User login
METHODS: A survey was mailed to primary care physicians in the University of California, San Francisco/Stanford Collaborative Research Net- work. This survey contained questions regarding treatment in response to 3 case vignettes, the use of opioids for CNMP in general, and the demographic characteristics of the physicians.
RESULTS: Among 230 physicians surveyed, 161 (70%) responded. Two percent of the respondents were never willing to prescribe schedule III opioids (eg, acetaminophen with codeine) as needed for patients with CNMP that persisted unchanged after exhaustive evaluation and attempts at treatment. Thirty-five percent were never willing to prescribe schedule II opioids (eg, sustained-release morphine) on an around-the-clock schedule for these patients. The most significant predictor of willingness to prescribe opioids for patients with CNMP was a lower level of concern about physical dependence, tolerance, and addiction.
CONCLUSIONS: Primary care physicians are willing to prescribe schedule III opioids as needed, but many are unwilling to use schedule II opioids around the clock for CNMP. Individual prescribing practices vary widely among primary care physicians. Concerns about physical dependence, tolerance, and addiction are barriers to the prescription of opioids by primary care physicians for patients with CNMP.
Opioids are effective analgesics that are widely accepted as therapy for cancer pain and pain related to other terminal illnesses.1-2 However, the use of opioids to treat chronic nonmalignant pain (CNMP) is controversial.3-8 Few clinical studies of opioids in the alleviation of CNMP have been conducted, and most have been small, retrospective, uncontrolled, or focused on patients seen in referral settings.9-21 Together these studies suggest that opioids may benefit certain patients with CNMP, though the results have not been conclusive.
In clinical practice the absence of definitive data on the risks and benefits of opioids for CNMP presents a dilemma. Decisions about potency, frequency, and duration of treatment must be made without the benefit of evidence-based guidelines and with the knowledge that state medical boards or other legal authorities may scrutinize opioid prescriptions. We conducted this study to learn more about attitudes, prescribing practices, and factors associated with the willingness of primary care physicians to prescribe opioids for their patients with CNMP.
Methods
Sample
The University of California, San Francisco/Stanford Collaborative Research Network (CRN) is a practice-based research network composed predominantly of family physicians practicing in Northern and Central California. In 1997 the CRN conducted this survey of all 230 primary care physician members who were not involved in designing our study. Up to 2 mailed reminders and 3 telephone calls were made to initial nonresponders to improve the response rate.
Instrument
The survey instrument was developed through a collaborative process involving 7 volunteer physicians from the CRN. It was pilot-tested and refined using focus groups of practicing non-CRN primary care physicians.
On the first page of the survey, CNMP was defined as pain lasting longer than 6 months that was not related to cancer or another condition expected to end a patient’s life within 6 months. The survey included 3 clinical vignettes Table 1 designed to evoke responses to a variety of patient characteristics, such as medical history, age, sex, and socioeconomic status. Each vignette was followed by a set of specific questions. The survey also contained questions unrelated to the vignettes, regarding general attitudes toward opioids and opioid prescribing practices. We asked about documentation practices, referral resources, and familiarity with state guidelines. The respondents were also queried about personal, patient, and practice characteristics.
Statistical Analysis
We conducted analyses using SAS software.22 Means and standard deviations (SDs) for continuous variables, and frequency distributions for categorical variables, were calculated to summarize physician respondent characteristics, estimates of the characteristics of their caseloads, and summaries of their responses to questions about the clinical vignettes. We used correlation coefficients to examine the strength of relationships between attitudes and practice. The results of the correlation coefficients were used to choose a set of independent variables that were most predictive of willingness to prescribe opioids for CNMP.
We examined with stepwise linear regression the association between willingness to prescribe opioid medications and specific physician characteristics, including year of medical school graduation, size of patient caseload, and concerns about physical dependence, tolerance, addiction, side effects, regulatory scrutiny, and diversion for illegal use. In selecting the final set of variables for the stepwise linear regression predicting willingness to prescribe opioids for CNMP, we found that concern about physical dependence, tolerance, and addiction were highly intercorrelated. Among these variables, concern about physical dependence was the most consistently predictive of willingness to prescribe opioids for CNMP. When concern about physical dependence was entered into stepwise models the variables measuring concern about tolerance and addiction dropped out. Therefore, we chose to use concern about physical dependence as a proxy for measuring generalized concerns about all 3 concerns taken together.
We constructed 3 models to predict willingness to prescribe opioids. In Model 1 the dependent variable to designate willingness to prescribe was constructed from the sum of responses to the 5-point Likert-scaled question asked after each of the 3 vignettes: “If the pain persisted unchanged, would you prescribe opioids for this patient on a long-term basis?” In Model 2 the dependent variable was defined according to the range of agreement on a 5-point Likert scale with the following statement: “For patients with CNMP that persists unchanged after exhaustive evaluation and attempts at treatment, I am willing to prescribe opioids not requiring triplicates (such as Tylenol with codeine) on an as-needed basis.” In Model 3, the dependent variable was defined according to the range of agreement with the following statement on a 5-point Likert scale: “For patients with CNMP that persists unchanged after exhaustive evaluation and attempts at treatment, I am willing to prescribe opioids requiring triplicates (eg, fentanyl patch, methadone, or sustained-release morphine) on a fixed, around-the-clock basis.”
Results
Physician and Practice Characteristics
A total of 161 of 230 physicians (70%) completed the survey. The demographic characteristics of the respondents are presented in Table 2. Table 3 shows physician estimates of patient demographics in their practices. As a group the CRN physicians were mostly white men, but they care for an ethnically, financially, and age-diverse population. The large SDs in Table 3 reflect the wide variety of practice types included in the CRN membership.
Physicians reported seeing an average of 280 patients (SD=157), including 18 CNMP patients (SD=26), per month. An average of 7 patients (SD=8) with CNMP were prescribed opioid analgesics per month, and 90% of the physicians reported prescribing opioids for CNMP at least once a month. The wide SDs again reflect broad variation in the number of patients seen, the number of patients encountered with CNMP, and the number of patients treated with opioid analgesics by different physicians.
Attitudes and Practices of Physicians
Only 15% of respondents agreed with the statement: “I enjoy working with patients who have CNMP.” However, only 15% also felt that daily opioids have no place in the treatment of CNMP. Only 7% agreed with the statement: “I never prescribe opioids for CNMP.”
Many physicians wait for their patients to bring up the subject of opioid treatment, as indicated by the fact that 41% of the respondents agreed that “most of my patients who get opioid prescriptions from me for CNMP requested an opioid before I suggested their use.” In addition, 37% responded that they rarely or never are the first physician to prescribe opioids to their patients with CNMP, possibly waiting for other specialists to take the initiative.
The responses to questions about the 3 clinical vignettes are presented in Table 4. Nearly all physicians felt that the vignettes were realistic, and most believed they were knowledgeable about evaluation and treatment for these patients. However, each case generated substantial variation of opinion with regard to the level of optimism about being able to help the patient, the need for specialty referral, and the willingness to treat with opioids. For each vignette respondents were generally more concerned about physical dependence, tolerance, and addiction than they were about diversion for illegal use, regulatory scrutiny, or side effects. However, physicians’ level of concern about each of these outcomes varied substantially for each vignette.
The physicians were asked general questions about situations in which they would never prescribe opioids. Although none of the respondents said that they had a policy of refusing opioids to patients aged older than 65 years, 19% said they would never prescribe opioids to a child younger than 18 years. In addition, 16% said they would never prescribe opioids to a previous substance abuser, and 42% said they would never prescribe opioids to a current substance abuser, even if recommended by an appropriate specialist.
