Alcoholics Anonymous (AA), 12-Step Movement, and Minnesota Model

Describes the nature of Alcoholics Anonymous (AA), other 12-Step programmes, and the Minnesota Model, how they developed, and the key assumptions that underlie their approach. (1,320 words)

Read More ➔

Stopping Heroin Use Without Treatment

Research by Patrick Biernacki reveals important insights into how people recover from heroin addiction. It also illustrates the major challenges that people with a heroin addiction face on their journey to recovery (2,283 words). 

Read More ➔

Factors That Facilitate Addiction Recovery

Describes the nature of addiction recovery and the factors that facilitate the process, using the Stories and quotes from our Storytellers and research conducted by members of the Wired In team. (12,698 words)

Read More ➔

The Drug Experience: Heroin, Part 9

People who have been addicted to heroin report experiencing cravings for the drug long after they have given up using. Many people who have relapsed and gone back to using the drug after a period of abstinence attribute their relapse to their cravings for the drug.

A craving for heroin is used to describe a strong desire or need to take the drug.  Craving is often brought about by the appearance of a cue that is associated with the past drug use. These may be cues associated with the withdrawal from heroin, or with the pleasurable effects of the drug.

Wikler has claimed that the relapse of abstaining heroin addicts can be attributed to conditioned withdrawal sickness. People who have stopped using heroin will crave the drug if they are exposed to certain stimuli that they have learned, as result of their past experiences with withdrawal sickness, to associate with actual acute withdrawal.

Thus, people returning to an area where they have previously used the drug, may experience symptoms of withdrawal, and as a result of these feelings and the accompanying discomfort, they begin to think about the drug again, obtain it, and then use.

Lindesmith has postulated that people who have used heroin to prevent the onset of withdrawal symptoms, learn to generalise withdrawal distress and come to use the drug in response to all forms of stress. When they become abstinent, they experience stress as a craving to use the addictive drug once again.

Despite these ideas, Biernarki reported that only a small number of people in his sample described their cravings as being linked to withdrawal distress. Though they sometimes reported that problematic life situations during abstinence led to thoughts about the drug, they did not report any specific symptoms of withdrawal.

The feelings of the cravings were commonly described as emanating from associations made in past experiences of using heroin and feeling the drug’s effects. The cravings were ‘experienced and interpreted as akin to a low-grade ‘high’. The person feels a ‘rush’ through the body and by feelings of nausea located in the stomach or throat, and he thinks about enhancing the feeling by using the addictive drug.’ Both the ‘rush’ and nausea are sometimes experienced when actually taking the drug.

This kind of craving was of short duration, generally 15-20 minutes, and rarely longer than an hour.  The frequency with which these cravings occurred diminished over time and generally appeared rarely, if at all, after about a year.

Biernacki pointed out that the cravings could be managed in two basic ways, that can be employed individually or together: drug substitution and a rethinking of their lives.

As described in our last Briefing, the initial step in breaking away from heroin use—to minimise temptations to use—commonly entails a literal or symbolic move away from the drug scene. However, this move does not preclude the possibility that the person will experience drug-related cues, since some may be noticed in any environment. Moreover, it does not necessarily help the person to manage the cravings once they do occur.

The first strategy used to overcome heroin cravings is simply to substitute some other non-opiate drug. The most popular substitutes in the Beirnacki study were marijuana, alcohol and tranquillisers such as valium. Whilst some of the sample subsequently developed serious problems with alcohol, most who adopted this strategy used other non-opiate drugs only on an occasional basis.

A second strategy used to manage cravings involved a ‘subjective and behavioural process of negative contexting and supplanting.’ Thus, when people experienced heroin cravings, they ‘reinterpreted their thoughts about using drugs by placing them in a negative context and supplanted them by thinking and doing other things.’

Biernacki emphasised that this is not just a mental process (e.g. the power of positive thinking), but it entailed subjective and social elements. ‘The substance for the negative contexting and supplanting of the drug cravings is provided by the new relationships, identities, and corresponding perspectives of the abstaining individuals.

To illustrate the above, some people who overcame their dependence on heroin became very health conscious and concerned about their physical well-being. When they experienced heroin cravings, they may place the thoughts about using the drug in a negative context by thinking about a physical illness that can arise from injecting the drug, e.g. hepatitis.

Then they may replace the thoughts of using the drug by thinking of the personal benefits that can be gained from some physical activity, such as cycling. The substance for these alternative thoughts comes from the social world of participatory sports. The person may then go cycling and the feeling aspect of the craving can be masked by the physical exertion or can be reinterpreted as an indication of exertion.

Biernacki provided examples, of other former users who became religious converts or who engaged in political activity. He emphasised that, ‘An effort such as this must be made each time the cravings appear, until the power of various cues to evoke the cravings diminishes and the cravings are redefined as the ex-addict becomes more thoroughly involved in social worlds that are not related to the use of addictive drugs.’

Recommended Reading:

Patrick Biernacki (1986) Pathways from heroin addiction: Recovery without treatment. Temple University Press, US.

