Physician attitudes and beliefs about use of morphine for cancer pain

Physician attitudes and beliefs about use of morphine for cancer pain

I~L 7No. 3 April 1992 oTournalof Pain and SymptomManagement 141 Original Article Physician Attitudes and Beliefs About Use of Morphine for Cancer P...

611KB Sizes 2 Downloads 21 Views

I~L 7No. 3 April 1992

oTournalof Pain and SymptomManagement 141

Original Article

Physician Attitudes and Beliefs About Use of Morphine for Cancer Pain Thomas E. Elliott, MD, and Barbara A. Elliott, PhD

Section of Medical Ontology and Hematology, The Duluth Clinic,Ltd. (E.E.E.); The Duluth Family Practice Center (B.A.E.); and The Univerd~ of 3Iinnesota, School of Medidne, Duluth, Minnesota (E.E.E., B.A.E.)

Abstract The recent literatureasserts that mistaken physician beliefs and attitudes are critical barriers to adequate cancerpain relief. To determine the prevalence of l 2 proposed myths or misconceptions about morphine use in cancerpain management (CPM), we survo'ed all pfffsicians engaged in directpatient care in Duluth, Minnesota ( N = 243). A 62% response was obtained. Many physidans misunderstood concepts of morphine tolerance, both to analgesia (51%) and to side effects (39%). Many were unaware of the use of adjuvant analgesics (29%), effcafy (oral morphine (27%), and nonexistent risk of addiction in CP3/I (20%). Analysis of result by physidan age and spedalty groups confirmed significant levels of misunderstanding in all subsets. Strategies to changepl~sician attitudes and beliefs regarding morphhze in CP~I should focus on tolerance concepts, dosing sclu'mes, safety, effcaqy, lack of addictive risk, use of drug combinations, and thefact that cancerpain can be relieved.J Pain Symptom Manage 1992;7:141-148.

Key Words Cancer pain, narcotic analgesics, attitudes, analgesia, pain control, morphine, physidan education.

Introduction Cancer pain management (CPM) is a major international health problem. In the United States alone, it is estimated that 510,000 cancer deaths occurred in L990.1 Of these, 60-80% had severe pain during the terminal phase of the illness. 2 With proper CPM, up to 95% of these patients could have had adequate relief of pain. a This realityhas not been achieved in the United States or in any other country.4

Addressreprintrequeststo:Thomas E. Elliott,MD, Sectionof Medical Oncologyand Hematology,the Duluth Clinic, Ltd., 400 East Third Street, Duluth, MN 55805. Azceptedforpublication."October 15, 1991. U.S. Cancer Pain ReliefCommittee, 1992 Published by Elsevier, New York, New York

Physician attitudes and beliefs about morphine in CPM have been cited as inadequate and mistaken.5-a This has been identified as one of the major causes ofworldwide undertreatment of cancer pain. 9 Even though the beliefs and myths that interfere with use ofmorphine in CPM have been widely described, litde published data exist about what physicians actually believe.6,7 The need to identify those attitudes that may be hindering analgesic therapy has been emphasized.l~ Few studies have evaluated physician attitudes and beliefs about the use of opioids in CPM. In 1986, Cleeland examined the attitudes and beliefs toward CPM in a sample of 91 Wisconsin physicians who were members of a statewide

9

0885-3924/92/S5.00

142

EUiott and Elliott

VoL 7 No. 3 April 1992

Tab/e 1 Myths About Morphine Use in Cancer Pain Management (CPM) I. Toleranceto pain relief."Patients need increasingdosesofmorphine to control the pain because theybecome tolerant to the

pain relief. II. Intoleranceto adverseo'mptoms: Patients usingmorphine for CPM remain intolerant of adverse side effects,e.g., sedated/cloudywith continueduse. III. Adjm.antdrugs:Reliefofcaneer pain does not involveregimens of multiple classes of drugs and coanalgesies. IV. Parenteraldrugs:CPM requires parenteral administrationwhen there is severe pain. V. Addiction:Morphine addiction is a dangerous risk ofnsing morphine in CPM. VI. In~table pain.. Cancer pain is an inevitab~e sympt~m ~f ~ancer whi~h cann~t be adequate~y relieved w~th nar~ti~s. VII. Ceilingdose:Morphine can onlybe prescribed belowcertain dosage ceilingsin CPM. Morphine has a narrow dosage

