AJPM Vol. 9 No. 1 January 1999
Alternative Medicine
THE EFFECT OF BIOFEEDBACK AND RELAXATION TRAINING ON
DEPRESSION IN CHRONIC PAIN PATIENTS
Philip Corrado, PhD, and Harold Gottlieb, PhD
Abstract. The combined effects of biofeedback and relaxation training on depression was assessed in chronic pain patients. Forty chronic pain patients were administered the Beck Depression Inventory (BDI) followed by random assignment to either a biofeedback and relaxation training group or a wait-list control group. Eight sessions of temperature and EMG-monitored biofeedback combined with relaxation training were given to the treatment group. Subjects used breathing, muscle relaxation, and autogenics to control temperature and reduce muscle tension. After training was complne4 both the treatment and control groups were again given the BDI. The treatment group reported significantly improved depression in comparison to the control group as measured by pre- and post-rest BDI scores.
Descriptors. Beck Depression Inventory, biofeedback chronic pain, depression
AJPM 1999; 9:18-21. Received:7-27-98: Accepted: 9-11-98
INTRODUCTION
Clinical depression is a significant problem for patients suffering from chronic pain (1). Some authors have estimated chat 10% to 60% of all chronic pain patients are clinically depressed (2). A sample of 100 patients from a structured pain program has revealed that over 70% had scores of 70 or greater on Scale II (the depression scale) of the Minnesota Multiphasic Personality Inventory (MMPI); this indicates clinical depression.
Philip Corrado. PhD. (Clinical Psychology) is certified in biofeedback and neuromuscular reeducation. He has worked in the area of pain management over the last eight years at the Pain & Rehabilitation Harold Gottlieb, PhD, is a Professor of Psychology at the California State University at Los Angeles. He is the Director of Behavioral Medicine Services at the Pain & Rehabilitation Medical Group. Address reprint requests to: Dr. Philip Corrode. 811 South Mayflower, Monrovia. CA 91016
Chronic pain patients, with evidence of pain past three months from point of origin become increasingly depressed the longer pain subsists unabated (3,4).
Kramlinger (5) examined the incidence of depression in 100 chronic pain patients in a pain management program. Based upon research diagnostic criteria, 25% of these patients were diagnosed with major depression, 39% were considered as marginally depressed, while the remaining 36% did not emit evidence of depression. Sternbach (3) and Bonica (4) documented behavioral and emotional disturbances of persons in chronic pain. These researchers found that there is a greater likelihood of depression as long as the pain persists.
Whether depression is antecedent to pain, or vice versa, is an important empirical question. It also seems clear that pain and depression are inextricably involved, making it often nearly impossible to separate which
AJPM Vol. 9 No. 1 January 1999
came first and which factor is causing the other. Nevertheless, there is clear evidence chat where depression exists, pain will be amplified. Craig (1) and Sternbach (3)
suggested chat depression can amplify pain. And Ellis (6) and Beck (7) showed the strong negative effects of depression on pain perception. They also found that depression can be generated by negative thinking and irrational beliefs. Thus. where causative factors serve to increase depression, they must serve to amplify depression and pain. Chronic pain patients who are depressed are also likely to engage in negative and catastrophic thinking, believing that the worst is always going to happen. These patients also report feeling helpless and hopeless and are likely to maintain the belief that no manor what they do, their own actions are not likely to change their pain dilemma.
In this regard, Bandura (8) and Seligman (9) showed that when people do not see themselves as instrumental agents in their own behalf, they are more likely to believe that they are not able to control aversive events. In general, such persons are more likely to score low on measures of self-efficacy and high on depression.
Low self-efficacy patients are likely to be depressed and, accordingly, are likely to experience pain of greater intensity than patients who are either high in self-efficacy and/or low in depression. This view is consistent with Beck's basic hypothesis, that patients who show a pessimistic thinking style are likely to exhibit helplessness and depression (10). These patients are also likely to transform uncomfortable bodily sensory cues to feelings of depression and pain. Thus, interventions that are designed to restore “power to the patient” (i.e., the perception of patient self-control over aversive environmental events) are likely to reduce depression and decrease pain perception.