> pdf document

> Part 10

Quitting Use of Cocaine

For a period of four years from 15th November 2004, I wrote a series of Background Briefings for Drink and Drugs News (DDN), the leading UK magazine focused on drug and alcohol treatment. I am slowly uploading these briefings on Recovery Stories as I have time.

Three of these briefings are focused on cocaine—The Drug Experience: Cocaine. In the first part, I explore the dynamic world of heavy cocaine use as revealed in a provocative, high-quality study by Dan Waldorf and colleagues. This research, conducted in the US in the 1980s, challenged many of the prevailing myths about cocaine.

In the second part, I point out that while cocaine is generally portrayed as having a very high addiction potential, the majority of people who use the drug do not have a problem. The research by Dan Waldorf and colleagues reveals a number of social and social psychological factors that influence how a person uses a drug.

Read More ➔

Alcohol Dependence

Here is an article I first wrote as a Background Briefing for Drink and Drugs News (DDN), the leading UK magazine focused on drug and alcohol treatment, in February 2005.

‘There has been a considerable scientific effort over the past four decades in to identifying and understanding the core features of alcohol and drug dependence. This work really began in 1976 when the British psychiatrist Griffith Edwards and his American colleague Milton M. Gross collaborated to produce a formulation of what had previously been understood as ‘alcoholism’ – the alcohol dependence syndrome.

The alcohol dependence syndrome was seen as a cluster of seven elements that concur. It was argued that not all elements may be present in every case, but the picture is sufficiently regular and coherent to permit clinical recognition.

Read More ➔

The Drug Experience: Cocaine, Part 3

Dan Waldorf and colleagues were ‘pleasantly surprised’ by the relative ease with which so many cocaine users managed to quit. Their research emphasises the importance of one’s personal and social identity in influencing drug use. (895 words)


In the last two Briefings, we focused on the most comprehensive ethnographic study of heavy cocaine users, conducted by Dan Waldorf and colleagues in Northern California. They interviewed 267 current and former heavy users of cocaine, a sample that did not include people in treatment programmes or in prison. Most of the respondents were ‘solidly working- or middle-class, fairly well-educated, and steadily employed.’

This research challenged many of the prevailing myths. In the present Briefing, we look at the process of giving up use of cocaine. Waldorf and colleagues interviewed 106 quitters—30 of these had received some form of treatment, whilst 76 stopped using cocaine without treatment.

When respondents were given a list of personal reasons for quitting, the most common (47% of sample) was given as health problems. The next most cited reasons were financial problems (41%), work problems (36%) and pressure from spouse and/or lover (36%). Only 7% cited actual arrest, although 28% cited fear of arrest as a reason for quitting.

Respondents were also given an open-ended summary question on the most important reason or reasons to quit. A total of 61% mentioned some form of psychological problem or stressful state caused by cocaine as the most important reason to quit. The next most common reasons were financial problems (23%), and severe or recurrent health problems or concerns (19%).

There was great diversity in actions that respondents took to quit using cocaine. Some made a number of attempts to stop before they actually succeeded. They despaired over the hold the drug had over them and had great difficulty in maintaining a resolve to stop using.

However, over a half of the sample stopped using on their first try, although this was not always easy. Two-thirds of the untreated cases stopped on their first attempt, whilst only one in five of treated cases did so.

More than 40% of all quitters reported making some sort of geographic move as part of their successful attempt to quit. Two-thirds of these people said they moved to another city or state, at least in part to help them stay away from cocaine.

The most frequently used strategies for stopping to use cocaine were social avoidance strategies. Nearly two-thirds of the quitters said they had stopped going to places where cocaine was being used, or had made conscious efforts to avoid seeing cocaine-using friends. Over 40% had also sought out new friends who did not use cocaine.

More than 75% of the sample became more concerned about their physical health whilst quitting, and acted upon these concerns. Two-thirds improved their eating habits, and a half undertook new programmes of physical conditioning.

Over half of the quitters sought out new interests, with 39% participating in sports to help them avoid using cocaine. Similarly, 55% of the sample used informal help, such as family or friends, to stop using cocaine.

Only 17% of the sample started using other drugs after quitting cocaine. Of those that did, the majority used only marijuana, which almost all had used before and during their cocaine use. Whilst 21% drank more alcohol, most drank less after giving up cocaine.

Most of this diverse sample had used cocaine heavily for a good number of years—but few were ever merely cocaine abusers. Moreover, their use had not led them to becoming stigmatised. The majority worked regularly, maintained homes, and were responsible citizens:

‘… a commitment to their everyday lives gave them a stake in normalcy and bonded them to the conventional world.’

The sample were different to heroin addicts in other studies, many of whom came from disadvantaged backgrounds, had been criminalised and stigmatised, and had few private resources (e.g. education, jobs).

For many of the present sample, prolonged use of cocaine stopped being fun and started disrupting, rather than enhancing, everyday lives. Since these lives had meaning and value, the difficulties caused by cocaine became powerful spurs for cessation.