range. VIII. Physicaldepender~e:When pain is gone, patients are physicallydependent on the drug and experiencewithdrawalif morphine dosingis tapered slowly. IX. PRNadmlnistration:Morphine in CPM shouldbe prescribed in a PRN dosingschedule. X. Low effwa~y:Cancer pain cannot be effectivelymanaged with morphine. XI. Respiratorydepression."Use of morphine seriouslydepresses respiration in patients with cancer and shortens life. XII. Prognosis:Use of morphine in CPM implies"givingup" on the patient.

oncology research group) 2 Misconceptions about tolerance to pain relief, intolerance to adverse drug effects, and "as needed" administration of analgesics were identified. Similar concerns about opioid tolerance and side effects were revealed in a preliminary analysis of the data collected in a recent survey of 1177 oncologists affiliated with the Eastem Cooperative Oncology Group)3,14 Another survey that evaluated beliefs about opioid analgesics in the context of acute, noncancer pain management in children explored the prevalence of myths regarding addiction and respiratory depression. 15Finally, several studies have examined the attitudes and beliefs about opioid analgesia among house staff/resident physicians536-19 or medical students. 2~ Only the Weissman and Dahl 2~ study of medical students and Charap's study 16of resident physicians examined attitudes and beliefs about cancer pain management. The present study explored the prevalence among practicing physicians of 12 proposed myths or misconceptions about the use of morphine in cancer pain management. Twycross6 has proposed 9 myths or misconceptions to explain the widespread reluctance of physicians to use morphine to relieve cancer pain, and Walsh 7 has proposed 11 similar misconceptions. The 12 myths examined in this study are derived in part from these earlier analyses (Table 1). We hypothesized that there would be significant adherence to these myths or misconceptions about morphine (supporting the literature's assertion that ph~'sicians' attitudes and beliefs may represent a barrier to appropri-

ate pain relief 5-8) and expected that there would be significant differences in the prevalence of these myths or misconceptions among physician specialty or age groups.

Methods A survey of all physicians engaged in direct patient care in Duluth,'Minnesota (3/'= 243) was conducted to determine the attitudes and beliefs they held regarding morphine use in CPM. Responses from 150 physicians were received (62% response). T h e 150 responding physicians had attended 47 different medical schools located in 31 states, Canada, and England. O f the responding physicians, 46% were graduates of the University of Minnesota Medical School. The respondents also represented 51 different postgraduate training programs located in 25 different states. No single postgraduate training program represented more than 12% of the survey sample. The physicians were divided into three age groups as foUows: medical school graduates prior to 1968 (older physicians, N = 41, or 27%), medical school graduates between 1968 and 1977 (middle-aged physicians, N = 53, or 35%), and medical school graduates after 1977 (younger physicians, N = 56, or 37%). The physicians were also divided into four specialty practice groups as follows: cancer specialists (N= 8, or 5%), primary care specialists (W = 90, or 60%), surgeons (W = 28, or 19%), and medical subspecialists (W= 24, or 16%). The survey presented a series of 22 attitudinal

IbL 7.No. 3 April 1992

Physicians' Attitudes About ~Iorphine Use

143

9Table 2

Physician C a n c e r Pain Attitude Questionnaire On the remaining questions, circle or mark the number that represents your opinion about each statement. Disagree Strongly

