IT WAS 1997. I had just been appointed director of a large, newly created addiction medicine service in the Swedish capital city of Stockholm. We had a staff of about four hundred, close to a hundred inpatient beds, an emergency room of our own in the city center, and clinics scattered across several locations in the most drug-infested part of the region. There was just one problem. Although we were affiliated with the famous Karolinska Institute, we had no idea what we were doing.
The patients were mostly poor, severely addicted, and socially marginalized people. Many were homeless or, as I would soon learn, in various stages of a homelessness progression that is not as clear as people with real homes tend to think. Sometimes patients would show up in our emergency room with severe withdrawal, delirium tremens, seizures, or overdoses. On other occasions it would be malnutrition, dehydration, or a kind of general medical derangement that is hard to pin down to a specific diagnostic code yet is probably a more important clinical finding than many well-defined conditions. And then there was the steady stream of people who spoke of being depressed and said they contemplated suicide. Sometimes, it was clear, they faked it to gain admission, when the night got too cold and the shelters were full. Could we really blame them? On other occasions they truly meant it right there and then but were fine after a night in a bed, a meal, and some talking. But I also knew that these were patients whose rates of completed suicide were about five times higher than those of the general population.
Events had thrown me into the job far too young and utterly unprepared. I had arrived at the Karolinska only two years earlier, fresh out of my psychiatry residency. Psychiatry and addiction had always fascinated me, but my research training was in basic neuropharmacology, in which I had obtained a doctorate and then spent two crazy, wonderful years as a postdoctoral fellow with George Koob at the Scripps Research Institute in La Jolla, California. When I was recruited to the Karolinska, it was to continue my basic science work while slowly building a small clinical research unit. I expected to be learning on the job and initially felt insecure and intimidated. A year into it, my anxieties had begun to ease. Having originally trained in a charming, medieval university town with a population of about 100,000 and little if any serious crime, I had been concerned that my clinical training might leave me unprepared to deal with the severe drug, alcohol, and crime problems of the big city. But the patients I had been seeing had been reasonably socially stable, with medical problems that for the most part seemed manageable. I had been able to start building the research program just like I had planned, train staff, obtain research grants, teach, and settle into a comfortable existence in academic medicine.
And then the hospital system decided to reshuffle everything. Several services were put together into fewer, larger departments in a way that seemed to serve only two purposes: to disguise large cutbacks in services for an already underserved patient category, and to give some bureaucrats power. I tried to work within the system, even as my best nurses one after another abandoned ship, but in the end I found that I couldn’t. I recall sitting in my office late one afternoon, after the last patient appointment, leafing through a medical journal and seeing an ad for a director’s job on the other side of town. I remember leaving it there, with the page open, for several days before the next peak of frustration prompted me to write the application letter, mostly as a way to release the frustration. George Koob, who not only had remained a great mentor but also become a good friend, came by for a talk at the Karolinska just as I was thinking whether I should accept the job offer; he told me, in his characteristically blunt way, “Markus, there is one biologically undeniable truth. It is always better to be in control than not.”
So I held on to the best thing the old hospital had offered, which was a beautiful, warmhearted nurse who was soon to become my wife, left everything else, and took the job—only to discover that rather than being in control, I was confronted with an entirely new and previously unimaginable level of feeling powerless. Yes, I could make many more decisions about what treatments to provide and how. But what were the right decisions to make? I was completely, utterly out of my depth.
The fundamental problem was this: We could—and did—debate which patients should be admitted, or how long a detox should be. We could argue over whether detox medications should be used,
1 and if so, which ones, at what doses, and for how long. After that, we could do all in our power to engage social workers and get them to help with discharge planning and organizing funding for treatment at a residential facility for twenty-eight days or maybe longer. This last step was viewed as the ultimate success, harder and harder to achieve as public finances worsened. But it took just a small step back from the day-to-day management, and a look with a slightly dispassionate eye, to see that the hard work of our staff made little significant dent in the disease process of addiction, which meanwhile was killing our patients. This is not to negate the value of a bed and a meal for a homeless, hungry addict. It is not to understate the power of a kind word or someone being ready to listen and take a person’s anguish seriously, even when it is expressed in the middle of the night, through intoxication, and by someone who does not necessarily smell nice. But as a physician, I wanted to make a difference. To take the Hippocratic oath in reverse, I wanted to comfort always, treat often, and preferably also cure sometimes.
It was clear to me that at best we were providing comfort. And I saw that even that was frequently challenging for our staff. Every relapse was taken as a sign either that the patient was not motivated or that we were powerless to change the course of addiction. Over the years, as they saw little of their efforts pay off, some treatment providers quit for less frustrating jobs, while others became hardened, jaded, and seemingly uncaring. We clearly had to change our ways or would continue to burn through our most valuable asset: our staff’s compassion and desire to help.