STUDY
The present study did not investigate the self-efficacy depression-pain connection, per se, but did investigate relaxation and biofeedback training as methods of imparting a sense of control to the patient and its subsequent relationship to depression.
Both relaxation and biofeedback training can constitute skills that the patient can acquire and use as needed for a wide range of medical and psychological conch-dons. This study tested the hypothesis that when patients are provided relaxation and biofeedback training in combination, nor as single modalities, they are able to alter skin temperature and muscle tension and arc more likely to demonstrate lower levels of depression as indicated by lower scores on The Beck Depression Inventory
(BDI).
METHODOLOGY
Subjects. Forty chronic pain patients from an outpatient
pain clinic participated in this study. Twelve females and 8 males were in the treatment group; and 13 females and
7 males were in the control group. Ages ranged from 30 to 56 years; the overall mean age of all subjects was 42.63 years (SD = 8.24). The mean age of the treatment group was 41 years (SD = 8.14). The mean age of the control group was 44.25 years (SD = 8.21). The treatment group contained 14 individuals who were Caucasian, 2 African-American, and 4 Hispanic. The control group was composed of 15 Caucasians, 2 African-Americans, and 3 Hispanics. Subjects were randomly assigned to either the treatment condition or a wait-list control group. In order to be included in this study, subjects had to have chronic pain of more than 6 month’s duration. Subjects were excluded if they had morphine pumps or spinal cord stimulators, or if they cook antidepressant medications.
Materials. The J&J I-330 physiological monitor was used in this study to record skin temperature, breathing, heart rate and muscle tension. Patients receiving biofeedback got both visual and auditory feedback.
The BDI was used to measure depression. This self-report questionnaire is widely used to assess clinical depression and has an established track record for reliability and validity. Research criteria used to indicate depression on the BDI a score of 0 to 10 equals no depression: a score of 11 to 18 equals mild depression; a score of 19 to 25 indicates moderate depression; and a score of 26 or greater indicates severe or major depression.
Procedure. Forty chronic pain patients were randomly assigned to one of two conditions: (i) a treatment group which received biofeedback and relaxation training or (ii) a wait-list control group. The patients in the control group were cold chat they would receive biofeedback and relaxation training, if they so desired, after the experiment. All subjects were told that they would have an equal chance of either getting biofeedback now or being placed on a waiting list where they would get biofeedback later. Subjects were given the BDI as a pretest before random assignment to treatment or control group.
The treatment group received eight sessions of biofeedback combined with relaxation training during which they were taught diaphragmatic breathing, progressive muscle relaxation and autogenics: trained to do six diaphragmatic breaths per minute; trained to increase their skin temperature and to reduce muscle tension in the frontalis and upper trapezius muscles to 2.0 uV; trained to lower their electrodermal response, a measure of sympathetic tone or stress, back to baseline at the presentation of a cognitive stressor; taught progressive muscle rein-scion, while receiving electromyographic (EMG) feedback of the frontalis muscle; and were taught to warm their hands by repeating autogenic phrases. During temperature training, the patients in the treatment group also
American Journal of Pain Management
AJPM Vol. 9 No. 1 January 1999
received audio feedback in the form of a tone that provided information about skin temperature. The physiologic monitor recorded data and provided subjects with information on how well they did at the end of the session.
The mastery model was employed in the treatment group (i.e., subjects were given clear training criteria and then were trained to meet these goals). For example, subjects were trained to increase fingertip temperature from 93o to 95o Fahrenheit. After the eight sessions of biofeedback and relaxation training, both groups were given the BDI as a post-test measure.
RESULTS
An independent T-test was calculated to determine whether the pretest BDI scores were significantly different for the treatment and control groups. The results indicated no significant difference between the groups, c
(38) = 0.73, P 0.47.
Depression scores were analyzed with a 2 x 2 mixed factorial ANOVA. The between-subject factor was condition (biofeedback and relaxation versus control). The within-subject factor was time (before versus after treatment).
The condition main effect was not significant F (1,38) = 0.00.
The time main effect was significant F
(1,38) = 15.261. p <.001.