Agree Strongly

I 1 1 1 I 1 1

2 2 2 2 2 2 2

3 3 3 3 3 3 3

4 4 4 4 4 4 4

1

2

3

4

1 1 1

2 2 2

3 3 3

4 4 4

1

2

3

4

1

2

3

4

1 1 1 1 1 1

2 2 2 2 2 2

3 3 3 3 3 3

4 4 4 4 4 4

1

2

3

4

1

2

3

4

1

2

3

4

Morphine is the drug of choice for treatment ofsevere chronic pain due to cancer. Morphine is a poor oral analgesic. Morphine has a narrow therapeutic dosage range. Morphine should be given on a regular (not PRN) schedule for chronic cancer pain. Pain is an inevitable symptom of advanced cancer. Cancer pain cannot be treated or relieved by narcotics. Oral morphine is dangerous because it causes clinically significant respiratory depression in patients with cancer pain. Morphine addiction (drug abuse or psychological dependence) is irrelevant and is very rarely seen in patients with cancer. Ifmorphine is prescribed too soon, then nothing else is leR to relieve pain at the end. The use ofmorphine signifies a hopeless prognosis and that ph)zicians have "given up." The increasing dose of morphine that is required in some cancer patients indicates an increase in pain and not tolerance to the drug. The appropriate dose of morphine for cancer pain is whatever dose relieves the pain as completely as possible; there is no ceiling dose for morphine. Heavy sedation and mental cloudiness are frequent side effects of chronic morphine use in cancer patients. Underdosing cancer patients with morphine is the major reason for unrelieved pain. "As required" or "PRN" morphine for chronic cancer pain is irrational. Long-term use ofanfiemetles with morphine is usually necessary. Severe cancer pain commonly requires parenteral morphine, even though the patient can swallow. Morphine-induced sedation is only a transient problem and will clear with continued use. Morphine doses sufficient to relieve cancer pain inevitably cause respiratory depression and thus shorten llfe or contribute to the death. The wlthdmwal or abstinence syndrome due to chronic morphine can be avoided by slowly tapering morphine when pain is gone. The relief of cancer pain often requires the addition of nonsteroidal antiinflammatory drugs, corticosterolds, and/or psychotropic drugs (e.g., antidepressants) to morphine. Patients whose pain is completely relieved do better than those whose rest and nutrition continue to be disturbed by continued pain.

statements (Table 2). T h e s e statements e x a m i n e d the degree to which the physician endorsed each o f t h e 12 proposed misconceptions a b o u t the use o f m o r p h i n e for C P M (Table 1). T h e respondents indicated the strength o f their a g r e e m e n t with each s t a t e m e n t u s i n g a scale o f l = strongly agree to 4 = strongly disagree. T h e data were analyzed to evaluate physician a g r e e m e n t with the 12 p r o p o s e d myths a b o u t m o r p h i n e use in C P M . Further analysis o f the data was done according to the year o f medical school graduation a n d specialty o f practice. Variances between these age a n d specialty groups were analyzed with Z 2. Further analyses were d o n e to determine the type o f continuing medical education a b o u t C P M desired by physicians a n d the type o f education they h a d experienced previously. T h e latter data were previously reported31

Results T h e physicians w h o returned the survey were representative o f the overall population. O f the 72 surveyed physicians w h o graduated before 1968, 5 1 % returned the questionnaire, 5 7 % Of the 85 physicians w h o g r a d u a t e d between 1968 a n d 1977 responded to the survey, and 74% o f the 73 physicians w h o g r a d u a t e d after 1977 returned their questionnaires. Except for the cancer specialists (all o f w h o m returned the survey), the respondents also represented the specialty practice groups proportionately; 6 4 % o f the p r i m a r y care physicians, 5 5 % o f the surgeons, a n d 5 5 % o f the medical subspecialists returned the survey. Although y o u n g e r physicians were overrepresented, the consistency in training backgrounds b e t w e e n those w h o responded and those w h o did not, as well as their c o m p a r a b l e graduation dates and specialty

144

Elliott and Elliott

Tab/e3

T h e results indicate that many physicians do support myths that could interfere M t h appropriate prescribing of morphine in C P M (Table 4). One-half of the physicians in the survey indicated a misunderstanding o f drug tolerance and adverse effects. T h e majority misunderstood that an increased need for morphine usually indicates an increase in pain. Also, nearly 40% misunderstood that sedation as a dose-limiting side effect usually disappears within a few days. Similarly, more than 20% believe that addiction is a serious concern in prescribing opioids for C P M or that parenteral administration is the only route to give morphine for severe pain. An equal proportion misunderstood the role of adjuvant drugs as part o f a pain control regimen. T h e r e were no statistically significant differences among the four physician specialty groups in response to the following myths: inevitability o f pain, P R N administration of morphine, and prognostic implications (Table 4); however, there were significant differences among the specialty groups with regard to the other 8 myths. Although the cancer specialists' data are limited by the small n u m b e r of respondents ( N = 8), they do reveal several strongly held and appropriate beliefs. T h e ~ancer specialists understood that regular, time-fixed dosing and adjuvant drugs help to control pain, and that addiction does not develop in this population.

