Strategies for improving prescribing practice

Geoffrey M. Anderson, MD, PhD; Joel Lexchin, MD, CCFP(EM), DABEM

Canadian Medical Association Journal 1996; 154: 1013-1017

Dr. Anderson is a senior scientist at the Institute for Clinical Evaluative Sciences in Ontario, North York, Ont. Dr. Lexchin is a staff physician in the Department of Emergency Medicine, Toronto Hospital, and an assistant professor in the Department of Family and Community Medicine, University of Toronto, Toronto, Ont.
Paper reprints of the full text may be obtained from: Dr. Geoffrey M. Anderson, Institute for Clinical Evaluative Sciences in Ontario, G-106-2075 Bayview Ave., North York ON M4N 3M5

Contents

The first article in this four-part series appeared in the Mar. 1 issue of CMAJ; the third and fourth articles will appear in the May 1 and June 1 issues.
See also:
  • Letter: Improving prescribing practices in nursing homes, C. MacKnight
  • Body and mind: Revisiting the O complex: urinary incontinence, delirium and polypharmacy in elderly patients

    Abstract

    Drug therapy is an integral component of modern medical care, and practising physicians are faced with the difficult task of keeping up with rapid changes in pharmacologic treatments. Recent evidence indicates that prescribing practice is often inconsistent with criteria for safety and effectiveness. Surveys indicate that community-based physicians are not satisfied with current sources of information on prescription drugs. The dissemination of printed material alone does not lead to improved prescribing practice, but specific education and feedback strategies can. To improve prescribing practice in Canada we need to systematically evaluate strategies to change prescribing behaviour, to design quality- assurance programs based on proven strategies and to develop collaboration and cooperation among providers, manufacturers, governments and the public.
    [Contents]

    Introduction

    If used properly, prescription drugs can be among the most effective and cost-effective forms of treatment; used inappropriately, they not only waste limited health care resources but also may do more harm than good. Drug therapy is perhaps the most active area of change in medicine. Drug manufacturers invest huge amounts in the development of new products,[1] and patterns of use shift rapidly.[2]

    In Canada[3-5] and other countries[6] a substantial proportion of prescribing practice is not consistent with criteria for appropriate care. This reflects a problem faced by physicians everywhere: how to keep up with the rapidly changing world of therapeutics. Pharmacology and therapeutics learned in medical school are theoretical and often rapidly outdated. Postgraduate training provides more relevant training but, unfortunately, drug choices learned during this period are often based on anecdotal information and are not always supported by research evidence. Once physicians are busily engaged in independent practice, they must find time to absorb and integrate new information into their practice patterns. For these reasons the appropriate use of prescription drugs, perhaps more than any other aspect of medical practice, requires access to new information and continuing education.

    We face the challenge of providing physicians with accurate, relevant and up-to-date information on the benefits, risks and costs of prescription drugs and of ensuring that this information is used to provide high-quality care.

    The purpose of this article is to:

    [Contents]

    Current sources of information

    Little research has been done on the question of where practising community-based physicians in Canada obtain their knowledge about therapeutics; what is available is difficult to interpret. Some studies we identified examined general sources of information on prescription drugs, some measured physicians' satisfaction with their knowledge of prescription drugs, others examined sources of information that drove changes in prescribing practice, and yet others examined sources of information used by physicians to answer specific questions about prescription drugs. Much of this research is dated or methodologically suspect. In this context, a systematic synthesis of current research on sources of information about prescription drugs is not feasible, and our review can only draw some tentative conclusions.

    There are two broadly defined sources of information on prescription drugs. The information provided by medical schools and specialty societies through continuing medical education (CME), scientific meetings and journal articles is distinguished from that obtained from commercial sources such as sales representatives, advertisements in journals and company-sponsored symposia. Physicians have been generally consistent in ranking professional sources as more important, useful and credible than commercial sources.[7-10] General practitioners[9] and high-volume prescribers[7] tend to rate commercial sources more highly than do other groups of physicians. Although physicians generally downplay the usefulness of commercial sources, in one national survey involving a random sample of 200 physicians, 82% of respondents indicated that information provided by drug manufacturers was always or sometimes a sufficient basis for informed decisions about the risks and benefits of drugs.[8]