Also, respondents expressed an increased reluctance to prescribe opioids to CNMP patients as the frequency and potency of the medication was increased. Although only 2% of physicians said they would never prescribe low-potency (schedule III) opioids on an as-needed basis, 35% said they would never prescribe high-potency (schedule II) opioids around the clock, even after exhaustive evaluation and attempts at treatment.
In addition, the willingness of respondents to prescribe opioids varied according to the medical condition being treated. Forty-two percent of respondents said they would never prescribe long-acting schedule II opioids to a patient with post-herpetic neuralgia; 57% would never prescribe them for chronic low back pain; and 75% would never prescribe them for chronic daily headache.
We asked about the use of specialists to assist in the evaluation and treatment of patients who may benefit from opioid treatment for CNMP. Fifty-two percent of the physicians reported always or usually requiring their patients to undergo evaluation by a specialist before prescribing opioids on an ongoing basis for CNMP. Yet only 55% felt they had adequate consultation and referral resources to assist with patients who have CNMP. In addition, only 29% felt they had adequate consultation and referral resources in their communities to assist them with patients who might be abusing or selling opioid prescriptions.
Familiarity with State Prescribing and Documentation Guidelines
In 1994, the Medical Board of California issued guidelines for prescribing opioids for CNMP that were designed to standardize referral and documentation practices and to reduce fear of regulatory scrutiny among physicians who prescribe opioids for CNMP. The guidelines were mailed to all licensed physicians in the state on 3 occasions between 1994 and 1996.23 We found that 39% of respondents remembered reading the guidelines 1 year after the third mailing. We also found that physicians varied in their self-reported compliance with recommended documentation practices. Ninety percent said they always or usually document a history and physical examination before prescribing opioids, and 86% document periodic reassessment of chronic pain. However, only 60% said that they always or usually document rules of use and misuse of opioid medications; 45% document treatment objectives; and 24% document informed consent. When asked about regulatory scrutiny, 40% of physicians agreed that fear of legal investigation tempers their use of opioids for patients with CNMP.
Predictors of Willingness to Prescribe Opioids
Three models were postulated to clarify the determinants of willingness to prescribe opioids for CNMP. The results of these analyses are presented in Table 5. The stepwise linear regression for each model generated a value for each variable (R2) that represents the proportion of the variance that can be explained by the given variable.
In all 3 models lower levels of concern about physical dependence in response to the vignettes were associated with greater willingness to prescribe opioids. Other variables that were significant predictors of willingness to prescribe opioids in 1 or more models were more recent graduation from medical school, enjoyment in working with chronic pain patients, less fear of regulatory scrutiny, and fewer total patients seen per month.
Discussion
Nearly all the physicians in our sample were willing to treat certain CNMP patients with schedule III opioids on an as-needed basis. However, a third of these physicians said they never use the more potent long-acting schedule II opioids for CNMP. There was also substantial disagreement about which patients would benefit from opioids and which might be likely to suffer adverse effects.
Concern about physical dependence appears to be among the most important barriers to the use of opioids for patients with CNMP. Whether this is always an appropriate concern is debatable. For example, in the case of using schedule III opioids on an as-needed basis, the lack of continuous exposure should limit the risk of physical dependence.
Our finding that physician concerns about physical dependence, tolerance, and addiction were highly intercorrelated raises the possibility that many physicians believe, correctly or incorrectly, that these 3 conditions are closely related effects of opioids. It is also possible that physicians are unclear about what distinguishes one of these outcomes from another. More research is needed to determine the root of physician concerns about physical dependence, tolerance, and addiction. Although all 3 of these outcomes can result when opioids are used around the clock, they nonetheless do not always occur together or necessarily all have equally serious implications when they occur.24 Only a slight majority of respondents felt that they had adequate consultation and referral resources in their community to assist with patients who have CNMP. Primary care physicians may benefit from more information about pain management resources in their communities. In addition, communities without these resources may benefit from the development of pain management centers that can assist primary care physicians with patients who suffer from CNMP.
More recent graduation from medical school was a predictor of increased willingness to prescribe opioids. Recently trained physicians may be more likely to have been exposed to an environment of more liberal use of opioids for CNMP. Conversely, the decreased willingness of more experienced physicians to prescribe opioids may be influenced by their clinical experiences with the complications of opioid use.
Fear of regulatory scrutiny also appeared to limit willingness to prescribe as-needed low-potency schedule III opioids. Recent laws and guidelines have attempted to reduce both the risk and fear of regulatory scrutiny when opioids are prescribed for chronic pain.23,25,26 However, it is not clear whether awareness of these guidelines would increase or decrease physician concern with regard to regulatory scrutiny, since many physicians reported that their documentation standards are not up to those recommended. In addition, we found no differences in willingness to prescribe opioids based on awareness of the guidelines in California.
We found that physicians who saw more patients were less likely to use more potent opioids. In California, schedule II opioids must be prescribed using triplicate forms purchased from the state. Physicians with high-volume practices may be less inclined to prescribe high-potency opioids because of the time required to complete triplicate forms. Other possible explanations are that these physicians have had more adverse experiences with the use of opioids for CNMP or that they feel more vulnerable to regulatory scrutiny because of their increased volume of patients who might receive opioid prescriptions.
We found that most physicians did not enjoy working with chronic pain patients, and this lack of enjoyment with treating CNMP was a significant barrier to willingness to prescribe opioids in 2 of our 3 models. More investigation of why most physicians do not enjoy working with these patients could further illuminate barriers to the use of opioids for CNMP.
Limitations
There are several limitations to our study. First, the physicians surveyed may not be representative of all practicing primary care physicians. However, CRN physicians are quite similar in many characteristics to family physicians practicing in California.27 Barriers to prescribing opioids in California may also be different from barriers faced by physicians in other parts of the country, so our results may not be easily generalized to other geographic regions. In addition, the data were generated by self-report, and actual practices may differ.
However, our findings are consistent with a 1991 survey of 90 Wisconsin physicians that concluded that concerns about addiction outweigh concerns about regulatory scrutiny for most physicians.28 A national survey of 1912 physicians from multiple specialties found, as we did, a high level of intercorrelation among physician concerns about physical dependence, tolerance, and addiction.29 Unfortunately, that study was not designed to elucidate the relative importance of factors that determine a physician’s willingness to prescribe opioids.
Another limitation of our study is that none of the models we postulated could explain more than a small proportion (24%) of the total variance in the willingness to prescribe opioids for CNMP. Clearly other factors, unmeasured in the current study, also influence physicians’ willingness to prescribe opioids for CNMP. For example, in a study of the prescribing habits at a referral center in Seattle, pain specialists were significantly influenced in their willingness to prescribe opioids for CNMP by a set of pain behaviors exhibited by the patient.30 These behaviors included distorted ambulation or posture, negative affect, facial and audible expressions of distress, and avoidance of activity. The nature of our study did not allow for such factors in our models of willingness to prescribe opioids, but these factors may be worthy of further investigation in direct observation studies of primary care.
Conclusions
Our results suggest that primary care physicians disagree about the relative risks and benefits of opioids in the treatment of individuals who suffer from CNMP. Also, these physicians function with limited reliable information or specialty resources to guide them in choosing which of these patients to treat with opioids. Concerns about addiction, tolerance, and physical dependence appear to be important barriers to the use of opioids by many physicians. More research is needed in primary care settings to determine appropriate uses for opioids in the treatment of CNMP and to further elucidate the concerns of physicians and barriers to more effective use.
Acknowledgments
This research was partially supported by grant #5D32PE19036-09 from the Department of Health and Human Services Health Resources Services Administration to support the establishment of a Department of Family Practice and by a grant from the California Academy of Family Physicians. We would like to acknowledge Dr Eric Sanford, Dr Lawrence Bruguera, Dr Charles Kano, Dr Joyce Hightower, and Dr Yeva Johnson for their assistance with data interpretation and preparation of this manuscript. In addition, we would like to acknowledge the work of Ms Catherine Brosnan and Ms Elizabeth Dito in assisting us with coordination and data collection for our study.
