Authors: Lance Dodes
In sharp contrast, a good psychiatric center provides frequent individual psychotherapy for every patient. This is administered by trained psychological professionals (rather than people recovering from psychiatric problems) and is unique to the specific issues of the patient. Good psychiatric hospitals also provide psychodynamic groups whose purpose is to explore the way people interact with others in a way that is designed to bring out the singular attitudes, concerns, and difficulties unique to each person. They are not formulaic groups organized around predetermined topics. Psychiatric centers also offer sophisticated psychological testing to better understand complex or covert psychological problems and to ferret out neurobiological issues such as learning or attention disorders.
How did standardized, education-like group therapy become the predominant mode of treatment at rehab centers, despite their having ample room and money to employ a higher standard of professional care? One clue might be to examine the credentials and expertise of the staff.
Many staff members at rehabilitation centers have extremely limited training. Although every program boasts of the presence of psychologists and psychiatrists in a consulting capacity, many of those who provide direct treatment are qualified mainly by being “in recovery.” This is not a terribly difficult credential to attain: Hazelden’s own website invites visitors to “Become an addiction counselor in as little as one year.” Training to be a clinical social worker, psychologist, or psychiatrist, by contrast, requires from three to eight years, and to be an excellent therapist takes years beyond the end of formal training.
Also noteworthy is that although many rehab programs have psychiatrists or psychologists on speed dial, these people virtually never become the primary therapists for patients, instead serving in supervisory or consultative roles. To this day there are no academic requirements for becoming a counselor or “therapist” in a drug-rehabilitation facility.
A 2007 exposé in the
Los Angeles Times
noted: “Promises and fierce rival Passages Addiction Cure Center make sweeping claims on their websites about their clinical successes and reputations, purporting to have few or no equals in the world. Addiction researchers say the boasts are virtually impossible to substantiate. In addition, Promises, Passages, and other Malibu rehab firms have identified on their websites a number of psychiatrists and other physicians as staff members, even though the centers are not licensed to provide medical care.”
9
The issue here goes deeper than the value of good training. There is considerable evidence to support the idea that counselors without professional backgrounds develop their own personal ideas about what constitutes appropriate treatment and philosophy. For example, non-professionally trained “recovering” addicts in AA, who often provide treatment for addiction in this country, tend to tell patients to do what they themselves did, since they have neither training nor experience with anything else.
Untrained counselors may do more harm than good. One study surveyed 317 staff members of hospital-based residential detoxification and rehabilitation programs, nonhospital detoxification and rehabilitation centers, outpatient and intensive outpatient drug-free clinics, methadone maintenance clinics, freestanding recovery houses, and several specialized inpatient and outpatient programs for adolescents, women, and women with children.
10
The authors found that “[i]ndividuals with more formal training tended to be less supportive of confrontation. . . . Support for the increased use of confrontation was strongest among staff members with the least formal education.” The significance of this is that confrontation is basically a nonprofessional stance, in contrast with understanding, or often even trying to understand, what drives people to behaviors they themselves wish they could stop. Too often, it is also an enactment of these untrained counselors’ frustration with patients. The authors appropriately deplored this finding, writing, “Perhaps staff rely upon confrontational approaches because they are unfamiliar with alternatives . . . beliefs about the utility of confrontation may be subject to change based on education . . . senior clinicians might be most easily enlisted to implement, and possibly help transfer, these less confrontational approaches.”
Another recent paper examined 592 treatment providers in the United States and United Kingdom and found that the belief that addiction is a disease was stronger among those who provide for-profit treatment, have stronger spiritual beliefs, and have had a past addiction problem.
11
One would hope that what treaters believe about addiction would arise not from these factors but from knowledge—just the thing professional training provides.