    The use of different sources of information does not always correlate with the perceived value of the information. In one survey CME was rated as a very important source of information by 64% of respondents; commercial seminars were given this rating by only 25%. However, these same physicians attended, on average, 1.5 CME events per year, as compared with 2.5 commercial seminars.[7] Another survey showed that sales representatives were considered a much less credible source of information than CME, but were none the less more frequently used.[9]

    As part of an Ontario Royal Commission a random sample of physicians was surveyed in 1989 about their drug knowledge.[7] Although 63% were satisfied with their knowledge of drug effectiveness and 41% were satisfied with their knowledge of the use of specific categories of drugs, only 15% were satisfied with their knowledge of drug costs. In a national survey 62% of respondents said that they could obtain all the information they needed about drugs, but only about one third felt that their prescribing knowledge was very up to date.[8] When asked to identify the factors that influenced changes in prescribing practice, a nonrandom sample of Canadian specialists rated journals as the most important initial source of information, followed by CME, discussions with colleagues and contact with sales representatives. The factors that actually precipitated changes in practice were anticipated benefits in patient care, discussion with peers, product availability, CME and information given in medical journals.[11] A proprietary survey of practitioners obtained similar findings with regard to the initial source of information leading to prescribing changes, but the precipitating factor was not identified.[12]

    The Compendium of Pharmaceuticals and Specialties (CPS)[13] is a widely available source of information on prescription drugs. A survey of family physicians showed that 90% used the CPS as their first choice for information on acute drug overdoses.[14] The CPS is used as a source of information not only on overdoses, but also on adverse drug reactions; it is also used as an aid in patient counselling.[8] To respond to questions about drugs that arose in the course of day-to-day practice, staff and residents in a family practice unit first consulted the CPS and then "other family practice journals." The CPS was considered to be the most available resource and the easiest to use.[15] It was also the most frequently used source of drug information among 159 Nova Scotia physicians surveyed in 1976.[16] Although the CPS can provide useful information on dosages and adverse effects, its product monographs are not regularly updated, and it is not designed to provide comparative information on the clinical effects of different drugs or the relative costs of different therapies.

    Although we do not have definitive information on the sources of information used by physicians to make decisions about the selection of prescription drugs, we can identify some important themes from current research. It appears that many community-based physicians are not satisfied with their current knowledge of prescription drugs and have difficulty keeping their knowledge up to date. They are faced with diverse sources of information and, although they rate commercial sources as being less useful than professional ones, they rely heavily on commercial sources and the CPS to make prescribing decisions.

    [Contents]

    Techniques for promoting appropriate prescribing

    Evaluation of the literature

    Little research has been done in Canada on initiatives to improve prescribing practice in primary care, but there is a substantial body of research from other countries on the effect of education and feedback on the quality of prescribing. In 1989, Soumerai, McLaughlin and Avorn[17] critically analysed the available experimental literature, reviewing [44] studies of techniques to improve prescribing in primary care, including 16 randomized controlled trials. They found that the results of inadequately controlled studies were not consistent with the results of randomized controlled trials. Given the problems with the validity of the results of nonrandomized controlled trials, we limited our review to randomized controlled trials. We located these studies using, among other sources, the Research and Development Data Base in CME, a comprehensive inventory of trials on physician behaviour change initially developed at McMaster University, Hamilton, Ont., and now maintained by the Office of Continuing Education at the University of Toronto.

    The review by Soumerai and associates[17] concluded that mailed educational materials alone may change knowledge but have little or no detectable effects on actual prescribing practice. Although these materials alone may not be sufficient to change prescribing behaviour, they can be an important component of other strategies such as feedback or face-to-face education. There were no studies of traditional CME techniques (e.g., large-group didactic presentations). Small-group learning showed some promise, but the most effective educational strategy was brief one-to-one education by specially trained pharmacists or physicians. Although these "academic detailing" strategies were expensive, they paid for themselves, if targeted correctly, by reducing prescribing costs. The review suggested that computerized reminders to physicians could reduce some prescribing errors; however, these systems had been tested only in academic group-practice settings rather than in private offices. The effect on practice quickly deteriorated after the reminder systems were removed, which suggests that reminders work as an administrative tool rather than by improving knowledge.