Related resources
- American Pain Society http://www.ampainsoc.org/
- American Academy of Pain Management http://www.aapainmanage.org
- American Pain Foundation http://www.painfoundation.org
1. Jadad AR, Browman GP. The WHO analgesic ladder for cancer pain management: stepping up the quality of its evaluation. JAMA 1995;274:1870-73.
2. Levy MH. Pharmacologic treatment of cancer pain. N Eng J Med 1996;46:128-38.
3. Schofferman J. Long-term use of opioid analgesics for the treatment of chronic pain of nonmalignant origin. J Pain Symptom Manage 1993;8:279-88.
4. Large RG, Schug SA. Options for chronic pain of nonmalignant origin: caring or crippling. Health Care Anal 1995;3:5-11.
5. Portenoy RL. Opioid therapy for chronic nonmalignant pain: a review of the critical issues. J Pain Symptom Manage 1996;11:203-17.
6. Kyriaki D, Pither CE, Wessely S. Medication misuse, abuse and dependence in chronic pain patients. J Psychosom Res 1997;43:497-504.
7. McQuay H. Opioids in pain management. Lancet 1999;353:2229-32.
8. Parrott T. Using opioid analgesics to manage noncancer pain in primary care. J Am Board Fam Pract 1999;12:293-306.
9. Portenoy RK, Foley KM. Chronic use of opioid analgesics in non-malignant pain: report of 38 cases. Pain 1986;25:171-86.
10. Kjaersgaard-Andersen P, Nafei A, Skov O, Madsen F, et al. Codeine plus paracetamol versus paracetamol in longer-term treatment of chronic pain due to osteoarthritis of the hip: a randomised, double-blind, multicentre study. Pain 1990;43:309-18.
11. Zenz M, Strumpf M, Tryba M. Long-term opioid therapy in patients with chronic nonmalignant pain. J Pain Symptom Manage 1992;7:69-77.
12. Kell M, Musselman D. Methadone prophylaxis of intractable headaches: pain control and serum opioid levels. Am J Pain Manage 1993;3:7-14.
13. Arkinstall W, Sandler A, Groghnour B, Babul N, Harsanyi Z, Darke AC. Efficacy of controlled-release codeine in chronic non-malignant pain: a randomized, placebo-controlled clinical trial. Pain 1995;62:169-78.
14. Moulin DE, Iezzi A, Amireh R, et al. Randomised trial of oral morphine for chronic non-cancer pain. Lancet 1996;347:143-47.
15. Gardner-Nix JS. Oral methadone for managing chronic nonmalignant pain. J Symptom Pain Manage 1996;11:321-28.
16. Simpson RK, Edmondson EA, Constant CF, Collier C. Transdermal fentanyl for chronic low back pain. J Pain Symptom Manage 1997;14:218-24.
17. Haythornthwaite JA, Menefee LA, Quatrano-Piacentini AL, Pappagallo M. Outcome of chronic opioid therapy for non-cancer pain. J Pain Symptom Manage 1998;15:185-94.
18. Sheather-Reid RB, Cohen ML. Efficacy of analgesics in chronic pain: a series of n-of-1 studies. J Pain Symptom Manage 1998;15:244-52.
19. Jamison RN, Raymond SA, Slawsby EA, Nedeljkovic SS, et al. Opioid therapy for chronic noncancer back pain. Spine 1998;23:2591-600.
20. Ytterberg SR, Mahowald ML, Woods SR. Codeine and oxycodone use in patients with chronic rheumatic disease pain. Arthritis Rheum 1998;41:1603-12.
21. Watson CP, Babul N. Efficacy of oxycodone in neuropathic pain: a randomized trial in postherpetic neuralgia. Neurology 1998;50:1837-41.
22. SAS Institute Inc. SAS System for Microsoft Windows, release 6.12. Cary, NC: SAS Institute Inc; 1996.
23. Medical Board of California. Action report: new, easy guidelines on prescribing. 1994; 51:1.
24. Sees KL, Clark HW. Opioid use in the treatment of chronic pain: assessment of addiction. J Pain Symptom Manage 1993;8:257-64.
25. Clark H. Opioids, chronic pain, and the law. J Pain Symptom Manage 1993;8:297-306.
26. Hill CS. Government regulatory influences on opioid prescribing and their impact on the treatment of pain of nonmalignant origin. J Pain Symptom Manage 1996;11:287-98.
27. Croughan-Minihane MS, Thom DH, Petitti DB. Research interests of physicians in two practice-based primary care research networks. West J Med 1999;170:19-24.
28. Weissman DE, Joranson DE, Hopwood MB. Wisconsin physicians’ knowledge and attitudes about opioid analgesic regulations. Wis Med J 1991;12:671-75.
29. Turk DC, Brody MC, Okifuji AE. Physicians’ attitudes and practices regarding the long-term prescribing of opioids for non-cancer pain. Pain 1994;59:201-208.
30. Turk DC, Okifuji A. What factors affect physicians’ decisions to prescribe opioids for chronic noncancer pain patients? Clin J Pain 1997;13:330-36.
METHODS: A survey was mailed to primary care physicians in the University of California, San Francisco/Stanford Collaborative Research Net- work. This survey contained questions regarding treatment in response to 3 case vignettes, the use of opioids for CNMP in general, and the demographic characteristics of the physicians.
RESULTS: Among 230 physicians surveyed, 161 (70%) responded. Two percent of the respondents were never willing to prescribe schedule III opioids (eg, acetaminophen with codeine) as needed for patients with CNMP that persisted unchanged after exhaustive evaluation and attempts at treatment. Thirty-five percent were never willing to prescribe schedule II opioids (eg, sustained-release morphine) on an around-the-clock schedule for these patients. The most significant predictor of willingness to prescribe opioids for patients with CNMP was a lower level of concern about physical dependence, tolerance, and addiction.
CONCLUSIONS: Primary care physicians are willing to prescribe schedule III opioids as needed, but many are unwilling to use schedule II opioids around the clock for CNMP. Individual prescribing practices vary widely among primary care physicians. Concerns about physical dependence, tolerance, and addiction are barriers to the prescription of opioids by primary care physicians for patients with CNMP.
Opioids are effective analgesics that are widely accepted as therapy for cancer pain and pain related to other terminal illnesses.1-2 However, the use of opioids to treat chronic nonmalignant pain (CNMP) is controversial.3-8 Few clinical studies of opioids in the alleviation of CNMP have been conducted, and most have been small, retrospective, uncontrolled, or focused on patients seen in referral settings.9-21 Together these studies suggest that opioids may benefit certain patients with CNMP, though the results have not been conclusive.
In clinical practice the absence of definitive data on the risks and benefits of opioids for CNMP presents a dilemma. Decisions about potency, frequency, and duration of treatment must be made without the benefit of evidence-based guidelines and with the knowledge that state medical boards or other legal authorities may scrutinize opioid prescriptions. We conducted this study to learn more about attitudes, prescribing practices, and factors associated with the willingness of primary care physicians to prescribe opioids for their patients with CNMP.
Methods
Sample
The University of California, San Francisco/Stanford Collaborative Research Network (CRN) is a practice-based research network composed predominantly of family physicians practicing in Northern and Central California. In 1997 the CRN conducted this survey of all 230 primary care physician members who were not involved in designing our study. Up to 2 mailed reminders and 3 telephone calls were made to initial nonresponders to improve the response rate.
Instrument
The survey instrument was developed through a collaborative process involving 7 volunteer physicians from the CRN. It was pilot-tested and refined using focus groups of practicing non-CRN primary care physicians.
