Post-acute withdrawal syndrome is a hypothetical condition in which a set of symptoms persist long after a person has undergone withdrawal from alcohol (or from other drugs, though alcohol will be our focus here). These symptoms reputedly include, among others, cognitive, emotional, physical and sleep impairments.
A quick online search reveals many attestations of PAWS, primarily by rehabs and recovery organizations. But PAWS (sometimes known as protracted alcohol withdrawal syndrome, or protracted withdrawal syndrome) has never made it into the Diagnostic and Statistical Manual of Mental Disorders (DSM), the diagnostic bible of the American Psychiatric Association. So to what extent, if any, is it real?
The idea of it was introduced by Terence Gorski and Merlene Miller in their 1982 book Counseling for Relapse Prevention, and they expanded upon it in Staying Sober: A Guide for Relapse Prevention (1986). Staying Sober became an extremely popular workbook in the addiction treatment industry, and untold numbers of patients were introduced to the concept of PAWS as though it were a scientifically proven reality.
It’s not hard to see why PAWS gained traction, in light of certain narratives that were dominant in the 1980s and have lingered since.
Neither Gorski nor Miller was a research scientist. Both were addiction counselors, and based their description of PAWS on certain preconceived ideas about “alcoholism” and on observation.
However, it’s not hard to see why PAWS gained traction, in light of certain narratives that were dominant in the 1980s and have lingered since.
The idea that drugs cause permanent brain damage fed into the propaganda of an era notorious for its “This is your brain on drugs” PSAs. PAWS also fits in well with Alcoholics Anonymous, when AA’s Step 2 indicates that people with alcohol addiction require “a Power greater than ourselves” to “restore us to sanity.”
I well remember my introduction to PAWS in the mid-1990s, when I was made to watch one of Gorski’s films in treatment. My first reaction was, “No wonder these people are depressed and confused, since they are all being forced to attend AA!”
It should be noted that PAWS is quite different from alcohol withdrawal syndrome (AWS). Acute AWS occurs shortly after long-term, daily, heavy drinkers cease consuming alcohol, and can last between two and seven days. Its duration and severity depend on how long the person has been drinking, how much they’ve been drinking, and genetic and physiological factors. Symptoms of acute AWS include tremors, sweating, rapid or uneven heartbeat, fever, seizures, hallucinations and delirium. In severe cases, if untreated, it can cause death. No one questions its existence, and the DSM duly includes it.
PAWS, however, is another story, and people have been debating its reality ever since Gorski and Miller proposed it. The kind of observation they relied on is highly susceptible to confirmation bias: People see what they expect to see and fail to see things which don’t conform with their preconceived ideas. Scientists use blinded randomized controlled trials in order to eliminate confirmation bias.
PAWS is postulated to occur after AWS subsides—and to last for months, years or even a lifetime.
According to Gorski and Miller, there are six types of PAWS symptoms:
1. Inability to think clearly
2. Memory problems
3. Emotional overreactions or numbness
4. Sleep disturbances
5. Physical coordination problems
6. Stress sensitivity
Gorski and Miller also gave a list of 37 warning signs of “relapse” which were a result of PAWS—including depression and a refusal to go to AA. Interestingly, University of New Mexico researcher William Miller (no relation) used Gorski and Miller’s warning signs to construct the 28-item Assessment of Warning-signs for Relapse (AWARE) scale; nine of Gorski and Miller’s warning signs were rejected as invalid by William Miller when tested experimentally.
Many of the symptoms ascribed to PAWS are also pre-existing symptoms which people may self-medicate with drugs.
The AWARE scale has proved a reliable tool for predicting resumption of drug use, but of course it tells us nothing about whether withdrawal produces prolonged psychiatric symptoms.
PAWS is difficult to study because many of the symptoms ascribed to it, such as insomnia, depression or anxiety, are also pre-existing symptoms which people may self-medicate with drugs. And unless these symptoms frequently co-occur in the same individuals, we are not justified in saying that they combine to form a syndrome.
According to DSM-5, anxiety, bipolar, depressive, neurocognitive, obsessive-compulsive, psychotic, sexual and sleep disorders can all be either substance-induced disorders or independent disorders. For all of these except neurocognitive disorder, DSM-5 uses the same criteria to distinguish the substance-induced disorders from the independent ones: If the disorder persists more than a month after substance use ceases, or if it existed prior to the substance use, it is to be diagnosed as an independent disorder. Otherwise, it is to be diagnosed as a substance-induced disorder.
Quite a number of scholars have raised objections to how DSM’s categorizations involve arbitrary hairsplitting to differentiate one diagnosis from another; that one-month time limit for substance-induced disorders is one such example.
The reason DSM-5 makes an exception for neurocognitive disorder is that Wernicke-Korsakoff syndrome, sometimes referred to as “wet brain,” is almost always classified as a substance-induced neurocognitive disorder. This rare condition results from the death of brain cells due to thiamine deficiency; in Western countries, such a deficiency is most often the result of long-term, heavy drinking. Wernicke-Korsakoff syndrome is largely irreversible and causes severe impairment. However, other less severe—and reversible—cognitive impairments sometimes occur in people who have ceased heavy drinking.
A review of substance-related cognitive impairment could easily take up several articles by itself; suffice it to say that this is a topic with many controversies and disagreements among researchers. I will limit myself to outlining a well-known 1990 paper by Fein et al. This found that in a clinical sample, one-half to two-thirds of people with alcohol addiction showed cognitive impairments in the first few months of abstinence.
Fein et al. found that distractibility, mild confusion and irritability generally subsided within two weeks after the subject ceased to drink alcohol. Problems with attention and concentration, reaction time, verbal learning ability, verbal abstract reasoning and verbal short-term memory lasted two-to-eight weeks.
