Placebo
The use of, and attitude towards, placebo according to general practitioners and patients
Abstract
Objectives
The aim of this study is to evaluate how often Dutch general practitioners prescribe both inactive and active placebos and the reasons for prescribing them. Also the attitude of patients towards placebo was investigated.
Methodology (Design Setting Participants)
A questionnaire was sent to 400 randomly drawn general practitioners. Another questionnaire was distributed among 150 patients visiting one local pharmacy. These questionnaires consisted of questions about the use of, and attitude towards, placebo and included four case histories.
Results
The inactive placebo was prescribed by 14% of the general practitioners at least once a year. The active placebo was prescribed by 87% of the general practitioners at least once a year. Of the general practitioners 71% considered a placebo effective at least in some cases. Mostly prescribed active placebos were antibiotics, physiotherapy and vitamins. The most important reason for prescribing a placebo was the demand for a specific medicine from a patient. Most patients namely 69% disliked receiving a placebo.
Conclusion
Even though the WGBO and evidence based medicine placebo use is still very common. It seems that the inactive placebo is prescribed a lot often then the inactive placebo. A fifth of the general practitioners thought they could test with placebos whether the disease is functional or organic. Proper schooling of general practitioners in the use and effect of placebos is required.
Introduction
Placebos can be divided into two groups. The original or so-called inactive placebo is a placebo with no pharmacological active ingredients. For example lactose pills. The active placebo has a pharmacological active ingredient however the ingredient is not relevant to the specific symptoms being treated. For example antibiotics for a viral infection. The placebo can have a positive effect on a variety of symptoms like pain, anxiety, seasickness etc. Beecher estimated a placebo effectiveness of 35.2% ± 2.2% in providing ≥50% relief in patients suffering from back-pain. This percentage holds for most human pharmacological studies. Placebo use, just like the use of any medicine, can lead to side effects, like headache, nausea, drowsiness, dry mouth and rash (1).
The use of the inactive placebo is little and declining. According to the SFK, a database recording data of nearly all pharmacies in the Netherlands(1.590) , the inactive placebo was prescribed approximately 1000 times in 2005 (2). However, there is little information on the use of the active placebo because most prescription databases do not contain information on the indication of prescribing. There are two reasons why the use of placebo has become more and more questionable lately. First treatments of patients should be based on the principle of evidence-based medicine. According to this principle pharmacological compounds should be prescribed of which it is proven in a randomized controlled trials that their beneficial effects are statistically significant and clinically relevant different from placebo. Second in the Netherlands there is the Law of medical treatment agreement (WGBO)in which it is stated that the patient has the right to know everything about his condition and the treatment (informed consent). So the general practitioner has the duty to inform the patient about his treatment, placebo or not. Because of these two developments one might expect that placebos become out of grace.
Several studies have been published on the use of and attitude towards placebo of which most of them are more than 25 years old. Furthermore in most of the previous surveys not a clear distinction was made between active and inactive placebos. These articles reported that between 51% and 86% of hospital based physicians, general practitioners or nurses had used placebos (3-6), although in one study only one in 74 general practitioners used a placebo (7). The aim of our study was to evaluate how often the active and inactive placebos are presently prescribed and what the opinion is of general practitioners and patients towards placebo use.
Materials and Methods
Study design
The study design was a cross-sectional study using questionnaires to obtain information.
Participants
A questionnaire was sent, by mail, to 400 Dutch general practitioners randomly drawn from a total population of about 8.400 general practitioners in the Netherlands. The questionnaire was accompanied by an information letter in which also the definitions of active and inactive placebos were explained. A pre-stamped return envelope was also included. Also 150 patients visiting a local pharmacy received a questionnaire. They also received information about the study and placebo and a pre-stamped return envelope.
Questionnaire
The general practitioners received a questionnaire that consisted of demographic, multiple choice and open-end questions about the use, and attitude towards, placebo. It also included four case histories, each followed by four statements. Response alternatives were graded on the following scale: “agree completely”, “agree for the most part”, “disagree for the most part”, and “completely disagree”. Three of these cases were copied from Lynöe N. et al (8).
The patients received a questionnaire that also consisted of demographic, multiple choice, and open-end questions about their opinion to the use of placebo. Furthermore they received the same four case histories as the general practitioners.
Data analysis
Data were analyzed with SPSS 10.0. Descriptive statistics were used to present question answers as numbers and percentages. For some questions multiple answers could be given so percentages may add up above 100%.
