Psychoeducational intervention

strategies offer significant

benefits in the treatment

of cancer-related pain.

Christopher Perkins. Taranaki, 1931. Oil on canvas, 508 × 914 mm. Courtesy of Auckland Art
Gallery Toi o Tamaki, purchased 1968.

Nonpharmacological Interventions With

Chronic Cancer Pain in Adults
Elizabeth M. Thomas, PsyD, PhD, and Sharlene M.Weiss, RN, PhD

Background: Pain is often poorly controlled in cancer patients. Chronic pain affects adult patients at all
stages of cancer management. Optimal pain management may require attention to psychosocial variables
and the inclusion of nonpharmacological techniques.
Methods: Three nonpharmacological strategies that are effective in reducing pain caused by cancer — patient
psychoeducation, supportive psychotherapy, and cognitive-behavioral interventions — are reviewed.
Recommendations for physicians to facilitate a mental health referral are also discussed.
Results: Effective treatment of cancer pain begins with assessing the severity, characteristics, and impact
of pain. Emotional distress (especially anxiety, depression, and beliefs about pain) has emerged as predictive
of patient pain levels. Appropriate pain management may require a multidisciplinary approach.
Conclusions: Patient psychoeducation has empowered patients to actively participate in pain control
strategies. Supportive psychotherapy can assist patients in managing the stressors associated with cancer,
and cognitive-behavioral therapy helps patients to recognize and modify the factors that contribute to physical
and emotional distress.

Introduction mates suggest that chronic pain affects 60% of adult

patients with newly diagnosed or intermediate-stage
Physical pain is perhaps one of the most feared cancer and up to 95% of patients with advanced dis-
consequences for patients with cancer. Available esti- ease.1,2 Indeed, the magnitude of the problem is so
great that some reports indicate that 25% of individuals
From the Sylvester Comprehensive Cancer Center at the University may actually die in significant pain.3
of Miami School of Medicine, Miami, Fla.
Address reprint requests to Sharlene M. Weiss, RN, PhD, at the Ninety percent of cancer patients are believed to
University of Miami School of Medicine, Sylvester Comprehensive be manageable with relatively simple medical interven-
Cancer Center, 1475 N.W.12th Avenue, Suite 3311, Miami, FL 33136.
tions.2 However, in practice, less than 50% experience
No significant relationship exists between the authors and the
companies/organizations whose products or services may be effective pain relief.3 To account for this discrepancy in
referenced in this article. pain management, a number of psychosocial factors

