Abstract
OBJECTIVE To explore the attitudes of FPs toward benzodiazepine (BZD) prescribing and the perceived barriers to nonpharmacologic approaches to managing stress, anxiety, and insomnia.
DESIGN A questionnaire including 32 statements about treatment of insomnia, stress, and anxiety.
SETTING Local quality groups for FPs in Belgium.
PARTICIPANTS A total of 948 Belgian FPs.
MAIN OUTCOME MEASURES Barriers to using nonpharmacologic approaches in family practice.
RESULTS We identified 3 different groups of FPs according to their attitudes about BZD prescribing. A first relatively big group of FPs (39%) were not really concerned about the risks of BZD prescribing. Those in the second group (17%) were aware of the problems associated with BZDs, but did not perceive it to be their role to use nonpharmacologic approaches in family practice. Those in the third group (44%) were concerned about BZD prescribing and found it to be a “bad solution,” but were faced with various barriers to applying nonpharmacologic approaches. Surprisingly, we found that nearly 97% of FPs thought that most people were eligible for nonpharmacologic approaches, but experienced implementation barriers at the level of the patient, the level of the FP, and the level of the health care system.
CONCLUSION Using different education and behavioural-change strategies for different FP groups seems important. A large group of FPs does not find prescribing BZDs to be problematic. Sensitizing and alerting FPs to this issue remains very important.
Family physicians are frequently consulted by patients suffering from stress, anxiety, or insomnia; FPs consider treatment of such problems to be part of their role,1 as they are ideally placed to recognize the strengths, resources, and vulnerabilities of their patients and to incorporate these factors into their patients’ treatment plans. A Dutch study showed that nearly all patients with sleeping disorders seeking treatment in family practices received psychotropic drugs, mainly benzodiazepines (BZDs).2 Family physicians perceiving BZDs to be a problem need more information about the effectiveness of nonpharmacologic approaches. The more motivated FPs need opportunities for training to advance their knowledge and to give them additional skills.
A cross-national study in the 1980s showed that Belgium was one of the countries with the highest use of anti-anxiety and sedative drugs.3 Since then, the use of BZDs has further increased.4,5 One out of 3 patients takes BZDs chronically and daily.6 European studies examining use over longer periods have produced similar findings of high use; for example, the European Study of the Epidemiology of Mental Disorders found that 9.8% of the population was using BZDs at some point over the past 12 months.7,8 Use is lower in the Canadian population, where the overall weighted frequency of use was 3.4%.9 However, once patients have started taking BZDs, a more or less consistent pattern of continued use for long periods of time is common, irrespective of countries’ professional standards.10 The benefits associated with sedative use are marginal and are outweighed by the risks, particularly in people older than 60 years of age.11,12 Acute administration of BZDs is associated with polypharmacy complications and impairments in cognition, memory, coordination, and balance. Long-term use, even at therapeutic dosages, has been associated with tolerance, dependence, and withdrawal effects.13–16 Apart from the risk of abuse and primary dependence, there is also the risk of low-dose dependence when BZDs are prescribed in clinically recommended oral doses; this is of particular importance in light of the high prevalence of long-term BZD treatment.17 Because of the risk of tolerance and the difficulty for patients to motivate themselves to stop taking BZDs,18 FPs should be careful when prescribing these drugs. Nonpharmacologic approaches should play an important role in the management of stress, anxiety, and insomnia, as they have been shown to be effective in the short-term and tend to have more durable effects.19–23
The best way to avoid dependence is to not initiate treatment with BZDs. Given the dearth of data on BZD prescribing practices in family practice, we performed a qualitative study24 to understand FPs’ views on initiation of BZD treatment and their perceptions about nonpharmacologic alternatives. One of the findings was that FPs resorted to BZD prescribing because of time constraints and a lack of usable alternatives. The FPs’ main concern was to help their patients and, therefore, they demonstrated their empathy by prescribing. Another study looked at first-time users’ attitudes and beliefs surrounding initiating BZDs and nonpharmacologic alternatives.25 The results showed that first-time BZD users asked for help with their distress, but placed the responsibility for solving their problems on their FPs. From both studies it seems that the attitude of the FPs toward BZDs and nonpharmacologic alternatives is an important factor in whether or not BZD treatment is initiated. It is clear that physicians’ knowledge and attitudes have an influence on their prescribing practices; further, physicians’ underlying beliefs, values, and perceptions of the benefits and risks of drugs seem to matter as well.