Education of Sample

Respon- Nonredents spondents Totals Graduation Be~re 1968 1968-1977 Since 1977

41 53 56

35 37 21

76 90 77

Specialties Primary care Surgeons Medical subspecialists Heme--oncology

90 28 24 8

51 23 19 0

141 51 43 8

Schools represented Number of medical schools 47 Number ofpostgraduate sites 51

30 51

79 102

VoL 7 No. 3 Apn'11992

choices (Table 3), indicate that t h e sample is representative of the medical community in Duluth, Minnesota. T h e community is private practice based, nonuniversity, and nongovernmental. T h e sample represented 23 different specialties and subspecialties--practitioners o f primary, secondary, and tertiary care. One-half of the responding physicians directly cared for 10 or more cancer patients per year, and 16 physicians (10.6%) managed or partly managed 100 or more cancer patients per year.

Tab/e 4 Myths Supported by P h y s i c i a n s mad P h y s i c i a n S p e c i a l i t y Groups Myths I. Tolerance:pain reliefu II. Tolerance:side effectsb III. Adjuvant drugsd IV. Parenteral drugs with severe painc V. Addiction b

VI. Inevitablepain VII. Morphine ceiling doseb VIII. Physicaldependenced IX. PRIg administration X. Morphine has low efficacy~ XI. Respiratory depressionc XII. Poor prognosis

Cancer specialists

% Misunderstanding Primary care Surgeons

Total

50 50 13 25

50 33 29 20

48 61 40 40

48 33 23 41

51 39 29 27

13 13 25

17 19 13

29 24 24

26 13 36

20 19 17

38 13 25

8 8 7

35 23 11

23 9 13

16 11 I0

25

7

0

11

7

0

3

8

4

4

The statisticalsignificanceof the differencesbetweenspedahygroupswithZ2and dr= 9: a p_< .05. b p < .01. c p < .001. a p < .0001.

Subspecialists

VoL 7 No. 3 April 1992

Phy~idans"Attitudes About Morphine Use

145

Tab/*5 Myths Supported by Physicians and Physic/an Age Groups % Misunderstanding

r,tvth I. Tolerance: pain relief~ VIII. Physical dependence* XII. Poor prognosis*

Older

hfiddle-aged

Young

Total

40 26 10

52" 15 4

56 11 0

51 16 4

The level of statistical significanceof the differencesbetween physicianage groups: " P = .10. b p < .001.

There were some opinions expressed by the cancer specialists, however, that reveal continuing misunderstanding as well; 50% misunderstood the concepts of tolerance to pain relief and side effects, 38% believed that physical dependence becomes a concern for patients with cancer pain, and 25% misunderstood ceiling doses, morphine efficacy, and respiratory depression issues. Primary care specialists had the most favorable profile in response to these myths. Other th~an the 33% who believed that tolerance to the adverse effects of morphine does not develop and the 50% who believed that significant tolerance to pain reliefoccurs, this group seemed fairly well oriented to the facts about morphine use in CPM. Surgeons were different from other specialty groups in their strong statement that tolerance to adverse effects of morphine does not develop (61%); however, they stated that respiratory depression was a minor issue with morphine in CPM. In addition to this incongruency, surgeons held other misunderstandings as follows: PRN dosing is preferred, only parenteral drugs are effective in severe pain, physical dependence and addiction are continuing concerns, adjuvant drugs are not necessary, and tolerance develops to pain relief. Medical specialists (not including cancer specialists) revealed a profile that is distinguished by their endorsement of several of the myths; 41% agreed with the myth that morphine is not an effective oral analgesic. Also, one in three believed that there is a ceiling dose for morphine in CPM, one-halfagreed that tolerance develops to pain relief, and one-quarter stated a concern with addiction issues. Further analysis of these findings demonstrated little difference between older and younger physicians (Table 5). Recent medical

