Harm Reduction: Defining It and Doing it Properly in Order to Reduce Harm


The term "harm reduction" is controversial and has been applied to a variety of different public health initiatives, treatment approaches, and law enforcement and public policy strategies relating to the addictions. Dr. Rosenthal maintained that present day harm reduction primarily focuses on harm to drug users and largely ignores harm to the rest of society. Dr. Tapert presented findings from a randomized clinical trial of an intervention designed to reduce harm from "binge drinking" and alcohol abuse among youth. Dr. Willenbring presented harm reduction as an alternative clinical approach that involves continuing work with patients who are unlikely to achieve permanent abstinence.

Symposium Chair: Roger D. Weiss, M.D., McLean Hospital, Belmont, MA

Harm to Whom?

Mitchell S. Rosenthal, MD, Phoenix House Foundation, New York, NY

As it is now generally understood, "harm reduction," when applied to drug abuse, refers to efforts that seek to reduce the dam-age done to addicts by both their use of illicit drugs and society’s prohibition of drug use. But concern for the damage done by addicts to the rest of society does not figure prominently in harm reduction rhetoric. The harm reduction position is characterized by the assumptions that legal prohibitions are more harmful than drugs and that widespread drug use is inevitable. It embodies the convictions that present substance abuse policies are fatally flawed; that addiction treatment rarely achieves abstinence; and that "controlled use" is the norm for drug users. It promotes policies and programs designed to diminish drug-related harm with-out requiring drug abstinence and fosters the perception of sub-stance abusers as "victims" of a condition they have little ability to control.

But is widespread use of drugs really inevitable? About 80 million Americans have tried drugs, but only about 13 million are involved in any kind of illicit use today. That’s just 6 percent of Americans over 12—down from twice that many two decades ago. Yet, while overall drug use has plunged, the hard core population—the most harmed and harmful drug abusers—has diminished hardly at all. So, the cost of drug abuse remains high, and it is not likely to be cut by liberalizing drug policies or accommodating drug use. There is ample proof, however, that these costs can be substantially lowered when drug use is reduced by restrictive policies consistently enforced, as has been successfully demonstrated in societies as different as those in Japan and in Sweden.

What about the notion that we can diminish drug related harm without requiring abstinence? In terms of treatment, this translates into what harm reduction advocates call "meeting drug users where they are," and there is nothing wrong with this, as long as we do not leave them where they are. Treatment must involve the expectation of change to rightly be called treatment. Substance abuse does not demand palliative care. It is behavior that is preventable and a condition from which recovery is not only possible, but largely predictable.

Harm reduction philosophy holds that we cannot force anyone to quit. But, quite often, we can, when we compel addicts to enter treatment—and those who are most dysfunctional and into denial rarely enter treatment any other way. Since drug abusers must play an active role in their own recovery, the first task of treatment is to overcome denial and generate the motivation necessary for success. Fortunately, there is treatment available today cap-able of accomplishing this, and research has repeatedly shown that drug abusers who enter treatment under duress are no less successful than those who enter voluntarily.

There is no argument that drug use does harm to drug abusers (e.g., overdose, HIV infection, tuberculosis, malnutrition, loss of in-come, disgrace, and imprisonment). But, to what degree is a society obliged to protect drug abusers from the consequences of their actions—and at what cost? When does society become the enabler and victim of addiction? Does not responsible public policy require us to do whatever reduces the most harm and protects the greatest number—whether they are victims of drug abuse, drug abusers, or drug abuse policy? To me, this would mean diminishing the overall impact of drug abuse in the most productive way possible—primarily by curtailing or reducing drug abuse among high risk, high cost drug abusers. Thus, we reduce child abuse, child poverty, and the number of at-risk children. We cut crime, welfare dependency, and the costs of health care and all the services and benefits subsidizing addiction.

Treatment is the most effective means of reducing the size of the hard core population—treatment that is supportive enough, demanding enough, and sufficiently prolonged to alter behavior and affect changes in brain function, restoring the addict’s control over impulsive and irrational behavior. This is treatment that will not allow addicts to remain at their own level, and it can rightly be called "harm reduction" because it will, in fact, reduce harm. But it will also involve prohibition as a policy and coercion as a means of keeping drug abusers in appropriate treatment. It will embrace the concept of individual responsibility and recognize the capacity of men and women for positive change and growth. It will promote abstinence as one aspect of a broader goal—the return of former substance abusers to society with both the desire and the ability to lead productive, drug-free lives.

Not much is to be lost by abandoning the present approach to harm reduction. Needle exchange has already won broad acceptance and is credited with reducing HIV infection among IV-heroin addicts, although there are other and more modest means of making syringes available to the injecting public. Methadone maintenance is well established as a valid mainstream intervention and has proven most effective when methadone is provided in appropriate doses with adequate counseling and other services. Whatever else constitutes harm reduction seems to amount to simply "giving up" on addicts, while sending the disturbing message of "responsible use" to youth, and pursuing policies that would shield drug abusers from injuries to themselves by exacerbating the harm they do to their families, communities, and society as a whole.