The condition by time was significant F
(1.38) = 5.45, p <.05.
The means for the groups are summarized in Table I.
Table I. Means and standard deviations for Beck Depression Inventory.
TimeConditionBefore Treatment After Treatment
biofeedback / M 19.35
SD 10.56 / M 12.60
SD 10.90
control / M 16.70
SD 12.29 / M 15.00
SD 11.50
marginal means / M 18.02
SD 11.39 / M 13.80
SD 11.13
The nature of the interaction is shown in Figure 1. The control group showed a slight reduction in BDI scores. However, the treatment group showed a significantly greater reduction in depressive scores, indicating that biofeedback combined with relaxation was effective in reducing depressive symptoms. Figure 1 shows the interaction effect of condition by time between the treatment and control groups.
DISCUSSION
The hypothesis tested by this study was that effective relaxation training and biofeedback training reduces depressive symptoms, and that there is a significant difference between the group receiving the biofeedback and relaxation training and a wait-list control group.
The results confirmed the hypothesis; biofeedback combined with relaxation is effective in reducing Beck Depression Inventory scores. There was a significant correlation between biofeedback and BDI scores, over time, with the treatment group demonstrating significantly lower scores on the BDI than did a wait-list control group.
It is clear that biofeedback training has a direct impact the course of depression and that some consideration for its incorporation in the treatment of depression is indicated. It is inferred that patients who learn self-regulatory skills, namely relaxation and biofeedback, also experience a reduction in depression, most likely as a consequence of a reduction in learned helplessness. Biofeedback and relaxation training may create a shift in locus of control to where patients begin to have faith that they can control their pain.
The present study had several limitations. A sham feedback group was lacking. The addition of a sham feedback group would control for the effect of attention. Future studies of this type should include a sham treatment group that receives eight sessions of false feedback
Another possible limitation was that individuals in the control group were told that they would be put on a wait list and receive biofeedback at a later time. This may have created an expectation within the control group that they would get treatment and, therefore, improve. It appears that this would only have made it harder to demonstrate a difference between treatment and control groups and would only have increased the probability of not rejecting the null hypothesis.
A future study should be designed using four groups:
one group would receive biofeedback without any relaxation training, a second would receive a combination of biofeedback and relaxation training, a third would receive relaxation training without biofeedback, and a fourth would be a wait-list control group. The purpose of such a study would be to assess the relative impacts of each component.
Since biofeedback was effective in reducing BDI scores, it would be interesting to compare biofeedback with antidepressant drugs and cognitive behavioral interventions. An experiment could be designed to explore these three treatments to determine the contribution of each factor in reducing depressive symptoms. Clearly, more research in this area is needed, and this study provides a first step.
AJPM Vol. 9 No. 1 January 1999
Figure 1. Beck depression scores over time.
REFERENCES
1.Craig Kb. Modeling and social learning factors in
chronic pain. In: Bonica JJ, editor. Advances in pain
research and therapy. New York: Raven Press, 1983:813.
2.France RD. Keefe FJ. Chronic pain. In: Houpt NJL,
Broody KH, editors. Consultation/liaison psychiatry and
behavioral medicine. New York: Lippincott. 1986:231-
242.
3. Sternbach RA. Pain patient traits and treatment. New
York: Academic Press, 1974.
4. Bonica JJ. Important clinical aspects of acute and
chronic pain. In: Beers RE, Bassett: EC, editors: Mechanisms of pain and analgesic compounds. New York: Raven Press, 1979:183.
5. Kramlinger KG, Swanson DW, Maruta T. Are patients with chronic pain depressed? Amer J Psy 1983;
140:747-749.
6. Ellis A. A new guide to rational living. North Hollywood (CA): Wilshire Books, 1975.
7.Beck A. Cognitive therapy of depression. New York,
Gilford Press, 1979.
8.Bandura A. The social foundations of thought and
action. Inglewood Cliffs (NJ): Prentice Hall, 1986
9. Seligman M. Helplessness. San Francisco: WH Freeman, 1975
10. Beck A. Cognitive therapy and emotional disorders.
New York: New American Library, 1979.