Predictors of prescription of morphine for severe cancer pain by physicians in Korea

Y. H. Yun1, S. M. Park1, K. Lee2, Y. J. Chang3, D. S. Heo4, S.-Y. Kim5, Y. S. Hong6 and B. Y. Huh2,*

1 Quality of Cancer Care Branch, Research Institute and Hospital, National Cancer Center, Goyang, Gyeonggi; 2 Department of Family Medicine, Seoul National University, Seoul; 3 Cancer Information Branch, Research Institute, National Cancer Center, Goyang, Gyeonggi; 4 Department of Internal Medicine, Seoul National University Hospital, Seoul; 5 Department of Internal Medicine, Kyunghee University Hospital, Seoul; 6 Department of Internal Medicine, The Catholic University of Korea, Seoul, Korea

* Correspondence to: Dr B. Y. Huh, Department of Family Medicine, Seoul National University Hospital, 28 Yungun-dong, Jongro-gu, Seoul, Korea. Tel: +82-2–760–3351; Email: bongyul{at}plaza.snu.ac.kr


    Abstract
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Background:: This study was undertaken to identify predictors of the prescription of strong opioids, which are important for the management of severe cancer pain, by Korean physicians.

Methods:: A questionnaire based on a hypothetical case designed to assess the prescription of morphine by physicians was administered to 800 specialists in the Korea Cancer Association, of whom 147 (18.4%) responded, and to 2200 specialists in the Korean Academy of Family Medicine, of whom 388 (17.6%) responded. We used a multidimensional approach to identify the predictors of prescription of morphine by physicians.

Results:: In the hypothetical case scenario, only 16.5% of the respondents stated that they would prescribe morphine for severe cancer pain. Multiple logistic regression analysis showed that physicians with a positive attitude regarding opioid addiction [odds ratio (OR) 2.62; 95% confidence interval (CI) 1.54–4.46], experience of pain assessment (OR 2.09; 95% CI 1.13–3.87), recent residency training (OR 2.27; 95% CI 1.30–4.0) and positive self-evaluation as an oncology specialist (OR 2.60; 95% CI 1.41–4.78) were more likely to prescribe morphine. None of the 13 variables in the knowledge dimension significantly predicted prescription of morphine for severe cancer pain.

Conclusions:: The results of the survey suggest that we need to develop strategies to develop a positive attitude toward opioids, to increase experience in pain assessment and to improve cancer pain management training among Korean physicians.

Key words: cancer pain, morphine, physician, prescription


    Introduction
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Pain is one of the most common and disturbing symptoms in patients with advanced cancer [1Go–5Go]. When cancer pain management (CPM) guidelines are utilized, 80–90% of cancer-related pain can be controlled by available drugs and other interventions [6Go–8Go]. However, it has been estimated that 50–60% of cancer patients with pain do not receive adequate analgesic therapy [4Go, 9Go, 10Go]. Despite extensive progress in the scientific understanding of pain and widely disseminated cancer pain guidelines, CPM still remains far below the desired standard and patients continue to suffer needlessly [11Go–13Go].

Even though approximately 10 000 new cases of cancer are diagnosed each year in Korea [14Go], and cancer is the leading cause of death, no national policy for the improvement of cancer pain relief, such as national clinical guidelines or training programs for managing cancer pain, has been established. In addition, supplies of various opioids are limited, as is insurance coverage for them.

Although countries differ in training requirements, legal restrictions and drug availability, recent studies of physicians' attitudes have found that the most important barriers to appropriate CPM are problems related to the physician, including inadequate knowledge of CPM, inadequate pain assessment experience, excessive concern about addiction, tolerance and the side-effects of pain medications, underestimation of the number of patients in pain and the low priority of pain management in cancer care [5Go, 13Go, 15Go–21Go]. In addition, many studies identified predictors associated with less frequent morphine prescription [15Go], earlier intervention with maximum analgesic therapy [15Go, 16Go, 20Go], a reluctance to prescribe morphine and the physician's knowledge of cancer pain [13Go, 18Go].