March/April 2000, Vol. 7, No.2 Cancer Control 157

have been implicated in the literature. While some bers of the intervention group were significantly more
researchers have focused on inadequacies related to likely to take their medications on the correct schedule
health care providers or health care systems (eg, an and at the correct dosage. They were also less likely to
emphasis on prolonging life or achieving cure rather cease taking the medicine when they felt better. While
than alleviating suffering),2,4 others have identified both groups were equally likely to experience medica-
issues involving the patients themselves.5-9 That is, tion side effects, the experimental group was more like-
patients experience difficulty assessing and communi- ly to believe that side effects could be prevented and
cating about pain, are reluctant to report pain, have lim- that they had some personal control over their pain
ited expectations for relief, and generally lack knowl- experience. Moreover, intervention participants were
edge about current therapeutic approaches. Even when significantly less fearful about the possibilities of toler-
pain is addressed medically, patients may be noncompli- ance and addiction. There was no difference in patient-
ant with treatment due to concerns about drug toler- reported pain relief, although there was a trend for the
ance, addiction, side effects, or respiratory depression. experimental participants to report less pain. In fact,
44% of subjects in the intervention group reported no
Of all of the psychosocial factors, emotional dis- or mild pain compared with 24% of control subjects.
tress (particularly anxiety, depression, and beliefs about The characteristics and results of the reviewed studies
pain) has consistently emerged as predictive of patient are summarized in Table 1.
pain levels. Spiegel and Bloom10 found that the site of
metastases in women with advanced breast cancer was Over the past decade, pain experts have discovered
not reliably associated with pain; rather, emotional dis- that a combination of pharmacological and nonphar-
tress and the belief that pain signaled a worsening of macological strategies provides the most effective pain
their condition predicted the reporting of pain. Simi- management.1,2,13,14 Subsequent case reports and stud-
larly, Daut and Cleeland11 found that cancer patients ies have thus begun to incorporate multiple compo-
who attributed their pain to a cause other than cancer nents into the educational process. Current compre-
reported the least interference with activities of daily hensive programs attempt to provide education about
living and pleasure. Further, Ahles et al12 compared the basic principles of pain assessment and control,
cancer patients with and without pain and demonstrat- pharmacological intervention, and nonpharmacologi-
ed that patients with pain scored higher on measures of cal treatments.6-9 With regard to pain assessment, pain-
depression, anxiety, hostility, and somatization. Thus, rating scales can be easily understood by patients. Pain
attention to the psychosocial variables of the cancer rating scales are measured on a Likert scale ranging
patient through nonpharmacological intervention from 0 (indicating no pain) to 10 or 100 (indicating the
seemingly provides an additional and viable avenue for worst possible pain imaginable). These pain scales
the treatment of cancer-related pain. have been found to be a reliable and clinically useful
means for patients to label and communicate about
This article focuses on three of the most common pain.15 Another useful instrument is the daily pain diary
approaches employed by mental health professionals: or log that documents the date and time of a pain expe-
patient psychoeducation, supportive psychotherapy, rience, pain severity, distress due to pain, actions taken,
and cognitive-behavioral therapy. The basic principles and severity of pain after one hour.6 Such a tool pro-
behind these methods are explained and relevant vides ready information concerning pain patterns and
research studies are discussed. Only those studies that the effectiveness of a pain management program. Phar-
have methodically strong designs are included in order macological pain education should include information
to critically examine the efficacy of these approaches. designed to counteract fears of drug addiction, toler-
ance, side effects, and respiratory depression. It may
also be useful to discuss the medical control of compli-
Patient Psychoeducation cating symptoms such as nausea and constipation.
Informing cancer patients about the side effects of
Educational efforts have attempted to empower treatment has not been found to increase the occur-
patients to actively participate in pain control strate- rence of these side effects or to have other negative
gies. Unfortunately, only a handful of empirical studies consequences.16 Finally, education about nonpharma-
have considered the role of education in changing cological interventions should stress the importance of
patients’ attitudes and beliefs in pain management. In these modalities as an adjunct to effective pain relief.
an early study, Rimer and colleagues5 investigated the Discussions of various modalities and referrals to appro-
effects of a 15-minute patient education program. priate personnel or resources should be provided.
Printed materials concerning pharmacological issues
were individualized and discussed with patients by a Two recent studies evaluated the effectiveness of
counselor. One month later, results indicated that mem- this type of multicomponent education. Ferrell and

158 Cancer Control March/April 2000, Vol. 7, No.2

 Table 1. — Summary of Characteristics and Results of Controlled Randomized Studies

Authors (ref) Study Design Significant Results

Rimer et al (5) • N = 230, mixed cancers Increase in knowledge of taking medications on correct
• 2 groups: experimental and control schedule and knowledge of taking medications at correct
dosage.
• 1 individual session, 15 minutes
Decrease in stopping medications when felt better, fear
of tolerance, and fear of addiction.

Ferrell et al (6) • N = 40 patients (mixed cancers) and family members Patients:

• Solomon Four-Group Design: 2 experimental Increase in knowledge of pain medications, use of pain
and 2 control drugs, and sleep.
• 3 sessions (either individual or with family caregiver) Decrease in fear of addiction, pain intensity, pain severity,
and anxious mood
Family caretakers:
Increase in providing medications consistently and at the
correct dosage.
Decrease in fear of addiction and fear of respiratory
depression.

de Wit et al (9) • N = 313, mixed cancers Increase in pain knowledge (taking medications on
• 4 groups: experimental with and without home nursing, correct schedule and at correct dosage).
control with and without home nursing Decrease in fear of addiction and pain intensity for those
• 1 individual session + 2 telephone contacts, without home nursing.
60-90 minutes

Spiegel et al (20) • N = 86, metastatic breast cancer Decrease in anxiety, depression, fatigue, confusion,
• 2 groups: experimental and control phobias, and negative coping responses.
• 52 group sessions, 90 minutes Increase in vigor.

Spiegel and Bloom (21) • N = 54, metastatic breast cancer patients with pain Both support and support + hypnosis groups:
• 3 groups: supportive psychotherapy, Decrease in pain sensation and suffering compared with
supportive psychotherapy + hypnosis, control control.
• 52 group sessions, 90 minutes of support, Support + hypnosis group:
5-10 minutes of hypnosis Decrease in pain sensation compared with both support
alone and control groups.