I realized that I was immersed in something quite different from other areas of medicine I had experience with, such as my second love, endocrinology. Evidence-based medicine had by then transformed clinical practice almost everywhere. The latest advances might not be implemented as fast as they should be anywhere, but a young physician could for the most part assume that clinical practice was based on data from controlled studies that showed measurable benefits for patients from specific clinical interventions. Here, I realized, it was quite possible that we were using considerable resources without helping patients with outcomes that mattered. The bitter divisions between different schools of thought alone were enough to show that we couldn’t all be practicing based on available evidence. And there was a clear possibility that some of our ways made things worse, not better.
In what little time there was between managing staff, budgets, and patients, haggling with politicians over funding, and trying to be a good enough father for our first child, I read as much as I could. Throughout my career one simple principle has helped me more than anything else, both as a scientist and as a physician: if you don’t know what you are doing, quickly find someone who does, and emulate their approach. Every science class should start with a reading of a single sentence by Sir Isaac Newton: “If I saw further than other men, it was because I stood on the shoulders of giants.” Clinical medicine is not much different.
I still didn’t know exactly what I was doing, but after a while a couple of things seemed clear. Reading the literature was one thing, but we could not just read our way to improving our methods. That would have required me, or someone like me, to integrate the published scientific reports and tell our staff what changes in practice the weight of the evidence implied. Even if I had felt capable of doing that, which I did not, this approach was not likely to get staff onboard, enthusiastic, and past their ideologically driven convictions. To them it would just be another voice articulating a belief, among other voices competing for their attention. But perhaps we could get people out of their entrenched positions if we could do the journey together, in a literal sense. Perhaps we could go on an expedition to a place where clinical needs guided the science, where the science guided the way clinical work was done, and where we might be able to absorb a culture as much as specific knowledge.
The list of places we were able to identify seemed surprisingly short, although at the time I assumed that was because I just did not know the field well enough. I wrote two people I had never met but whose names often appeared on research papers that were as interesting and scientifically stringent as they were relevant for clinical issues of addiction medicine: Charles P. O’Brien and Tom McLellan at the University of Pennsylvania Treatment Research Center in Philadelphia. To my amazement, Chuck, as I soon learned he was called by the entire field, quickly answered, welcomed us, and had someone put together a program for us. It was unheard of to take a whole group of nurse managers on a trip abroad and pay for it out of hospital funds. But I asked that someone show me what regulations we would be breaking, and no one could come up with any. I guess this was one of those situations where being in control indeed was better than not. In Philadelphia we stayed at the cheapest possible hotel on Chestnut Street, the elevator only worked intermittently, and even some of the more hardnosed nurses turned pale when shots were heard in the neighborhood one night.
But it was all worth it: we were in for a treat. To hear Chuck tell the story of how his team discovered naltrexone, the first modern alcoholism medication, was enough to make the visit worthwhile, and I will revisit that story in some detail later. But it was Tom McLellan, in cowboy boots, getting up on the table and explaining the concept of addiction as a chronic, relapsing disease that probably made the greatest impression on our nursing staff.
So you have a patient with hypertension. You give him a beta blocker, and his blood pressure comes down. Everyone says, “Look, treatment works!” Then for some reason the patient decides to stop the medication. Within weeks, blood pressure is out of control again. Everybody says, “Didn’t we tell you—this treatment definitely works, and the patient really needs it!” Now, instead, you have a patient with alcohol addiction. The patient goes into treatment, stops drinking, and improves in every aspect of his life. Then treatment is discontinued. After a while the patient starts using alcohol again, and things fall apart. At which time family members, the patient’s employer and friends all say, “Didn’t we know it. Alcoholics really are hopeless. And treatment is a waste of money. People just relapse.” Doesn’t this all strike you as a bit odd?
The message was one to which we had not previously given much thought. Addiction is inherently a chronic, relapsing disease, not much different from diseases we already have successful management models for, such as hypertension, diabetes, or asthma. Similar to those diseases, the risk for developing addiction has a strong genetic component. Likewise, the development and course of addiction are determined by an intricate interplay among genetic risk factors, environmental influences, and behavioral choices. Addiction cannot currently be cured but can be managed with a degree of success that is sufficient to allow patients to live a good life. If not managed, on the other hand, it disables, kills, and leads to costs and suffering that are hard to fathom. In none of these aspects does addiction differ from other common, complex illnesses of a chronic relapsing nature. Yet for these conditions, long-term disease management that combines pharmacological and behavioral approaches is an undisputed norm, and success is hardly assessed by the number of people completely cured of their ailment. At the same time, we continue to debate whether addiction really is a medical condition, focus on short-term fixes such as detoxification or twenty-eight-day residential programs, see anything other than complete abstinence as a failure, and frequently view the harm caused by addiction as self-inflicted.
A couple of years after our visit to Philadelphia, in 2000, Tom McLellan, Chuck O’Brien, and several other leaders in the field of addiction research forcefully laid out this tragic paradox in a paper in a top medical journal.