To improve CPM in Korea, we believe that it is important to identify the predictors of prescription of morphine for severe cancer pain by physicians. We endeavoured to identify these predictors by conducting a national survey in based on a hypothetical case scenario.


    Methods
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Subjects
Survey forms and postage-paid return envelopes were mailed to physician members of the Korea Cancer Association and the Korean Academy of Family Medicine. To ensure participant anonymity no names were put on the forms.

We received answers from 147 (18.4%) of the 800 specialists in the Korea Cancer Association at university or general hospitals and 388 (17.6%) of the 2200 specialists in the Korean Academy of Family Medicine. The overall response rate was 17.8% and 514 valid questionnaires were analyzed. Specialties included family medicine (71.4%), internal medicine (13.8%), surgery (5.3%), radiation oncology (3.3%) and others (6.3%).

Questionnaire
To evaluate physician attitudes to the prescription of morphine for severe cancer pain, we included the following hypothetical scenario in the questionnaire.

A 40-year-old man is hospitalized with severe untreated back pain of more than 1 month duration attributable to bone metastases without vertebral collapse. He weighs 70 kg; he has no cardiovascular or respiratory problems and his prognosis is <12 months. He has no history of medication allergies and is opioid naive. What would be your recommendation for an initial analgesic drug for the patient?

Physicians who indicated that they would use a strong opioid such as morphine were regarded as having answered the question correctly and in keeping with appropriate CPM practices.

The instrument, which comprised components of previously reported questionnaires [16Go, 18Go], was designed to assess participating physicians on situational factors, attitudes, knowledge and personal experience (Table 1). Attitude questions used a scale, ranging from 1 (not at all concerned) to 5 (extremely concerned). Knowledge questions used a Likert scale, with 1 being ‘strongly agree’ and 5 being ‘strongly disagree’. We considered a desired response to be one that indicated few knowledge deficits and appropriate attitudes.


View this table:
[in this window]
[in a new window]
 
Table 1. Multidimensional assessment to identify the predictors of the prescription of morphine for severe cancer pain and the outcome variables

 
Statistical analysis
Descriptive statistics were reported for each response. Because not all respondents answered every question completely, we included the numbers that consisted of the basis for analysis of each item with the reported responses.

We used a three-step multidimensional approach to identify the predictors of prescription of morphine for severe cancer pain. In the first step, for the categorical variables within each dimension, we used {chi}2 tests to screen the predictors significantly associated with the prescription of morphine In the second step, to identify potentially important predictors, we performed multiple logistic analyses with one-dimensional approaches. We then selected variables with significant associations for one-dimensional multiple stepwise logistic regression analysis. In the last step, to identify potentially important predictors, we performed multiple logistic analyses with a multidimensional approach from the second step. We used SPSS program version 11.0 and defined statistical significance as P <0.05.


    Results
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Physician characteristics
The characteristics of the physicians are summarized in Table 2. Their mean age was 45.3 years, and most were men. Their mean length of clinical practice was 19.2 years, and most had no experience of using the Visual Analog Scale (VAS) or the Numeric Rating Scale (NRS).


View this table:
[in this window]
[in a new window]
 
Table 2. Characteristics of the study subjects

 
Physician practices regarding cancer pain management
Using the hypothetical case scenario of severe cancer pain, 16.5% of the respondents said that they would prescribe a strong opioid such as morphine, 53.8% said that they would prescribe codeine or a non-steroidal analgesic and 13.2% said that they would prescribe demerol.

Univariate analysis and one-dimensional predictors of prescription of morphine for severe cancer pain
In univariate analysis for situational dimension, age, practice setting, medical specialty, years in medical specialty and self-evaluation as an oncology specialist were significantly related to prescription of morphine for severe cancer pain (Table 3). One-dimensional multiple regression analysis with these five variables suggested that years in medical specialty, self-evaluation as an oncology specialist and practice setting were independent predictors (Table 4).