graduates did appear to have more appropriate beliefs about physical dependence and prognostic implications in the use of morphine in CPM. On the other hand, older physicians reflected a better understanding of tolerance concepts. Regarding drug tolerance, the youngest physicians were more likely to inappropriately agree that tolerance develops to the pain relief (older physicians, 40%; middle-aged, 52%; young, 56%; P < .001); however, more older physicians inappropriately responded that physical dependence on morphine is a concern with cancer patients (older, 26%; middle-aged, 15%; young, 1 I%; P = .10). '~nally, more older physicians agreed with the myth that using morphine in CPM is "giving up on the patient" (older, 10%; middle-aged, 4%; young, 0%; P = .10). Despite the prevalence of misconceptions about morphine in cancer pain management, most physicians (90%) agreed that morphine is the drug ofchoice for treatment ofsevere Chronic pain due to cancer. There were no significant differences among physician age or specialty groups agreeing with this statement.

Discussion This study reports the extent to which practicing physicians endorse myths and misconceptions that may interfere with appropriate morphine use in CPM. Physicians' beliefs and attitudes (in combination with their knowledge base and skills) impact on their CPM behaviors. These data do not assess the quality or validity of the physicians' CPM behaviors or skills, nor do they assess how their beliefs affect their CPM behaviors. Numerous published reports have documented a general lack of physician knowledge and s"ldll in CPM, but no other study has assessed the beliefs of practicing, communitybased physicians about morphine use in CPM.

146

Elliott and Elliott

This study is also the first communitywide evaluation of practicing physicians in which the respondent sample represents all specialties and ages, and the attitudes explored refer to most of the myths posed in the literature as barriers to adequate cancer pain relief. Because most cancer patients with pain are managed in their home communities (as opposed to tertiary cancer centers) by primary care physicians (as opposed to oncologists), the data collected in this study may be a more valid representation of physician attitudes and beliefs about morphine in CPM than other published studies. As with all surveys, the present study is limited by the sample size, the respondents' characteristics, the wording of the statements, and the self-report nature of the data. Further limitations are that the data were obtained from a single community, many respondents care for few or no cancer patients, and there was no assessment of CPM behaviors. The recognized cancer specialists (5 medical oncologists and 3 radiotherapists) in this survey represent only 8 respondents of 150; any conclusions drawn from this subset are tentative and very limited. Of the 12 myths that were proposed to explain the inappropiate use of morphine in CPM, nearly one-half were endorsed by 20% of the physicians or more. It is likely these myths do interfere with morphine use in CPM. Fewer than 20% ofthese physicians endorsed one-half of the proposed myths, and, at least in this community, they may not be interfering with CPM. The misconceptions endorsed by physicians in this study indicate traditional attitudes toward opioid use. To counter these misunderstandings, innovative educational activities are needed to change the beliefs, attitudes, and knowledge of physicians regarding the use of morphine in CPM. Such a change may alter physician CPM behavior and improved cancer pain relief may occur. Fishbein and Ajzen22 have asserted that attitudes are critical in predicting prescribing behaviors. Data from this study regarding the physicians' wish to acquire CPM kno~dedge were previously reported. 21 We propose that the myths that were endorsed by 2 0 % or more of the physicians represent a serious barrier to appropriate CPM, and should be the focus ofeducational efforts. By integrating the physician continuing medical education (CME) preferences from the previous