Of course, there is one more difference between rehab and traditional 12-step programs: money. Most rehabilitation centers are extraordinary financial enterprises, charging more in a few months than the most expensive universities charge for a full year of tuition. Even those that are legally nonprofits seem somehow to justify large monthly rates. Hazelden charges around $28,300 a month and notes that “additional services such as counseling for other issues, prescriptions, etc. are charged separately when needed.” The Betty Ford Center charges $32,000 for thirty days, not counting detoxification. Sierra Tucson starts its residential program at $39,000 for thirty days, but the price leaps to $2,300 a day ($69,000 a month) for residents in the “Medical Assessment and Stabilization Unit.” Promises Malibu’s prices range from $55,000 for a shared room and up to $90,000 a month for a private suite. Passages Malibu starts at $88,500 for a twenty-eight-day stay.
One of the principal ways that these facilities justify their price tags is with outsized claims of effectiveness. Yet, the industry regularly does not provide this data. I made a direct inquiry to Dr. A. Thomas McClellan, the chief executive officer of the Treatment Research Center that has for years done research for Betty Ford. He replied: “We have done work for them for quite a while but there is to my knowledge no follow up study—at least in the past ten years.” The response to the same inquiry put to a different rehab, Sierra Tucson, was that they had no outcome data at all. As one addiction researcher put it in the
Los Angeles Times
, “Anybody can make any claim they want and get away with it. It’s essentially an unregulated industry.”
12
McClellan told the
New York Times
, “It doesn’t really matter if you’re a movie star going to some resort by the sea or a homeless person. The system doesn’t work well for what for many people is a chronic, recurring problem.” The
New York Times
put it this way in 2008:
Very few rehabilitation programs have the evidence to show that they are effective. The resort-and-spa private clinics generally do not allow outside researchers to verify their published success rates. The publicly supported programs spend their scarce resources on patient care, not costly studies.
And the field has no standard guidelines. Each program has its own philosophy; so, for that matter, do individual counselors. No one knows which approach is best for which patient, because these programs rarely if ever track clients closely after they graduate. Even Alcoholics Anonymous, the best known of all the substance-abuse programs, does not publish data on its participants’ success rate.
13
Rehab programs thrive in this gray area.
Hazelden is a slight exception, having been far more forthcoming than many other rehab programs in describing and studying its own outcomes. On its website, Hazelden has reported that at one month following discharge, over 20 percent of patients said that they had resumed drinking; at six months, that number had risen to over 40 percent; and at one year, almost 50 percent of patients had resumed drinking.
14
Although there is no data beyond one year, the downward slope of this outcome suggests that fewer than half of these former inpatients remained abstinent after the first year. This is a troubling result. But as it turns out, even these findings have been inflated.
Hazelden’s abstinence figures are taken from a 1998 article published in the
Journal of Addictive Behaviors
.
15
The study involved people who had gone through the Hazelden rehab and then followed up at one, six, and twelve months. None of these patients were interviewed in person; instead, all were sent questionnaires by mail. If patients did not return the questionnaire, they were called on the telephone. All of the data was therefore captured from self-report or the reports of people whom patients had chosen to reply for them.
The limitations of this method are obvious: people who are not doing well often will not reveal the extent of their return to addictive behavior because of shame, an unwillingness to acknowledge that they have not succeeded in the caller’s program, or hostility to telling the truth to somebody they don’t know. Consequently, it is far easier to get false positive results from this sort of study design (people claiming to be doing better than they are) than false negatives (people claiming to be doing worse than they are). But even more problematic than the tendency of self-reports to underreport bad results is what the researchers did with those who failed to respond at all. Unlike Hazelden’s summary, the paper itself reveals that the authors ignored the critically large attrition rate in their subjects. They reported the results of only the people they could contact, and did not count those who dropped out. It is well known in survey-based research that those who drop out tend to be those who fared worst; indeed, the paper’s authors made this very point, even though they fell into this error:
Outcome figures may be considered to represent the upper limits of outcome . . . the self-administered mailed questionnaires were completed and returned by those individuals who could easily be contacted and who were willing to complete the questionnaire. If someone did not respond to the mailing, then the telephone follow-up method was initiated. These individuals may not have responded to the mailing because they did not want to report that they had used alcohol during the follow-up period. These results are corroborated by other studies showing that easy-to-contact subjects have better outcomes than do difficult-to-contact subjects. Therefore, those clients who were not contacted with either follow-up method are more likely to have poorer outcomes, as a group than those who were contacted.
