Biofeedback and Self-Regulation, Vol. 7, No. 4, 1982

Self-Control Procedures in Biofeedback: A Review of Temperature Biofeedback in the Treatment of Migraine 1 Jan Johansson 2 and Lars-Gi~ran Ost University of Uppsala, Sweden

It is argued that in order to optimize the achievement of self-control and to evaluate the clinical effects of biofeedback three skills should be included in training and assessment, namely: (1) the ability of voluntary control with external feedback, (2) the ability of voluntary control without external feedback, and (3) the ability to apply the self-control skill in critical situations in everyday life. A review of the literature concerning temperature-biofeedback in the treatment of migraine headaches shows that the research from this point of view is in a rather poor state of affairs and that no definite conclusion can in fact be drawn about the degree of self-control which has been achieved and hence o f the ultimate clinical value of biofeedback.

Ever since biofeedback was introduced there has been a great optimism about the clinical value of this psychological and psychophysiological treatment method. There have even been " . . . many proponents suggesting that biofeedback is a 'cure-all' therapeutic t e c h n i q u e . . . " (Gatchel & Price, 1979). The last few years these very positive claims have to a large extent been challenged and a more modest and even pessimistic view of the clinical value of biofeedback has emerged. For example, Gatchel and Price (1979)

~This study was supported by grant no. 80/66 from the Bank of Sweden Tercentenary Foundation. 2Address all correspondence to Jan Johansson, Psychiatric Research Center, Ullerhker Hospital, S-750 17 Uppsala, Sweden. 435 0363-3586/82/1200-0435503.00/0 © 1982PlenumPublishingCorporation

436

Johansson and 0st

in summarizing reviews of different areas of clinical applications of biofeedback concluded that " . . . it is quite clear that the claims for the therapeutic efficacy for biofeedback have been grossly exaggerated and some times even wrong" (p. 231). Silver and Blanchard (1978) in their review of research comparing biofeedback to regular relaxation training concluded that "there is no consistent advantage for one form of treatment over the other with any psychophysiological disorder for which a comparison has been made" (p. 217). There have also been reported rather varied degrees of clinical effects, even when the same type of biofeedback has been used for the same kind of psychophysiological disorder, and also low (or nonexisting) correlations between the degree of change of the psychophysiological parameter studied and the clinical improvements (cf. Holroyd, Andrasik & Westbrook, 1977; Martin & Mathews, 1978)--all of which also casts doubts on specific effects of biofeedback. The purpose of this review is to show that there is a possibility that much of the research on biofeedback might be of limited use in estimating the (absolute and relative) clinical value of biofeedback, that biofeedback has not yet been fairly evaluated and that there still might be a point for optimism about the clinical value of biofeedback. The basic argument for such a rather strong claim is that in many studies, even when they are experimentally very sound, the procedures of training (when analyzed in terms of self-control) might not be optimally designed and that often very little is known about the degree of self-control which have been achieved through the training.

B I O F E E D B A C K FOR S E L F - C O N T R O L

According to Brener (1974) voluntary control of a physiological function is dependent upon the ability to discriminate changes in that parameter. Normally such an ability does not exist, and hence normally no voluntary control of physiological functions exists. Through the use of biofeedback an external feedback loop is formed and through this it becomes possible to discriminate changes in the physiological function in question, and to change it voluntarily. That this can be achieved has been shown in years of research on biofeedback, and it is, of course, in itself a very important and interesting result. In view of clinical applicability it is, however, not enough. Most of the physiological (autonomic) parameters and physiological disorders for which biofeedback has been used are mainly phasic, i.e., the degree of activation or severity is changing. In consequence then, what should be the ultimate goal of biofeedback in clinical applications is the ability of the client to discriminate critical changes (for the disorder in

