Family Practice Vol. 18, No. 6, 597-601
© Oxford University Press 2001
Original Paper |
Stopping long-term drug therapy in general practice. How well do physicians and patients agree?
Section for General Practice/Family Medicine, Department of General Practice and Community Medicine, University of Oslo, N-0318 Oslo and
a Department of Public Health and Primary Health Care, University of Bergen, N-5009 Bergen, Norway.
Straand J and Sandvik H. Stopping long-term drug therapy in general practice. How well do physicians and patients agree? Family Practice 2001; 18: 597–601.
Received 18 December 2000; Revised 1 May 2001; Accepted 9 July 2001.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background. GPs have a particular responsibility to limit the occurrence of polypharmacy and adverse drug reactions, and to discontinue unnecessary drug therapy.
Objective. The aim of the present study was to measure the extent to which patients and physicians agree upon information communicated in a consultation when a drug is withdrawn.
Methods. A total of 272 Norwegian GPs and 272 patients filled in questionnaires after a consultation in which a long-term drug therapy had been discontinued. Their answers were compared and the agreement measured by kappa statistics.
Results. There was 100% concordance between physicians and patients as to what drug had been discontinued. Most of the drugs (72%) were cardiovascular. There was fair agreement as to whether the drug was to be stopped abruptly or gradually withdrawn (
0.61) and whether a follow-up appointment had been scheduled ( 0.41). Physicians were not able to judge patients' satisfaction accurately ( 0.20). Most patients (73%) were satisfied or very satisfied with the decision to withdraw the drug, and many commented that good communication and close follow-up is a prerequisite for successful withdrawal of long-term drug treatment.
Conclusions. Discontinuation of drug treatment was welcomed by most patients. Physicians and patients agreed completely as to what drug was to be discontinued, and fairly well about other factual aspects, but physicians were not able to judge patients' satisfaction accurately.
Keywords. Drug discontinuation, family medicine, general practice, patient satisfaction, questionnaire survey.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Drug prescribing occurs in ~60% of all contacts between GPs and patients.1 As major drug prescribers, GPs also have a particular responsibility for discontinuing unnecessary drug therapy.
In a study of hospitalized elderly patents, it was found that 12% were taking drugs that were contraindicated, and 27% of the prescribed drugs were deemed to be unnecessary.2 According to that study, the use of inappropriate drugs was in fact a leading cause (comprising almost half of all) for adverse drug reactions demanding hospital admissions for elderly people.
It is a common belief among doctors that patients are sceptical about drug withdrawal.3 Guidelines and textbooks rarely focus on drug discontinuation. In medical practice, however, indications for long-term drug treatment commonly are not life long. The illness may subside, the initial indication may be wrong, patients may get new illnesses, and adverse drugs reactions and drug–drug or drug–disease interactions may occur.
Thirty years ago, Balint et al. described a "repeat prescription syndrome" in which drugs were prescribed by telephone or by the receptionist for years without revision by the physician.4 From a Norwegian survey, it has been reported that more than half of all psychotropic drugs in general practice are issued during indirect doctor–patient contacts and that almost 90% of these prescriptions are repeats.5 Under these circumstances, indications for prescribing are probably rarely re-evaluated.6,7
As long as the patient stays well, it may be tempting to ascribe absence of symptoms to the ongoing drug treatment. However, a discontinuation trial may be the only way to reveal that the patient has recovered and manages well without the drugs. Nevertheless, only a few withdrawal studies have been performed on specific drugs8 or diagnoses.9
Drug prescribing practice and medication-taking behaviour are both complex processes where the decision making is heavily influenced by the physician–patient relationship. Virtually nothing is known about the patients' perceptions of stopping long-term medication. We undertook this study to measure the extent to which patients and physicians agreed upon factual information when a drug was withdrawn.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The study was conducted in Norway during 1996. A postal inquiry was sent to a random sample of 1500 GPs who were invited to join in an observational, multipractice study describing their cessation practice of any long-term drug treatment.
The physicians were to complete two questionnaires, one of which mapped background demographic variables about themselves, how long they had been in general practice, practice settings and some experiences and attitudes towards drug withdrawal in general.