    We identified nine randomized controlled trials, published after Soumerai and associates' review, that examined the impact of education or feedback on prescribing practice in primary care. Four dealt with the impact of face-to-face education provided by an academic detailer. A US study showed that visits by a clinical pharmacist to physicians reduced prescription drug costs for patients treated in a general medicine outpatient clinic, compared with costs for patients treated by a control group of physicians that did not receive any information on costs or by a group that received weekly feedback on their overall prescription drug costs and those of their peers.[18] A second US trial, conducted in a Health Maintenance Organization,[19] used a 10-minute, face-to-face educational meeting between prescribers and a pharmacist to achieve a more appropriate use of anti-ulcer medications. Although differences between the control and intervention groups were significant after 1 month, these differences had disappeared by the second month after the intervention. A British study targeted prescribing of nonsteroidal anti-inflammatory drugs in an intervention involving visits by a specially trained pharmacist.[20] Each community general practitioner in the intervention group received a single visit. The intervention group showed an improvement in the appropriateness of prescribing that lasted for at least 5 months. Finally, an Australian study showed that a group of community general practitioners who received mailed educational materials, followed by a visit from a pharmacist, had better compliance rates than a control group for prescribing antibiotics in tonsillitis.[21]

    In the five remaining trials feedback was provided on prescribing practice along with recommendations for changes in practice; three of these dealt with the issue of polypharmacy. Of the three, one compared a group of physicians who received feedback (involving chart review, two letters and individualized suggestions for reducing polypharmacy), a group that received a single letter identifying patients who had been given 10 or more prescriptions and recommending a reduction in the number of medications, and a control group that received no intervention.

    Polypharmacy was reduced in both intervention groups compared with the control group, but there was no statistically significant difference between the effects of the two interventions.[22] A second trial, conducted at an internal medicine clinic, showed that individualized oral and written recommendations to house staff regarding changes in medications resulted in a reduction in polypharmacy in the intervention group compared with a control group that received no feedback.[23] The final study on polypharmacy showed that pharmacists' consultations with elderly patients and their physicians before and after discharge from hospital resulted in more appropriate use of prescription drugs than when the patients and physicians received no such consultations.[24]

    The other two trials that investigated the effect of feedback focused on drug costs in outpatient general medical clinics. In the first, interns in the intervention group received a manual of comparative drug prices annotated with prescribing advice, two feedback reports and weekly cost-oriented prescribing reminders; the control group participated in a manual-based educational program on cholesterol management.[25] In the second, residents in both the intervention and control groups received monthly computer printouts summarizing the total number of prescriptions that they had written and the total cost of their prescriptions. The intervention group also received a monthly letter addressing different therapeutic topics.[26] In the first trial the intervention group prescribed less expensively than the control group;[25] in the second, the experimental and control groups did not have significantly different outpatient pharmacy charges.[26]

    Findings from the literature review

    The existing research indicates that the dissemination of printed material alone does not lead to improvements in practice, but specific educational and feedback strategies can improve the quality of care. Successful educational strategies involve face-to-face contact between an expert and the physician. Feedback that involves not simply a description of current practice but, rather, includes specific recommendations for change in the use of medications can also improve practice. Although both face-to-face education and feedback with recommendations are costly to provide, preliminary analysis suggests that both strategies can be very cost-effective and may reduce costs while improving the quality of care. In a broader context, research on provider behaviour has suggested that interventions that combine education and feedback are more successful than interventions that rely on a single strategy.[27]

    Limitations of current studies

    Aside from trials using academic detailing, nearly all published studies have been conducted in academic family practice units, large group practices or outpatient clinics. These may not be generalizable to community settings in Canada, which are dominated by solo and small-group practice. A second limitation is a lack of data on patient outcomes resulting from the interventions. Soumerai and associates[17] identified only three trials in which the effect on patients was measured, and our subsequent search did not find any additional outcome studies. Finally, few studies have examined a comprehensive approach to improving prescribing practice. Most have examined only limited components of prescribing, such as prescription costs,[18,25,26] polypharmacy,[22,23] the use of specific drugs such as antibiotics,[21] nonsteroidal anti-inflammatory drugs[20] and anti-ulcer agents,[19] or the treatment of specific problems such as hypertension or urinary-tract infections.[17] Only Lipton and collaborators[24] looked at a broader range of quality-of-care issues, including inappropriate choice of therapy, underdosage and overdosage, scheduling, drug-drug interactions, therapeutic duplication, allergies and the omission of necessary drug therapies.