Self-Control in Biofeedback

437

question) in specified physiological parameters and the ability to voluntarily control these parameters when these critical changes occur in daily life (Epstein & Blanchard, 1977). In the process of achieving these goals the ability to voluntarily change a physiological function in the presence of external feedback can only be viewed as a first (and of course necessary) step. Even if such control has been achieved we should not take for granted that self-control in the absence of external feedback also has been established (Epstein & Blanchard, 1977; Friar & Beatty, 1976). There are data which indicate that this often is not the fact, i.e., that self-regulation without external feedback ispoor even when control in the presence of external feedback is evident (cf Epstein, Hersen, & Hemphill, 1974; Epstein & Abel, 1977; Perski, 1977). Even if a subject in fact has learned to voluntarily change a physiological parameter in the desired direction without external feedback, the question remains, as regards to clinical applications, if the person is capable of applying this skill in daily life when critical changes in physiological parameters occur. In many of the physiological disorders where biofeedback has been used, these "critical" changes occur in relation to stress (tension headache, migraine, Raynaud's disease, hypertension, etc.). In consequence then, in order to be clinically useful the skills have to be applied in situations which, at the least, are very different from "the reclining-chair in the dimly lit, sound-protected, temperature-controlled laboratory" where the regular biofeedback training usually has been carried out. Possibly, to be maximally clinically effective, biofeedback has to include some kind of training in application of the self-regulation skills in daily active life, and even in stressful situations.

A REVIEW OF THE LITERATURE

It has been argued above that in order to optimize and evaluate the effects of biofeedback, three areas should be included in training and assessment: (1) the ability of voluntary control with external feedback, (2) the ability of voluntary control without external feedback, and (3) the ability to apply the self-control skills in daily life. Whether and when the clients in fact apply the self-regulation skills in order to counteract their psychophysiological disorder should also ideally be assessed at the end of and after treatment. In the following presentation group-studies concerning clinical applications of temperature-biofeedback in the treatment of migraine headaches have been reviewed. Table I shows to what extent these studies have included such training and assessment procedures of self-control skills which have been outlined above.

438

Johansson

.o

= = 0

ooo.~

0

i~ .= o~

+

ql

I I I 1 ~ 1 1 1

Iq

+ + + + + + + l

II

I l l l t l t l

++

+ + + + + + + +

,x

I ~'

I

I

I

I

I +

k

t

I

¢

"~ I t I +

I +

I +

+ +

I I ~

I ~,,.~+

I I

I + + +

• .-, 0

's

E 0

z

and Ost

Self-Control in Biofeedback

439

Biofeedback by definition consists of training in self-regulation with the aid of external feedback, and of course all the reviewed studies have included this form of training. As can be seen in Table I though, in four of these studies the degree of achieved control in the presence of feedback has not been reported. Regarding those studies the reader can thus not decide whether the clients have gained any control of finger-temperature at all. Only one study (Reading & Mohr, 1976) reported the degree of control achieved in the absence of external feedback. This means that in the other studies the possibility exists that the clients were not capable of self-regulation of finger-temperature at all when feedback changes were not presented. Furthermore, only three studies mentioned any training of self-regulation without feedback. In two of these (Mitch, McGrady & Iannone, 1976; Sargent, Waiters, & Green, 1973) this training was done at home and in the third (Blanchard, Theobald, Williams, Silver, & Brown, 1978) it was only reported that part of the treatment-sessions were devoted to "attempts at self-regulation of the physiological responses" (p. 584). Apart from the description by Mitch et al. (1976) that the subjects practiced "with the machine off and checked estimates of their temperature during the session by turning it on periodically" (p. 269), it is unclear whether any specific training procedures were instituted. Regular hometraining of self-regulation of finger-temperature is most often carried out once or twice a day for 10-30 minutes while the client rests alone at home lying in bed or sitting in a reclining chair. In all except two studies instructions seem to have been given regarding such training sessions. Three of those (Andreychuk & Skriver, 1975; Lake, Rainey, & Papsdorf, 1979; Sargent et al., 1973) reported that the clients were expected to record if and when those instructions in fact were followed, but only Lake et al. (1979) presented data from these recordings. Regarding the application of the learned skills of temperature-regulation in order to abort or avoid migraine headaches, it first of all seems that none of the 13 reviewed studies included any kind of specific training of self-control in daily life or stressful situations. Furthermore, in as many as six of the studies it is unclear whether the clients had even been instructed to apply the learned skills to counteract headaches during daily life. Only one study (Sargent et al., 1973) reported that the clients were expected to record attempts at control of headaches through self-regulation and the estimated success in doing so, but unfortunately no data from these recordings were presented. DISCUSSION The above review aimed at giving an indication of whether specific training procedures for the achievement of self-regulation skills of

