Family Practice Vol. 21, No. 5 © Oxford University Press 2004, all rights reserved.
Perceptions of patient expectation for an antibiotic: a comparison of walk-in centre nurses and GPs
a School of Pharmacy and Pharmacology, University of Bath, Bath BA2 7AY, b Primary Care Nursing Research Unit, School of Nursing and Midwifery, James Clerk Maxwell Building, King's College, 57 Waterloo Road, London SE1 8WA, and c Division of Primary Care, University of Bristol, Cotham House, Cotham Hill, Bristol BS6 6JL, UK
E-mail: m.weiss{at}bath.ac.uk
Received 4 July 2003; Revised 16 February 2004; Accepted 17 May 2004.
Weiss MC, Deave T, Peters TJ and Salisbury C. Perceptions of patient expectation for an antibiotic: a comparison of walk-in centre nurses and GPs. Family Practice 2004; 21: 492–499.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionDeclarationAppendix 1 QuestionnairesReferences |
|---|
Background. Patient expectation for a prescription is a recognized influence on GPs' prescribing, particularly in relation to the prescribing of antibiotics. Nurses are now able to supply antibiotics under a Patient Group Direction (PGD) in NHS walk-in centres and may experience similar pressures in this new role.
Objectives. Our aim was to compare walk-in centre nurses' and GPs' perceptions of the influence of patient expectation on their supply of an antibiotic to patients with an acute respiratory tract infection presenting with a sore throat or cough.
Methods. Between June and December 2001, all patients presenting with a sore throat or cough at six walk-in centres and six nearby general practices were eligible to participate in the study. After the health professional–patient consultation, the health professional and the patient each completed a questionnaire.
Results. There were 472 health professional (181 GPs and 291 walk-in centres) and 160 (34%, 160 out of 472) patient questionnaires returned. GPs were more likely to report that the patient expected an antibiotic than nurses (72% of 181 versus 13% of 291, P < 0.001). GPs were also less likely to report that an antibiotic was indicated than nurses (88% of 136 versus 97% of 194, P < 0.001). There was a trend for doctors to prescribe more frequently than nurses, in 74% of 180 patients versus 66% of 291 patients (P = 0.06). GPs were likely to report that the patient expected an antibiotic when the patient reported wanting a prescription (60% of 68, P = 0.05) and to report that the patient expected an antibiotic if the patient thought an antibiotic would be beneficial (62% of 68, P = 0.001). There was a much weaker relationship between nurse perceptions of patient expectation for an antibiotic and, either patient desire for a prescription or the patient's affirmative belief that an antibiotic would be beneficial.
Conclusions. Nurses may be compensating for a lack of security in their new role as antibiotic suppliers by not acknowledging the influence of patient expectation on their decision making. The acknowledgement of the influence of patient expectation might be beneficial to nurses' development as suppliers of medicines by giving them greater understanding of the consultation process and in the need to discuss patient expectations explicitly in the consultation.
Keywords. Antibiotics, patient demand, patient group directions, prescribing decisions.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1 QuestionnairesReferences |
|---|
The creation of supplementary prescribers is the latest in a series of government initiatives to advocate the extension of prescribing rights to nurses and other health care professionals.1 Other mechanisms include the use of Patient Group Directions (PGDs) which enable the supply or administration of prescription-only medicines by health professionals who are not medically qualified. To comply with legislation, these PGDs must specify a broad range of information to include the name and dose of the medicine, the clinical condition or situation to which the PGD applies, and relevant warnings.2 From a public health perspective, a growth in the range of professionals able to supply or prescribe an antibiotic is a potentially worrying development given the concerns about increasing community antimicrobial resistance.3 Such concerns are intensified in the context of research evidence suggesting that the benefits of antibiotics are modest in acute self-limiting conditions such as sore throat4 or cough.5
PGDs are a key method of supplying prescription-only medicines by nurses in new venues of health care delivery such as walk-in centres. Walk-in centres form part of the UK government's plan for modernization of the National Health Service (NHS) and are meant to complement the traditional method of accessing the health service, the GP.6 Key features of NHS walk-in centres include greater accessibility through wide opening hours and walk-in access (without appointments), the provision of information and treatment for minor conditions and the development of services that meet the needs of the local population.7 As of September 2001, 39 walk-in centres had opened.7 The range, number and type of medication supplied on PGD at each walk-in centre varies greatly.
