Observations placeholder

Beyond Biofeedback – Drs Elmer and Alyce Green

Tension Headache

In the control of muscle-contraction headache, more commonly known as tension headache, the earliest and best-known studies are those of Budzynski, Stoyva, and associates at the University of Colorado (1970, 1973).

The work of two British researchers in the late 1950s had demonstrated, through the use of a then-novel EMG integrative circuit, that the resting levels of the muscles of the forehead are higher in tension-headache patients than in normals. Since the usual immediate cause of pain in tension headache is sustained contraction of the muscles of scalp and neck, Budzynski and Stoyva hypothesized that if patients could learn to relax these muscles the pain would be alleviated.

Their early studies indicated that individuals can learn to lower the tension level of the forehead muscle, the frontalis, through EMG biofeedback training, and that the relaxation tends to generalize to other muscle groups, particularly in the head and neck.

A pilot study with five tension-headache patients using EMG feedback from the frontalis muscle resulted in a reduction of headache frequency and severity.

The researchers decided to do a controlled study (1973) to rule out the possibility that the positive results were mainly due to placebo effects or the effects of suggestion. Eighteen patients who had suffered severe headaches for from six to nine years were selected for the controlled study: two men and sixteen women, ranging in age from twenty-two to forty-four. A base-line headache level was established for each patient, and pre-training EMG levels were assessed in two no-feedback sessions. The patients were then randomly divided into three groups.

• Group A patients were given EMG biofeedback training, receiving auditory feedback from the frontalis in the form of clicks. The click rate was proportional to forehead tension: Increased tension produced a faster rate and reduced tension produced a slower rate.
• Group B received what Budzynski et al. describe as "pseudo-feedback." Muscle-tension signals produced by Group A patients were tape-recorded and played back to Group B patients.
• Members of Group C received no training, but were asked to continue keeping headache charts.

Instructions to Group A included an explanation of tension headache; a statement of the goal of the study (to learn to relax so tension would no longer cause headaches); an explanation of the clicks that would be used to give information about the level of tension in their foreheads; and suggestions for using the information to achieve relaxation.

Instruction to Group B was the same except the patients were not told that the clicks they would hear would give them information about their tension. They were told only that keeping their attention focused on the clicks heard through the earphones would help to keep out intruding thoughts, and so help them relax. Clearly, Group B did not get biofeedback, nor in actuality can it be described as "false feedback" or "pseudo-feedback"; the clicks were used simply as an attention-focusing device. Biofeedback by definition is the feedback of (one's own) biological information. Group B patients, told simply to listen to the clicks, were not told that the clicks reflected (fed back) tension levels in their foreheads. This was because "they could easily have determined that this was not true" (emphasis added).

This statement by the researchers raises an interesting point.

We believe it is not possible to design a satisfactory "false-feedback group" in biofeedback research. Subjects can be fooled for a short period of time but not for a long-term training period. This also raises a question of ethics on two counts.

• First, to indicate to a patient that you are feeding back information about his or her body process when you are not is to lie to that person.
• Secondly, biofeedback is not like a sugar pill; it is not a "nothing." If a person is attempting to change a physiological function and the information which is fed back to him or her about that function is unrelated to what is really happening, this will lead to confusion, dissatisfaction with the process, and-if the subject suspects the truth-distrust.

To return to the study, Groups A and B received sixteen sessions of training. They were also asked to practice relaxation outside the laboratory twice a day for fifteen to twenty minutes each time, in the same manner as in the lab but without instruments.

Persons in Group C (the no-treatment group) were told their training would begin in two months.

Training sessions were followed by a three-month period during which Group A and B patients kept charts of their daily headache activity. At the end of the three-month follow-up they returned for three no-feedback sessions to test their ability to produce low EMG levels.

The mean frontalis EMG level for the A group was 3.92 microvolts and for the B group 8.43 microvolts (on a time-integrated scale).

Drug usage had declined dramatically in all Group A patients in the three-month follow-up period. Decreases in medication were reported by only two Group B patients.

At the eighteen-month follow-up only four of the six Group A patients could be contacted. Three who had shown significant decline in headache activity during training had also maintained a very low level. The fourth patient had not shown a significant reduction during training but nevertheless reported a reduced level of headache activity.

 Only one member of Group B reported a significant decline in number and severity of headaches.

Eight members of the B and C control groups accepted the delayed offer to receive EMG training. Overall results indicated a 75 percent decline in headache activity.

Budzynski et al. make an interesting observation, based on verbal reports, that patients passed through several stages in their ability to use a trained relaxation response to reduce headache.

• In Stage 1 the patients were unable to prevent or abort headaches;
• in Stage 2 they became more aware of tension preceding a headache and could relax to some degree, but could not abort the headache;
• in Stage 3 patients increased their awareness of tension, were more able to relax, and could abort light-to-moderate headaches;
• in Stage 4 patients could relax automatically in the face of stress. This ability became a habit, resulting in a changed life style, in which headache activity was greatly reduced or eliminated.

Among many others who have used EMG feedback in the treatment of tension headache are Wickramasekera (1972) and Gladman and Estrada (1974).