The other questionnaire was to be completed in relation to their next patient in whom they withdrew any drug regimen that had lasted for at least 3 months. Switching from one compound to another for the same indication was not to be included in the survey. The physician was asked which drug was discontinued, who originally started the therapy and for what indication. Furthermore, the physician filled in information about the patient's age and gender, how long he/she had been a GP for this patient, who first suggested withdrawal of the drug and how problematic the decision to stop was for the physician. It was also reported whether the drug was to be stepped down gradually or discontinued abruptly, and whether a new appointment was scheduled for follow-up. Finally, the physician was to give his/her opinion as to whether the patient seemed satisfied or not with the decision to withdraw the drug.
The patients who were included for drug withdrawal were given a corresponding questionnaire by their GP. This was to be filled in after the consultation and returned separately by mail to one of us (JS). The questionnaires were anonymous, but physician–patient pairs could be identified through a common registration number.
The patients were asked what drug had been discontinued, and if they knew the reason (diagnosis) why this drug had been instituted originally. Other questions addressed if the physician had explained to them why the drug was to be withdrawn, and if the medication was to be stepped down or discontinued abruptly. Furthermore, they reported if a new appointment for follow-up had been scheduled, and whether they were satisfied or not with the decision to withdraw the drug.
Both patients and physicians were invited to make comments in their own words. These data consisted of short statements (a few words) up to comprehensive texts with more than 250 words. The analysis of these texts was performed by repeatedly reading all the material to obtain an overall impression, identifying units of meaning, and condensing and abstracting the meaning within each of the groups.
Statistics
Strength of agreement between information given by physicians and patients to corresponding questions was measured by kappa statistics. > 0.75 signifies excellent, and > 0.40 good agreement beyond chance.10
Material
Of the 1500 GPs who were invited, 530 accepted to participate; 439 (83%) of them returned the questionnaire which revealed that 349 (66%) had recruited one patient in whom a long-term medication was discontinued. A total of 278 patients returned questionnaires, six of whom did not have a corresponding questionnaire from their physician. Thus, we had 272 pairs of questionnaires for analysis, and they comprise the material for this study.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Of the 272 physicians, 218 (80%) were male, their mean age was 43 years (median 42, range 28–68) and they had been working in general practice for a mean of 13 years (median 12, range 1–40). They had known the corresponding patient for a mean of 7 years (median 5, range 0–34 years). Of the 272 patients, 162 (60%) were female and their mean age was 64 years (median 67, range 18–93).
The drug which was stopped had been ordinated initially by the same physician in 121 cases (45%), by a hospital physician in 64 cases (24%) and by another physician in 75 cases (28%). In 3% of the cases, the physician did not know who had started this therapy. Median length of the drug treatment was 2 years and 24% had lasted <1 year.
In 18% of the cases, it was the patient who first suggested discontinuation of the drug. Almost all (88%) patients stated that their physician had explained to them why the drug ought to be discontinued. The physicians thought it easy to discontinue the drug in 74% of the cases, whereas they felt it problematic in 16% (10% neutral).
There was 100% concordance between physician and patient as to which drug had been discontinued. Cardiovascular drugs, anatomical therapeutic chemical (ATC) main group C, were discontinued most commonly and comprised 197 of the 272 cases (72%). The second most common were drugs in ATC main group N (central nervous system), which constituted 32 cases (12%). Only nine cases were withdrawals of potentially addictive drugs (mostly benzodiazepines).
In 200 cases (74%), the physician and patient agreed on the diagnostic indication, i.e. why the drug originally was instituted. According to the physicians, the withdrawn drugs most frequently had been ordinated for hypertension (49% of all), angina pectoris (9%), cardiac arrhythmias (6%), depression (4%), heart failure (4%) and arthritis (4%). According to the patients, the most frequent indications were hypertension (47%), angina pectoris (7%), cardiac arrhythmias (5%), depression (5%) and arthritis (4%).
In 18% of the cases, the physicians and patients disagreed about whether the drug was to be stopped abruptly or stepped down gradually [0.61, 95% confidence interval (CI) 0.52–0.71]. In 9% of the cases, they disagreed about whether a new appointment had been scheduled for follow-up (0.41, 95% CI 0.22–0.60).