In general practice, there is a considerable body of research which has explored the influences on GPs' prescribing and found that GPs' perceptions of patient expectation for a prescription is one of the strongest predictors of prescribing.8–11 This provokes an interesting question as to whether nurses, working in walk-in centres and recently able to supply antibiotics under a PGD, are similarly influenced by patient expectation. The aim of this study was to compare nurses' and GPs' perceptions of the influence of patient expectation on their supply of an antibiotic when patients presented with the symptoms of an acute respiratory tract infection involving a cough or sore throat.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1 QuestionnairesReferences |
|---|
From the 20 walk-in centres that had antibiotic PGDs in February 2001, six were selected for inclusion in the study. At the time, many walk-in centres were new, so the sites eligible for inclusion had to have been using their antibiotic PGDs for a period of at least 3 months. Walk-in centre sites were selected for their breadth of PGDs, covering a wide range of antibiotics likely to be used in acute respiratory tract infections presenting with a cough or sore throat, and/or because they recorded a high level of supply of these antibiotics. All six walk-in centres approached agreed to participate in the study. All walk-in centres were asked to recruit patients with acute respiratory tract infections, presenting with either sore throat (upper respiratory tract infection) or cough (lower respiratory tract infection). However, two walk-in centres had PGDs restricting their supply of antibiotics to clinical situations involving sore throat only. The rationale for these clinical inclusion criteria have been discussed by previous authors.12,13 The clinical inclusion criteria are shown in Table 1 (sore throat)12 and Table 2 (cough).13 These criteria were printed at the top of the health care professionals' questionnaires.
|
|
The six GPs were recruited from six different geographical areas in close proximity (3 miles) to the local walk-in centres. Five GPs were near to a walk-in centre participating in the study. The sixth GP was recruited from a non-participating walk-in centre area as a GP in close proximity to the participating walk-in centre could not be recruited. Between June and December 2001, 1200 matched health care professional–patient questionnaires were distributed, 100 for each GP and walk-in centre site. GPs and walk-in centre nurses were asked to complete the health care professional questionnaire for each patient meeting either the cough or sore throat inclusion criteria. They were also asked to give the patient a questionnaire at the end of the consultation. Patients were asked to return the questionnaire to the researchers in the stamped addressed envelope provided. For ethical reasons, the researchers had no access to the patient names that met the inclusion criteria as identified by the health care professionals, and therefore it was not possible to send out reminders. GPs were offered an incentive for their participation—£5 for every health care professional questionnaire returned. Neither walk-in centre nurses nor patients were offered a similar incentive. GP and walk-in centre sites were telephoned and sent reminder letters, including the recruitment progress, to maintain interest in the study.
The five questions on the health care professional questionnaire for GPs and walk-in centres nurses were the same (Appendix 1). These questions were identical for both sore throat and cough patients. The patient was asked for a description of their symptoms, any prior self-treatment, their expectations for a prescription, whether they thought they would benefit from an antibiotic, their views about whether antibiotics were necessary and their satisfaction with the consultation (Appendix 1). Four-point responses in the health care professional questionnaire were later dichotomized for the analysis. Analysis was conducted using the statistical software package, Intercooled Stata version 8.014 Descriptive statistics, including likelihood ratio chi-squared statistics, were performed.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1 QuestionnairesReferences |
|---|
From the health professionals, there were 181 (mean 30, range 0–54) and 291 (mean 49, range 8–85) questionnaires returned, from GP and walk-in centre sites, respectively, a total of 472 questionnaires. From patients, there were 160 questionnaires returned, with an additional 24 returned (not analysed further) without a matching health professional questionnaire. These 24 patient questionnaires were evidently distributed by the health professional in the consultation but the professional neglected to return their questionnaire to the investigators. It is difficult to determine an accurate response rate as the number of eligible consultations in each setting is not known. However, assuming the health care professionals distributed patient questionnaires in all consultations for which they completed a questionnaire themselves, the return of patient questionnaires is only 34% (160 out of 472).