    [Contents]

    Other potential techniques

    We have limited our review to published trials that examined the effects of various interventions on prescribing practice. Other strategies for improving prescribing practice have been suggested but not adequately tested. For example, there is evidence that expressing the results of drug trials in terms of relative rather than absolute risk reduction,[28] or expressing outcomes as averaged rather than stratified gains in life expectancy,[29] can change attitudes toward the use of drugs. Similarly, informing physicians about patients' drug-plan coverage and the costs of various therapies can alter hypothetical drug choices.[30] However, only prescribing intentions, which may be different from actual behaviour, have been measured to date.[31,32] Other strategies include the use of practice formularies,[33] pharmacy-enforced treatment protocols[34] and peer review.[35-37] Because patient expectations may influence prescribing,[38] providing patients with more complete information on the risks and benefits of drugs is another educational approach that requires further research.

    The use of online computerized systems to provide community-based physicians with rapid information and feedback could potentially improve office-based prescribing practice. However, we could not identify any published studies that systematically evaluated this strategy; to be feasible, such an approach will have to await the more widespread use of online computerized systems in physicians' offices.

    [Contents]

    Conclusion

    The appropriate use of prescription drugs by community-based physicians is a key element of high-quality primary care. We need to develop new ways to successfully provide up-to-date, accurate and relevant information to physicians on the everchanging and increasingly complex world of therapeutics. Three key issues need to be addressed.

    First, we need to undertake more research on strategies to improve office-based prescribing practice. This research should recognize the important practical and financial issues related to developing a comprehensive intervention. Second, we need to establish a mechanism that will draw on both existing and future research to design the complex and multifaceted interventions most likely to improve the quality and cost-effectiveness of care. Third, we need to promote the collaboration of governments, providers, manufacturers and the public that is necessary to ensure that Canadians receive the best care possible.

    We thank Mr. Craig Martin, Mr. Brad Buxton, Ms. Pat Lemay, Dr. William Pryse-Phillips, Dr. Donna Radmanovich, Ms. Barbara Shea, Dr. James Wright and Dr. Jeff Poston for their valuable contributions to discussions on the content of this paper.

    [Contents]