Objective
After describing how prescriptions for BZDs come about24,25 and how barriers to nonpharmacologic approaches are perceived, we investigated the prevalence of such perceptions in a large group of FPs. This is important information if decisions on interventions and training are going to be made to achieve a more rational use of BZDs and avoid dependence.
METHODS
Design
A standardized questionnaire was used. The questionnaire included 32 statements that had to be answered on a 5-point Likert scale that ranged from “fully disagree” to “fully agree.” The content of the questionnaire was based on 2 published qualitative studies24,25 and comprised 5 different sections: 1 for FPs’ general attitudes of toward BZD prescribing, 1 for attitudes toward recommending nonpharmacologic approaches, and 3 for recommending nonpharmacologic approaches for the management of stress, anxiety, and insomnia.
Setting and participants
Data were gathered from 948 FPs before they attended a training module on the rational use of BZDs. The training was funded by the Belgium Federal Public Service for Health. Ethics approval was granted by the Ethics Committee of Ghent University Hospital.
In Belgium, the 10 116 accredited FPs receive continuing medical education in local quality groups. Belgian FPs are obliged to attend at least 2 meetings a year for renewal of their certification. An opportunistic sample of FPs attending these training sessions on the rational prescribing of BZDs was used. The questionnaire was completed before the session started. The sample of FPs under consideration was weighted in order to be representative of the entire population of accredited FPs (data derived from e-mail communication with Belgian’s National Institute for Health and Disability Insurance, December 2009). No stratification or clustering was involved in the sample design.
Statistical methods and data analyses
To explore the data, cross-tabulations with χ2 statistics were performed using SAS 9.1 for Windows. Rao-Scott and Wald χ2 statistics were computed. The statistically significant level was established at P < .05. We also looked at the dichotomy of “non-believers” and “believers” in nonpharmacologic approaches by collapsing FPs who answered “agree” or “fully agree” for the statement “Nonpharmacologic approaches are the role of the FP” into one group (believers) and comparing them with FPs who answered “neither agree nor disagree,” “disagree,” or “fully disagree” for this statement (non-believers). For all complaints, we grouped the items on “knowledge,” “motivation,” “being eligible,” and “self-confidence” together. This reduction of information was validated by Cronbach α’s of .79 (knowledge), .74 (motivation), .75 (being eligible), and .77 (self-confidence) for the items under consideration.
RESULTS
Participant characteristics
The 948 FPs who completed the survey were representative of FPs practising in Belgium in terms of sex—one-third of the respondents (34%) were female. The median age of Belgian FPs is 51 (data derived from e-mail communication with Belgian’s National Institute for Health and Disability Insurance, December 2009); in our sample the median age was 48 (Table 1). Younger FPs (ie, younger than 40 years) are overrepresented in our sample, and older FPs (ie, older than 60 years) are underrepresented (
Age of the study population compared with national figures for accredited Belgian FPs
Attitudes toward BZDs
Forty percent of the FPs did not perceive BZDs to be a problem in family practice. Almost 1 out of 4 FPs thought chronic use of BZDs was justified as long as the patient functioned better and did not experience any side effects.
More than 2 out of 3 respondents (71%) felt justified in initiating BZD treatment for a week. Nearly all FPs (96%) thought it was important to inform patients about BZD dependence. Family physicians’ attitudes toward BZD prescribing are influenced by their perceptions of patients’ attitudes. Almost half of the FPs (47%) thought that patients expected prescriptions, and as many as 18% thought that not writing prescriptions would threaten the doctor-patient relationship. Sixty percent acknowledged difficulties motivating patients to stop taking BZDs, yet only 56% believed they knew how to manage a patient’s withdrawal from BZDs.
Table 2 shows significant differences in attitudes according to age (P < .01). Younger FPs were more resistant to prescribing BZDs; older FPs put more emphasis on the role of perceived patient expectations. Within the age categories there were no significant effects relating to sex of the FPs.
Participants’ attitudes toward BZDs
Attitudes and barriers to nonpharmacologic approaches
As shown in Table 3, only 26% of FPs agreed with the statement that nonpharmacologic approaches needed to be supported with medication. Nearly half of FPs perceived prescribing or recommending nonpharmacologic approaches to be too time-consuming, and more than half of FPs (55%) found referral to be too expensive and motivating the patient to be too difficult. One out of 4 (24%) thought that patients did not feel that their FPs were taking them seriously if they did not receive prescriptions. For all barriers, there was a significant effect of age: older FPs were more likely to experience these barriers (P < .05).