To evaluate differences, in responses, between general practitioners and patients on the four cases the chi-square test was used. (р ≤ 0,05 was considered significant).
The influence of age (continuous) on the frequency of prescription of inactive and active placebo was tested with Oneway ANOVA. The influence of sex and practice size on the frequency of prescription of inactive and active placebo was tested by the Kruskal-Wallis Test.
Results
Participants
Of the 400 questionnaires distributed to the general practitioners 116 (29%) were returned. Eight general practitioners were retired, 1 wasn’t a general practitioner and 14 were blank. 93 (23%) questionnaires were completed. 48 (52%) general practitioners were male and 45 (48%) were female.
Of the 150 questionnaires distributed to the patients 72 (48%) were returned. 34 (47%) patients were male and 38 (53%) were female.
There were no significant differences in prescription habits, among respondents with respect to age, sex and practice size). Except for the influence of age on the prescription of inactive placebos. Older general practitioners prescribe significant more inactive placebos in regard to their younger colleagues (P ≤0.05).
Placebo use
Of the general practitioners 86% never prescribed inactive placebos, while 14% stated to prescribe inactive placebos at least once a year. The active placebo was prescribed by 87% of the general practitioners at least once a year. Thirteen percent of GP’s prescribed active placebos daily, 32% once a weak, 32% once a month and 10% once a year. Of the general practitioners 6% thought to prescribe placebos more often then their colleagues, 37% thought their prescribing was similar, 32% thought they prescribed them less often, and 26% had no idea how often their colleagues prescribe placebos.
Of the patients 22% did not know whether they had ever received a placebo, 57% knew for certain they never received a placebo, 19% thought they probably never received a placebo, and not more then 1 patient thought he/she had ever received a placebo. Sixty nine percent of the patients wouldn’t appreciate it when they had ever been prescribed a placebo, 24% indicated that they did not care and 7% had no opinion.
In response on the question how often the patient thinks a GP prescribes placebos 21% thought daily, 22% once a week, 21% once a month, 20% once a year and 16% thought a general practitioner never prescribes a placebo.
Active placebos
Of the general practitioners that prescribed active placebos (87%) antibiotics were prescribed by 75%, physiotherapy by 67%, vitamins by 48%, homeopathic products by 32%, analgesics in too low doses by 15%, and 13% stated to use other kinds of treatments like inhibin, acupuncture, bisolvon and duspatal.
Circumstances of use
The main reasons for a general practitioner to prescribe a placebo were: in 68% to avoid a conflict when the patient asked for a specific treatment, in 52% to deliberately make use of the placebo effect, in 33% to please the patient, in 21% to differentiate between a functional or organic disease, in 15% to keep his or her reputation high and keep the trust of the patient, in 11% to avoid a confrontation with the patient to tell hem/her there are no treatment options anymore, and 8% had other reasons for prescribing a placebo
On the question for which conditions placebos may be prescribed 33% allowed it in case there is no treatment for the disease, 25% allowed it when all other treatments had failed, 21% allowed it when a patient asks for it, 10% allowed it to treat an “annoying” patient, 19% allowed it for non serious diseases, 14% allowed it for terminally ill patients to avoid adverse drug reactions and 21% of the patients were in all circumstances against the prescribing of placebos. Only a few times were mentioned to allow it in a trail(n=4), to allow it in case a patient has a imaginary illness(n=3) and 1 patient had the opinion that the use of placebo is always allowed.
Information given to the patient by the general practitioner
The information about the placebo treatment given to the patient was as follows: 57% told the patient it was a medicine, 4% said it was a placebo, 14% said it was a medicine without a specific effect, 6% said nothing at all and 38% said something different. Typical statements were ‘I don’t think it will help but I will just give it to you because you want it’, ‘it has a good chance to work’, ‘let see what happens’, ‘your symptoms will go away with time, with this drug maybe somewhat faster’ or ‘I have good experience with this treatment’.
Effectiveness of the placebo
Of the general practitioners that prescribed placebos 25% considered them to be effective in most cases, 46% considered them to be effective in some cases, 25% had no idea about effectiveness and 4% considered them to be ineffective.