    References

    1. Pharmaceutical R & D: Costs, Risks and Rewards (cat no OTA-H-522), Office of Technology Assessment, Washington, 1993
    2. Anderson GM: An overview of the use of acute care hospitals, physician and diagnostic services and prescription drugs. In Naylor CD, Anderson GM, Goel V (eds): Patterns of Health Care in Ontario: ICES Practice Atlas, Canadian Medical Association, Ottawa, 1994: 21-45
    3. Einarson TR, Segal HJ, Mann JL: Drug utilization in Canada: an analysis of the literature. J Soc Admin Pharm 1989; 6: 69-82
    4. Tamblyn RM, McLeod PJ, Abrahamowicz M et al: Questionable prescribing for elderly patients in Quebec. CMAJ 1994; 150: 1801-1809
    5. Hogan DB, Campbell NRC, Crutcher R et al: Prescription of nonsteroidal anti-inflammatory drugs for elderly people in Alberta. CMAJ 1994; 151: 315-322
    6. Wilcox SM, Himmelstein DU, Woolhandler S: Inappropriate drug prescribing for the community-dwelling elderly. JAMA 1994; 272: 292-296
    7. Williams AP, Cockerill R: Report on the 1989 Survey of the Prescribing Experiences and Attitudes Toward Prescription Drugs of Ontario Physicians. Prepared for the Pharmaceutical Inquiry of Ontario (Lowy Commission), Toronto, 1989
    8. Health Professional Attitudinal Survey on the Use of Medicines and Drug Information Needs [report to the Canada Drug Regulatory Affairs Division, Health and Welfare Canada], Decima Research, Ottawa, 1992
    9. Credibility and the Marketing Mix, Angus Reid Group, Toronto, 1991
    10. Information Overload, Angus Reid Group, Toronto, 1993
    11. Parboosingh J, Lockyer J, McDougall G et al: How physicians make changes in their clinical practice: a study of physicians' perception of factors that facilitate this process. Ann R Coll Physicians Surg Can 1984; 17: 429-435
    12. Putnam RW, Fox RD, Mazmanian PE: An Analysis of Drug Prescribing Changes by Physicians [unpublished report to Pharmaceutical Manufacturers Association of Canada], PMAC, Ottawa, 1984
    13. Krogh CME (ed): Compendium of Pharmaceuticals and Specialties, 30th ed, Canadian Pharmaceutical Association, Ottawa, 1995
    14. Parker WA: The Compendium of Pharmaceuticals and Specialties as a drug information resource for treatment of acute drug overdose. Can Fam Physician 1979; 25: 211-215
    15. Whelan AM, Nagpal S, Burge F: Drug information resources in a family medicine residency training program. Can Pharm J 1994; 127 (9): 41-43, 46
    16. Hall KW, Parker WA: Physician's view of the pharmacist's professional role. Can Pharm J 1976; 109: 311-314
    17. Soumerai SB, McLaughlin TJ, Avorn J: Improving drug prescribing in primary care: a critical analysis of the experimental literature. Milbank Q 1989; 67: 268-317
    18. Steele MA, Bess DT, Franse VL et al: Cost effectiveness of two interventions for reducing outpatient prescribing costs. DICP 1989; 23: 497-500
    19. Raisch DW, Bootman JL, Larson LN et al: Improving antiulcer agent prescribing in a Health Maintenance Organization. Am J Hosp Pharm 1990; 47: 1766-1773
    20. Newton-Syms FAO, Dawson PH, Cooke J et al: The influence of an academic representative on prescribing by general practitioners. Br J Clin Pharmacol 1992; 33: 69-73
    21. De Santis G, Harvey KJ, Howard D et al: Improving the quality of antibiotic prescription patterns in general practice: the role of educational intervention. Med J Aust 1994; 160: 502-505
    22. Meyer TJ, van Kooten DV, Marsh S et al: Reduction of polypharmacy by feedback to clinicians. J Gen Intern Med 1991; 6: 133-136
    23. Kroenke LTCK, Pinholt EM: Reducing polypharmacy in the elderly: a controlled trial of physician feedback. J Am Geriatr Soc 1990; 38: 31-36
    24. Lipton HH, Bero LA, Bird JA et al: The impact of clinical pharmacists' consultations on physicians' geriatric drug prescribing. Med Care 1992; 30: 646-655
    25. Frazier LM, Brown JT, Divine GW et al: Can physician education lower the cost of prescription drugs? A prospective, controlled trial. Ann Intern Med 1991; 115: 116-121
    26. Hershey CO, Goldberg HI, Cohen DI: The effect of computerized feedback coupled with a newsletter upon outpatient prescribing charges. Med Care 1988; 26: 88-93
    27. Davis DD, Thomson MA, Oxman AD et al: Evidence for the effectiveness of CME: a review of 50 randomized trials. JAMA 1992; 268: 1111-1117
    28. Hux JE, Levinton CM, Naylor CD: Prescribing propensity: influence of life expectancy gains and drug costs [working paper series no 009], Institute for Clinical Evaluative Sciences in Ontario, Toronto, 1993
    29. Bobbio M, Demichelis B, Giustetto G: Completeness of reporting trial results: effect on physicians' willingness to prescribe. Lancet 1994; 343: 1209-1211
    30. Hux JE, Naylor CD: High prices and third party payment: Do they influence medication selection? [working paper series no 13], Institute for Clinical Evaluative Sciences in Ontario, Toronto, Aug 1993
    31. Manning PR, Lee PV, Clintworth WA et al: Changing prescribing practices through individual continuing education. JAMA 1986; 256: 230-232
    32. Jones TV, Gerrity MS, Earp J: Written case simulations: Do they predict physicians' behavior? J Clin Epidemiol 1990; 43: 805-815
    33. Lexchin J: General practice formularies: The way ahead? Can Fam Physician 1989; 35: 1569-1570
    34. McCombs JS, Nichol MB: Pharmacy-enforced outpatient drug treatment protocols: a case study of Medi-Cal restrictions for ceflacor. Ann Pharmacother 1993; 27: 155-161
    35. Reilly PM, Patten MP: An audit of prescribing by peer review. J Coll Gen Pract 1978; 28: 525-530
    36. Grol R, Mokkink H, Schellevis F: The effects of peer review in general practice. J Coll Gen Pract 1988; 38: 10-13
    37. Von Ferber L, Luciano A, Koster I et al: Drug utilization research in primary care as exemplified by physicians' quality assessment groups. Int J Clin Pharmacol Ther Toxicol 1992; 30: 453-455
    38. Schwartz RK, Soumerai SB, Avorn J: Physician motivations for nonscientific drug prescribing. Soc Sci Med 1989; 28: 577-582
    GO TO CMAJ: Apr. 1, 1996