Barriers to nonpharmacologic approaches
Table 4 describes responses to items about barriers to nonpharmacologic approaches for complaints of stress, insomnia, or anxiety. Most respondents considered nonpharmacologic approaches applicable for a range of people suffering from stress, insomnia, or anxiety, and many people are eligible for this kind of treatment. However, only 1 in 3 FPs believed that their knowledge of nonpharmacologic approaches was sufficient, and less than half felt self-confident. For insomnia, more FPs were confident in recommending nonpharmacologic approaches and thought they had sufficient knowledge to do so. Yet FPs found it more difficult to motivate patients with complaints of anxiety and stress to try nonpharmacologic treatment. In the case of insomnia, most FPs thought that patients were aware of what sleep hygiene entailed. In particular, older FPs (41% compared with 14% of the younger FPs) did not find sleep hygiene useful, “as the patient does not want to change his behaviour.”
Attitudes and barriers to nonpharmacologic approaches to the treatment of stress, insomnia, or anxiety
Perceived barriers for “believers” versus “non-believers”
Family physicians’ perceptions of nonpharmacologic approaches varied for the different complaints under consideration: 65% of FPs saw these types of approaches as part of their role in treating cases of insomnia, 60% thought they were part of their role in treating stress, and 58% thought they were part of their role in treating patients with anxiety. Forty-one percent of respondents said they would consider nonpharmacologic approaches for all 3 complaints.
To identify barriers that might account for differences in the extent to which FPs recognize the importance of and are open to using nonpharmacologic approaches in family practice, FPs were divided in 2 groups: “believers” and “non-believers.” For the barriers at the patient level, we found 2 out of 3 items were significantly different (P < .05) between “believers” and “non-believers” (Table 5). These differences illustrate that FPs who thought that nonpharmacologic approaches were part of their role in family practice reported fewer barriers to such approaches than the “non-believers.” The cost of a nonpharmacologic approach was a barrier for all FPs, but “non-believers” found it more difficult to motivate patients. With respect to barriers at the physician level, not feeling self-confident (P < .05), lack of knowledge about nonpharmacologic approaches (P < .001), and the perceived time investment (P < .001) were all significant.
Perceived barriers to nonpharmacologic approaches according to “believers” vs “non-believers”: “Believers” agreed or fully agreed with the statement “Nonpharmacologic approaches are the role of the FP”; “non-believers” neither agreed nor disagreed, disagreed, or fully disagreed with this statement.
DISCUSSION
Our study examined FPs’ attitudes toward the role of BZDs and nonpharmacologic approaches to managing insomnia, stress, and anxiety, and identified barriers that might account for differences in the extent to which FPs recognize the importance of and are open to using nonpharmacologic approaches in family practice. A striking finding was that nearly 40% of the respondents did not perceive BZDs as a problem in family practice, and 2 out of 3 physicians did not have a problem with prescribing BZDs for 1 week. On their own these do not really pose a problem, as guidelines suggest that in some cases BZDs can be useful for a short period of time.26,27 Unfortunately, BZDs are often used for months at a time for the management of insomnia, stress, and anxiety. A Dutch-Swedish study compared an 8-year follow-up pattern of long-term use with respect to the characteristics of those with and without continued BZD use over the whole period of 8 years.10 Both populations showed that 2 out of 3 patients continued taking BZDs in the first year of the follow-up period. At the end of the 8 years, approximately 1 in 3 patients from the initial cohort were still receiving BZDs.10 Once patients have started taking BZDs, a more or less consistent follow-up pattern of continued use for long periods of time can be anticipated. The first prescription for BZDs can be the start of a long-lasting experience with BZD use.
Despite the physicians’ recognition of the importance of nonpharmacologic approaches and the fact that they found many patients to be eligible for these types of strategies, it is important to point out that a substantial number of physicians remained sceptical about the role of the FP in nonpharmacologic approaches. This scepticism seems to be related to different perceptions of the issue.
First, we have a group of mostly older FPs who do not appear to be concerned about the risks of BZD prescribing. This is a difficult group to reach in terms of motivating patients to withdraw from BZDs and changing prescribing behaviour, owing to a lack of motivation,24 and improvement of knowledge and skills alone might not solve the problem.24,28 It is now widely accepted that BZD prescribing has many risks, including tolerance, dependence, and misuse, as well as BZD-induced depression, cognitive impairment, and psychomotor impairment.3,14,29–31 This group of FPs will have to be informed about the potential consequences of chronic use of BZDs before any other intervention is planned.