Declining use of the inactive placebo
According to the general practitioners important reasons for the declining use of the inactive placebos are the fact that the healthcare companies think it is unethical (52%), that it is important to follow the principles of evidence based medicine (48%) and the fact that the recently introduced WGBO states that patients should be fully informed about their treatment (43%). Other reasons were the easy access for patients to medical information (34%) the emancipation of patients (27%), replacement of the inactive placebo by the active placebo (19%), availability of more and better treatments for patients (9%), fear for law suits (2%), the fact that placebos have no leaflets, that they are not available anymore, that a doctor nowadays is more inclined to tell the patient that there are no treatment options anymore and finally that it is unfair to cheat patients.
Cases
Case 1:
The majority of both general practitioners and patients thought that a general practitioners not always has to oblige the wishes of the patient. If taking the placebo effect into account ±50% of both groups thought that it is acceptable to prescribe a active placebo in this case. Most general practitioners and patients stated that the procedure is unacceptable because it goes against science and proven experience. Half of the patients considered the replacement of the active placebo for a inactive placebo, while a majority of general practitioners found this unacceptable (Table 1).
Case 2:
Both general practitioners and patients thought it is unacceptable to give placebos to an terminally ill patient, even though the risk is small that she discovers the misconception. Prescribing a placebo became more acceptable for the statement that it preserves the patients hope without making her final time unbearable. The majority of general practitioners and patients were against this procedure because the patient has not been given the adequate information and the faith in the healthcare system might be shaken when this becomes common knowledge (Table 2).
Case 3:
The majority of general practitioners and patients thought that the general practitioner has to persuade the patient but also that the general practitioner has to respect the patients desire to refuse the ordinary treatment. Most patients thought iscador has to be injected in order not to reduce the placebo effect, while most general practitioners thought not to agree to the patient because it goes against science and proven experience (Table 3).
Case 4:
A large majority of patients thought it was acceptable if the general practitioner prescribes a placebo to make use of the placebo effect. Placebo prescription is also acceptable for most of the patients because the general practitioner knows what’s best for the patient or to test whether a benzodiazepine is necessary. The majority of general practitioners had problems to accept this procedure for these reasons. Both general practitioners and patients thought it is unacceptable to prescribe a placebo because the patient hasn’t been told she is receiving a placebo (Table 4).
Discussion
Although the WGBO and evidence based medicine is of great importance nowadays this study shows that placebo medication still takes place. It seems that the inactive placebo is not that popular by the younger generation of general practitioners. It seems that older general practitioners prescribe more inactive placebos then their younger colleagues. Every year some of these older general practitioners goes with pension. This might explain the declining use of the inactive placebo. Only 14% of the Dutch general practitioners prescribe the inactive placebo at least once a year. The declining use of this placebo occurs because of various reasons such as the WGBO, evidence based medicine and because of the ethics. However the active placebo is prescribed a lot often. 87% of the general practitioners prescribe the active placebo at least once a year and almost 50% of the general practitioners state to prescribe the active placebo at least once a weak. So we can say that the active placebo use is common among the Dutch general practitioners. Even though this common use of placebos still only one of the questioned patients thinks he/she received a placebo at some time. The rest of the patients either think they never had any placebo or think to know it for sure. This is to be expected because: around 50% of the general practitioners prescribe a active placebo at least once a weak. For a Dutch general practitioner it is normal to see about 200 patients a weak. So roughly 1:400 patients that visits a general practitioner receives a active placebo. This would be slightly more because the general practitioners that prescribe placebos as often as once a month and once a year are left out of the equation.
We found that especially antibiotics, physiotherapy, vitamins and homeopathic medicines where given as active placebo treatment. Most of these placebos are harmless except maybe for the antibiotics. If used unnecessary much, resistance of various bacteria can become problematic. So with this in mind it seems that the active placebo has even more disadvantages in regard to the inactive placebo. In this light it is strange that the active placebo is more frequently prescribed then the inactive placebo.
The most important reason for the prescription of placebos is the demand for a specific medicine by patients. Of the general practitioners 62% state to prescribe an active placebo because the patient asks for it. This might explain the higher frequency of active placebo prescription.
Even while placebo use is common not every general practitioner knows what a response to a placebo treatment means for the condition of the patient. According to the questionnaire 19% of the general practitioners think they can test with a placebo whether a condition is functional or organic. A good example is the general practitioner that prescribes a placebo and later on confronts the patient with it and tells him the disease is not real because he responded to the placebo. A response to a placebo doesn’t mean that the disease is imaginary. It is proven that the placebo effect takes place in many conditions that have a real organic origin.