Syrjala et al (27) • N = 67, bone marrow transplant recipients Hypnosis group:

• 4 groups: treatment as usual control, therapist contact Decrease in pain compared with control.
control, hypnosis, and cognitive-behavioral skills
(imagery excluded)
• 12 semiweekly individual sessions
(2 pre-hospitalization and 10 during hospitalization)

Syrjala et al (28) • N = 94, bone marrow transplant recipients Both relaxation + imagery and
• 4 groups: treatment as usual, therapist support, cognitive-behavioral groups:
relaxation + imagery, cognitive-behavioral coping skills Decrease in pain in comparison to control.
package (includes relaxation + imagery)
• 12 semiweekly individual sessions (2 pre-hospitalization
and 10 during hospitalization)

Sloman et al (31) • N = 60, mixed cancers Both live and taped treatment groups:
• 3 groups: control, live relaxation + imagery, and taped Decrease in pain intensity, overall severity of cancer pain
relaxation + imagery over past week, and as-needed nonopioid analgesic intake
• 4 individual sessions over 2 weeks compared with control.
Live treatment group:
Decrease in pain sensation compared with control.

March/April 2000, Vol. 7, No.2 Cancer Control 159

associates6 implemented a pain package in three sepa- knowledge may not be the only variable responsible for
rate sessions and included verbal, written, and audio- a decrease in pain intensity. Indeed, for those with
taped instruction to maximize learning. Emotional dis- more complex or extensive pain, education alone may
tress, cancer treatment medications, and the physical not be sufficient. More rigorous intervention, such as
pain itself have been discussed as impairing patient supportive or cognitive-behavior psychotherapy in
retention of information.3,14 As a consequence, combination with pharmacotherapy and education,
researchers have suggested that information conveyed may be needed for adequate pain relief. To ensure
in small chunks and taught through multiple educa- proper transmission, use, and success of pain assess-
tional modalities may reduce learning interference.8 ment and treatment, education should begin when
Additionally, the authors of this study provided the pain analgesics become warranted in patient care and
program to family members who were caregivers. before activities of daily living significantly decline.14
Because pain management is now performed largely on
an outpatient basis,9,17 family members play an integral
role in patient care. They frequently oversee the admin- Supportive Psychotherapy
istration of medication and any adjuvant therapies.
Since family members are also subject to a lack of Supportive psychotherapy with cancer patients is
knowledge and misconceptions concerning cancer performed within several formats, including individual,
pain and its control, they should be included in any family, couples, and group. Selection of a format may be
educational intervention. Consistent with past find- dictated by patient variables such as physical condi-
ings, results from the Ferrell study indicated that inter- tion, preference, or coping style. Supportive psy-
vention subjects reported an increase in both knowl- chotherapy generally refers to the use of a supportive-
edge and usage of medication as well as a decrease in a expressive model in which a therapist provides emo-
fear of addiction. Anxiety levels declined, and sleep was tional support, encourages expression of feelings and
enhanced. More importantly, study subjects evidenced thoughts, and assists with strengthening and develop-
a decrease in pain intensity and the perception of pain ing coping skills.13,18-20 The purpose of supportive psy-
severity. They also reported positive attitudes towards chotherapy is to manage the limitations associated with
nondrug remedies, and they utilized methods such as cancer while continuing to live life meaningfully and
heat and massage to decrease pain levels. Caregiver establishing smaller, more obtainable goals.18,19
outcomes revealed similar and significantly positive
changes in knowledge, including a reduced fear of Supportive therapeutic work tends to be non-ana-
addiction and respiratory depression. In addition, lytical, focusing on present circumstances rather than
changes were noted in caregiver behaviors such as the examining the past, as is the case with a more tradi-
provision of adequate doses of medication and the tionally oriented approach.18 Emphasis is therefore
medication of patients on a consistent basis. placed on communication with others instead of intro-
spection.13,18-20 Nonetheless, discussion of current
More recently, de Wit and colleagues9 evaluated a stressors and reactions does sometimes involve explor-
comprehensive pain program that followed a similar ing and gaining insight into past issues. Supportive psy-
structure and individualized format as the Ferrell study. chotherapy also tends to be unstructured in that no
In this study, however, changes were measured over specific techniques are regularly employed. However,
multiple time points, and differences between subjects many therapists augment their sessions by teaching
were more closely inspected. First, patients were clas- concrete strategies to control cancer pain. Specific
sified as to whether they received at-home nursing. techniques, along with their theoretical framework,
They were then randomly assigned to either interven- are discussed in the next section.
tion (eg, pain education) or a control condition. Results
revealed that the pain education program was effective While many studies have examined the use of sup-
in improving pain knowledge and attitudes over time. portive psychotherapy with cancer patients, measures
Again, changes were noted with regard to drug mis- have primarily assessed affective distress and quality of
conceptions and the regular use of medication as an life issues. Only one controlled study has examined the
effective means to control cancer pain. While inter- effects of supportive psychotherapy on pain in cancer
vention participants reported less pain than control patients. Spiegel et al10,20,21 randomly assigned women
subjects, this decrease was observed primarily in those with metastatic breast cancer to weekly group psy-
intervention subjects who did not receive at-home chotherapy or a control condition that received only
nursing. The authors hypothesized that patients with routine oncologic care. In the psychotherapy group,
at-home nursing experienced more complex pain prob- members explored existential and practical issues relat-
lems. This difference between patients with and with- ed to living with cancer, including fears surrounding
out home nursing suggests that improvements in pain the death and dying process, family adjustment, and

