INTRODUCTION
Physician impairment from drug and alcohol abuse is a well-known national problem.1 Equally problematic but less well researched are physicians who misprescribe controlled substances, principally narcotics and benzodiazepines.2–5 General internists and family practitioners seem particularly vulnerable.5
Physicians cited for prescribing violations by a state medical board generally fall into two distinct groups: those who abuse (disabled) and/or sell drugs (dishonest), and physicians whose misprescribing results from a variety of other causes (dated, duped). The former group faces prosecution, loss of licensure and/or mandated treatment for addiction.6 Education and counseling may be more appropriate for the second group whose misprescribing results from being dated in their knowledge, duped by drug seekers, or disorganized in their practice management. Others are deficient in documentation, dysfunctional (can’t say no to drug seekers) due to family of origin or personality traits and/or overly concerned about regulatory oversight.5, 6 Unfortunately, few Continuing Medical Education (CME) programs target these needs, and even fewer have been evaluated for their impact.7, 8
Many state medical board credentialing committees are understandably reluctant to suspend privileges or revoke medical licenses of doctors who misprescribe drugs. They are equally reluctant to do nothing. Experience suggests that misprescribers may benefit from a mid-level intervention, i.e., a negotiated agreement that, at least in part, requires the physician to attend a CME program devoted to appropriate prescribing.5,7
This article describes a short-term evaluation of the behavioral and practice-related changes physicians have made as a result of attending one such course. Barriers that interfered with intended changes are described.
PROGRAM DESCRIPTION
The course’s history and contents are described elsewhere.5,7,8 Course registration is limited to no more than 12 participants. The small-group experience gives participants opportunities to vent anger at being accused of wrongdoing, to learn that they are neither alone nor being singled out by a state medical board, to examine the reasons that might underlie misprescribing, and to learn strategies for preventing recurrence.
Consistent with the literature on promoting physician behavior change, the three-day course combines traditional didactic instruction, confrontation, personal examination, practical planning, and problem solving.9–13 Course objectives aim to help participants:
constructively re-channel anger associated with being remanded to the course
identify personal reasons for misprescribing
review pharmacology of Schedule II narcotics and the benzodiazepines
demonstrate new skills in substance abuse identification and prescribing practice
develop workable plans for changing behaviors that contributed to their misprescribing
identify resources for implementing those changes
Lectures, case studies, role playing, and further discussions help participants gain some sense of the interpersonal issues, competing priorities, and practice habits that influence their decision making and help them develop plans for change.
PROGRAM EVALUATION
Subjects: Fifty-four physician participants were distributed among seven course offerings from January 1997 through January 1999. The participants were predominantly male (88%) and ranged in age from 32 to 74 years of age (mean 41 years). Most specialties were represented (Table 1). However, family practice, general practice, and internal medicine specialties accounted for 69% of the participants.
Participant Speciality (N=54)
The referral source for the physicians attending the course is depicted in Table 2. A small number of the physicians were self-referred. The majority were referred by their state’s state medical board or physician health program. Most physicians came to the attention of the state medical board or physician health program for the following reasons:
reported by a local pharmacist
following treatment or evaluation for substance dependence or abuse
patient complaints
advice of counsel
record review
Referral Source (N=54)
The reasons participants gave for attending the courses are summarized in Table 3. Several physicians attended for more than one reason. Concerns about overprescribing and/or underprescribing were primary.
Reason for Referral (N=54) *
COMMITMENT TO CHANGE PROCEDURE
Sessions were evaluated by means of a “commitment to change” procedure, which has been validated for use in assessing CME program outcomes.14,15 Following informed consent procedures approved by Vanderbilt University’s Committee for the Protection of Human Subjects, all participants completed a response form on which they identified 3 to 5 changes they intended to make over the next three months.
Physician participants were urged to make plans for change that were reasonable and had a chance of being implemented successfully. They were to choose the plans for change that they believed were important to their personal and professional health. The evaluation focused on which plans were successfully implemented, and if not completed, what barriers they encountered.
Three months later a cover letter, the original response form, and a follow-up questionnaire were sent to participants. Participants were asked to indicate the extent to which they had actually implemented each intended behavior change: 100% meant full implementation or fulfillment, 50% meant implemented about half way or about half the time. Participants were asked to identify obstacles to full implementation.
