Alan H. Roberts

The behavioral management of chronic pain: long-term follow-up with comparison groups.

&NA; To assess the long‐term efficacy of an operant inpatient treatment program for severely disabled chronic pain patients, 26 treated patients were compared with 20 rejected for treatment by a clinic team and 12 who refused treatment. At follow‐up ranging from 1 to 8 years, 77% of treated participants were leading normal lives without medication for pain compared to one patient in the other two groups. At time of evaluation, unsuccessfully treated patients used more medications and were higher on MMPI measures of paranoia and lower on ego‐strength than successfully treated patients. Spouses of unsuccessfully treated patients had higher MMPI scores on hypochondriasis and hysteria than spouses of successfully treated patients.

The power of nonspecific effects in healing: Implications for psychosocial and biological treatments

Abstract We evaluate the hypothesis that the power of nonspecific effects may account for as much as two thirds of successful treatment outcomes when both the healer and the patient believe in the efficacy of a treatment. Five medical and surgical treatments, once considered to be efficacious by their proponents but no longer considered effective based upon later controlled trials, were selected according to strict inclusion criteria. A search of the English literature was conducted for all studies published for each treatment area. The results of these studies were categorized, where possible, into excellent, good, and poor outcomes. For these five treatments combined, 40 % excellent, 30 % good, and 30 % poor results were reported by proponents. We conclude that, under conditions of heightened expectations, the power of nonspecific effects far exceeds that commonly reported in the literature. The implications of these results in evaluating the relative efficacy of biological and psychosocial treatments is discussed.

The behavioral management of chronic pain: A response to critics

&NA; Common criticisms of behavioral treatment programs for chronic pain are summarized. Some criticisms are based on conceptual misunderstandings; therefore, basic concepts and goals of behavioral programs are presented. Other criticisms question the effectiveness of these programs; therefore, the role of social reinforcers in maintaining or reducing pain behaviors is reviewed. The failure to isolate specific treatment variables is alleged; this is acknowledged, along with the practical and ethical questions making this virtually impossible. Finally we describe the need to change the thinking about ‘pain’ from the pathological or disease model, appropriate to acute pain, to a learning model when discussing the excess disability and suffering of chronic pain patients.

Skin Temperature Biofeedback and Migraine Headaches A Double-Blind Study ~

To assess the relative contribution of specific and nonspecific effects of skin temperature biofeedback upon migraine headache, 11 migraine patients were taught to increase the temperature of their hand. Training to decrease the skin temperature of the hand served as a control for 12 other migraine patients. An additional 11 control subjects were not trained but kept records of migraine activity. Under carefully controlled double-blind procedures, migraine patients who learned to raise finger temperatures showed statistically significant and clinically therapeutic improvement during a 6-week follow-up period. However, they were not significantly better than those trained to lower finger temperatures, those who did not meet a learning criterion, or those receiving no training. While these groups did show some significant improvement when compared to subjects who learned to decrease finger temperature, the results are most parsimoniously explained through nonspecific rather than specific factors. The necessity of using double-blind procedures in evaluating therapeutic effectiveness is again stressed.

Skin temperature biofeedback for Raynaud's disease: A double-blind study

The lack of control procedures inherent in most of the experiments conducted to assess the effectiveness of skin temperature biofeedback in the treatment of Raynauds disease renders the results inconclusive. In this study, control groups and a double-blind approach are adopted. Thirty-six patients, carefully screened for a diagnosis of primary Raynauds disease, were assigned to a skin temperature increase group (N=12), to an EMG relaxation control group (N=12), or to a notreatment control group (N=12). All patients kept records of their symptoms for the duration of the study. Each subject in the two training groups received 20 sessions, the last 2 conducted under cold stress. Data analysis according to original group assignment, as well as following regrouping of subjects according to several learning criteria, showed that while all patients reported a marked decrease in the number of vasospastic attacks, no significant differences were found among the three groups on the clinical measures used to assess symptomatic relief. The general improvement reported must therefore be attributed to nonspecific factors.