Harm Reduction for Binge Drinking in Young Adults: Four-Year Follow-up Results From a Randomized Clinical Trial

G. Alan Marlatt, PhD, University of Washington (Presented by Susan Tapert, PhD, University of California, San Diego)

A definition of harm reduction used by Alan Marlatt1 at the University of Washington is "a collection of policies and programs that are designed to reduce the harm to both self and others that are associated with addictive behaviors." Harm reduction principles include: 1) a public health alternative to some of the criminal and disease models of dealing with addictive behaviors, 2) a practical and humane approach to addiction as opposed to a more idealized approach where people will no longer have addiction problems, 3) a "low threshold" access to care, 4) abstinence as an ideal outcome, but not the only one, 5) harm reduced by degrees along a continuum, and 6) integration of user feedback.

The key to harm reduction methods is to reduce some of the consequences of drug use. Behavior change only happens if there is a negative consequence. However, some negative consequences can involve permanent and life-long changes. Specific methods to reduce some of the consequences of drugs include changing the route of administration of drugs; suggesting safer substances; and reducing the frequency, intensity, and duration of use. These methods move people toward abstinence and cessation of the harmful behavior—with the primary goal of reducing permanent problems and harm to other people.

Before describing a clinical trial at the college level, background on youth alcohol use must be understood. Adolescent and college alcohol use often involves binge drinking—drinking with the goal of getting drunk. Some risks for young people from binge drinking are problems in school, unplanned or unprotected sex, and physical injury—a significant public health problem. Some studies show that about half of college students participating binge drank at least one time in the previous two weeks. The question is: "Is this a normal part of adolescent and young adult development or are these young people going to develop long term problems that need preventing?" It is important to know the best policies and pro-grams for young people. Making it illegal to drink under 21 hasn’t been entirely effective. Factors that influence youth drinking include parent’s drinking behavior, parental monitoring, peers, perspectives on how much peers drink, availability, and price.

The following describes a randomized controlled study looking at a brief harm reduction intervention for college students. It is cal-led the "Life Styles Project" and was con-ducted at the University of Washington by Alan Marlatt, Ph.D. This study evaluated a brief individually administered motivational enhancement intervention, designed to reduce the harmful consequences of drinking. Subjects were youth entering the Univ. of Washington, age 19 or under in fall of 1990. They were followed over four years.

The study identified 350 high-risk youth who drank quite a bit in high school, may have shown signs of conduct disorder, and may have had a family history of alcohol abuse. One hundred youth were selected randomly for a normative comparison sample group. The high-risk group was randomized into an intervention and a control group. The intervention included personalized feedback on drinking compared to the average, discussions about drinking to build a cognitive dissonance or disparity between their current level of drinking and where they would like to be. This intervention had a small effect during early follow-up, but increased over time. Since the intervention involved only one face-to-face interview, it is quite remarkable how the effect persisted over the college years.

In summary, drinking behaviors change throughout college for everyone. Youth that are low-risk when they enter college are very unlikely to develop alcohol problems in college, but for high-risk youth the quantity and frequency becomes more stable and consequences for most decline over time. The high-risk youth who received the intervention were more likely to resolve problem drinking and have fewer problems at follow-up, and the favorable effects persisted over time. The intervention was cost-effective.

The clinical aspects of this intervention were based on motivational interviewing techniques developed by Miller and Rollnick.2 This strategy is very compatible with the philosophy of harm reduction. It involves ex-pressing empathy, developing discrepancies between the person’s behavior and where they would like to be, avoiding arguments, rolling with the resistance, and supporting self-efficacy by building on success. Steps devised to help with change include: Feedback, Responsibility, Advice, Menu of Options, Empathy, Self-Efficacy (FRAMES). In addition, the Stages of Change Model by Prochaska and DiClemente3 fits very well with motivational interviewing. These stages include precontemplation, contemplation, determination or planning, action, maintenance, and relapse.

In summary, cost-effective harm reduction strategies can produce long-term benefits in young adult drinkers. Additionally, motivational interviewing techniques work well in populations that may initially have little interest in changing behavior.


  1. Marlatt GA, Baer JS, Larimer M. Preventing alcohol abuse in college students: A harm-reduction approach. In GM Boyd, J Howard, RA Zucker (Eds). Alcohol Problems Among Adolescents: Current Directions in Prevention Research. Hillsdale, NJ: Erlbaum Publishers, 1995, p. 172.
  2. Miller WR, Rollnick S. Motivational Interviewing: Preparing People to Change Addictive Behavior. New York, NY: Guilford, 1991.
  3. Prochaska JO, DiClemente CD. Stages and processes of self-change of smoking: Toward an integrative model of change. J Consult Clin Psychol. 1983;51:390-395.