160 Cancer Control March/April 2000, Vol. 7, No.2

communication with physicians. Psychosocial mea- and an increase in the perception of pain. Intervention
sures were taken on all study participants every 4 attempts to modify behaviors, cognitions, or a combi-
months for 1 year. Following the yearlong intervention, nation of the two. By changing thoughts or behaviors
those in the experimental group reported being signif- in a positive manner, feeling and pain states are pre-
icantly less tense, depressed, fatigued, and confused. sumed to be naturally and similarly affected.
They reported more vigor, fewer maladaptive coping
responses (eg, overeating, drinking, smoking), and Cognitive-behavioral intervention is composed of
fewer phobias than the control subjects. Differences numerous techniques that may be used singularly or
were noticeable at 4 and 8 months but were not signif- collectively in a treatment package. Behavioral strate-
icant until month 12. With regard to pain over the year gies include progressive muscle relaxation, relaxation
period, control subjects reported a sizeable increase in training, and hypnosis. Cognitive strategies include
the sensation of pain, while those in the treatment sam- guided imagery, autogenic training, distraction, thought
ple reported no change in pain levels. In fact, by the monitoring, coping self-statements, and problem solv-
end of the year, reports of pain in the treatment group ing. Descriptions of the techniques utilized with can-
were half that of the control group. In terms of suffer- cer pain patients are provided in Table 2.
ing associated with the pain, the control patients expe-
rienced an increase and the treatment group a Cognitive-behavioral strategies have been practiced
decrease. Interestingly, there were no differences extensively in the treatment of chemotherapy-related
between the treatment and control groups in terms of
either the frequency or duration of pain experiences, Table 2. — Definitions of Cognitive-Behavioral Strategies
suggesting that those aspects of pain that can be influ-
enced by psychological variables may have been signif- Technique Name Description
icantly affected by the group intervention. In fact,
Progressive Actively tensing and relaxing various muscle
changes related to pain were found to be significantly
Muscle Relaxation groups, one at a time, to differentiate between
related to changes in mood states. For the treatment the muscle in its tense and usually wakeful
participants, improvements in overall mood were asso- state, and the muscle in its goal state of
ciated with a decrease in the sensation of pain. For con- relaxation.
trol group members, however, increases in pain dura- Relaxation Relaxation in which attention and imagination
tion were significantly related to increases in overall are focused on the dissipation of tension in
mood disturbance and specifically to increases in anxi- successive muscle groups.
ety, depression, or fatigue.
Hypnosis Achieving an intense state of relaxation,
or trance, and receiving suggestions to alter
sensations, behavior, feelings or thoughts.
Cognitive-Behavioral Therapy Guided Imagery Using mental imagery, usually of a neutral
or positive nature in which the person is led
Cognitive-behavioral therapy is another theoretical through a particular scene. Imagery may be
model that has been employed in the treatment of visual, auditory, kinesthetic or a combination.
cancer-related pain. As with supportive psychotherapy, Frequently used in combination with relaxation
it is practiced in a number of formats, with individual or or hypnosis.
group sessions tending to be more common than fami- Autogenic Focusing on internal bodily states and
ly or couples. At its core, the cognitive-behavioral Training transforming sensations through imagery
model suggests that a person’s distressing physical and (eg, a threatening sensation such as pain
is imagined to be instead a soothing
mental symptoms are partially a consequence of mal-
sensation such as warmth).
adaptive thoughts, feelings, or behaviors.22,23 This per-
spective thus focuses on recognizing and modifying the Distraction Diverting attention away from the sensation
thoughts, feelings, and behaviors that contribute to of pain to a neutral or pleasant stimulus.
physical and emotional distress. Cognitive Monitoring and evaluating negative
Restructuring thought patterns in an effort to create
Several researchers have adapted the cognitive- more realistic and adaptive cognitions.
behavioral perspective to specifically address cancer- Coping Stating specific positive affirmations such as
related pain.23-25 That is, cancer pain is reputed to con- Self-statements “I can do this” or “I am strong enough to
tain an objective component (the pain stimulus) and a handle this.”
subjective component (the perception of the pain stim- Problem Solving Labeling of a problem and generation of
ulus). The subjective portion of pain is postulated to be possible solutions utilizing a cost-benefit
influenced by distorted or irrational thoughts or behav- analysis.
iors that, in turn, generate exaggerated feeling states