On the first page of the survey, CNMP was defined as pain lasting longer than 6 months that was not related to cancer or another condition expected to end a patient’s life within 6 months. The survey included 3 clinical vignettes Table 1 designed to evoke responses to a variety of patient characteristics, such as medical history, age, sex, and socioeconomic status. Each vignette was followed by a set of specific questions. The survey also contained questions unrelated to the vignettes, regarding general attitudes toward opioids and opioid prescribing practices. We asked about documentation practices, referral resources, and familiarity with state guidelines. The respondents were also queried about personal, patient, and practice characteristics.
Statistical Analysis
We conducted analyses using SAS software.22 Means and standard deviations (SDs) for continuous variables, and frequency distributions for categorical variables, were calculated to summarize physician respondent characteristics, estimates of the characteristics of their caseloads, and summaries of their responses to questions about the clinical vignettes. We used correlation coefficients to examine the strength of relationships between attitudes and practice. The results of the correlation coefficients were used to choose a set of independent variables that were most predictive of willingness to prescribe opioids for CNMP.
We examined with stepwise linear regression the association between willingness to prescribe opioid medications and specific physician characteristics, including year of medical school graduation, size of patient caseload, and concerns about physical dependence, tolerance, addiction, side effects, regulatory scrutiny, and diversion for illegal use. In selecting the final set of variables for the stepwise linear regression predicting willingness to prescribe opioids for CNMP, we found that concern about physical dependence, tolerance, and addiction were highly intercorrelated. Among these variables, concern about physical dependence was the most consistently predictive of willingness to prescribe opioids for CNMP. When concern about physical dependence was entered into stepwise models the variables measuring concern about tolerance and addiction dropped out. Therefore, we chose to use concern about physical dependence as a proxy for measuring generalized concerns about all 3 concerns taken together.
We constructed 3 models to predict willingness to prescribe opioids. In Model 1 the dependent variable to designate willingness to prescribe was constructed from the sum of responses to the 5-point Likert-scaled question asked after each of the 3 vignettes: “If the pain persisted unchanged, would you prescribe opioids for this patient on a long-term basis?” In Model 2 the dependent variable was defined according to the range of agreement on a 5-point Likert scale with the following statement: “For patients with CNMP that persists unchanged after exhaustive evaluation and attempts at treatment, I am willing to prescribe opioids not requiring triplicates (such as Tylenol with codeine) on an as-needed basis.” In Model 3, the dependent variable was defined according to the range of agreement with the following statement on a 5-point Likert scale: “For patients with CNMP that persists unchanged after exhaustive evaluation and attempts at treatment, I am willing to prescribe opioids requiring triplicates (eg, fentanyl patch, methadone, or sustained-release morphine) on a fixed, around-the-clock basis.”
Results
Physician and Practice Characteristics
A total of 161 of 230 physicians (70%) completed the survey. The demographic characteristics of the respondents are presented in Table 2. Table 3 shows physician estimates of patient demographics in their practices. As a group the CRN physicians were mostly white men, but they care for an ethnically, financially, and age-diverse population. The large SDs in Table 3 reflect the wide variety of practice types included in the CRN membership.
Physicians reported seeing an average of 280 patients (SD=157), including 18 CNMP patients (SD=26), per month. An average of 7 patients (SD=8) with CNMP were prescribed opioid analgesics per month, and 90% of the physicians reported prescribing opioids for CNMP at least once a month. The wide SDs again reflect broad variation in the number of patients seen, the number of patients encountered with CNMP, and the number of patients treated with opioid analgesics by different physicians.
Attitudes and Practices of Physicians
Only 15% of respondents agreed with the statement: “I enjoy working with patients who have CNMP.” However, only 15% also felt that daily opioids have no place in the treatment of CNMP. Only 7% agreed with the statement: “I never prescribe opioids for CNMP.”
Many physicians wait for their patients to bring up the subject of opioid treatment, as indicated by the fact that 41% of the respondents agreed that “most of my patients who get opioid prescriptions from me for CNMP requested an opioid before I suggested their use.” In addition, 37% responded that they rarely or never are the first physician to prescribe opioids to their patients with CNMP, possibly waiting for other specialists to take the initiative.
The responses to questions about the 3 clinical vignettes are presented in Table 4. Nearly all physicians felt that the vignettes were realistic, and most believed they were knowledgeable about evaluation and treatment for these patients. However, each case generated substantial variation of opinion with regard to the level of optimism about being able to help the patient, the need for specialty referral, and the willingness to treat with opioids. For each vignette respondents were generally more concerned about physical dependence, tolerance, and addiction than they were about diversion for illegal use, regulatory scrutiny, or side effects. However, physicians’ level of concern about each of these outcomes varied substantially for each vignette.
The physicians were asked general questions about situations in which they would never prescribe opioids. Although none of the respondents said that they had a policy of refusing opioids to patients aged older than 65 years, 19% said they would never prescribe opioids to a child younger than 18 years. In addition, 16% said they would never prescribe opioids to a previous substance abuser, and 42% said they would never prescribe opioids to a current substance abuser, even if recommended by an appropriate specialist.
Also, respondents expressed an increased reluctance to prescribe opioids to CNMP patients as the frequency and potency of the medication was increased. Although only 2% of physicians said they would never prescribe low-potency (schedule III) opioids on an as-needed basis, 35% said they would never prescribe high-potency (schedule II) opioids around the clock, even after exhaustive evaluation and attempts at treatment.
In addition, the willingness of respondents to prescribe opioids varied according to the medical condition being treated. Forty-two percent of respondents said they would never prescribe long-acting schedule II opioids to a patient with post-herpetic neuralgia; 57% would never prescribe them for chronic low back pain; and 75% would never prescribe them for chronic daily headache.
We asked about the use of specialists to assist in the evaluation and treatment of patients who may benefit from opioid treatment for CNMP. Fifty-two percent of the physicians reported always or usually requiring their patients to undergo evaluation by a specialist before prescribing opioids on an ongoing basis for CNMP. Yet only 55% felt they had adequate consultation and referral resources to assist with patients who have CNMP. In addition, only 29% felt they had adequate consultation and referral resources in their communities to assist them with patients who might be abusing or selling opioid prescriptions.
Familiarity with State Prescribing and Documentation Guidelines
In 1994, the Medical Board of California issued guidelines for prescribing opioids for CNMP that were designed to standardize referral and documentation practices and to reduce fear of regulatory scrutiny among physicians who prescribe opioids for CNMP. The guidelines were mailed to all licensed physicians in the state on 3 occasions between 1994 and 1996.23 We found that 39% of respondents remembered reading the guidelines 1 year after the third mailing. We also found that physicians varied in their self-reported compliance with recommended documentation practices. Ninety percent said they always or usually document a history and physical examination before prescribing opioids, and 86% document periodic reassessment of chronic pain. However, only 60% said that they always or usually document rules of use and misuse of opioid medications; 45% document treatment objectives; and 24% document informed consent. When asked about regulatory scrutiny, 40% of physicians agreed that fear of legal investigation tempers their use of opioids for patients with CNMP.
Predictors of Willingness to Prescribe Opioids
Three models were postulated to clarify the determinants of willingness to prescribe opioids for CNMP. The results of these analyses are presented in Table 5. The stepwise linear regression for each model generated a value for each variable (R2) that represents the proportion of the variance that can be explained by the given variable.
In all 3 models lower levels of concern about physical dependence in response to the vignettes were associated with greater willingness to prescribe opioids. Other variables that were significant predictors of willingness to prescribe opioids in 1 or more models were more recent graduation from medical school, enjoyment in working with chronic pain patients, less fear of regulatory scrutiny, and fewer total patients seen per month.