A second group of FPs are aware of the problems associated with BZDs, but do not perceive it to be their role to use nonpharmacologic approaches in family practice. In our survey, approximately 1 in 2 physicians shared this opinion. The older the FP, the more resistance they showed to (exclusively) nonpharmacologic approaches. It has been demonstrated that younger FPs are more inclined toward psychosocial assessment of patients’ health,32–34 and that they have higher ideals and are more enthusiastic in keeping with them.35 In our previous qualitative study,24 FPs expressed that they were uncertain of the psychosocial services to which they could refer patients; for this group of FPs, information and understanding of how and where to refer their patients to psychosocial services could be helpful as a minimal intervention. It is up to professional groups of FPs to discuss whether nonpharmacologic approaches should be a core concern of FPs.
A third group of FPs do take responsibility for nonpharmacologic approaches; however, they experience hindrances at different levels. This group needs to be made aware of studies that have looked at the feasibility of nonpharmacologic interventions suitable for primary care settings.36–38 To help motivate FPs to overcome these difficulties, we have to remove barriers at 3 different levels.
Barriers at the level of the FP
Our respondents did not have enough knowledge of nonpharmacologic approaches and did not feel confident in these matters. Even those who supported nonpharmacologic approaches believed that their knowledge was not sufficient. Yet, studies have found that if a patient thinks that the FP is taking his or her problem seriously, the patient will open up to the FP.25 The patient does not expect the FP to be an expert in all matters.19 A study by Lang39 showed that patients feeling distressed often think that understanding the cause of their distress would be most helpful. Also, patients express more satisfaction with medical care when information is given to them.40,41 Providing FPs with practical tools and training in counseling and different nonpharmacologic approaches is very important to strengthening their confidence and is a first essential step in the process of rational prescribing of BZDs.24,42
Barriers at the level of the patient
An interesting finding was the FPs’ perception that patients expected prescriptions. This does not seem to correspond to the findings of research examining patients’ points of view of treatment by their personal physicians for emotional distress.43 Most patients in that study preferred that their physicians provided counseling, not drugs, while only 23% wanted medication and 11% desired referrals. Perceived patient expectations are a strong predictor of the decision to prescribe.44 Doctors’ assessments of patients’ expectations are often based on an intriguing variety of cues.45,46 By asking patients about their experiences and beliefs, physicians can open dialogue and provide the patient with information.46,47 A recent study has shown that exploring and clarifying ideas, concerns, and expectations might lead to fewer new medication prescriptions.48
Some FPs want to appear “scientific” or independent in their decision making49; this could be seen as socially accepted behaviour. They deny that patient demands for medications substantially influence their prescribing behaviour.50 Also, perceived patient expectations of drug treatment can be seen as a “rationalization” for FPs’ own lack of knowledge of nonpharmacologic approaches. Many FPs thought that referral was too expensive for patients or that it was too difficult to motivate patients to see therapists, but most FPs did not feel confident enough in their own knowledge of nonpharmacologic approaches.
In our survey, more than 25% of FPs said that sleep hygiene was not useful because patients would not want to change their behaviour. This is not in accordance with the results of a study of patients’ treatment preference for chronic insomnia, which showed that psychological treatment was preferred to pharmacologic treatment, and that of all treatment components, sleep hygiene was rated as the most liked and most useful.19
Barriers at the level of the health care system
One out of 2 respondents reported that a lack of time limited their ability to integrate nonpharmacologic alternatives into their patient care practices. This barrier is important in countries with fee-for-service models. Family physicians in fee-for-service systems tend to spend more time on profitable activities than on “costly” time-consuming interventions.51 More appropriate reimbursement, similar to the situation in Switzerland for longer psychosocial consultations, might be a solution for this.52 The increasingly restricted time physicians have to spend with their patients results in prescribing because FPs do not have sufficient time to adequately address the psychosocial domain of patients lives.53 Yet, another study showed that a doctor’s decision about whether or not to prescribe for psychosocial problems had no relationship to the consultation length.54 Nonpharmacologic approaches can be as simple as educating patients about stress, insomnia, or anxiety. Another solution to time pressures could be actively referring patients to psychosocial services, with regular debriefing sessions with the patients.
Strengths and limitations
A self-reported questionnaire was chosen because we considered it to be consistent with the purpose of the study. Self-reported measures are essential when the purpose is to obtain subjective assessments of experiences.55 We did not use a validated questionnaire because none was available. Content validity was achieved by developing the questionnaire based on results from previous studies24,25; this should ensure the relevance of the items.