An alternative perspective on biofeedback efficacy studies: A reply to Steiner and Dince

Clinical applications of biofeedback have proliferated and considerable lore surrounding the application of these techniques has evolved. Many assertions about the effectiveness of biofeedback training are based on findings of the least well-controlled studies, while many of the better controlled studies have failed to show that biofeedback directly mediates target symptoms or is superior to other treatments. Steiner and Dince (1981) suggest that the failure of these controlled studies is primarily attributable to methodological deficiencies. We believe that the question of whether or not there is a specific effect of biofeedback training is still frequently confused with the question of whether or not the treatment package as a whole has therapeutic value. Biofeedback is often therapeutic; however, evidence is often lacking that its effectiveness is due to biofeedback-trained changes in a target physiological process.

All cats are not gray; all pain programs are not alike.

The recent article by Gallon (Pain, 37 (1989) 67-75) was interesting and provided some useful data concerning long-term outcomes for a pain treatment program at a medical center in Bangor, ME. Quality long-term follow-up studies of this kind are desirable and should be encouraged. His report, however, calls attention to a common tendency for researchers and reviewers alike to think about and write about all pain programs as if they were the same. In comparing outcomes among pain programs, reports seem to assume that all pain programs use the same methods, that chronic pain patients constitute a homogeneous population to be treated similarly, and that the quality of the treatment provided is similar among programs. It is obvious, when so stated, that these assumptions are not correct, but many who compare outcomes among programs seem to make these assumptions implicitly. Gallon, for example, compares the long-term outcomes for his 300 chronic back pain patients with reports from other programs treating different populations of patients by methods which probably differ from his, either qualitatively or quantitatively. Pain is a symptom, not a disease; chronic pain is not a disease either, it is a syndrome. The etiology of chronic pain varies from patient to patient; thus, its treatment may also need to be varied from patient to patient. When reviewing reports of treatment programs it becomes clear that some attempt to confine interventions to operant behavioral techniques (‘behavioral programs’). Other programs incorporate or emphasize a variety of other interventions (‘multi-modal programs’) which may include biofeedback, physical therapy modalities (ultrasound, massage, etc.), or psychological treatments (group therapy, coping skills training, cognitive therapy, hypnotherapy, relaxation training, etc.). Some are medically oriented and provide techniques from anesthesiology and other medical or surgical specialties. Some programs work with families, others do not. Some are inpatient and others outpatient. Length of treatment may vary from a few days to eight weeks or longer. It has been speculated that the use of some treatment modalities may reinforce pain behaviors and thus reduce program effectiveness. The population treated also varies from program to program. Sources of referral may differ. Pain etiologies may differ. Some patients may request treatment while others may be sent with variable motivation. Pending litigation or compensation remains an unresolved issue (partly because all compensation/litigation patients are not the same just as all pain patients are not the same) and this too may differ from program to program. Some programs accept all who apply; others are selective and will accept only those they believe will profit. Behavioral program outcomes are extremely sensitive to staff composition, communication skills and the ability of staff to consistently apply behavioral principles to patients who are sometimes angry or manipulative. Skilled social workers teaching families to apply behavioral principles in

Volitional vasomotor lability and vasomotor control.

This experiment tested the hypothesis that skin temperature variability during instructions to attempt skin temperature self-control without feedback (volitional vasomotor lability) predicts the acquisition of vasomotor control through biofeedback training. Skin temperature was recorded from the hands of 232 volunteers during a screening session. Twenty-three labile and 17 stabile subjects were chosen to participate in a 16-session training program under double blind conditions. Visual and auditory feedback were used to train subjects to produce temperature differences between the two hands in a specified direction. Comparisons between the labile and stabile groups revealed statistically significant differences in the predicted direction on measures of performance and learning. These findings provide preliminary support for the hypothesized positive relationship between volitional temperature variability and voluntary vasomotor control. Fruitful directions for future research are suggested.This experiment tested the hypothesis that skin temperature variability during instructions to attempt skin temperature self-control without feedback (volitional vasomotor lability) predicts the acquisition of vasomotor control through biofeedback training. Skin temperature was recorded from the hands of 232 volunteers during a screening session. Twenty-three labile and 17 stabile subjects were chosen to participate in a 16-session training program under double blind conditions. Visual and auditory feedback were used to train subjects to produce temperature differences between the two hands in a specified direction. Comparisons between the labile and stabile groups revealed statistically significant differences in the predicted direction on measures of performance and learning. These findings provide preliminary support for the hypothesized positive relationship between volitional temperature variability and voluntary vasomotor control. Fruitful directions for future research are suggested.