Discussion
Nearly all the physicians in our sample were willing to treat certain CNMP patients with schedule III opioids on an as-needed basis. However, a third of these physicians said they never use the more potent long-acting schedule II opioids for CNMP. There was also substantial disagreement about which patients would benefit from opioids and which might be likely to suffer adverse effects.
Concern about physical dependence appears to be among the most important barriers to the use of opioids for patients with CNMP. Whether this is always an appropriate concern is debatable. For example, in the case of using schedule III opioids on an as-needed basis, the lack of continuous exposure should limit the risk of physical dependence.
Our finding that physician concerns about physical dependence, tolerance, and addiction were highly intercorrelated raises the possibility that many physicians believe, correctly or incorrectly, that these 3 conditions are closely related effects of opioids. It is also possible that physicians are unclear about what distinguishes one of these outcomes from another. More research is needed to determine the root of physician concerns about physical dependence, tolerance, and addiction. Although all 3 of these outcomes can result when opioids are used around the clock, they nonetheless do not always occur together or necessarily all have equally serious implications when they occur.24 Only a slight majority of respondents felt that they had adequate consultation and referral resources in their community to assist with patients who have CNMP. Primary care physicians may benefit from more information about pain management resources in their communities. In addition, communities without these resources may benefit from the development of pain management centers that can assist primary care physicians with patients who suffer from CNMP.
More recent graduation from medical school was a predictor of increased willingness to prescribe opioids. Recently trained physicians may be more likely to have been exposed to an environment of more liberal use of opioids for CNMP. Conversely, the decreased willingness of more experienced physicians to prescribe opioids may be influenced by their clinical experiences with the complications of opioid use.
Fear of regulatory scrutiny also appeared to limit willingness to prescribe as-needed low-potency schedule III opioids. Recent laws and guidelines have attempted to reduce both the risk and fear of regulatory scrutiny when opioids are prescribed for chronic pain.23,25,26 However, it is not clear whether awareness of these guidelines would increase or decrease physician concern with regard to regulatory scrutiny, since many physicians reported that their documentation standards are not up to those recommended. In addition, we found no differences in willingness to prescribe opioids based on awareness of the guidelines in California.
We found that physicians who saw more patients were less likely to use more potent opioids. In California, schedule II opioids must be prescribed using triplicate forms purchased from the state. Physicians with high-volume practices may be less inclined to prescribe high-potency opioids because of the time required to complete triplicate forms. Other possible explanations are that these physicians have had more adverse experiences with the use of opioids for CNMP or that they feel more vulnerable to regulatory scrutiny because of their increased volume of patients who might receive opioid prescriptions.
We found that most physicians did not enjoy working with chronic pain patients, and this lack of enjoyment with treating CNMP was a significant barrier to willingness to prescribe opioids in 2 of our 3 models. More investigation of why most physicians do not enjoy working with these patients could further illuminate barriers to the use of opioids for CNMP.
Limitations
There are several limitations to our study. First, the physicians surveyed may not be representative of all practicing primary care physicians. However, CRN physicians are quite similar in many characteristics to family physicians practicing in California.27 Barriers to prescribing opioids in California may also be different from barriers faced by physicians in other parts of the country, so our results may not be easily generalized to other geographic regions. In addition, the data were generated by self-report, and actual practices may differ.
However, our findings are consistent with a 1991 survey of 90 Wisconsin physicians that concluded that concerns about addiction outweigh concerns about regulatory scrutiny for most physicians.28 A national survey of 1912 physicians from multiple specialties found, as we did, a high level of intercorrelation among physician concerns about physical dependence, tolerance, and addiction.29 Unfortunately, that study was not designed to elucidate the relative importance of factors that determine a physician’s willingness to prescribe opioids.
Another limitation of our study is that none of the models we postulated could explain more than a small proportion (24%) of the total variance in the willingness to prescribe opioids for CNMP. Clearly other factors, unmeasured in the current study, also influence physicians’ willingness to prescribe opioids for CNMP. For example, in a study of the prescribing habits at a referral center in Seattle, pain specialists were significantly influenced in their willingness to prescribe opioids for CNMP by a set of pain behaviors exhibited by the patient.30 These behaviors included distorted ambulation or posture, negative affect, facial and audible expressions of distress, and avoidance of activity. The nature of our study did not allow for such factors in our models of willingness to prescribe opioids, but these factors may be worthy of further investigation in direct observation studies of primary care.
Conclusions
Our results suggest that primary care physicians disagree about the relative risks and benefits of opioids in the treatment of individuals who suffer from CNMP. Also, these physicians function with limited reliable information or specialty resources to guide them in choosing which of these patients to treat with opioids. Concerns about addiction, tolerance, and physical dependence appear to be important barriers to the use of opioids by many physicians. More research is needed in primary care settings to determine appropriate uses for opioids in the treatment of CNMP and to further elucidate the concerns of physicians and barriers to more effective use.
Acknowledgments
This research was partially supported by grant #5D32PE19036-09 from the Department of Health and Human Services Health Resources Services Administration to support the establishment of a Department of Family Practice and by a grant from the California Academy of Family Physicians. We would like to acknowledge Dr Eric Sanford, Dr Lawrence Bruguera, Dr Charles Kano, Dr Joyce Hightower, and Dr Yeva Johnson for their assistance with data interpretation and preparation of this manuscript. In addition, we would like to acknowledge the work of Ms Catherine Brosnan and Ms Elizabeth Dito in assisting us with coordination and data collection for our study.
Related resources
- American Pain Society http://www.ampainsoc.org/
- American Academy of Pain Management http://www.aapainmanage.org
- American Pain Foundation http://www.painfoundation.org
METHODS: A survey was mailed to primary care physicians in the University of California, San Francisco/Stanford Collaborative Research Net- work. This survey contained questions regarding treatment in response to 3 case vignettes, the use of opioids for CNMP in general, and the demographic characteristics of the physicians.
RESULTS: Among 230 physicians surveyed, 161 (70%) responded. Two percent of the respondents were never willing to prescribe schedule III opioids (eg, acetaminophen with codeine) as needed for patients with CNMP that persisted unchanged after exhaustive evaluation and attempts at treatment. Thirty-five percent were never willing to prescribe schedule II opioids (eg, sustained-release morphine) on an around-the-clock schedule for these patients. The most significant predictor of willingness to prescribe opioids for patients with CNMP was a lower level of concern about physical dependence, tolerance, and addiction.
CONCLUSIONS: Primary care physicians are willing to prescribe schedule III opioids as needed, but many are unwilling to use schedule II opioids around the clock for CNMP. Individual prescribing practices vary widely among primary care physicians. Concerns about physical dependence, tolerance, and addiction are barriers to the prescription of opioids by primary care physicians for patients with CNMP.
Opioids are effective analgesics that are widely accepted as therapy for cancer pain and pain related to other terminal illnesses.1-2 However, the use of opioids to treat chronic nonmalignant pain (CNMP) is controversial.3-8 Few clinical studies of opioids in the alleviation of CNMP have been conducted, and most have been small, retrospective, uncontrolled, or focused on patients seen in referral settings.9-21 Together these studies suggest that opioids may benefit certain patients with CNMP, though the results have not been conclusive.
In clinical practice the absence of definitive data on the risks and benefits of opioids for CNMP presents a dilemma. Decisions about potency, frequency, and duration of treatment must be made without the benefit of evidence-based guidelines and with the knowledge that state medical boards or other legal authorities may scrutinize opioid prescriptions. We conducted this study to learn more about attitudes, prescribing practices, and factors associated with the willingness of primary care physicians to prescribe opioids for their patients with CNMP.
Methods
Sample
The University of California, San Francisco/Stanford Collaborative Research Network (CRN) is a practice-based research network composed predominantly of family physicians practicing in Northern and Central California. In 1997 the CRN conducted this survey of all 230 primary care physician members who were not involved in designing our study. Up to 2 mailed reminders and 3 telephone calls were made to initial nonresponders to improve the response rate.