A total of 263 pairs of physicians and patients had indicated whether the patient seemed/was very satisfied, satisfied, neutral or sceptical towards discontinuing this drug. The variance was somewhat larger among patients than among physicians. Patients' answers were over-represented among those who were very satisfied and among those who said they were sceptical (Table 1
). Altogether, 193 of the patients (73%) were satisfied or very satisfied. Likewise, 194 of the physicians (74%) thought that their patients were satisfied or very satisfied (0.20, 95% CI 0.11–0.29). Subgroup analyses (e.g. patient and physician of same gender, or doctor–patient relationship of long duration) did not reveal significant changes in concordance.
|
Some of the patients and the physicians also made comments in their own words. A selection of patients' statements, grouped by content categories, are listed in Table 2
. Of special interest are cases in which corresponding physician and patient both had made their own comments.
|
A 49-year-old woman who was to discontinue the use of a hypnotic wrote: "I am not very pleased with this decision because I fear I will get more pain due to lack of sleep. In addition to the fact that my physician wants me to stop taking Rohypnol (flunitrazepam), I have also taken my own decision to stop using this drug." The corresponding physician wrote: "It is important to prepare the patient for this decision over some time, saying "I’m telling you this today, think about it, and then we'll discuss it more the next time. This goes for this very chronic patient."
A physician who discontinued a thiazide diuretic in a 66-year-old man, commented that the patient had experienced serious side effects of other drugs before, and that he in general was not happy with taking drugs on a regular basis. The patient himself wrote: "I have changed to homeopathic medicine, and after that my blood pressure has come down."
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Although the participating GPs may be representative for Norwegian GPs with regard to their age and gender, it is possible that the sample is biased in terms of attitudes towards drug use and discontinuation. Therefore, the results presented here should not be regarded as an epidemiological description of the frequency of discontinuation of different drugs in general practice. However, for the present purpose, i.e. measurement of agreement between GPs and patients, this is probably of less importance.
The fact that all physicians and patients agreed upon which drug was to be discontinued is an indication that the internal validity of the data in this study is good. With respect to other aspects of the process, the agreement was less than perfect. This is to be expected, as the facts may not be entirely clear. For example, the original indication, i.e. why the drug was instituted originally, may be difficult to establish. In a study of withdrawal of diuretics, it was not possible to establish why the drug originally had been initiated in 30% of the cases.8 However, while factual information was communicated reasonably well, it seems to be more difficult for the physician to estimate ‘softer’ aspects such as patients' satisfaction or scepticism.
Some other studies have tried to measure the level of agreement between doctor and patient as to what actually happened during a given consultation. Armstrong et al. investigated 263 pairs of questionnaires about the need for reattendance, and found a kappa of 0.41.11 This is exactly the same result as we found on this particular question. The agreement tends to be lower when asking about ‘softer’ data. For example, Fagerberg et al. found low kappa values when asking GPs and patients about sufficiency of time (0.04), quality of communication ( 0.13), urgency of the consultation ( 0.36) and the number of problems raised during the consultation ( 0.39).12 In a study of discontinuation of recently prescribed antihypertensive medications, Suarez et al. reported only fair agreement (0.35) when physicians and patients were asked about their treatment status.13
Some of the GPs who declined to participate in this study, or who were not able to find suitable patients, stated that they hardly ever discontinued long-term drug treatment in their patients. It is a common belief among doctors that patients are sceptical about stopping long-term drug therapy.3 However, only a small minority of the patients were truly sceptical.
A total of 73% of the patients were satisfied or very satisfied with the decision to discontinue the drug. A similar number of physicians indicated that their patient seemed to be satisfied. However, the individual agreement between GPs and patients as measured by kappa statistics was rather poor. In a study of 250 encounters from an out-patient clinic in the USA, a similar level of patient satisfaction was found (78% highly satisfied). However, 49% of all patients were less satisfied than the physician believed them to be, whereas 9% were more satisfied.14
In a qualitative UK study of 20 GPs, it was found that misunderstandings in prescribing decisions occurred in 28 out of 35 consultations and that most patients had agendas that were not voiced during the consultation.15 In our study, many patients emphasized the need for open communication, adequate time, thorough information and follow-up consultations.
For some patients, there was clearly an element of anxiety involved in discontinuing medication they had got used to. Some had been told by the physician who initiated the treatment that they probably would have to take this drug for the rest of their lives. Others were just relieved to let go of drugs that might cause uncomfortable side effects, or they just did not like taking drugs. Even if there were some patients who wanted to leave the decisions entirely to the doctor, most comments from the patients stressed that they were pleased to participate in the decision making during the consultation.