View this table:
[in this window]
[in a new window]
 
Table 3. Univariate analysis of clinically significant predictors of prescription of morphine for severe cancer pain in situational and experience dimensions

 

View this table:
[in this window]
[in a new window]
 
Table 4. One-dimensional multiple logistic regression analyses of clinically significant predictors of prescription of morphine for severe cancer pain

 
Univariate analysis for experience dimensions found that physicians significantly more likely to prescribe a strong opioid for severe cancer pain were those more experienced in using a pain scale or the World Health Organisation (WHO) three-step ladder and those with a higher percentage of cancer patients in pain (Table 3). Multiple regression analyses showed that only the use of a pain scale was a statistically significant predictor of the prescription of an opioid analgesic (Table 4).

Univariate analysis for attitude dimension showed that factors significantly related to prescription of morphine included regarding the patient as being the best judge of pain severity, lack of concern about the patient or a family member becoming addicted to opioids and earlier intervention with maximum analgesic therapy (Table 5). Multiple regression analysis also showed that these variables, except for lack of concern about a family member becoming addicted to opioids, were statistically significant predictors (Table 4).


View this table:
[in this window]
[in a new window]
 
Table 5. Univariate analysis of clinically significant predictors of prescription of morphine for severe cancer pain in knowledge and attitudes dimensions

 
In univariate analysis for the knowledge dimension, eight variables were significantly related to prescription of morphine (Table 5). However, in one-dimensional multiple logistic regression analysis using these variables, the only independent predictor was preference for a regular dose schedule (Table 4).

Multidimensional multivariate analyses of prescription of morphine for severe cancer pain
To conduct the final multidimensional logistic regression analysis, we combined the eight significant variables identified by one-dimensional multivariate analysis (Table 6). The three-step multidimensional multiple logistic analyses showed that four categories of physicians would prescribe a strong opioid for severe cancer pain: in the situation dimension, physicians with <10 years in their medical specialty and those who thought of themselves as oncology specialists; in the experience dimension, those who had used a pain scale; in the attitude dimension, those rarely concerned about patient addiction to opioids. No variable in the knowledge dimension significantly predicted prescription of morphine for severe cancer pain.


View this table:
[in this window]
[in a new window]
 
Table 6. Multidimensional multiple stepwise logistic regression analysis of clinically significant predictors of prescription of morphine for severe cancer pain

 

    Discussion
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
This study, in which we surveyed family physicians and oncologists in Korea, showed that most physicians engaged in suboptimal CPM practices, a finding in agreement with previous studies [13Go, 17Go, 19Go]. There have been few reports regarding the influence of demographics, experience and attitude variables on physician pain management practices. This study shows that a positive attitude regarding opioid addiction, experience in pain assessment, recent residency training and self-evaluation as an oncology specialist were positive predictors of prescription of morphine for severe cancer pain.

In response to a hypothetical severe cancer pain scenario, only 16.5% of respondents stated that they would prescribe a strong opioid such as morphine. Almost half the sample would use only codeine or non-steroidal analgesics, and almost a quarter would wait until the patient had end-stage disease. These findings were unfavorable compared with those of the Eastern Cooperative Oncology Group [17Go], the Radiation Therapy Oncology Group [20Go] and an Israeli study [13Go], in which 41–52% of physicians stated that they would provide an opioid for the treatment of severe cancer pain. Furthermore, in contrast with previous surveys [13Go, 16Go, 17Go, 20Go], which showed that few physicians were concerned with the possibility of opioid addiction, we found that this was a concern with many family physicians and oncologists in Korea.

Physicians who had received training more recently were more likely to prescribe morphine for severe cancer pain. Other studies have also found that younger physicians had better CPM concepts [16Go, 17Go, 22Go]. In addition, self-evaluation as an oncology specialist was associated with prescription of morphine for severe cancer pain, but specialty and the number of cancer patients treated had no influence. Therefore our results indicated that recent training and more experience in specialized oncology were more important than physician age, specialty and patient volume [15Go, 22Go].