16
How many people were unaccounted for? The authors again: “The outcome figures are based on 1-month, 6-month, and 1-year follow-up response rates of 79%, 76%, and 71% of the sample, respectively. . . . About one-quarter of the sample remains unaccounted for in terms of follow-up outcome data.” The authors also found that the people who didn’t return the mailed questionnaire and had to be contacted by phone showed poorer outcomes than those who did return the questionnaire, giving further support to the notion that the people most eager to respond were those with the best outcomes.
What happens to the data if everyone is included? We cannot know whether all the dropouts resumed drinking, but as all researchers (including the authors) agree, it is likely that they did worse as a group. Let’s start by assuming that all the dropouts resumed drinking. Then, using the reported percent of dropouts at each measuring point, here are the results with everyone counted:
At one month, nearly 40 percent of
patients resumed drinking. At six months, about 55 percent resumed drinking. At one year, 63 percent of patients had resumed drinking
.
These results paint an even grimmer picture than Hazelden’s presentation of the data. But, just as Hazelden’s numbers overestimate its success, these numbers may overstate its failure. So let’s recalculate with an optimistic assumption in Hazelden’s favor. Let us assume that, instead of all of the dropouts resuming drinking, they had only mildly worse outcomes, say 25 percent worse than the measured group. Then the numbers look like this:
At one month, about 27 percent of patients had resumed drinking. At six months, the number rises to about 44 percent. By one year, 51 percent resumed drinking
.
The correct numbers are probably somewhere between these two results. But even with this more optimistic reading of the data, one year after a rehabilitation treatment whose stated aim is the AA goal of abstinence, most patients had returned to drinking. Given the downward direction of the data, we can reasonably conclude that if the study were continued beyond one year, the outcome would continue to worsen. For a lifetime problem, this is a serious deficiency.
Hazelden opted to present its data another way as well, reporting findings about the percentage of days abstinent (PDA), rather than complete abstinence. This actually makes a good deal of sense from a treatment perspective, since patients can improve without being continuously abstinent. Looking at days without drinking, Hazelden’s expatients reported significant improvement at all the follow-ups within the first year, although as with continuous abstinence, the improvement declined over time. Hazelden shows a graph of these apparently excellent results on its website, though the reference for this graph is not given. Without seeing the data behind the graph, we cannot know what the PDA data mean in real terms. But we know that the numbers Hazelden used for its presentation are averages; they don’t include breakdowns of how many patients were drinking large amounts, how many were drinking medium amounts, and how many patients were abstinent. And given the way data was gathered and treated in the
Journal of Addictive Behaviors
paper just cited, we cannot know whether these are all self-reports by questionnaire or whether the dropouts in the study were counted.
However, we can look at the same PDA data from the
Journal of Addictive Behaviors
paper just examined, since PDA results were also described there. That paper found that at one year 16 percent of the non-dropout patients drank at least one day weekly. Once again, however, the authors do not account for the 420 of the original 1,083 people who dropped out of this portion of the study. If we again assume they did less well than the people who responded and drank at least one day weekly, the percentage of patients who were weekly drinkers after one year more than triples—to 49 percent. If we recalculate with the same optimistic view of the dropout group that we used before (that the dropouts’ weekly drinking was only 25 percent more than the measured group), the weekly drinking figure rises to over 18 percent. Somewhere between 18 percent and 49 percent represents a dismaying proportion of people who were drinking every week twelve months after leaving rehab. And this number, of course, is derived from data biased toward favorable self-reports. Finally, this result doesn’t include all those (the majority, as we saw above) who are drinking, even if not every single week.