March/April 2000, Vol. 7, No.2 Cancer Control 161

nausea and vomiting.26 With the advent of new in intensity and of a shorter duration. There were no
antiemetic drugs, however, nausea is now better con- differences among the four groups in terms of opioid
trolled, and these techniques are less utilized.17 Cogni- usage, suggesting that decreased pain report in the
tive-behavioral treatment also has a solid history with hypnosis group was not simply a function of addition-
chronic pain syndromes,26 whereas cancer pain litera- al pain medication. As the authors suggest, the superi-
ture has been slowly evolving. Overall, one of the most ority of the hypnosis group over the cognitive-behav-
widely used and espoused techniques for cancer related ioral skills program implies that the guided imagery
pain is hypnosis. In the Spiegel study described above, component may be pivotal to effective treatment.
intervention participants who were experiencing pain However, this result may have been influenced by the
were further subdivided to examine the effects of hyp- extraordinary degree of pain associated with oral
nosis. That is, half of the intervention subjects received mucositis and transplantation. In support of this, the
the group psychotherapy condition, and the remaining article does comment that the patients who received
treatment subjects received group therapy plus a 5- to the cognitive-behavioral skills package began to refuse
10-minute hypnosis exercise for pain control. Results at sessions. Those patients engaging in hypnosis did not
one year revealed that intervention members who rebuff intervention, but they required active, engaging
received group psychotherapy plus hypnosis reported imagery to stay involved. Thus, patients experiencing
less pain sensation than those who received only group severe levels of pain may require an intensely distract-
psychotherapy. Differences between the intervention ing approach to pain management such as that provid-
and control groups were significant, suggesting that the ed by guided imagery. Additionally, as noted by the
addition of a hypnosis procedure may produce a cumu- researchers, the lack of success with the cognitive-
lative effect on the reduction of cancer pain. behavioral skills training also may have been compro-
mised by the number of techniques used, which may
In another study by Syrjala and colleagues,27 can- have surpassed what patients could master in such a
cer patients with oral mucositis pain undergoing short period of time.
bone marrow transplantation were randomly as-
signed to one of four groups: (1) routine treatment, In a subsequent study by many of the same
(2) a therapist attentional control, (3) hypnosis (ie, researchers,28 bone marrow transplant patients were
relaxation and imagery of a visual, auditory and again assigned to several conditions: (1) treatment as
kinesthetic nature), or (4) a cognitive-behavioral skills usual, (2) therapist support, which comprised a psy-
package. The cognitive-behavioral skills package was choeducation component and reassurance but not the
quite extensive and included progressive muscle training of new coping skills, (3) relaxation, imagery,
relaxation, autogenic training, cognitive restructuring, and autogenic training (called hypnosis in the previous
distraction, coping self-statements, problem solving, study), and (4) a cognitive-behavioral skills program.
and exploration of the patients’ interpretations of This time, the package of cognitive-behavioral
their illnesses and treatments. Additionally, psychoed- techniques was more limited in scope. It included the
ucation specific to transplantation pain was provided. relaxation program provided to group 3 as well as the
Guided imagery, however, was specifically excluded techniques of coping self-statements, distraction, and
from the cognitive-behavioral skills package. Patients problem solving. Patient training and therapy adminis-
assigned to the hypnosis and the cognitive-behavioral tration were identical to the companion study.
groups participated in two individualized verbal train-
ing sessions prior to the transplant procedure, and As noted, the relaxation training in this study was
they received written and audiotaped instructions to a near duplicate of the hypnosis procedure in the prior
practice their skills prior to hospital admission. Ther- study. Apparently, the authors had chosen to use a dif-
apy sessions to reinforce training were provided ferent label to increase patient acceptance of the pro-
twice a week for the first five weeks of hospitaliza- cedure. Indeed, there has been some inconstancies in
tion. Patients in the therapist attention control con- terminology in the literature, with researchers labeling
dition met with a mental health professional to dis- identical procedures differently. According to Jay and
cuss general, non–pain-related topics for the equiva- associates,29 this has been due not only to patient resis-
lent amount of time and session frequency as the hyp- tance, but also to a lack of clear definition of terms and
nosis and cognitive-behavioral groups. standardization of procedures, making similar tech-
niques (eg,“relaxation with guided imagery” and “hyp-
Results indicated that only the hypnosis-alone nosis”) and their associated outcomes in studies diffi-
group reported significantly less posttransplant pain cult to compare. As a consequence, the strategies of
than that reported by controls. This was particularly relaxation with guided imagery and hypnosis with can-
true during weeks 2 and 3 posttransplant. Indeed, cer pain have not been proven to differ empirically at
reported peak pain for the hypnosis group was lower this time.30

