Think back to Christine and the dead mouse at the beginning of the chapter. What if, instead of exposure and response prevention, Christine had been assigned psychoanalysis or psychosurgery? Could these alternatives have been just as effective (and justified) for treating her OCD? Through this chapter, we have explored various psychological and biological treatments that may help people with psychological disorders. But do these treatments actually work, and which ones work better than the others?

As you learned in the Methods in Psychology chapter, pinning down a specific cause for an effect can be a difficult detective exercise. The detection is made even more difficult because people may approach treatment evaluation very unscientifically, often by simply noticing an improvement (or no improvement, or even a worsening of symptoms) and reaching a conclusion based on that sole observation. Determination of a treatment’s effectiveness can be misdirected by illusions that can only be overcome by careful, scientific evaluation.

657

Imagine you are sick and the doctor says, “Take a pill.” You follow the doctor’s orders, and you get better. To what do you attribute your improvement? If you are like most people, you reach the conclusion that the pill cured you. That is one possible explanation, but there are at least three others: Maybe you would have gotten better anyway; maybe the pill was not the active ingredient in your cure; or maybe after you are better, you mistakenly remember having been more ill than you really were. These possibilities point to three potential illusions of treatment: illusions produced by natural improvement, by placebo effects, and by reconstructive memory.

Natural improvement is the tendency of symptoms to return to their mean or average level. The illusion in this case happens when you conclude mistakenly that a treatment has made you better when you would have gotten better anyway. People typically turn to therapy or medication when their symptoms are at their worst. When this is the case, the client’s symptoms will often improve regardless of whether there was any treatment at all; when you are at rock bottom, there is nowhere to move but up. In most cases, for example, depression that becomes severe enough to make individuals candidates for treatment will tend to lift in several months no matter what they do. A person who enters therapy for depression may develop an illusion that the therapy works because the therapy coincides with the typical course of the illness and the person’s natural return to health. How can we know if change was caused by the treatment or by natural improvement? As discussed in the Methods in Psychology chapter, we could do an experiment in which we assign half of the people who are depressed to receive treatment and the other half to receive no treatment, and then monitor them over time to see if the ones who got treatment actually show greater improvement. This is precisely how researchers test out different interventions, as described in more detail below.

Recovery could be produced by nonspecific treatment effects that are not related to the specific mechanisms by which treatment is supposed to be working. For example, the doctor prescribing the medication might simply be a pleasant and hopeful individual who gives the client a sense of hope or optimism that things will improve. Client and doctor alike might attribute the client’s improvement to the effects of medication on the brain, whereas the true active ingredient was the warm relationship with the doctor or an improved outlook on life.

Simply knowing that you are getting a treatment can be a nonspecific treatment effect. These instances include the positive influences that can be produced by a placebo, an inert substance or procedure that has been applied with the expectation that a healing response will be produced. For example, if you take a sugar pill that does not contain any painkiller for a headache thinking it is Tylenol or aspirin, this pill is a placebo. Placebos can have profound effects in the case of psychological treatments. Research shows that a large percentage of individuals with anxiety, depression, and other emotional and medical problems experience significant improvement after a placebo treatment.

One recent study compared the decrease in symptoms of depression seen in 718 patients randomly assigned to receive either antidepressant medication or pill placebo (Fournier et al., 2010). Participants receiving medication showed a dramatic decrease in symptoms over the course of treatment. However, so did those taking placebo. Closer examination of the data revealed that for those with mild or moderate depression, placebo is just as effective as antidepressant medication at decreasing a person’s symptoms, and it is only in instances of severe depression that antidepressants seem to work better than placebo (see FIGURE 16.7).

658

A third treatment illusion can come about when the client’s motivation to get well causes errors in reconstructive memory for the original symptoms. You might think that you have improved because of a treatment when in fact you are simply misremembering: mistakenly believing that your symptoms before treatment were worse than they actually were. This tendency was first observed in research examining the effectiveness of a study skills class (Conway & Ross, 1984). Some students who wanted to take the class were enrolled, but others were randomly assigned to a waiting list until the class could be offered again. When their study abilities were measured afterward, those students who took the class were no better at studying than their wait-listed counterparts. However, those who took the class said that they had improved. How could this be? Those participants recalled their study skills before the class as being worse than they actually had been. This motivated reconstruction of the past was dubbed by Conway and Ross (1984), “Getting What You Want by Revising What You Had.” A client who forms a strong expectation of success in therapy might conclude later that even a useless treatment had worked wonders by recalling past symptoms and troubles as worse than they were and thereby making the treatment seem effective.