440

Johansson and Ost

physiological or autonomic responses, as well as assessment of these skills, have been included in studies of clinical application of biofeedback. At least, concerning the reviewed studies on temperature biofeedback in the treatment of migraine headache, this has been done to a very small extent. Very few of the reviewed studies seem to have used any specific procedures for training of voluntary control without feedback in the laboratory, and none has included any specific training in the application of self-regulation skills outside the laboratory in everyday life. Concerning assessment of t h e achieved degree of self-control, it is evident that some studies have not even presented data on the degree of control that was achieved in the presence of feedback, and that only one presented data on the degree of control without feedback. No data have, in the reviewed studies, been presented regarding whether the clients have the ability of self-regulation in every-day life and/or stressful situations, and to what extent they apply these skills to counteract migraine headaches. It can thus be concluded that from the literature reviewed here we know almost nothing about whether the clients have learned to self-regulate finger-temperature without the aid of external feedback, whether they have the ability of increasing their finger temperature outside the laboratory, and whether/when they in fact try to raise the temperature to avoid or abort headaches. It seems then that the conclusion drawn by Epstein and Blanchard (1977) in their theoretical analysis of self-control and self-management in biofeedback still holds true: " . . . biofeedback researchers have paid little attention to specific procedures designed to produce self-control but rather have assumed self-control is a natural product of biofeedback" (p. 202). Whether these short-comings are present in research in other areas of clinical applications of biofeedback will not be discussed here. A few examples from other areas where the question of self-control has been cared for will be described. For example, Engel and co-workers have in their treatments of cardiac arrhythmias (Weiss & Engel, 1971) often introduced successively longer periods of no-feedback towards the end of treatment. The same procedure has successfully been used by Engel, Nikoomanesh & Schuster (1974) in the treatment of fecal incontinence and Cox, Freundlich, and Meyer (1975) in the treatment of tension headaches. Perski (1977) studied the degree of transfer of learned control of heartrate from periods with feedback to periods without feedback. The training procedure where the greatest transfer of effects were found consisted of introduction of periods without feedback from the very beginning of training, a successive increase in time without feedback in favor of time with feedback during every training session and "terminal feedback" (i.e., summary) of achieved results after every period without feedback.

Self-Control in Biofeedback

441

Concerning the applications of self-regulation skills in stressful situations, Budzynski (1973) has suggested that one way of preparing for this might be to have the client imagine stressful (or anxiety-arousing) situations while at the same time trying to self-regulate the physiological response with or without the aid of feedback. Furthermore, training in application of self-regulation skills might be done by cueing Short attempts of self-regulation to some high-frequency behavior (e.g., looking at the watch, sitting down) occurring during the day (Budzynski, 1973; Johnsson & Ost, Jerremalm, & Johansson, 1981). Drury, DiRisi, and Liberman (1979) have used other "generalization procedures" including engaging the subjects in active behaviors during training sessions and increasing the intervals between sessions towards the end of treatment. In summary then, given the "state of affairs" described, it is evident that the effects of biofeedback have not yet been fairly evaluated and that no firm conclusions can be drawn about the absolute and relative clinical value of biofeedback in the treatment of migraine headaches. Much research seems to be needed to refine the training procedures of biofeedback, especially procedures concerning the optimization of learning of self-control (without feedback) and the application of self-control skills outside the laboratory. REFERENCES Andreychuk, T., & Skriver, C. Hypnosis and biofeedback in the treatment of migraine headache. The International Journal of Clinical and Experimental Hypnosis, 1975, XXIII, 172-183. Attfield, M., & Peck, D. F. Temperature self-regulation and relaxation with migraine patients and normals. Behavior Research and Therapy, 1979, 17, 591-595. Blanchard, E. B., Theobald, D. E., Williams, D. A., Silver, B. V., & Brown, D. A. Temperature biofeedback in the treatment of migraine headaches. Archives of General Psychiatry, 1978, 35, 581-588. Brener, J. A general model of voluntary control applied to the phenomena of learned cardiovascular change. In: P. A. Obrist, A, M. Black, J. Brener, & L. V. DiCara (Eds,), CardiovascularPsychophysiology. Chicago: Aldine, 1974. pp. 365-391. Budzynski, T. H. Biofeedback procedures in the clinic. Seminars in Psychiatry, 1973, 5, 537547. Cox, D. J., Freundlich, A., & Meyer, R. G. Differential effectiveness of electromyographic feedback, verbal relaxation instructions, and medication placebo with tension headaches. Journal of Consulting and Clinical Psychology, 1975, 43, 892-898. Drury, R. L., DeRisi, W. J., & Liberman, R. P. Temperature biofeedback treatment for migraine headache: A controlled multiple baseline study. Headache, 1979, 5, 278-284. Engel, B. T., Nikoomanesh, P., & Schuster, M. M. Operant conditioning of rectosphincteric responses in the treatment of fecal incontinence. New England Journal of Medicine, 1974, 290, 646-649. Epstein, L. H., & Abel, G. G. An analysis of biofeedback training effects for tension headache patients. Behavior Therapy, 1977, 8, 37-47, Epstein, L. H., & Blanchard, E. B. Biofeedback, self-control, and self-management. Biofeedback and Self-Regulation, 1977, 2, 201-211.