In conclusion, this study showed that discontinuation of drug treatment was welcomed by most patients. Physicians and patients agreed completely as to what drug was to be discontinued, and fairly well about other factual aspects, but physicians were not able to judge patients' satisfaction accurately. Good communication allowing patients to express their ideas and feelings and close follow-up are prerequisites for successful withdrawal of long-term drug treatment that may no longer be needed.
|
Acknowledgments |
|---|
This study was funded by a grant from the Fund for Quality Assurance in Primary Health Care by the Norwegian Medical Association
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 Rokstad K, Straand J, Fugelli P. General practitioners' drug prescribing practice and diagnoses for prescribing: The Møre & Romsdal Prescription Study. J Clin Epidemiol 1997; 50: 485–494.[Web of Science][Medline]
2
Lindley CM, Tully MP, Paramsothy V, Tallis RC. Inappropriate medication is a major cause of adverse drug reactions in elderly patients. Age and Ageing 1992; 21: 294–300.
3 Strømme HK, Botten G. Physicians' perceived drug-related problems and preventive strategies concerning old patients in general practice. J Soc Adm Pharm 1992; 9: 172–178.
4 Balint M, Hart J, Joyce D, Marinker M, Woodcock J. Treatment and Diagnosis: A Study of Repeat Prescriptions in General Practice. London: Tavistock Publications, 1970.
5 Rokstad K, Straand J. Drug prescribing during direct and indirect contacts with patients in general practice. A report from the Møre & Romsdal Prescription Study. Scand J Prim Health Care 1997; 15: 103–108.[Web of Science][Medline]
6 Cantrill JA, Dowell J, Roland MO. General practitioners' views on the appropriateness of their long-term prescribing. Eur J Gen Pract 2000; 6: 5–9.
7 Dijkers F. Towards optimisation of repeat prescribing. Eur J Gen Pract 2000; 6: 4.
8
Straand J, Fugelli P, Laake K. Withdrawing long-term diuretic treatment among elderly patients in general practice. Fam Pract 1993; 10: 38–42.
9 Aylett M, Creighton P, Jachuck S, Newrick D, Evans A. Stopping drug treatment of hypertension: experience in 18 British general practices. Br J Gen Pract 1999; 49: 977–980.[Web of Science][Medline]
10 Fleiss JL. The measurement of interrater agreement. In Fleiss JL (ed.). The Measurement of Proportions and Rates. New York: Wiley & Sons, 1981: 212–236.
11 Armstrong D, Glanville T, Bailey E, O'Keefe G. Doctor-initiated consultations: a study of communication between general practitioners and patients about the need for reattendance. Br J Gen Pract 1990; 40: 241–242.[Web of Science][Medline]
12 Fagerberg CR, Kragstrup J, Støvring H, Rasmussen NK. How well do patient and general practitioner agree about the content of consultations? Scand J Prim Health Care 1999; 17: 149–152.[Web of Science][Medline]
13 Suarez AP, Staffa JA, Fletcher P, Jones JK. Reason for discontinuation of newly prescribed antihypertensive medications: methods of a pilot study using computerized patient records. Pharmacoepidemiol Drug Safe 2000; 9: 405–416.[Medline]
14 Probst JC, Greenhouse DL, Selassie AW. Patient and physician satisfaction with an outpatient care visit. J Fam Pract 1997; 45: 418–425.[Web of Science][Medline]
15
Britten N, Stevenson FA, Barry CA, Barber N, Bradley CP. Misunderstandings in prescribing decisions in general practice: qualitative study. Br Med J 2000; 320: 484–488.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  F. Legare, A. M. O'Connor, I. D. Graham, G. A. Wells, and S. Tremblay Impact of the Ottawa Decision Support Framework on the Agreement and the Difference between Patients' and Physicians' Decisional Conflict Med Decis Making, July 1, 2006; 26(4): 373 - 390. [Abstract] [PDF]
|
|
|
|
 |
|
 H. M. Holmes, D. C. Hayley, G. C. Alexander, and G. A. Sachs Reconsidering medication appropriateness for patients late in life. Arch Intern Med, March 27, 2006; 166(6): 605 - 609. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||