RESULTS OF COMMITMENT TO CHANGE PROCEDURE
Initial mailing and follow-up netted a response rate of 83% (45 out of 54 responded). Non-responders’ demographic profiles were not different from responders’. One physician returned the forms marked “not applicable,” explaining that her/his duties no longer involved prescribing. The results are based on the 44 who retained prescribing privileges throughout the period.
Respondents committed themselves to make an average of 3.9 behavior changes. Straightforward behaviors ranged from reducing prescriptions for benzodiazepines, to being more careful with prescription pads, and better documenting/justifying prescriptions for narcotics. Other intended changes were more ambitious, including confronting patients more on problems the physicians identify, joining a (health professional) support group, hiring an additional physician, and spending more time with their children.
The commitments totaled 172. No progress on 13 of these commitments in the three-month interval was reported. Forty percent (69 of the 172) commitments were reported to have been fully implemented (90–100% of the time). The overall implementation rate was 73%. If a behavior change was implemented at all, some change was reportedly made 80% of the time.
To better understand the nature of the participants’ intentions and subsequent rates of fulfillment, each intention was assigned to one of 3 categories (Table 4). The three categories include those physician behavior factors that were observed to influence the prescribing of controlled drugs.6 “Core personality traits” are those behaviors that improve a physician’s approach to the practice of medicine and the particular difficulties encountered in prescribing to demanding patients. These new behaviors included: joining a support group, refining refusal skills to patients inappropriately requesting narcotics, and any other activity focused on personal growth (taking time off, spending time with one’s children). The second category was defined under the term “pharmacological knowledge” which includes plans to use medications that have appropriate indications for use in pain and anxiety, refining techniques of recognizing the drug-seeking patient, reading and learning more about addiction in its various forms. The third category, “practice management,” includes maintaining complete records of patient encounters, reducing daily patient census to allow time to provide optimum care, protecting prescription pads, removing the DEA number from prescription pads, and learning of referral resources for the chronic pain patient.
Intention to Change (N=44)
Why were the behavior changes to which attendees committed themselves not universally adopted? The most commonly cited reason was a version of: lack of time, anxiety about leaving unfinished work, or continued heavy workload. Familiar barriers such as fatigue, staff turnover, too costly [to hire additional staff], and patients becoming angry when they are told “no,” were noted by several physicians. Other important barriers reflected ongoing concerns about diagnostic, therapeutic, and communication skills. Comments included:
“I believe I may now be under-prescribing.”
“A part of me still believes that some drug-seeking patients may truly be in pain.”
“The three months that have passed [since the course] can’t quickly change a lifetime of people-pleasing behavior.”
DISCUSSION
The physicians referred to the CME course provided an unusual opportunity to observe physicians away from the hectic pace of their practices and the demands of their personal lives. In the small group of colleagues in each course they were free to express their anxiety, anger, and frustration over being referred to the course by a state medical board or the physician wellness program in their state. Once they were assured that they were in a safe place and confidentiality would be honored, they became very vocal and honest about their practice organization, difficulties with demanding patients, and the treadmill of practice that put daily demands on their personal and professional health. They were frank and open about their need for knowledge to treat pain properly, to deal with the anxious patient and to know how to recognize the drug-seeking patient.
It is not uncommon for attendees to see more than 50 patients a day. Approximately 50% of those patients came from towns of less than 50,000. The participants were willing to reexamine their practice behaviors, particularly their inadequate record keeping. Many were prescribing controlled drugs without a minimum history or physical, laboratory data, or progress notes. Many physicians under the weight of large practices had ignored the basic training of keeping good records, which is vital to proper patient care.
The proportions of reported behavior changes are similar to those obtained by others15,16 who have argued that while participants’ responses were not objective proof of change, they at least implied an intention to engage in a new pattern of practice and/or lifestyle. Such outcomes are important since intentions are good predictors of actual behavior.17 Indeed, it was shown that actual prescription-writing for certain non-controlled prescription drugs was consistent with physicians’ intentions immediately following a CME program.16
Nearly two-thirds of participants intended to undertake one or more “self-improvement projects,” but fulfilling them proved challenging. Competing priorities and lack of time were common barriers. To promote greater change, participants may need help reworking their schedules (i.e., their priorities) to make time for these important pursuits. Reported barriers to making practice changes suggested that most participants needed additional information, practice, and feedback. For greater change, courses may need to be supplemented with periodic “booster shots” in the form of individualized drug utilization reviews and new sources for consultation, referral, interpersonal feedback, and support.18,19
The study has potentially important limitations. There is reason to be skeptical of self-report behavior change data. Despite the procedures used to ensure confidentiality, some participants may have simply presented what they thought the course developers wanted to hear. Others may have mistakenly believed that they had made greater progress than had actually been achieved. On the other hand, participants’ responses included reports of substantial difficulty, even complete failure with intended behavior changes. This suggests a level of candor that was not purely self-serving.