VoL 7 No. 3 April 1992

study with the results of this study, focused CME can be designed to target deficiencies in physician beliefs and attitudes about CPM. Primary care physicians endorsed fewest of the beliefs (4 of 12) that interfere with CPM. As 15reviously reported, primary care doctors prefer to learn through local CME presentations, consultations, and hospice contacts. 21 Educational efforts should focus on these approaches. In addition, 41% of these physicians reported an interest in learning from the literature; strategically placed articles will reach many of them. Surgeons inappropriately endorsed more myths (9 of 12) than the other specialty groups. Their endorsement of misconceptions about tolerance, addiction, physical dependence, administration and dosing of morphine, adjuvant therapies, and inevitable pain reflect beliefs that could interfere with their management of cancer pain. Surgeons, like primary care physicians, indicated a preference to learn about CPM in local staff conferences, through consultations, and in contact with hospices. ~1 Fewer surgeons reported an interest in learning from the literature (28%), and this resource may be less likely to reach surgeons. Subspecialists (other than cancer specialists) endorsed 7 of the 12 beliefs that have been proposed as myths regarding morphine use in CPM; 3 were supported by about one-quarter of the subspecialists; the other 4 were endorsed by at least 40% of the physicians. The attitudes needing the most attention include misconceptions about tolerance and the administration and dosing of morphine. Some attention also needs to be focused on issues of addiction and physical dependence. Subspecialists expressed interest in learning about CPM through local CME conferences, consultations, and the literature (65%). 21 These presentation should be especially directed toward correction of the most prevalent misconceptions, and articles in the subspecialty literature are likely to reach two-thirds of these practitioners. Fewer of these physicians (48%) indicated an interest to learn CPM from hospice contacts. Although many subspecialists may not be providing primary care for their patients, attention to CPM issues will add to their understanding of the proper use of opioids. The 8 cancer specialists in this sample inappropriately endorsed 7 ofthe 12 myths. Radiotherapists were more likely to express these misunderstandings; hematology/oncology spe-

VoL 7 No. 3 April 1992

Physicians"AttitudesAbout Morphine Use

cialists endorsed only 2 of the 12 myths. The issues for the medical oncologists include misunderstanding of tolerance and the use of parenteral administration for severe pain. The misunderstanding of opioid tolerance in GPM is consistent with the recently published study of 1,177 EGOG oncologists.13,14 Those myths that were endorsed by the radiotherapists, in addition to these 2, were misconceptions about side effects, a limiting ceiling dose, efficacy, respiratory depression, and physical dependence. Despite the small sample size, the differences in attitudes and beliefs between the medical oncologists and radiotherapists are striking, with the medical oncologists reflecting a more appropriate understanding of the use of morphine in CPM. As reported previously, these physicians indicated an interest to learn about GPM in conferences (local, regional, and national) and through the literature. 21 Indeed, 88% wished to learn through the literature. Conferences on CPM need to be available for cancer specialists, and articles need to be published in the oncology literature. Analyzing the data by physician age groups revealed only 1 myth which varied to a statistically significant level: the misconception about developing tolerance to pain relief. It was interesting to find a more favorable belief regarding tolerance among older physicians than younger ones. An explanation for this finding is not apparent. We can only speculate that older physicians have learned through experience that drug tolerance in CPM is not a significant problem. In summary, we draw four specific conclusions from this survey. First, not all of the 12 myths believed to interfere with CPM are endorsed by physicians in our community. Only one-half of the myths were inappropriately endorsed and are likely interfering with CPM behaviors. Second, many physicians believe that tolerance to morphine is a confounding issue in GPM. Specifically, physicians believe that patients do develop tolerance to the pain relief produced by morphine and that tolerance to sedation and other side effects does not routinely develop. Third, younger, more recently trained physicians do not necessarily have more appropriate beliefs and attitudes about morphine use in CPM. Fourth, surgeons are more likely than medical subspecialists and primary c~ire physicians to hold beliefs that interfere with appropri-

147

ate morphine use in CPM. Primary care physicians are least likely to have attitudes that interfere with GPM. The following recommendations can be proffered: 1. Innovative medical education at all levels (medical school, residency fellowships, and continuing medical education) is needed to present new knowledge about opioid use in CPM, with special emphasis on concepts of tolerance, pathophysiology of pain, pharmaeulogy ofmorphine and other opioids, and use of adjuvant therapies in chronic cancer pain. 2. Interventions need to be designed which will change attitudes, beliefs and behaviors of physicians treating cancer pain. 3. The National Institutes of Health should increase funding toward research, training programs, and demonstration projects in CPM. 4. The American Society of Clinical Oncology, American Cancer Society, and other cancer organizations and agencies, editors, and editorial boards need to become sensitized to the issue of CPM and provide education, service, funding, conferences, and ahicles about it. 5. Cancer specialists need to become well trained to act as consultants and role models regarding CPM in their communities. 6. State cancer pain initiatives need to be established and supported to study CPM outcomes, and to design and implement interventions to improve CPM. 7. Further study is needed to learn what medical schools, residencies, and fellowships are now teaching about CPM.