2 They started by noting that the effects of addiction
3 on society have helped shape a widely held view that addiction is primarily a social problem, not a health problem. They then examined evidence that addiction is in fact indistinguishable from chronic medical illnesses. They reviewed the scientific literature and compared the role of genetic and environmental factors, disease mechanisms, and treatment outcomes such as adherence to treatment and relapse between addiction, on one hand, and type 2 diabetes mellitus, hypertension, and asthma, on the other. They found that genetic risk factors, personal choice, and environmental factors are comparably involved in the etiology and course of all these disorders. Furthermore, just like consumption of an unhealthful diet produces lasting cardiovascular changes in diabetes, heavy alcohol or drug use produces significant and lasting changes in brain chemistry and function. Despite widespread perceptions to the contrary, effective medications are available for treating several addictions, such as those for nicotine, alcohol, or opiates. Medication adherence and relapse rates are very similar in addictive disorders and the other illnesses with which the comparison was made.
A fundamental problem, the authors concluded, is that addiction generally has largely been treated as if it were an acute illness. Their review instead showed that, similar to the other chronic relapsing illnesses, long-term care strategies of medication management and continued monitoring are the ones that produce lasting benefits. Because of this, addiction should be insured, treated, and evaluated like any other chronic illness. This, they stated, stands in stark contrast with the public view that drug dependence is primarily a social problem that requires interdiction and law enforcement, rather than being a health problem that requires prevention and treatment, a view that many physicians apparently share. Most physicians still fail even to screen for alcohol or drug use during routine examinations, viewing such screening efforts as a waste of time. A survey of general practice physicians and nurses indicated that most believed no available medical or health care interventions would be appropriate or effective in treating addiction.
McClellan and his coauthors noted that 40 to 60 percent of patients treated for alcohol or other drug addiction indeed return to active substance use within a year following treatment discharge. They outlined two distinct possibilities to account for this fact, with different implications. On one hand, it is possible that these disappointing results indeed confirm the widely held belief that addiction is not a medical illness and is therefore not significantly affected by medical interventions. On other hand, maybe the problem is instead related to current treatment strategies and outcome expectations, which view addiction as a curable, acute condition. Yet if addiction is more like diabetes than a broken leg, then the appropriate standards for treatment and outcome expectations would be those applied to other chronic illnesses and would focus on reduction rather than elimination of relapse, improved function, and quality of life.
In their concluding comments, the authors wrote that
similarities in heritability, course, and particularly response to treatment raise the question of why medical treatments are not seen as appropriate or effective when applied to alcohol and drug dependence. One possibility is the way drug dependence treatments have traditionally been delivered and evaluated.
Many drug dependence treatments are delivered in a manner that is more appropriate for acute care disorders. Many patients receive detoxification only. Others are admitted to specialty treatment programs, where the goal has been to rehabilitate and discharge them as one might rehabilitate a surgical patient following a joint replacement. Outcome evaluations are typically conducted 6 to 12 months following treatment discharge. The usual outcome evaluated is whether the patient has been continuously abstinent after leaving treatment.
Imagine this same strategy applied to the treatment of hypertension. Hypertensive patients would be admitted to a 28-day hypertension rehabilitation program, where they would receive group and individual counseling regarding behavioral control of diet, exercise, and lifestyle. Very few would be prescribed medications, since the prevailing insurance restrictions would discourage maintenance medications.
4 Patients completing the program would be discharged to community resources, typically without continued medical monitoring. An evaluation of these patients 6 to 12 months following discharge would count as successes only those who had remained continuously normotensive for the entire postdischarge period.
In this regard, it is interesting that relapse among patients with diabetes, hypertension, and asthma following cessation of treatment has been considered evidence of the effectiveness of those treatments and the need to retain patients in medical monitoring. In contrast, relapse to drug or alcohol use following discharge from addiction treatment has been considered evidence of treatment failure.
The paper is a passionate, elegant, and incontrovertible argument for why addiction treatment must focus on continuing care and disease management that involve behavioral interventions as well as relapse preventive medications, and that may require interventions and monitoring throughout the patient’s lifespan. It explains why outcome expectations must reflect the chronic relapsing nature of the disease, rather than narrowly focusing on abstinence alone. It seems impossible for anyone involved in the treatment of patients with addictive disorders to read the paper without inferring that the entire treatment enterprise in this area needs to be fundamentally revamped. The paper is one of the most influential in the field of addiction in decades and has so far been cited over a thousand times by other scientists. These days I rarely review a research grant application that does not start out by stating that “addiction is a chronic, relapsing illness.” By these measures, the paper has been as successful as a university professor can ever hope for when writing an academic piece. Devoting most of a chapter to reviewing its analysis and conclusions over a decade later may seem like a waste of time.
There is just one little problem. Despite everything, more than a decade later, very little has changed in the real world of addiction treatment. But before addressing the lack of progress, I must deal with another fundamental issue.