Instrument
The survey instrument was developed through a collaborative process involving 7 volunteer physicians from the CRN. It was pilot-tested and refined using focus groups of practicing non-CRN primary care physicians.
On the first page of the survey, CNMP was defined as pain lasting longer than 6 months that was not related to cancer or another condition expected to end a patient’s life within 6 months. The survey included 3 clinical vignettes Table 1 designed to evoke responses to a variety of patient characteristics, such as medical history, age, sex, and socioeconomic status. Each vignette was followed by a set of specific questions. The survey also contained questions unrelated to the vignettes, regarding general attitudes toward opioids and opioid prescribing practices. We asked about documentation practices, referral resources, and familiarity with state guidelines. The respondents were also queried about personal, patient, and practice characteristics.
Statistical Analysis
We conducted analyses using SAS software.22 Means and standard deviations (SDs) for continuous variables, and frequency distributions for categorical variables, were calculated to summarize physician respondent characteristics, estimates of the characteristics of their caseloads, and summaries of their responses to questions about the clinical vignettes. We used correlation coefficients to examine the strength of relationships between attitudes and practice. The results of the correlation coefficients were used to choose a set of independent variables that were most predictive of willingness to prescribe opioids for CNMP.
We examined with stepwise linear regression the association between willingness to prescribe opioid medications and specific physician characteristics, including year of medical school graduation, size of patient caseload, and concerns about physical dependence, tolerance, addiction, side effects, regulatory scrutiny, and diversion for illegal use. In selecting the final set of variables for the stepwise linear regression predicting willingness to prescribe opioids for CNMP, we found that concern about physical dependence, tolerance, and addiction were highly intercorrelated. Among these variables, concern about physical dependence was the most consistently predictive of willingness to prescribe opioids for CNMP. When concern about physical dependence was entered into stepwise models the variables measuring concern about tolerance and addiction dropped out. Therefore, we chose to use concern about physical dependence as a proxy for measuring generalized concerns about all 3 concerns taken together.
We constructed 3 models to predict willingness to prescribe opioids. In Model 1 the dependent variable to designate willingness to prescribe was constructed from the sum of responses to the 5-point Likert-scaled question asked after each of the 3 vignettes: “If the pain persisted unchanged, would you prescribe opioids for this patient on a long-term basis?” In Model 2 the dependent variable was defined according to the range of agreement on a 5-point Likert scale with the following statement: “For patients with CNMP that persists unchanged after exhaustive evaluation and attempts at treatment, I am willing to prescribe opioids not requiring triplicates (such as Tylenol with codeine) on an as-needed basis.” In Model 3, the dependent variable was defined according to the range of agreement with the following statement on a 5-point Likert scale: “For patients with CNMP that persists unchanged after exhaustive evaluation and attempts at treatment, I am willing to prescribe opioids requiring triplicates (eg, fentanyl patch, methadone, or sustained-release morphine) on a fixed, around-the-clock basis.”
Results
Physician and Practice Characteristics
A total of 161 of 230 physicians (70%) completed the survey. The demographic characteristics of the respondents are presented in Table 2. Table 3 shows physician estimates of patient demographics in their practices. As a group the CRN physicians were mostly white men, but they care for an ethnically, financially, and age-diverse population. The large SDs in Table 3 reflect the wide variety of practice types included in the CRN membership.
Physicians reported seeing an average of 280 patients (SD=157), including 18 CNMP patients (SD=26), per month. An average of 7 patients (SD=8) with CNMP were prescribed opioid analgesics per month, and 90% of the physicians reported prescribing opioids for CNMP at least once a month. The wide SDs again reflect broad variation in the number of patients seen, the number of patients encountered with CNMP, and the number of patients treated with opioid analgesics by different physicians.
Attitudes and Practices of Physicians
Only 15% of respondents agreed with the statement: “I enjoy working with patients who have CNMP.” However, only 15% also felt that daily opioids have no place in the treatment of CNMP. Only 7% agreed with the statement: “I never prescribe opioids for CNMP.”
Many physicians wait for their patients to bring up the subject of opioid treatment, as indicated by the fact that 41% of the respondents agreed that “most of my patients who get opioid prescriptions from me for CNMP requested an opioid before I suggested their use.” In addition, 37% responded that they rarely or never are the first physician to prescribe opioids to their patients with CNMP, possibly waiting for other specialists to take the initiative.
The responses to questions about the 3 clinical vignettes are presented in Table 4. Nearly all physicians felt that the vignettes were realistic, and most believed they were knowledgeable about evaluation and treatment for these patients. However, each case generated substantial variation of opinion with regard to the level of optimism about being able to help the patient, the need for specialty referral, and the willingness to treat with opioids. For each vignette respondents were generally more concerned about physical dependence, tolerance, and addiction than they were about diversion for illegal use, regulatory scrutiny, or side effects. However, physicians’ level of concern about each of these outcomes varied substantially for each vignette.
The physicians were asked general questions about situations in which they would never prescribe opioids. Although none of the respondents said that they had a policy of refusing opioids to patients aged older than 65 years, 19% said they would never prescribe opioids to a child younger than 18 years. In addition, 16% said they would never prescribe opioids to a previous substance abuser, and 42% said they would never prescribe opioids to a current substance abuser, even if recommended by an appropriate specialist.
Also, respondents expressed an increased reluctance to prescribe opioids to CNMP patients as the frequency and potency of the medication was increased. Although only 2% of physicians said they would never prescribe low-potency (schedule III) opioids on an as-needed basis, 35% said they would never prescribe high-potency (schedule II) opioids around the clock, even after exhaustive evaluation and attempts at treatment.
In addition, the willingness of respondents to prescribe opioids varied according to the medical condition being treated. Forty-two percent of respondents said they would never prescribe long-acting schedule II opioids to a patient with post-herpetic neuralgia; 57% would never prescribe them for chronic low back pain; and 75% would never prescribe them for chronic daily headache.
We asked about the use of specialists to assist in the evaluation and treatment of patients who may benefit from opioid treatment for CNMP. Fifty-two percent of the physicians reported always or usually requiring their patients to undergo evaluation by a specialist before prescribing opioids on an ongoing basis for CNMP. Yet only 55% felt they had adequate consultation and referral resources to assist with patients who have CNMP. In addition, only 29% felt they had adequate consultation and referral resources in their communities to assist them with patients who might be abusing or selling opioid prescriptions.
Familiarity with State Prescribing and Documentation Guidelines
In 1994, the Medical Board of California issued guidelines for prescribing opioids for CNMP that were designed to standardize referral and documentation practices and to reduce fear of regulatory scrutiny among physicians who prescribe opioids for CNMP. The guidelines were mailed to all licensed physicians in the state on 3 occasions between 1994 and 1996.23 We found that 39% of respondents remembered reading the guidelines 1 year after the third mailing. We also found that physicians varied in their self-reported compliance with recommended documentation practices. Ninety percent said they always or usually document a history and physical examination before prescribing opioids, and 86% document periodic reassessment of chronic pain. However, only 60% said that they always or usually document rules of use and misuse of opioid medications; 45% document treatment objectives; and 24% document informed consent. When asked about regulatory scrutiny, 40% of physicians agreed that fear of legal investigation tempers their use of opioids for patients with CNMP.
Predictors of Willingness to Prescribe Opioids
Three models were postulated to clarify the determinants of willingness to prescribe opioids for CNMP. The results of these analyses are presented in Table 5. The stepwise linear regression for each model generated a value for each variable (R2) that represents the proportion of the variance that can be explained by the given variable.