The different cases showed that patients have no problems if the general practitioner should prescribe a placebo whether the patients asks for it or not. But when asked what the patients would think if they received a placebo almost all patients filled in that they wouldn’t appreciate it.
Strengths and (weaknesses/limitations)
Some general practitioners may find active placebo hard to define and may mistake it with the use of a medicine for a condition not yet proven for. If happened this leads to an over estimation of the real active placebo use. To prevent this as much we could we gave the example of antibiotics for a viral disease.
Sometimes a general practitioner cant be 100% sure of a diagnose or prescribes something just in case. So in some cases a viral infection can lead to a prescription of an antibiotic. Not that it cures the viral infection it rather prevents a later contamination with bacteria. So prescribing a antibiotic for a viral infection can either be done to make use of the placebo effect or as a protective measure against a super infection by bacteria. If this mistake is made it leads to an over estimation of the real placebo use. This isn’t likely to be occurred in this survey because the general practitioner would rather give the answer he thinks is most likely. Whether it is a placebo or a treatment.
It isn’t likely that neither the general practitioners nor the patients gave social desirable answers. This can be seen because of the great number of general practitioners stating to prescribe placebos while this cant be seen as social desirable at all. For the patients it would be in their best interest to answer the questions like they think it is best.
Selection bias is always a problem when questionnaires are used. We can’t be sure that our sample is representative for the whole population of general practitioners. This because general practitioners that never prescribe placebos might be less interested to a questionnaire about placebo and is so less likely to cooperate.
Other studies
It seems that a great percentage (87%) of the Dutch general practitioners prescribe active placebos. This percentage lays in the upper limit of the percentage of their foreign colleagues (51-86%) (3-6). The Dutch general practitioners prescribe active placebos at a higher frequency then their foreign colleagues. 77% of the Dutch general practitioners prescribe a active placebo at least once a month. While only 37-48% of their foreign do so (3, 6).
We found that there are still general practitioners who believe they can test with a placebo whether the patient has an ‘imaginary’ disease or not. Previous studies in different countries found the same results (5, 9, 10).
Conclusion
This survey showed that, even though the WGBO and evidence based medicine, placebo use still occurred on a great scale. While the use inactive placebo was declining the use of the active placebo was still very relevant. Active placebos were prescribed by more general practitioners and by a much higher frequency than inactive placebos.
With this many placebos being prescribed it is problematic that almost a fifth of the general practitioners don’t know that you cant test whether a condition is imaginary or not. General practitioners should be more properly schooled in the use and effect of placebos. So they can harvest the positive effects of placebo and don’t draw the wrong conclusions.
Authors’ note: We thank Svetlana Belitser for her valuable help on the statistics.
Geschreven door: Martijn E. F. van Kesteren, MSc; Nina A. Winters, PhD; Anthonius de Boer, PhD (Universiteit Utrecht, Farmaceutische Wetenschappen)
Referentielijst:
1. Beecher HK. The powerful placebo. J Am Med Assoc 1955;159(17):1602-6.
2. SFK. Nep-pillen via de apotheek. In: SFK; 2004.
3. Nitzan U, Lichtenberg P. Questionnaire survey on use of placebo. Bmj 2004;329(7472):944-6.
4. Gray G, Flynn P. A survey of placebo use in a general hospital. Gen Hosp Psychiatry 1981;3(3):199-203.
5. Goldberg RJ, Leigh H, Quinlan D. The current status of placebo in hospital practice. Gen Hosp Psychiatry 1979;1(3):196-201.
6. Hrobjartsson A, Norup M. The use of placebo interventions in medical practice–a national questionnaire survey of Danish clinicians. Eval Health Prof 2003;26(2):153-65.
7. Berger JT. Placebo medication use in patient care: a survey of medical interns. West J Med 1999;170(2):93-6.
8. Lynoe N, Mattsson B, Sandlund M. The attitudes of patients and physicians towards placebo treatment–a comparative study. Soc Sci Med 1993;36(6):767-74.
9. Thomson RJ, Buchanan WJ. Placebos and general practice: attitudes to, and the use of, the placebo effect. N Z Med J 1982;95(712):492-4.
10. Goodwin JS, Goodwin JM, Vogel AV. Knowledge and use of placebos by house officers and nurses. Ann Intern Med 1979;91(1):106-10.