Multidimensional multiple regressions showed that inappropriate concerns regarding patient addiction were associated with inadequate prescription of morphine for severe cancer pain. In contrast, while regarding the patient as the best judge of pain severity and being willing to intervene early with a maximum dose of analgesics were associated with appropriate prescription of morphine for severe cancer pain in one-dimensional multiple regression, these factors lost their independence in multidimensional multiple regression. It should be noted that univariate analysis showed that knowledge regarding tolerance, adverse effects, risk of addiction, adjuvant and effectiveness of opioids were not significantly associated with a physician's ability to respond correctly to questions on the management of severe cancer pain. Although univariate analysis found that treatment of pain by the clock was a predictor in the knowledge dimension, it lost independence in multivariate analyses. Thus, overall, our results support previous findings that inappropriate attitudes regarding opioid addiction are a more important barrier to adequate CPM than physician knowledge about CPM [17Go], and that improved basic knowledge is not sufficient to improve cancer pain practice [8Go, 11Go, 15Go].

We also found that experience with pain assessment was a strong predictor of prescription of morphine for severe cancer pain, confirming that poor pain assessment is the single most important barrier to optimal CPM [13Go, 18Go, 20Go]. Moreover, our findings and those of other studies indicate that new effective strategies and professional education regarding optimal CPM must address the attitudes of physicians toward opioid addiction [16Go, 17Go, 19Go, 23Go, 24Go]. Because physician judgment plays a major role in the administration of morphine for severe cancer pain, incorporation of the principles of CPM into therapeutic protocols and residency training, improvement of health professionals' evaluation of pain as a vital sign during routine medical practice, and monitoring pain assessment and relief as indicators of quality of care may improve CPM [19Go, 20Go, 23Go–25Go].

There were limitations to our study. The first was the low response rate (17.8%), perhaps indicating that physicians who responded were more interested in the issue of pain control and thus introducing a self-selection bias. Therefore our results may underestimate the percentage of physicians who appropriately prescribed morphine for cancer pain control. Secondly, although it may be regarded as questionable to use a hypothetical case scenario to assess prescription of morphine by physicians for severe cancer pain, this technique has been used elsewhere to evaluate physician pain management practice [16Go–18Go].

In conclusion, our results revealed that, in Korea, positive attitudes toward opioid addiction, experience in pain assessment, more recent training and self-evaluation as an oncology specialist were predictors of prescription of morphine for severe cancer pain and were more important than knowledge of CPM. This suggests that we need to develop strategies to develop a positive attitude toward opioids, to increase experience in pain assessment and to improve CPM training among Korean physicians.


    Acknowledgements
 
This study was supported by a grant from the 2001 Korean National Cancer Control Program by the Ministry of Health and Welfare, Korea.

Received for publication October 3, 2004. Revision received January 9, 2005. Accepted for publication January 10, 2005.


    References
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
1. Foley KM. The treatment of cancer pain. N Engl J Med 1985; 313: 84–95.[Abstract]

2. Greenwald HP, Bonica JJ, Bergner M. The prevalence of pain in four cancers. Cancer 1987; 60: 2563–2569.[ISI][Medline]

3. Daut RL, Cleeland CS. The prevalence and severity of pain in cancer. Cancer 1982; 50: 1913–1918.[ISI][Medline]

4. Yun YH, Heo DS, Lee IG et al. Multicenter study of pain and its management in patients with advanced cancer in Korea. J Pain Symptom Manage 2003; 25: 430–437.[CrossRef][ISI][Medline]

5. Rawal N, Hylander J, Arner S. Management of terminal cancer pain in Sweden: a nationwide survey. Pain 1993; 54: 169–179.[CrossRef][ISI][Medline]

6. World Health Organization. Cancer Pain Relief. Geneva: World Health Organisation 1986.

7. Jacox A, Carr DB, Payne R. New clinical-practice guidelines for the management of pain in patients with cancer. N Engl J Med 1994; 330: 651–655.[Free Full Text]

8. Jacox A, Carr DB, Payne R et al. Management of Cancer Pain: Clinical Practice Guideline No. 9. Agency for Health Care Policy and Research (AHCPR) Publication 94–0592. Rockville, MD: AHCPR, US Department of Health and Human Services, Public Health Service, 1994.