It is difficult to determine the extent to which the attitudes and patterns we identified are representative of the larger population of FPs. A key limitation is our use of a nonrepresentative sample of FPs who already had enough interest in the topic to attend a meeting on the rational use of BZDs; however, we have weighted our sample with the percentages of the total accredited Belgian FPs. Moreover, it is reasonable to think that the general group of physicians might be even less likely to recognize their role in nonpharmacologic approaches and the importance of such strategies, as there was already considerable scepticism among respondents in our sample. Our goal was to explore different barriers to prescribing or recommending nonpharmacologic approaches, and in this group it was interesting to find strategies that can be applied to other groups of FPs. In addition we did not measure the FPs’ actual behaviour, but their perception of it. Another limitation is that we focused on the physician’s decision to start BZD treatment and did not measure attitudes toward discontinuation of long-term use; evidence suggests that many physicians do not attempt tapering. Yet BZD tapering is feasible for FPs and can be followed by improved psychomotor and cognitive functioning for patients.56–59
Our results do provide an insight into the attitudes of FPs toward a difficult area of practice and should help us in developing potential strategies to help FPs and interventions to change FP behaviour.
Conclusion
Among a 9.5% convenience sample of Belgian FPs, nearly half did not have a problem with BZD prescribing or did not find it to be problematic. When aiming to reduce BZD prescribing, one has to make sure that the target group is aware of the problems of prescribing before applying interventions to change behaviour. Sensitizing and alerting FPs to this issue remains very important.
If BZDs are perceived to be a problem, then FPs need to be aware that there are other options available. More information is needed about the effectiveness of nonpharmacologic approaches. Family physicians need more training to advance their knowledge and to give them more skills.
Acknowledgments
We are indebted to the doctors who participated in the study and to the trainers who made this study possible. Funding for the study was received from the Federal Public Service for Health, Food Chain Safety and Environment.
Notes
EDITOR’S KEY POINTS
-
The benefits associated with benzodiazepine (BZD) use are marginal and are generally outweighed by the risks. Short-term use is associated with polypharmacy complications and impairments in cognition, memory, coordination, and balance. Long-term use, even at therapeutic dosages, has been associated with tolerance, dependence, and withdrawal effects. There is also a risk of low-dose dependence when BZDs are prescribed in clinically recommended oral doses, which is important in light of the high prevalence of long-term BZD treatment.
-
Because of the risk of tolerance and the difficulty for patients to motivate themselves to stop taking BZDs, FPs should be careful when prescribing these drugs. Nonpharmacologic approaches should play an important role in the management of stress, anxiety, and insomnia, as they have been shown to be effective in the short-term and tend to have more durable effects.
-
Family physicians perceiving BZDs to be a problem need more information about the effectiveness of nonpharmacologic approaches. The more motivated FPs need opportunities for training to advance their knowledge and to give them additional skills. Those who do not find BZDs to be problematic need further education to be sure they are aware of the problems surrounding BZD prescribing.
POINTS DE REPÈRE DU RÉDACTEUR
-
Les avantages de l’utilisation des benzodiazépines (BZD) sont marginaux par rapport aux risques qu’ils entraînent. Leur usage a court terme se complique de polymédication et de problèmes de cognition, de mémoire, de coordination et d’équilibre. Leur usage à long terme, même en doses thérapeutiques, s’accompagne de tolérance, de dépendance et de problèmes de sevrage. Il existe aussi un risque de dépendance aux faibles doses lorsque les BZD sont prescrits aux doses orales cliniquement recommandées, ce qui s’avère important vue la forte prévalence de l’administration de BZD à long terme.
-
À cause du risque de tolérance et parce que les patients ont de la difficulté à cesser de consommer des BZD, le MF devrait prescrire ces médicaments avec prudence. Les approches non pharmacologiques devraient jouer un rôle important dans le traitement du stress, de l’anxiété et de l’insomnie puisque leur efficacité à court terme a été démontrée et que leurs effets tendent à être plus durables.
-
Les MF qui considèrent que les BZD sont problématiques ont besoin d’être mieux renseignés sur l’efficacité des approches non pharmacologiques. On devrait offrir aux plus motivés des occasions de parfaire leurs connaissances à ce sujet et leur enseigner des habiletés additionnelles. Ceux qui ne trouvent pas que les BZD sont problématiques devraient avoir davantage de formation pour s’assurer qu’ils sont au fait des problèmes entourant la prescription des BZD.
Footnotes
-
This article has been peer reviewed.
-
Cet article a fait l’objet d’une révision par des pairs.
-
Contributors
All authors contributed to the concept and design of the study; data gathering, analysis, and interpretation; and preparing the manuscript for submission.
-
Competing interests
None declared
- Copyright© the College of Family Physicians of Canada