How can we make sure that we are using treatments that actually work and not wasting time with procedures that may be useless or even harmful? Research psychologists use the approaches covered in the Methods in Psychology chapter to create experiments that test whether treatments are effective for the different mental disorders described in the previous chapter.

Treatment outcome studies are designed to evaluate whether a particular treatment works, often in relation to some other treatment or a control condition. For example, to study the outcome of treatment for depression, researchers might compare the self-reported symptoms of two groups of people who were initially depressed: those who received treatment for 6 weeks and a control group that had also been selected for the study but were assigned to a waiting list for later treatment and were simply tested 6 weeks after their selection. The outcome study could determine whether this treatment had any benefit.

659

Researchers use a range of methods to ensure that any observed effects are not due to the treatment illusions described earlier. For example, the treatment illusions caused by natural improvement and reconstructive memory happen when people compare their symptoms before treatment to their symptoms after treatment. To avoid this, a treatment (or experimental) group and a control group need to be randomly assigned to each condition and then compared at the end of treatment. That way, natural improvement or motivated reconstructive memory cannot cause illusions of effective treatment.

But what should happen to the control group during the treatment? If they simply stay home waiting until they can get treatment later (a wait-list control group), they will not receive the placebo effects. So, ideally, a treatment should be assessed in a double-blind experiment, a study in which both the participant and the researcher/therapist are uninformed about which treatment the participant is receiving. In the case of drug studies, this is not hard to arrange because active drugs and placebos can be made to look alike to both the participants and the researchers during the study. Keeping both participants and researchers “in the dark” is much harder in the study of psychological treatments; in fact, in most cases it is not possible. Both the participant and the therapist can easily notice the differences in treatments such as psychoanalysis and behaviour therapy, for example, so there is no way to keep the beliefs and expectations of both participant and therapist completely out of the picture in evaluating psychotherapy effectiveness. Nevertheless, by comparing treatments either to no treatment or to other active interventions (such as other psychological treatments or medications), researchers can determine which treatments work, and which are most effective for different disorders.

The German-born British psychologist Hans Eysenck (1916–1997) reviewed the relatively few studies of psychotherapy effectiveness available in 1957 and raised a furor among therapists by concluding that psychotherapy—particularly psychoanalysis—not only was ineffective but seemed to impede recovery (Eysenck, 1957). Much larger numbers of studies have been examined statistically since then, and they support a more optimistic conclusion: The typical psychotherapy client is better off than three quarters of untreated individuals (Seligman, 1995; Smith, Glass, & Miller, 1980). Although critiques of psychotherapy continue to point out weaknesses in how clients are tested, diagnosed, and treated (Baker, McFall, & Shoham, 2009; Dawes, 1994), strong evidence generally supports the effectiveness of many treatments (Nathan & Gorman, 2007), including psychodynamic therapy (Shedler, 2010). The key question then becomes: Which treatments are effective for which problems (Hunsley & Di Giulio, 2002)?

One of the most enduring debates in clinical psychology concerns how the various psychotherapies compare to one another. Some psychologists have argued for years that evidence supports the conclusion that most psychotherapies work about equally well. In this view, common factors shared by all forms of psychotherapy, such as contact with and empathy from a professional, contribute to change (Luborsky et al., 2002; Luborsky & Singer, 1975). In contrast, others have argued that there are important differences among therapies and that certain treatments are more effective than others, especially for treating particular types of problems (Beutler, 2002; Hunsley & Di Giulio, 2002). How can we make sense of these differing perspectives?

660

In 1995, the American Psychological Association (APA) published one of the first attempts to define criteria for determining whether a particular type of psychotherapy is effective for a particular problem (Task Force on Promotion and Dissemination of Psychological Procedures, 1995). The Canadian Psychological Association (CPA) followed, generally endorsing the APA guidelines, and elaborating upon them in a Canadian context (Task Force on Empirically Supported Treatments, 1998). The official criteria for empirically validated treatments in the APA report defined two levels of empirical support: well-established treatments, those with a high level of support (e.g., evidence from several randomized controlled trials), and probably efficacious treatments, those with preliminary support. After these criteria were established, a list of empirically supported treatments was published by the APA (Chambless et al., 1998; Woody & Sanderson, 1998). A recent review of such treatments highlighted several specific psychological treatments that have been shown to work as well as, or even better than, other available treatments, including medication (Barlow et al., 2013). TABLE 16.3 lists several of these treatments.