9. Cleeland CS, Gonin R, Hatfield AK et al. Pain and its treatment in outpatients with metastatic cancer. N Engl J Med 1994; 330: 592–596.[Abstract/Free Full Text]

10. Larue F, Colleau SM, Brasseur L et al. Multicentre study of cancer pain and its treatment in France. BMJ 1995; 310: 1034–1037.[Abstract/Free Full Text]

11. Weissman DE. Cancer pain education for physicians in practice: establishing a new paradigm. J Pain Symptom Manage 1996; 12: 364–371.[CrossRef][ISI][Medline]

12. Foley KM. Pain relief into practice: rhetoric without reform. J Clin Oncol 1995; 13: 2149–2151.[Free Full Text]

13. Sapir R, Catane R, Strauss-Liviatan N et al. Cancer pain: knowledge and attitudes of physicians in Israel. J Pain Symptom Manage 1999; 17: 266–276.[CrossRef][ISI][Medline]

14. Central Cancer Registry Center in Korea, Ministry of Health and Welfare. Annual Report of the Central Cancer Registry in Korea. Seoul: Central Cancer Registry Center in Korea 2004.

15. Elliott TE, Murray DM, Elliott BA et al. Physician knowledge and attitudes about cancer pain management: a survey from the Minnesota Cancer Pain Project. J Pain Symptom Manage 1995; 10: 494–504.[CrossRef][ISI][Medline]

16. Larue F, Colleau SM, Fontaine A et al. Oncologists and primary care physicians' attitudes toward pain control and morphine prescribing in France. Cancer 1995; 76: 2375–2382.[ISI][Medline]

17. Von Roenn JH, Cleeland CS, Gonin R et al. Physician attitudes and practice in cancer pain management: a survey from the Eastern Cooperative Oncology Group. Ann Intern Med 1993; 119: 121–126.[Abstract/Free Full Text]

18. Ger LP, Ho ST, Wang JJ. Physicians' knowledge and attitudes toward the use of analgesics for cancer pain management: a survey of two medical centers in Taiwan. J Pain Symptom Manage 2000; 20: 335–344.[CrossRef][ISI][Medline]

19. Weinstein SM, Laux LF, Thornby JI et al. Physicians' attitudes toward pain and the use of opioid analgesics: results of a survey from the Texas Cancer Pain Initiative. South Med J 2000; 93: 479–487.[ISI][Medline]

20. Cleeland CS, Janjan NA, Scott CB et al. Cancer pain management by radiotherapists: a survey of radiation therapy oncology group physicians. Int J Radiat Oncol Biol Phys 2000; 47: 203–208.[CrossRef][ISI][Medline]

21. Elliott TE, Elliott BA. Physician attitudes and beliefs about use of morphine for cancer pain. J Pain Symptom Manage 1992; 7: 141–148.[ISI]

22. Cleeland CS, Cleeland LM, Dar R et al. Factors influencing physician management of cancer pain. Cancer 1986; 58: 796–800.[ISI][Medline]

23. Weissman DE, Griffie J, Gordon DB et al. A role model program to promote institutional changes for management of acute and cancer pain. J Pain Symptom Manage 1997; 14: 274–279.[CrossRef][ISI][Medline]

24. Bookbinder M, Coyle N, Kiss M et al. Implementing national standards for cancer pain management: program model and evaluation. J Pain Symptom Manage 1996; 12: 334–347.[CrossRef][ISI][Medline]

25. American Pain Society Quality of Care Committee. Quality improvement guidelines for the treatment of acute pain and cancer pain. JAMA 1995; 274: 1874–1880.[Abstract]





This Article
Abstract
Full Text (PDF)
All Versions of this Article:
16/6/966    most recent
mdi180v1
E-letters: Submit a response
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Disclaimer
Request Permissions
Google Scholar
Articles by Yun, Y. H.
Articles by Huh, B. Y.
PubMed
PubMed Citation
Articles by Yun, Y. H.
Articles by Huh, B. Y.