162 Cancer Control March/April 2000, Vol. 7, No.2

 Data analysis from the second Syrjala study techniques were effective.33 Further research is need-
revealed that patients in the relaxation/imagery/auto- ed to delineate the effect of these techniques on can-
genic training group and in the cognitive-behavioral cer-related pain.
skills group reported significantly less pain than those
in the treatment-as-usual control group. However, there
were no differences between the relaxation/imagery/ Conclusions
autogenic training group and the cognitive-behavioral
skills group in terms of pain levels. Thus, findings sug- Cancer pain is a complex phenomenon, affected by
gest that the addition of cognitive-behavioral tech- both medical and psychosocial variables. Effective treat-
niques to relaxation/imagery/autogenic training did not ment of cancer pain begins with comprehensive assess-
further reduce pain levels. In addition, those partici- ment. In this regard, researchers suggest that physicians
pants who received therapist support also reported less inquire about the existence of pain as a standard part of
pain than the treatment-as-usual controls. However, the treatment and use patient self-report as the primary
difference was a trend and did not reflect a statistically source of information.1,2 At a minimum, Cleeland and
significant effect. Again, no differences were detected Syrjala34 indicate that pain assessment should focus on
among the groups in terms of opioid use. As psycho- pain severity (usually rated on a 0 to 10 verbal rating
logical distress was measured only prior to transplant, scale), pain characteristics (location, temporal pattern,
no data were available to examine changes in this vari- quality, and response to treatment) and pain impact on
able that occurred during intervention. However, the quality of life (mood, physical functioning, social inter-
authors did report that emotional distress prior to actions, and concurrent symptoms). A formalized and
transplantation was found to be a significant predictor routine assessment of pain assists in not only develop-
of subsequent pain reports and opioid usage. ing a common patient-physician language for the expe-
rience of pain, but also orienting the patient towards
One other methodically sound study supports the active participation and compliance in treatment.
idea that relaxation and guided imagery produce sig-
nificant effects on cancer-related pain. Sloman et al31 While patients tend to be compliant with cancer
randomly assigned hospitalized cancer patients who therapies, adherence to taking symptom-related medica-
were experiencing physical pain to one of three condi- tions is less common.14 Patients may eliminate, skip, or
tions for a 2-week regimen of (1) routine care (ie, a con- reduce medications because of misconceptions regard-
trol condition), (2) progressive muscle relaxation and ing such issues as side effects, tolerance, and addiction.
guided imagery by audiotape, or (3) progressive muscle Indeed, pain and its medical treatment appear to be sig-
relaxation and guided imagery by live nurse instruc- nificantly affected by the fears and faulty belief systems
tion. Subjects in the audiotaped and live intervention of patients and their family members.5-9 A review of the
groups received two relaxation and imagery sessions results of several psychoeducational interventions
each week, and they were directed to practice twice a reveals that simple education of pharmacological issues
day. In comparison to controls, results indicated that has the potential to not only increase patient and family
both of the intervention groups reported a significant adherence to a medication regimen, but also substan-
reduction in the intensity and overall severity of pain. tially decrease reported pain levels.
The live instruction group also reported less pain sen-
sation than the control group, suggesting that live inter- Moderate to severe cancer-related pain may be
vention may yield some additional benefits. Lastly, par- more complicated to treat because pain is likely to sig-
ticipants in the audiotaped and live-instruction groups nificantly impact a patient’s quality of life. Available evi-
required less as-needed nonopioid medication than did dence suggests that cancer patients who experience
the control subjects. pain report significantly more affective distress, typical-
ly depression and anxiety, than those without pain.4,10,12
Psychological research suggests that relaxation Further, the more severe the experience of emotional
with guided imagery (ie, hypnosis) is an effective treat- distress, the more severe the experience of pain.10,21
ment strategy for the relief of cancer pain. In fact, in a Proper management may therefore require a multidis-
meta-analysis that examined cancer pain, relaxation ciplinary approach with the inclusion of a mental
interventions consistently produced a positive and health professional knowledgeable in pain strategies.
large effect on cancer pain.32 However, the analysis did Under these circumstances, therapeutic models such as
not compare relaxation with and without imagery. supportive or cognitive-behavioral psychotherapy may
Data are less clear on the efficacy of cognitive-behav- be useful adjuncts to pharmacotherapy. Research indi-
ioral techniques without the benefit of relaxation with cates that these psychotherapeutic interventions can
imagery. A meta-analysis of cognitive-behavioral strate- offer significant benefits such as increased relief from
gies utilized for non–cancer-related pain found that all pain and associated emotional distress.

