Some functions in the body occur automatically, outside of conscious control (such as heart rate and blood pressure). Biofeedback is a method of making those “involuntary” processes something you can do at will.
The basic method is quite simple. In biofeedback, a machine gives you direct information regarding the bodily process in question (the “feedback” part of the term “biofeedback”). Given this information, you can find a way to control it, just like you can learn to wiggle your ears if you try hard enough.
For example, your blood pressure might be displayed on a screen. Blood pressure naturally goes up and down from time to time. When it goes down, you’ll notice that and feel pleased; when it goes up, you’ll feel displeased. Pleasure and displeasure act like the reward and punishment technique used for training animals. When a rat in a maze is rewarded with food for going the right way and given an electric shock for going the wrong way, it will soon learn to go the right way. Similarly, the unconscious parts of the nervous system figure out a way to get a "reward" instead of receive "punishment." In the case just described, this means reducing blood pressure.
The display screen provides the feedback because normally we can’t detect our own blood pressure. Using a machine to provide that information allows the person to achieve conscious control. This process generally works, at least to a modest extent. After a number of sessions, most people reach a place where they can lower their blood pressure simply by thinking, “I want my blood pressure to fall.” They don’t know how they're doing it (any more than ear wigglers know how they’ve accomplished that); nonetheless, they can cause the desired effect.
In addition to measuring blood pressure and heart rate, there are biofeedback machines in fairly common use that measure muscle tension, skin temperature, skin resistance to electricity, and brain wave activity.
What Is the Scientific Evidence for Biofeedback?
Although many studies have evaluated biofeedback, most of them suffer from inadequate design.
Only one form of study can truly prove that a treatment is effective: the
double-blind, placebo-controlled trial
. (For more information on why such studies are so crucial, see
Why Does This Database Rely on Double-blind Studies?
However, it is somewhat tricky to fit biofeedback into a study design of this type. The main problem is finding a placebo for biofeedback treatment. In the best-designed studies of biofeedback, people in the placebo group practice biofeedback with a machine that produces carefully garbled information. Study participants in this group believe they are practicing biofeedback, but in fact they are not learning any conscious control over the body process in question.
Many biofeedback studies do not use placebo biofeedback; they compare biofeedback to no treatment. Studies of this type cannot provide reliable evidence about the efficacy of a treatment. If a benefit is seen, there is no way to determine whether biofeedback caused it or whether it was caused generically by attention. (Attention alone will almost always produce some reported benefit.)
Other trials used intentionally neutral therapies, such as the use of a home diary. These are better than studies with a no-treatment control group. However, when the placebo is so different in form than the treatment under study, any apparent differences in outcome could simply represent differences in the power of suggestion in each approach.
Still other studies simply involved giving people biofeedback and seeing whether they improved. Such trials are almost completely meaningless; numerous studies have shown that both participants and examining physicians will frequently think they observe improvement in people given a treatment, regardless of whether the treatment does anything on its own. For example, early studies of biofeedback for stroke rehabilitation that did not use blinding or a control group reported miraculous successes equivalent to the “throw down your crutches and walk” cliché.
However, when controlled trials were performed, it turned out that biofeedback did not provide much more than marginal benefit, if any. For this reason, we do not even report uncontrolled studies below.
Given these caveats, the following is a summary of what science knows about the medical benefits of biofeedback.
Possible Effects of Biofeedback
Of all the medical conditions for which biofeedback has been advocated, the best studied is
. However, a review of the literature published in 2003, which found 22 controlled trials of acceptable quality,
concluded that real biofeedback is not more effective than fake biofeedback. A study published subsequent to this review did report benefits, but it was poorly designed.
In addition, a 2010 review of 36 trials involving 1,660 patients found no consistent evidence that biofeedback's effectiveness for hypertension compared to other behavioral therapies, drug therapy, placebo, or no intervention.
Prehypertension is a condition where blood pressure is elevated above what is considered normal, but
not the level of a hypertension diagnosis. A small randomized trial involving 43 patients with prehypertension showed that heart rate variability feedback reduced systolic blood pressure compared to slow abdominal breathing and control groups.
Biofeedback has been studied for other medical conditions as well. At least one controlled study supports the use of biofeedback for each of the following:
chronic low-back pain,
female stress incontinence,
incontinence associated with radical prostatectomy,
and, possibly, rehabilitation from
Note that the evidence of benefit with biofeedback is not definitive for any of these conditions, and in many cases there are also studies with negative outcomes.
There is mixed evidence for the effectiveness of biofeedback for recurrent headaches, both
In a detailed review of multiple controlled studies, however, researchers concluded that biofeedback is useful for tension headaches, particularly when combined with other relaxation therapies.
And, in another review of 94 studies, researchers concluded that biofeedback is capable of significantly reducing the frequency for both migraine and tension-type headaches, among other benefits.
However, it is important to note that not all of the studies they used to arrive at this conclusion were randomized, placebo-controlled trials.
The balance of the evidence suggests that biofeedback is not effective for
and no more than marginally effective for
Evidence is mixed regarding biofeedback’s effectiveness for
and fecal incontinence.
However, a number of small trials have found biofeedback to be more effective than standard therapies alone for constipation related to pelvic floor dysfunction.