In all 3 models lower levels of concern about physical dependence in response to the vignettes were associated with greater willingness to prescribe opioids. Other variables that were significant predictors of willingness to prescribe opioids in 1 or more models were more recent graduation from medical school, enjoyment in working with chronic pain patients, less fear of regulatory scrutiny, and fewer total patients seen per month.
Discussion
Nearly all the physicians in our sample were willing to treat certain CNMP patients with schedule III opioids on an as-needed basis. However, a third of these physicians said they never use the more potent long-acting schedule II opioids for CNMP. There was also substantial disagreement about which patients would benefit from opioids and which might be likely to suffer adverse effects.
Concern about physical dependence appears to be among the most important barriers to the use of opioids for patients with CNMP. Whether this is always an appropriate concern is debatable. For example, in the case of using schedule III opioids on an as-needed basis, the lack of continuous exposure should limit the risk of physical dependence.
Our finding that physician concerns about physical dependence, tolerance, and addiction were highly intercorrelated raises the possibility that many physicians believe, correctly or incorrectly, that these 3 conditions are closely related effects of opioids. It is also possible that physicians are unclear about what distinguishes one of these outcomes from another. More research is needed to determine the root of physician concerns about physical dependence, tolerance, and addiction. Although all 3 of these outcomes can result when opioids are used around the clock, they nonetheless do not always occur together or necessarily all have equally serious implications when they occur.24 Only a slight majority of respondents felt that they had adequate consultation and referral resources in their community to assist with patients who have CNMP. Primary care physicians may benefit from more information about pain management resources in their communities. In addition, communities without these resources may benefit from the development of pain management centers that can assist primary care physicians with patients who suffer from CNMP.
More recent graduation from medical school was a predictor of increased willingness to prescribe opioids. Recently trained physicians may be more likely to have been exposed to an environment of more liberal use of opioids for CNMP. Conversely, the decreased willingness of more experienced physicians to prescribe opioids may be influenced by their clinical experiences with the complications of opioid use.
Fear of regulatory scrutiny also appeared to limit willingness to prescribe as-needed low-potency schedule III opioids. Recent laws and guidelines have attempted to reduce both the risk and fear of regulatory scrutiny when opioids are prescribed for chronic pain.23,25,26 However, it is not clear whether awareness of these guidelines would increase or decrease physician concern with regard to regulatory scrutiny, since many physicians reported that their documentation standards are not up to those recommended. In addition, we found no differences in willingness to prescribe opioids based on awareness of the guidelines in California.
We found that physicians who saw more patients were less likely to use more potent opioids. In California, schedule II opioids must be prescribed using triplicate forms purchased from the state. Physicians with high-volume practices may be less inclined to prescribe high-potency opioids because of the time required to complete triplicate forms. Other possible explanations are that these physicians have had more adverse experiences with the use of opioids for CNMP or that they feel more vulnerable to regulatory scrutiny because of their increased volume of patients who might receive opioid prescriptions.
We found that most physicians did not enjoy working with chronic pain patients, and this lack of enjoyment with treating CNMP was a significant barrier to willingness to prescribe opioids in 2 of our 3 models. More investigation of why most physicians do not enjoy working with these patients could further illuminate barriers to the use of opioids for CNMP.
Limitations
There are several limitations to our study. First, the physicians surveyed may not be representative of all practicing primary care physicians. However, CRN physicians are quite similar in many characteristics to family physicians practicing in California.27 Barriers to prescribing opioids in California may also be different from barriers faced by physicians in other parts of the country, so our results may not be easily generalized to other geographic regions. In addition, the data were generated by self-report, and actual practices may differ.
However, our findings are consistent with a 1991 survey of 90 Wisconsin physicians that concluded that concerns about addiction outweigh concerns about regulatory scrutiny for most physicians.28 A national survey of 1912 physicians from multiple specialties found, as we did, a high level of intercorrelation among physician concerns about physical dependence, tolerance, and addiction.29 Unfortunately, that study was not designed to elucidate the relative importance of factors that determine a physician’s willingness to prescribe opioids.
Another limitation of our study is that none of the models we postulated could explain more than a small proportion (24%) of the total variance in the willingness to prescribe opioids for CNMP. Clearly other factors, unmeasured in the current study, also influence physicians’ willingness to prescribe opioids for CNMP. For example, in a study of the prescribing habits at a referral center in Seattle, pain specialists were significantly influenced in their willingness to prescribe opioids for CNMP by a set of pain behaviors exhibited by the patient.30 These behaviors included distorted ambulation or posture, negative affect, facial and audible expressions of distress, and avoidance of activity. The nature of our study did not allow for such factors in our models of willingness to prescribe opioids, but these factors may be worthy of further investigation in direct observation studies of primary care.
Conclusions
Our results suggest that primary care physicians disagree about the relative risks and benefits of opioids in the treatment of individuals who suffer from CNMP. Also, these physicians function with limited reliable information or specialty resources to guide them in choosing which of these patients to treat with opioids. Concerns about addiction, tolerance, and physical dependence appear to be important barriers to the use of opioids by many physicians. More research is needed in primary care settings to determine appropriate uses for opioids in the treatment of CNMP and to further elucidate the concerns of physicians and barriers to more effective use.
Acknowledgments
This research was partially supported by grant #5D32PE19036-09 from the Department of Health and Human Services Health Resources Services Administration to support the establishment of a Department of Family Practice and by a grant from the California Academy of Family Physicians. We would like to acknowledge Dr Eric Sanford, Dr Lawrence Bruguera, Dr Charles Kano, Dr Joyce Hightower, and Dr Yeva Johnson for their assistance with data interpretation and preparation of this manuscript. In addition, we would like to acknowledge the work of Ms Catherine Brosnan and Ms Elizabeth Dito in assisting us with coordination and data collection for our study.
Related resources
- American Pain Society http://www.ampainsoc.org/
- American Academy of Pain Management http://www.aapainmanage.org
- American Pain Foundation http://www.painfoundation.org
1. Jadad AR, Browman GP. The WHO analgesic ladder for cancer pain management: stepping up the quality of its evaluation. JAMA 1995;274:1870-73.
2. Levy MH. Pharmacologic treatment of cancer pain. N Eng J Med 1996;46:128-38.
3. Schofferman J. Long-term use of opioid analgesics for the treatment of chronic pain of nonmalignant origin. J Pain Symptom Manage 1993;8:279-88.
4. Large RG, Schug SA. Options for chronic pain of nonmalignant origin: caring or crippling. Health Care Anal 1995;3:5-11.
5. Portenoy RL. Opioid therapy for chronic nonmalignant pain: a review of the critical issues. J Pain Symptom Manage 1996;11:203-17.
6. Kyriaki D, Pither CE, Wessely S. Medication misuse, abuse and dependence in chronic pain patients. J Psychosom Res 1997;43:497-504.
7. McQuay H. Opioids in pain management. Lancet 1999;353:2229-32.
8. Parrott T. Using opioid analgesics to manage noncancer pain in primary care. J Am Board Fam Pract 1999;12:293-306.
9. Portenoy RK, Foley KM. Chronic use of opioid analgesics in non-malignant pain: report of 38 cases. Pain 1986;25:171-86.
10. Kjaersgaard-Andersen P, Nafei A, Skov O, Madsen F, et al. Codeine plus paracetamol versus paracetamol in longer-term treatment of chronic pain due to osteoarthritis of the hip: a randomised, double-blind, multicentre study. Pain 1990;43:309-18.
11. Zenz M, Strumpf M, Tryba M. Long-term opioid therapy in patients with chronic nonmalignant pain. J Pain Symptom Manage 1992;7:69-77.
12. Kell M, Musselman D. Methadone prophylaxis of intractable headaches: pain control and serum opioid levels. Am J Pain Manage 1993;3:7-14.