Some have questioned whether treatments shown to work in well-controlled studies conducted at university clinics will work in the real world. For instance, some have noted that most treatment studies reported in the literature do not have large numbers of participants who are of ethnic minority status, and so it is unclear if these treatments will work with ethnically and culturally diverse groups. One recent, comprehensive review of all available data suggests that although there are gaps in the literature, where there are data available, results suggest that current evidence-based psychological treatments work as well with ethnic minority clients as with white clients (Miranda et al., 2005).

Even trickier than the question of establishing whether a treatment works is whether a psychotherapy or medication might actually do harm. The dangers of drug treatment should be clear to anyone who has read an American magazine advertisement for a drug and studied the fine print with its list of side effects, potential drug interactions, and complications. Many drugs used for psychological treatment may be addictive, creating long-term dependency with serious withdrawal symptoms. The strongest critics of drug treatment claim that drugs do no more than trade one unwanted symptom for another: depression for sexual disinterest, anxiety for intoxication, or agitation for lethargy and dulled emotion (e.g., see Breggin, 2000).

The dangers of psychotherapy are more subtle, but one is clear enough in some cases that there is actually a name for it. Iatrogenic illness is a disorder or symptom that occurs as a result of a medical or psychotherapeutic treatment itself (e.g., Boisvert & Faust, 2002). Such an illness might arise, for example, when a psychotherapist becomes convinced that a client has a disorder that in fact the client does not have. As a result, the therapist works to help the client accept that diagnosis and participate in psychotherapy to treat that disorder. Being treated for a disorder can, under certain conditions, make a person show signs of that very disorder—and so an iatrogenic illness is born.

There are cases of clients who have been influenced through hypnosis and repeated suggestions in therapy to believe that they have dissociative identity disorder (even coming to express multiple personalities) or to believe that they were subjected to traumatic events as a child and “recover” memories of such events when investigation reveals no evidence for these problems prior to therapy (Acocella, 1999; McNally, 2003; Ofshe & Watters, 1994). There are people who have entered therapy with a vague sense that something odd has happened to them and who emerge after hypnosis or other imagination-enhancing techniques with the conviction that their therapist’s theory was right: They were abducted by space aliens (Clancy, 2005). Needless to say, a therapy that leads clients to develop such bizarre beliefs is doing more harm than good.

661

Just as psychologists have created lists of treatments that work, they also have begun to establish lists of treatments that harm. The purpose of doing so is to inform other researchers, clinicians, and the public which treatments they should avoid. Many are under the impression that although every psychological treatment may not be effective, some treatment is better than no treatment. However, it turns out that a number of interventions intended to help alleviate people’s symptoms actually make them worse! Did your high school have a DARE (Drug Abuse and Resistance Education) program? Have you heard of Scared Straight, and boot-camp programs for young offenders? They all sound like they might work, but careful scientific experiments have determined that people who participate in these interventions are actually worse off after doing so (see TABLE 16.4) (Lilienfeld, 2007; Ontario Ministry of Children and Youth Services, 2010)!

To regulate the potentially powerful influence of therapies, psychologists hold themselves to a set of ethical standards for the treatment of people with mental disorders (American Psychological Association, 2002; Canadian Psychological Association, 2000). Adherence to these standards is required for membership in the American and Canadian Psychological Associations, and provincial and territorial regulatory bodies also monitor adherence to ethical principles in therapy. These ethical standards in Canada are based on four principles: (1) respect for the dignity of persons, which includes protecting the welfare of vulnerable/people and seeking fairness in treatment and avoiding biases; (2) responsible caring, which includes striving to benefit clients and taking care to do no harm; (3) integrity in relationships, by promoting accuracy, honesty, and truthfulness; and (4) responsibility to society, with means that when psychological knowledge is used in the development of social policies, it will only be used for beneficial purposes (i.e., those that reflect the first three principles). When people suffering from mental disorders come to psychologists for help, adhering to these guidelines is the least that psychologists can do. Ideally, in the hope of relieving this suffering, they can do much more.

662