442

Johansson and Os!

Epstein, L. H., Hersen, M., & Hemphill, D. P. Music feedback in the treatment of tension headache: An experimental case study. Journal of Behavior Therapy and Experimental Psychiatry, 1974, 4, 59-63. Friar, L. R., & Beatty, J. Migraine: Management by trained control of vasoconstriction. Journal of Consulting and Clinical Psychology, 1976, 44, 46-53. Gatchel, R. J,, & Price, K. P. (Eds.) ClinicalApplications of Biofeedback: Appraisal & Status. New York: Pergamon Press, 1979. Holroyd, K. A., Andrasik, F., & Westbrook, T. Cognitive control of tension headache. Cognitive Therapy and Research, 1977, 1, 121-133. Johansson, J., & Ost, L-G. Applied relaxation in the treatment of "cardiac neurosis": A systematic case study. Psychological Reports, 1981, 48, 463-468. Lake, A., Rainey, J., & Papsdorf, J. D. Biofeedback and rational-emotive therapy in the management of migraine headache. Journal of Applied Behavior Analysis, 1979, 12, 127-140. Largen, J. W., Mathew, R. J., Dobbins, K., Meyer, J. S., Sakai, F., & Claghorn, J. L. The effects of direction of skin temperature self-regulation on migraine activity and cerebral blood flow. Paper presented at 10th Annual Meeting of the Biofeedback Society of America, San Diego, California, Feb., 1979. Martin, P. R., & Mathews, A. M. Tension headaches: Psychophysiological investigation and treatment. Journal of Psychosomatic Research, 1978, 22, 389-399. Medina, J. L., Diamond, S., & Franklin, M. A. Biofeedback therapy for migraine. Headache, 1976, 16, 115-118. Mitch, P. S., McGrady A., & Iannone A. Autogenic feedback training in migraine: A treatment report. Headache, 1976, 15, 267-270. Mullinix, J. M., Norton, B. J., Hack, S., & Fishman, M. A. Skin temperature biofeedback and migraine. Headache, 1978, 17, 242-244. Ost, L-G., Jerremalm, A., & Johansson, J. Individual response patterns and the effects of different behavioral methods in the treatment of social phobia. Behavior Research and Therapy, 1981, 19, 1-16. Perski, A. Voluntary control of heartrate: Optimization o f acquisition and transfer of learning. Abstracts of Uppsala Dissertations from Faculty of Social Sciences, 17, 1977. Reading, C., & Mohr, P. D. Biofeedback control of migraine: A pilot study. British Journal of Social and Clinical Psychology, 1976, 15, 429-433. Sargent, J. D., Waiters, E. D., & Green, E. E. Psychosomatic self-regulation of migraine headaches. Seminars in Psychiatry, 1973, 5, 415-428. Silver, B. V., & Blanchard, E. B. Biofeedback and relaxation training in the treatment of psychophysiological disorders: Or are machines really necessary? Journal of Behavioral Medicine, 1978, 1, 217-239. Sovak, M., Kunzel, M., Strenbach, R, A., & Dalessio, D. J. Is volitional manipulation of hemodynamics a valid rationale for biofeedback therapy of migraine. Headache, 1978, 18, 197-202. Turin, A., & Johnson, W. G. Biofeedback therapy for migraine headaches. Archives of General Psychiatry, 1976, 33, 517-519. Weiss, J. M., & Engel, B. Operant conditioning of heart rate in patients with premature venctricular beats. Psychosomatic Medicine, 1971, 33, 301-321. Werbach, M. R., & Sandweiss, J. H. Peripheral temperatures of migraineurs undergoing relaxation training. Headache, 1978, 18, 211-214.