March/April 2000, Vol. 7, No.2 Cancer Control 163

 15. Syrjala KL, Chapman CR. Measurement of clinical pain: a
Ideally, psychosocial resources should be intro- review and integration of research findings. In: Benedetti C, Chap-
duced when a patient is first diagnosed with cancer, man CR, Moricca G, eds. Advances in Pain Research and Therapy.
and the patient should be reminded or encouraged to Vol 7. New York, NY: Raven Press; 1984:71-101.
16. Howland JS, Baker MG, Poe T. Does patient education cause
participate again as different issues or crisis periods side effects? A controlled clinical trial. J Fam Pract. 1990;31:62-64.
materialize. Data suggest that many patients may be 17. Du Pen AR, Niles R, Hansberry J, et al. Clinical strategies for
more open to mental health services than physicians outpatient cancer pain management. Qual Life-Nurs Challenge.
1997;4:95-103.
realize. For example, one group of researchers found 18. Spiegel D,Yalom ID. A support group for dying patients. Int
that 24 of 27 cancer patients stated they would likely J Group Psychother. 1978;28:233-245.
participate in a psychosocial group if referred by their 19. Wellisch DK. Intervention with the cancer patient. In:
Prokop C, Bradley AA, eds. Medical Psychology: Contributions to
oncologists.35 However, the number dropped dramati- Behavioral Medicine. New York, NY: Academic Press; 1981:223-240.
cally to 7 of 24 patients who indicated they would join 20. Spiegel D, Bloom JR,Yalom I. Group support for patients with
a group without first consulting their oncologist. metastatic cancer: a randomized outcome study. Arch Gen Psychia-
try. 1981;38:527-533.
21. Spiegel D, Bloom JR. Group therapy and hypnosis reduce
Thus, physicians are in a position to influence metastatic breast carcinoma pain. Psychosom Med. 1983;45:333-
their patients by addressing the very psychosocial fac- 339.
22. Beck AT, Rush AJ, Shaw BF, et al. Cognitive Therapy of Depres-
tors responsible for adequate pain relief. As noted by sion. New York, NY: Guilford Press; 1979.
Loscalzo,36 some patients may resist the concept or a 23. Jacobsen PB, Hann DM. Cognitive-behavioral interventions.
direct referral to a mental health professional because In: Holland JC, ed. Psycho-oncology. New York, NY: Oxford Univer-
sity Press; 1990:717-729.
of a fear that the physician has given up hope. It is 24. Fishman B. The cognitive behavioral perspective on pain
therefore incumbent on the physician to frame the management in terminal illness. In: Turk DC, Feldman CS, eds. Non-
referral in a way that facilitates patient involvement. invasive Approaches to Pain Management in the Terminally Ill.
New York: Haworth Press; 1992:73-88.
For example, a physician can state that problems with 25. Loscalzo M, Jacobsen PB. Practical behavioral approaches to
pain require adjustment, and mental health profession- the effective management of pain and distress. J Psychosoc Oncol.
als are specially trained to talk with the patient and 1990;8:139-169.
26. Compas BE, Haaga DAF, Keefe FJ, et al. Sampling of empiri-
generate additional coping strategies. In this way, tra- cally supported psychological treatments from health psychology:
ditional medical services could be combined with psy- smoking, chronic pain, cancer, and bulimia nervosa. J Consult Clin
chosocial support as an example of true comprehen- Psychol. 1998;66:89-112.