On the other hand, problems with nonverbal abstract reasoning, visuospatial abilities, mental flexibility and nonverbal short-term memory, they found, could last between two months and five years.
But it is important to note that clinical samples—i.e., subjects taken from hospitals or treatment facilities—almost invariably show far higher co-occurrence of disorders than do cohorts taken from the general population. I have not found any kind of reliable estimate of the prevalence of cognitive impairment in a general-population sample of people with alcohol use disorder (AUD) who quit drinking.
Meanwhile, a 1997 study by Schuckit, et al. looked at depressive disorders in 2,945 subjects with a diagnosis of alcohol dependence. This, again, was a clinical sample.
Schuckit et al. looked for depressive episodes which occurred during a three-month period of abstinence, rather than the one-month period given by DSM-5. The majority of their subjects, 58.4 percent, had no major depressive episodes, whereas 26.4 percent met criteria for substance-induced major depression, and 15.2 percent met criteria for independent major depression. According to the authors, substance-induced major depression typically goes away by itself, without treatment, after a few days or weeks of abstinence.
Interestingly, the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), which was a general-population sample, found quite different results. Using the NESARC data, Grant et al. found that among subjects with a past-year substance use disorder, 15.15 percent had co-occurring major depression, but only 0.65 percent had co-occurring substance-induced major depression. Similar results were found with anxiety disorders.
Turning to another alleged PAWS symptom, estimates of insomnia/sleep problems in clinical samples of people with AUD range from 36-91 percent. I have found no good estimates of the prevalence of insomnia/sleep problems in people with AUD in a general-population sample; however, logic would point to its being much lower.
Using the NESARC data, Brower and Perron found that 31.7 percent of people in the general population who were undergoing acute alcohol withdrawal reported short-term insomnia. However, NESARC says nothing about post-acute withdrawal insomnia. Insomnia also frequently precedes AUD; people with insomnia who use alcohol as a sleep aid may develop AUD as a result. But alcohol-induced insomnia/sleep problems are common, too, after one ceases drinking, and can last for weeks or months. Sleep aids and sleep hygiene are both important in treating insomnia and sleep problems which co-occur with AUD, whether induced or independent.
Some evidence supports alcohol withdrawal as the cause of some symptoms ascribed to PAWS. But whether these can justifiably be clumped into a syndrome is not conclusively answered.
Overall, this seems to be what we can say with certainty: Cognitive impairment following alcohol withdrawal is common in clinical samples, but we simply don’t know if this is true of people with AUD in the general population. Alcohol-induced cognitive impairment tends to resolve itself with abstinence or moderate drinking; some cognitive impairments go away in weeks, while others take months or even longer. Cognitive impairment is more common and severe in former heavy daily drinkers than in those who drank less frequently or heavily.
People with AUD also often have a co-occurring depressive disorder: perhaps one-third of those in clinical samples, and perhaps one-sixth of those in general-population samples. But in most cases, the depression was present before the AUD. In some cases, however, the depression seems to be caused by drinking, in which case it tends to go away by itself within a few weeks after quitting.
People with AUD often have co-occurring insomnia, too: both independent and alcohol-induced, in both clinical and general populations. Alcohol-induced insomnia can resolve itself after a period of abstinence.
But besides depression, insomnia and cognitive impairment, the literature suggests that other alcohol-induced psychiatric disorders are really quite rare. Although anxiety, for example, is common during acute withdrawal, it appears that this is overwhelmingly independent anxiety, rather than alcohol-induced.
So to differing degrees, some evidence supports alcohol withdrawal as the cause of some of the symptoms ascribed to PAWS. But the question of whether these disorders of sleep, mood and cognition can justifiably be clumped together to form a syndrome—rather than being viewed as independent symptoms linked only by a common cause—is not conclusively answered by existing evidence.
And although it’s beyond the scope of this article to discuss substance-induced psychiatric disorders for drugs other than alcohol, it is worth noting that a 1995 paper by Heather Ashton estimated that only about 10-15 percent of subjects withdrawn from benzodiazepines, the best-evidenced drug association, manifested PAWS.
Based on all of this, I would argue that the DSM is right to exclude PAWS. There aren’t enough grounds to establish it as a multi-symptom syndrome induced by withdrawal from alcohol (and most other drugs). And doing so would carry serious risks.
At the same time, we should not invalidate anyone’s experiences by telling them that what they are experiencing is not real, including if they ascribe it to PAWS. As always, we need to meet people “where they’re at” in helping them to deal with whatever issues they have.
PAWS has gone on to lend itself, frequently, to scare tactics. On current evidence, nobody should be hyping it.
Gorski and Miller based their idea of PAWS on the now largely debunked notion that alcohol causes permanent loss of brain function, and claimed that 75-95 percent of people who had been addicted to alcohol or other drugs were suffering from brain damage.
For Gorski and Miller, AA was the cure for everything, including PAWS—and refusing to go to AA was itself a symptom of the brain damage behind PAWS.
PAWS has gone on to lend itself, frequently, to scare tactics: Telling people that they can’t trust their own brains in order to make them subservient to an AA sponsor or to treatment.
Telling people who have just become abstinent that they are doomed to suffer from years of depression, insomnia and brain dysfunction is not only broadly untrue, but also potentially harmful. Suggestion is powerful, and may become self-fulfilling. It is better to deal with problems as they crop up than to induce them iatrogenically.
On current evidence, nobody, least of all treatment and recovery organizations, should be hyping PAWS to add to the fears of stigmatized people who use alcohol and other drugs.