In conclusion, health professionals who misprescribe controlled substances may do so because they lack particular knowledge, problem-solving skills, and/or practice management skills. A CME course appears to help misprescribers overcome at least some of these barriers, at least in the short run.
Acknowledgments
This work was supported in part by a grant from The Robert Wood Johnson Foundation, Princeton, NJ. The authors thank Bridgette Owen for her assistance with course preparation, logistics, and evaluation services.
References
- 1.↵O’ConnorPG, SpickardA,Jr. Physician impairment by substance abuse. Med Clin N Am.1997; 81: 1037– 52.
- 2.↵HollisterAS. Patterns of scheduled drug prescribing abuse by Tennessee physicians. Substance Abuse.1990; 11( 2): 69– 78.
- 3.ParranT,Jr. Prescription drug abuse: A question of balance. Med Clin N Amer.1997; 81: 967– 78.
- 4.WessanDR, SmithDE. Prescription drug abuse: Patient, physician and cultural responsibilities. West J Med.1990; 152: 613– 6.
- 5.↵SpickardAJr, DoddD, SwiggartW, DixonG, PichertJ. Physicians who misprescribe controlled substances: A CME alternative to sanctions. Fed Bull: The Journal of Medical Licensure and Discipline.1998; 85: 8– 19.
- 6.↵Council on Scientific Affairs Drug abuse related to prescribing practices. JAMA.1982; 247: 864– 6.
- 7.↵SpickardAJr, DoddD, SwiggartW, DixonG, PichertJ. Prescribing controlled substances in Tennessee: Progress, not perfection. Southern Med J.1999; 92: 51– 4.
- 8.↵SwiggartW, PichertJW, ElasyTA, DixonGL, SpickardAJr. Continuing medical education courses on proper prescribing of controlled substances in the United States. Fed Bull: The Journal of Medical Licensure and Discipline.1999; 86: 20– 8.
- 9.↵DavisD, O’BrienMA, FreemantleN, WolfFM, MazmanianP, Taylor-VaiseyA. Impact of formal continuing medical education: Do conferences, workshops, rounds, and other traditional continuing education activities change physician behavior or health care outcomes? JAMA. 1999; 282: 867– 74.
- 10.EisenbergJM. Doctors’ decisions and the cost of medical care: The reasons for doctors’ practice patterns and ways to change them. Ann Arbor: Health Administration Press, 1986.
- 11.CohenSJ, HalvorsonHW, GosselinkCA. Changing physician behavior to improve disease prevention. Prev Med.1994; 23( 3): 284– 91.
- 12.SchaffnerW, RayWA, FederspielCF, MillerWO. Improving antibiotic prescribing in the office practice: A controlled trial of three educational methods. JAMA. 1983; 250: 1728– 32.
- 13.↵CabanaMD, RandCS, PoweNR, WuAW, WilsonMH, AbboudP-AC . Why don’t physicians follow clinical practice guidelines?JAMA1999; 282: 1458– 65.
- 14.↵PurkisIE. Commitment for changes: An instrument for evaluating CME courses. J Med Educ1982; 57: 61– 3.
- 15.↵CurryL, PurkisIE. Validity of self-reports of behavior changes by participants after a CME course. J Med Educ1986; 61: 579– 84.
- 16.↵FishbeinM, AzjenI. Belief, attitude, intention and behavior: An introduction to theory and research. NY: Addison-Wesley, 1975.
- 17.↵SleathB, CollinsT. Physician responses to an educational intervention on improving their long-term prescribing of sedatives. Pt Educ Counsel 1997; 31: 215– 22.
- 18.↵SantosoB, SuryawatiS, PrawaitasariJE. Small group interventions vs. formal seminar for improving appropriate drug use. Soc Sci Med1996; 42: 1163– 8.
- 19.↵AndersonJF, McEwanKL, HrudeyWP. Effectiveness of notification and group education in modifying prescribing of regulated analgesics. CMAJ1996; 154: 31– 39