27. Syrjala KL, Cummings C, Donaldson GW. Hypnosis or cogni-
sive and integrative care. tive behavioral training for the reduction of pain and nausea during
cancer treatment: a controlled clinical trial. Pain. 1992;48:137-146.
28. Syrjala KL, Donaldson GW, Davis MW, et al. Relaxation and
References imagery and cognitive-behavioral training reduce pain during cancer
treatment: a controlled clinical trial. Pain. 1995;63:189-198.
1. Bonica JJ. Cancer pain. In: Bonica JJ, ed. The Management 29. Jay SM, Elliot C,Varni JW. Acute and chronic pain in adults and
of Pain. 2nd ed. Philadelphia, Pa: Lea & Febiger; 1990:400-460. children with cancer. J Consult Clin Psychol. 1986;54:601-607.
2. Management of Cancer Pain: Guideline Overview. Agency 30. Syrjala K. Relaxation techniques. In: Bonica JJ, ed. The Man-
for Health Care Policy and Research, Rockville, MD. J Natl Med Assoc. agement of Pain. 2nd ed. Philadelphia, Pa: Lea & Febiger; 1990:
1994;86:571-573, 634. 1742-1750.
3. Twycross RG, Lack SA. Symptom Control in Far Advanced 31. Sloman R, Brown P, Aldana E, et al. The use of relaxation for
Cancer. London, England: Raven Pitman; 1983:3-14. the promotion of comfort and pain relief in persons with advanced
4. Breitbart W. Psychiatric management of cancer pain. Cancer. cancer. Contemp Nurse. 1994;3:6-12.
1989;63:2336-2342. 32. Devine EC, Westlake SK. The effects of psychoeducational
5. Rimer B, Levy MH, Keintz MK, et al. Enhancing cancer pain care provided to adults with cancer: meta-analysis of 116 studies.
control regimens through patient education. Patient Educ Couns. Oncol Nurs Forum. 1995;22:1369-1381.
1987;10:267-277. 33. Fernandez E,Turk DC. The utility of cognitive coping strate-
6. Ferrell BR, Rhiner M, Ferrell BA. Development and imple- gies for altering pain perception: a meta-analysis. Pain. 1989;38:123-
mentation of a pain education program. Cancer. 1993;72(suppl 11): 135.
3426-3432. 34. Cleeland CS, Syrjala KL. How to assess cancer pain. In: Turk
7. Ferrell BR, Ferrell BA,Ahn C, et al. Pain management for elder- DC, Melzack R. eds. Handbook of Pain Assessment. New York, NY:
ly patients with cancer at home. Cancer. 1994;74(suppl 7):2139-2146. Guilford Press; 1992:362-387.
8. Ferrell BR, Rivera LM. Cancer pain education for patients. 35. Leis AM, Haines CS, Pancyr GC. Exploring oncologists’ beliefs
Semin Oncol Nurs. 1997;13:42-48. about psychosocial groups: implications for patient care and
9. de Wit R, van Dam F, Zandbelt L, et al. A pain education pro- research. J Psychosoc Oncol. 12:77-87
gram for chronic cancer pain patients: follow-up results from a ran- 36. Loscalzo M. Psychological approaches to the management of
domized controlled trial. Pain. 1997;73:55-69. pain in patients with advanced cancer. Hematol Oncol Clin North
10. Spiegel D, Bloom JR. Pain in metastatic breast cancer. Can- Am. 1996;10:139-155.
cer. 1983;52:341-345.
11. Daut RL, Cleeland CS. The prevalence and severity of pain in
cancer. Cancer. 1982;50:1913-1918.
12. Ahles TA, Blanchard EB, Ruckdeschel JC. The multidimen-
sional nature of cancer-related pain. Pain. 1983;17:277-288.
13. Breitbart W, Payne DK. Pain. In: Holland JC, ed. Psycho-
oncology. New York, NY: Oxford University Press; 1998:450-467.
14. Syrjala KL, Abrams J. Cancer pain. In: Gatchel RJ, Turk DC,
eds. Psychosocial Factors in Pain. New York, NY: Guilford Press;
1999:301-314.

164 Cancer Control March/April 2000, Vol. 7, No.2