Acknowledgments The authors wish to thank the physicians of their community who participated in this study. They wish to thank Colleen Renier for data analysis and Hospice Duluth for its assistance in the study.

References I. Silvcrberg E, Boring CG, Squires TS. Cancer statistics, 1990. CA 1990;40:9-26. 2. Bonlca jET. Cancer pain: a major national health problem. Cancer Nuts 1978;1:313-316.

148

Elliott and Elliott

VoL 7 No. 3 April 1992

3. Twycross RB, Lack SA. Symptom control in far advanced cancer:, pain relief. London: Pitman, 1983.

management survey by ECOG [abstract]. Proc Am Soc Cfin Oncol 1991;10:326.

4. Angell M. The quality of mercy. N Engl J Med 1982;306:98-99.

14. Von RoennJH. Many oncologists still say no to pain meds. Oncol Times 1991:5.

5. Marks RM, Sachar EJ. Undertreatment of medical inpatients with narcotic analgesics. Ann Intern Med 1973;78:173-18 I.

15. Schechter NL, Allen D. Physicians' attitudes toward -pain in children.J Dev Behav Pediatr 1986;7:350-354.

6. Twycross RG. Ethical and clinical aspects of pain treatment in cancer patients. Acta Anaesthesiol Scand 1982;74(Suppl):83-90. 7. Walsh'I'D. Common misunderstandings about the use of morphine for chronic pain in advanced cancer. CA 1985;35:164-169. 8. DahlJL, Joranson DE, Engber D, DoschJ. A report on the Wisconsin Cancer Pain Initiative. J Pain Symptom Manage 1988;3(Suppl):52-55. 9. Stjernsw~dJ. Cancer pain reliefi an important global public heaith issue. Ctin J Pain 1985; 1:95-97. 10. National Institutes of Health Consensus Development Conference. The integrated approach to the management of pain. J Pain Symptom Manage 1987;2:35-44. I 1. Cleeland CS. The impact ofpain on the patient with cancer. Cancer 1984;54:2635-2641. 12. Cleeland CS, Cleeland I_~l, Dar R, Rinehardt LC. Factors influencing physician management of cancer pain. Cancer 1986;58:796-800. 13. Von RoennJH, Cleeland CS, Gorin R, Pandya KJ. Results of physician's attitude toward cancer pain

16. Champ AD. The knowledge, attitudes and experience of medical personnel treating pain in the terminally ill. Mt SinaiJ Med (NY) 1978;45:561-580. 17. Morgan JP, Pleet DL. Opiophobia in the US: The undertreatment of severe pain. In" Morgan JP, Kagan DV, eds. Society and medication: conflicting signs for prescribers and patients. Lexington, MA: Lexington Press, 1983:313-326. 18. Weis OF, Sriwatanakul K, AflozaJL, Weintraub M, Lasagna L Attitudes of patients, housestaff, and nurses toward postoperative analgesia care. Anesth Analg 1983;62:70-74. 19. Sriwatanakul K, V~reis OF, Alloza jL, Kelvle W, Weintraub M, Lasagna L. Analysis of narcotic analgesic usage in the treatment of postoperative pain. JAMA 1983;250:926-929. 20. Wcissman DE, DahlJL. Attitudes about cancer pain: a survey of Wisconsin's first-year medical students.J Pain Symptom Manage 1990;5:345-349. 21. EUiott TE, Elliott BA. Physician acquisition of cancer pain management knowledge. J Pain Symptom Manage 1991 ;6:224-229. 22. Fishbein M, Ajzen I. Belief, attitude, intention and behavior. An introduction to theory and research. Reading, MA: AddisonAVesley, 1975.