13. Arkinstall W, Sandler A, Groghnour B, Babul N, Harsanyi Z, Darke AC. Efficacy of controlled-release codeine in chronic non-malignant pain: a randomized, placebo-controlled clinical trial. Pain 1995;62:169-78.
14. Moulin DE, Iezzi A, Amireh R, et al. Randomised trial of oral morphine for chronic non-cancer pain. Lancet 1996;347:143-47.
15. Gardner-Nix JS. Oral methadone for managing chronic nonmalignant pain. J Symptom Pain Manage 1996;11:321-28.
16. Simpson RK, Edmondson EA, Constant CF, Collier C. Transdermal fentanyl for chronic low back pain. J Pain Symptom Manage 1997;14:218-24.
17. Haythornthwaite JA, Menefee LA, Quatrano-Piacentini AL, Pappagallo M. Outcome of chronic opioid therapy for non-cancer pain. J Pain Symptom Manage 1998;15:185-94.
18. Sheather-Reid RB, Cohen ML. Efficacy of analgesics in chronic pain: a series of n-of-1 studies. J Pain Symptom Manage 1998;15:244-52.
19. Jamison RN, Raymond SA, Slawsby EA, Nedeljkovic SS, et al. Opioid therapy for chronic noncancer back pain. Spine 1998;23:2591-600.
20. Ytterberg SR, Mahowald ML, Woods SR. Codeine and oxycodone use in patients with chronic rheumatic disease pain. Arthritis Rheum 1998;41:1603-12.
21. Watson CP, Babul N. Efficacy of oxycodone in neuropathic pain: a randomized trial in postherpetic neuralgia. Neurology 1998;50:1837-41.
22. SAS Institute Inc. SAS System for Microsoft Windows, release 6.12. Cary, NC: SAS Institute Inc; 1996.
23. Medical Board of California. Action report: new, easy guidelines on prescribing. 1994; 51:1.
24. Sees KL, Clark HW. Opioid use in the treatment of chronic pain: assessment of addiction. J Pain Symptom Manage 1993;8:257-64.
25. Clark H. Opioids, chronic pain, and the law. J Pain Symptom Manage 1993;8:297-306.
26. Hill CS. Government regulatory influences on opioid prescribing and their impact on the treatment of pain of nonmalignant origin. J Pain Symptom Manage 1996;11:287-98.
27. Croughan-Minihane MS, Thom DH, Petitti DB. Research interests of physicians in two practice-based primary care research networks. West J Med 1999;170:19-24.
28. Weissman DE, Joranson DE, Hopwood MB. Wisconsin physicians’ knowledge and attitudes about opioid analgesic regulations. Wis Med J 1991;12:671-75.
29. Turk DC, Brody MC, Okifuji AE. Physicians’ attitudes and practices regarding the long-term prescribing of opioids for non-cancer pain. Pain 1994;59:201-208.
30. Turk DC, Okifuji A. What factors affect physicians’ decisions to prescribe opioids for chronic noncancer pain patients? Clin J Pain 1997;13:330-36.
1. Jadad AR, Browman GP. The WHO analgesic ladder for cancer pain management: stepping up the quality of its evaluation. JAMA 1995;274:1870-73.
2. Levy MH. Pharmacologic treatment of cancer pain. N Eng J Med 1996;46:128-38.
3. Schofferman J. Long-term use of opioid analgesics for the treatment of chronic pain of nonmalignant origin. J Pain Symptom Manage 1993;8:279-88.
4. Large RG, Schug SA. Options for chronic pain of nonmalignant origin: caring or crippling. Health Care Anal 1995;3:5-11.
5. Portenoy RL. Opioid therapy for chronic nonmalignant pain: a review of the critical issues. J Pain Symptom Manage 1996;11:203-17.
6. Kyriaki D, Pither CE, Wessely S. Medication misuse, abuse and dependence in chronic pain patients. J Psychosom Res 1997;43:497-504.
7. McQuay H. Opioids in pain management. Lancet 1999;353:2229-32.
8. Parrott T. Using opioid analgesics to manage noncancer pain in primary care. J Am Board Fam Pract 1999;12:293-306.
9. Portenoy RK, Foley KM. Chronic use of opioid analgesics in non-malignant pain: report of 38 cases. Pain 1986;25:171-86.
10. Kjaersgaard-Andersen P, Nafei A, Skov O, Madsen F, et al. Codeine plus paracetamol versus paracetamol in longer-term treatment of chronic pain due to osteoarthritis of the hip: a randomised, double-blind, multicentre study. Pain 1990;43:309-18.
11. Zenz M, Strumpf M, Tryba M. Long-term opioid therapy in patients with chronic nonmalignant pain. J Pain Symptom Manage 1992;7:69-77.
12. Kell M, Musselman D. Methadone prophylaxis of intractable headaches: pain control and serum opioid levels. Am J Pain Manage 1993;3:7-14.
13. Arkinstall W, Sandler A, Groghnour B, Babul N, Harsanyi Z, Darke AC. Efficacy of controlled-release codeine in chronic non-malignant pain: a randomized, placebo-controlled clinical trial. Pain 1995;62:169-78.
14. Moulin DE, Iezzi A, Amireh R, et al. Randomised trial of oral morphine for chronic non-cancer pain. Lancet 1996;347:143-47.
15. Gardner-Nix JS. Oral methadone for managing chronic nonmalignant pain. J Symptom Pain Manage 1996;11:321-28.
16. Simpson RK, Edmondson EA, Constant CF, Collier C. Transdermal fentanyl for chronic low back pain. J Pain Symptom Manage 1997;14:218-24.
17. Haythornthwaite JA, Menefee LA, Quatrano-Piacentini AL, Pappagallo M. Outcome of chronic opioid therapy for non-cancer pain. J Pain Symptom Manage 1998;15:185-94.
18. Sheather-Reid RB, Cohen ML. Efficacy of analgesics in chronic pain: a series of n-of-1 studies. J Pain Symptom Manage 1998;15:244-52.
19. Jamison RN, Raymond SA, Slawsby EA, Nedeljkovic SS, et al. Opioid therapy for chronic noncancer back pain. Spine 1998;23:2591-600.
20. Ytterberg SR, Mahowald ML, Woods SR. Codeine and oxycodone use in patients with chronic rheumatic disease pain. Arthritis Rheum 1998;41:1603-12.
21. Watson CP, Babul N. Efficacy of oxycodone in neuropathic pain: a randomized trial in postherpetic neuralgia. Neurology 1998;50:1837-41.
22. SAS Institute Inc. SAS System for Microsoft Windows, release 6.12. Cary, NC: SAS Institute Inc; 1996.
23. Medical Board of California. Action report: new, easy guidelines on prescribing. 1994; 51:1.
24. Sees KL, Clark HW. Opioid use in the treatment of chronic pain: assessment of addiction. J Pain Symptom Manage 1993;8:257-64.
25. Clark H. Opioids, chronic pain, and the law. J Pain Symptom Manage 1993;8:297-306.
26. Hill CS. Government regulatory influences on opioid prescribing and their impact on the treatment of pain of nonmalignant origin. J Pain Symptom Manage 1996;11:287-98.
27. Croughan-Minihane MS, Thom DH, Petitti DB. Research interests of physicians in two practice-based primary care research networks. West J Med 1999;170:19-24.
28. Weissman DE, Joranson DE, Hopwood MB. Wisconsin physicians’ knowledge and attitudes about opioid analgesic regulations. Wis Med J 1991;12:671-75.
29. Turk DC, Brody MC, Okifuji AE. Physicians’ attitudes and practices regarding the long-term prescribing of opioids for non-cancer pain. Pain 1994;59:201-208.
30. Turk DC, Okifuji A. What factors affect physicians’ decisions to prescribe opioids for chronic noncancer pain patients? Clin J Pain 1997;13:330-36.