Psychology 101: Intro to Psychology

How can you measure the effectiveness of various psychological treatments?

If you're sick with a sore throat, it's fairly easy to tell if a particular treatment has helped: your throat stops hurting. If you take antibiotics and you feel better, there is little debate about whether a treatment was effective or not. Even with something more complex, like cancer, there are concrete measures that can be used to determine whether a treatment is working; a tumor might shrink, or the cancer might stop spreading. But with psychological disorders, such clear measures are often unavailable. In some cases, improvement might be obvious. For example, if a woman who was unemployed due to severe depression was able to get a job after attending therapy sessions for a while, therapy could be judged as effective in her case, at least in relation to one symptom of her illness. In other cases, even determining whether someone has a mental illness in the first place is difficult; you can imagine, then, how difficult it can be in some cases to tell if a treatment is actually working.

There are three main ways in which treatment effectiveness is measured: the patient's own impression of wellness, the therapist's impression, and some controlled research studies. Note that in the realm of physical medicine, the first two methods would be largely irrelevant; a doctor wouldn't simply offer an opinion on whether the patient was better, then she'd show real concrete evidence of it. But in the realm of psychology, these methods can be valuable, though flawed, tools for evaluating treatment effectiveness.

Obviously if a patient feels better, that's great. So in one sense, a patient's impressions are extremely important--the goal of therapy is, after all, to restore her to mental and emotional well-being. But for the purposes of determining which treatments are most effective in which situations, there are several problems with a patient's own impressions of her progress. The first is simply that people in distress tend to get better. This is known as regression to the mean, or average, and it's when people have a tendency to move toward an average level of functioning or happiness from whatever state they are in. If you're really happy, you're most likely to get sadder, and if you're really sad, you're most likely to get happier. People spend most of their time feeling average, so moods that are above or below average are likely to return to this average. Since people usually enter treatment because they're feeling especially bad, they're likely to get better over time not because of anything the therapist is doing, but simply because they're regressing to the mean.

Secondly, people in therapy expect to get better. You might have heard of the placebo effect. For a quick summary, imagine you have a headache: what do you usually do to fix it? Let's say you take aspirin or some other painkiller, and usually the headache goes away. You expect that when you take aspirin, your head will feel better. Now let's say your friend, playing a practical joke on you, secretly replaces your aspirin with similar-looking sugar pills. The placebo effect predicts that when you take these sugar pills for your headache, you will still feel at least a little better because you expect to; some of the effect of the painkiller is that you think you'll get better when you take one. This is especially true of therapy's effect on mental illness, since the symptoms are often much more subjective than a headache to begin with. If patients expect to get better, they probably will, at least in some ways. On a related note, patients sometimes feel like they should be getting better as a way of justifying the effort involved in seeking treatment; going to see the therapist, paying money for sessions or for drugs. It would be difficult to put in this kind of effort while thinking it was all worthless; therefore, some patients decide to believe in treatment.

Therapists' evaluations of patients are subject to all of the same problems as patients' evaluations. They, too, may mistake regression to the mean for positive effects of treatment. They also rely on patient testimony to a certain extent, which may be influenced by the placebo effect or by the justification of effort. Additionally, when therapy is finished, the therapist may continue to hear from patients who are doing well, but not from patients who are dissatisfied; because therapists are reminded only of the positive results, they overestimate how often their patients have positive results. This is known as the availability heuristic.

A non-subjective way to figure out which treatments are best for which disorders is to conduct studies comparing treatments to each other or to placebos--like the sugar pill, though you can probably imagine that the therapy equivalent of 'sugar pill' might be a little more complicated. These kinds of studies have shown that for depression and panic disorders, cognitive therapy is most effective, potentially because these disorders are in part caused by the kind of negative thinking directly addressed by cognitive therapy. For obsessive-compulsive disorderand behavioral therapymedication works the best to combat the compulsive behaviors and anxiety that characterize these disorders. For specific phobias, systematic desensitization really does help patients face their fears.

Studies also isolated certain characteristics of the patient and the therapist that affect how well any of these treatments will work. The patients most likely to get better tend to be motivated to improve and have personalities that lead them to attack problems actively. They are also likely to have strong support from family and friends. Patients least likely to get better tend to think negatively and behave hostilely. For reasons therapists don't thoroughly understand, personality disorders and psychotic disorders, like schizophrenia, tend to have lower rates of recovery in general.

Regardless of the strategy they use, therapists who are warm and empathetic tend to have the highest rates of success with their patients. On the other hand, therapists who behave inappropriately can hinder therapeutic progress, or even do more harm than good. Therapists who act with prejudice, or without understanding of cultural differences between them and their patients, can end up making the patient distrustful of the therapist and of therapy in general. Those who, in a Freudian model, try to produce false memories of past trauma can end up setting a patient back in recovery. Finally, it should be obvious that a sexual relationship between a patient and therapist could be harmful to recovery; still, it happens, and is a serious ethical violation.

Therapy can only be effective if patients participate; many feel that there is a stigma associated with people who see therapists, or that therapy is just too expensive. In general, women are more likely to seek help than men. Cultural differences may lead some people to seek more informal advice from friends and family than to see a therapist. In the case of a disorder like schizophrenia, which usually requires medication to successfully manage, cultural barriers can be disruptive to therapy. Additionally, for other disorders, informal support networks may work well.

Assessing treatment effectiveness is important for developing standards and giving therapists some guidance about what to try with their patients. While patient and therapist reports are important--after all, improvement that is 'subjective' or the result of a placebo is still improvement--they are not as useful as controlled research studies for determining which treatments work best for certain disorders. Regardless of method, therapists who are warm and encouraging and patients who have large support networks tend to have the most success.