Practical Harm Reduction in the Clinical Setting

Mark L. Willenbring, MD, Minneapolis VA Medical Center, Minneapolis, MN

I would respectfully disagree with Dr. Rosenthal concerning the current national drug policy. In my opinion, the war on drugs is a destructive element in society and a dismal failure. Unfortunately, the rhetoric around the term "harm reduction" as national policy has gotten in the way of doing the right thing clinically. The definition of harm reduction (HR) used here is a drug policy option that seeks to minimize the collective harm to society from both drug use and drug policy. It is distinguished from the war on drugs which seeks to reduce harm from drug use, with less regard to the harm of policy (such as excessive and un-just imprisonment of African-Americans). Another term for HR is the public health model. It is a pragmatic strategy based on the assumption that drug use will be with us for a long time and will not be eliminated completely. Therefore, it focuses on attainable results.

How is this applied to treatment? What does this have to do with clinical care? Clinically, HR is a pragmatic policy that is based on the idea that we cannot cure or eliminate the disorder. We have to learn to manage it. In this way, HR clinically applied, is a new word for palliative care, care provided for incurable conditions (not only end-of-life care). Most medical and mental health care is in fact palliative. The term used in the new Veterans Administration (VA) practice guideline for the management of substance use disorders is "care management". Based on what Dr. Rosenthal said, I believe that although he and I disagree about national drug policy, we would probably do very similar things for patients.

Extant treatment models are not helpful in this regard, however. These models of treatment emphasize the need for total and permanent abstinence from substance use. Anything short of that is a failure. Furthermore, any treatment outside of specialized units has been considered inadequate or worse, enabling. Clients often have very severe and complex problems for which currently available pro-grams either do not work, will not accept these clients or the clients simply refuse referral. If we act, we are accused of enabling, doing too much without results; if we do not act, we are accused of neglect, or doing too little. This has had an inhibiting effect on clinicians. It is not enabling to address the addiction and en-courage movement toward abstinence. There will always be care for chronic addiction so why not do it more compassionately and efficiently.

Care Management of Chronic Addictions (CMCA) is a model that accepts chronicity, recognizes limits of treatment methods, is palliative (non-curative) in nature, stresses long term management and treats addiction like other chronic diseases such as bipolar disorder or diabetes. This is a model that addresses "extreme" substance dependence. Factors to consider when deciding who should have CMCA include past treatment history, motivation level and goals, ability to participate in rehabilitation, available programming to match patient needs, and prognosis of coexisting conditions. CMCA is indicated when either the patient refuses referral to rehab, but seeks med/psych care; or has a serious comorbidity precluding participation in rehab; or has repeated engagement in rehab with minimal response. Very broadly, any professional with common sense skills, without specialized training, can do CMCA.

The goals of CMCA include engaging the patient, coordinating care, reducing suffering, treating complications, improving motivation to change, inducing remission whenever possible, preventing or limiting relapse, slowing the rate of deterioration or extending meaningful life, reducing use of expensive, and ineffective services.

The principles of CMCA incorporate supportive, engaging approaches; documenting substance use systematically at every visit; ad-dressing substance use or relapse risk at each visit; providing motivational support; and coordinating care (case management). This is an active treatment where an ideal goal is abstinence. In many people this is an unrealistic proximal goal, but CMCA tries to move people in that direction. Other principles involve treating complications and co-existing disorders if possible; addressing co-existing social problems (housing, income support, socialization); engaging and using social net-works, treating nicotine dependence, and pro-viding opiate agonist therapy if appropriate.

Some patients refuse treatment except when in crisis. If so, then provide episodic care, crisis intervention, case management, and continue attempts to engage the patient. Involuntary treatment or asset management may be needed in extreme circumstances. Periodically it is also necessary to reassess goals and refer to rehab when goals change.

Does it work? One study showed that a group receiving a one-hour intervention by a nurse was more likely to be improved a year later. Another studied a sample of middle- aged Swedish men who had been drinking heavily. A nurse provided a monthly intervention and urged quitting or cutting down. Over the next two to five years the intervention group showed less than half the sick days/ person/year and hospital days were about one third less than the control group. They also reduced mortality significantly in the intervention group. Additionally, a recent study at the Minneapolis VAMC of the application of CMCA to medically ill alcoholics, showed significant reductions in drinking and improved engagement for CMCA patients com-pared to a control group. Survival was 82% vs. 68% (intervention to control) after two years in a first quasi-experimental study and 81% vs. 70% in a second study. Another study showed a gradual reduction in alcohol and drug problems of about 50% over 36 months in a group of patients with serious and persistent mental illness.

To summarize, care management is a pragmatic approach to treating substance use disorders as a chronic illness. Any professional in any setting can apply it and the evidence to date shows a better long-term prognosis than is commonly appreciated.


2000 Proceedings

2000 AAAP Annual Meeting Proceedings Copyright 2001 AAAP