Health Organisation, Geneva 1993.
WHO (1992) The International Classification of Diseases, ICD10 World Health Organisation Geneva.