The researchers were ‘pleasantly surprised’ by the relative ease with which so many cocaine users managed to quit. Their strategies were in general fairly common-sensical social avoidance strategies, designed simply to put distance between themselves and the drug.

Most of the quitters were able to manage the cravings they experienced after stopping cocaine use. They realised that cravings were only transitory—distractions caused them to subside. New interests and activities provided such distractions. Many quitters found cravings:

‘… little different from yearnings one might feel for an old lover – one feels the desire, but with time it subsides and one thinks of him or her less and less.’

These findings emphasise the importance of one’s personal and social identity in influencing drug use. A commitment to a conventional identity and everyday life helps form the social-psychological and social-organizational context within which control and cessation of drug use is possible.

It is commonly stated that drugs come to dominate identities and lives. This was true in the most problematic cases in the Waldorf study.  However, for the bulk of the sample, identities and lives usually dominated drug use. This is a critical fact that must be remembered when we try to help people overcome problems caused by drugs and alcohol.

Recommended reading:

Cocaine Changes: The Experience of Using and Quitting by Dan Waldorf, Craig Reinarman and Sheigla Murphy. Temple University Press, USA.

> pdf document

Factors Facilitating Recovery: Overcoming Withdrawal Symptoms

People who decide to stop taking drugs or drinking alcohol after using or drinking for long periods of time, need to be aware that they might experience withdrawal effects which can be irritating, debilitating and even life-threatening.

Many of these withdrawal signs, which can be psychological and physical in nature, are generally opposite to the effects the person experienced when the drug was being taken. For example, abrupt withdrawal from long-term use of Valium (diazepam) and other benzodiazepines, drugs which are prescribed to alleviate anxiety and insomnia, can lead to pronounced anxiety, insomnia, agitation, intrusive thoughts and panic attacks.

In addition, people withdrawing from benzodiazepines can experience physical withdrawal signs, such as burning sensations, feeling of electric shocks, and full-blown seizures. The duration and strength of these withdrawal signs is in part dependent on the amounts of drug having been used and the duration of time the person has been using the drug. 

Read More ➔

Pathways from Heroin Addiction: Recovery Without Treatment, Part 3

I continue my series of blog posts on Patrick Biernacki’s research from the mid-1980s focused on natural recovery from heroin addiction.

People who have been addicted to heroin report experiencing cravings for the drug long after they have given up using. Many people who have relapsed and gone back to using the drug after a period of abstinence attribute their relapse to their cravings for the drug.

A craving for heroin is used to describe a strong desire or need to take the drug.  Craving is often brought about by the appearance of a cue that is associated with the past drug use. These may be cues associated with the withdrawal from heroin, or with the pleasurable effects of the drug.

Read More ➔

Alcohol Dependence

Looks at the cluster of seven elements that make up the template for which the degree of alcohol dependence is judged. (900 words)

Read More ➔

Addiction and trust: Marc Lewis at TEDxRadboudU 2013

A former drug addict himself, Lewis now researches addiction. In order to get over ones addiction, he explains, self-trust is necessary.

Unfortunately, self-trust is extremely difficult for an addict to achieve. There are two factors that make it so difficult to get over an addiction: lack of self-control and an inability to put off reward. An addict wants his fix and he wants it now, despite the risk of losing out on a happier, healthier future.

The way to build self-trust, Lewis explained, and get over an addiction is for the addict to begin an internal dialogue with his future self to convince his present self that it can, in fact, live without its addiction.

Read More ➔

‘Talking to the Dalai Lama about addiction’ by Marc Lewis

rsz_dalai_lamaCan you imagine it, talking to the Dalai Lama about addiction? Well, it happened to some lucky people in the field recently, including Marc Lewis. I suggest that you read Marc’s blog directly, since you can see some photos and comments.

‘I got back yesterday around noon. What a relief it was to be home! India is overwhelming in so many ways, with poverty and raw need topping the list. To get back to this calm, orderly place was a reprieve and a pleasure, tinged with guilt at leaving all that suffering behind.

For anyone just tuning in now, I was at a week-long “dialogue” with the Dalai Lama on the theme of “Desire, craving, and addiction.” I was one of eight presenters, each of whom gave a talk to His Holiness (as he is called) and to the surrounding experts, monks, movie stars and what have you. All the talks are posted here. My talk is here. I want to tell you about two of the talks I found most fascinating and most relevant for people struggling with addiction.

Read More ➔

Brad’s Moment(s) of Clarity

stories-04Here’s the second of our Moment of Clarity series, taken from Brad’s Recovery Story

‘At this time, I thought willpower is what I needed to stop drinking, but I soon found out that this wasn’t the case. I was lacking a true willingness and desire to get well. I daydreamed and dreamt about stopping drinking, but I think that’s all it was at that stage. There was no real consideration of the work that would be involved in stopping.

Anyway, I decided I needed a break from the booze. I retired to bed and began going through the terror of a full-blown rattle, something I hope I never have to go through again. Five days later, I was physically dry. I then decided to see how long I could abstain from alcohol. After six weeks of no alcohol, I still wanted a drink. In fact, my desire for alcohol was worse than ever.

Read More ➔