As there were a larger number of health care professional returns (472), these were compared descriptively with the smaller sample (160) for which matching health professional–patient data were available. This was done to determine if the larger sample of health professional questionnaires was broadly similar to the sample for which there were matching patient data. In the matched sample, 94 (59%) were from walk-in centres and 66 (41%) from GPs (Table 3). This compares with the unmatched or ‘remainder’ sample of 312 (472–160) health professional questionnaires of which 197 (63%) and 115 (37%) were from walk-in centres and GPs, respectively. This suggests that the sample with matching health professional–patient questionnaires may be slightly different from the sample of health professional questionnaires with no matching patient data. In the sample as a whole, when asked about the extent to which the professional felt the patient expected an antibiotic, GPs were more likely than nurses to report that the patient expected an antibiotic (72% of 181 versus 13% of 291, P < 0.001). As shown in Table 4, this relationship held when examined in the matched data set (68% of 66 versus 18% of 94, P < 0.001).
|
|
Table 4 also shows that GPs were less likely than nurses to report an antibiotic was indicated (88% of 136 versus 97% of 194, P < 0.001; matched data set 90% of 52 versus 98% of 66, P = 0.047) when asked how certain they were about using an antibiotic. Similarly, GPs were more likely than nurses to report being influenced by patient expectation for an antibiotic (13% of 134 versus 3% of 193, P < 0.001). Although this relationship was broadly maintained in the matched data set (10% of 52 versus 3% of 67, P = 0.13), it was not statistically significant given the smaller sample size. There was a tendency (P = 0.06) for doctors to supply an antibiotic more frequently than nurses, in 74% of 180 patients compared with 66% of 291 patients for walk-in centre nurses. If a GP reported that a patient expected an antibiotic, then they were more likely to prescribe (83% of 130, P < 0.001) compared with walk-in centre nurses whose supply of an antibiotic was less influenced by their perceptions of patient expectation (57% of 38, P = 0.28).
In the smaller sample with matched patient data, patient expectations for a prescription were comparable between the two sites (84% of patients seeing GPs expected a prescription, while 83% of patients attending walk-in centres expected a prescription). It should be noted that health care professionals were asked the extent to which they felt the patient expected an antibiotic, while patients were asked two questions (i) whether they expected a prescription and (ii) whether they thought they would benefit from an antibiotic. As shown in Table 5, GPs were more likely to report that the patient expected an antibiotic when the patient reported wanting a prescription (60% of 68, P = 0.05). Even more clearly, GPs were more likely to report that the patient expected an antibiotic if the patient felt an antibiotic would be beneficial (62% of 68, P = 0.001). Conversely, there was a much weaker relationship between nurse perceptions of patient expectation for an antibiotic and either patient desire for a prescription or the patient's affirmative belief that an antibiotic would be beneficial. For example, of the 68 patients seeing a GP who thought an antibiotic would be beneficial, the GP thought 42 (62%) expected an antibiotic. In contrast, of the 97 patients seeing a walk-in centre nurse who thought an antibiotic would be beneficial, the nurse thought only 17 (18%) expected an antibiotic. A comparable situation exists with the decision to supply an antibiotic. As shown in Table 5, of the 55 patients who felt they would benefit from an antibiotic who saw a GP, 47 (85%) actually received one. Likewise, of the 88 patients who felt they would benefit from an antibiotic who saw a walk-in centre nurse, 63 (72%) actually received one.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1 QuestionnairesReferences |
|---|
This study has confirmed previous findings about the importance of GPs' perceptions of patients' expectations as an influence on their prescribing. However, it also raises questions about the extent to which nurses are aware of these non-clinical or social factors as an influence on prescribing. Although the study is limited by the low response rate, this research raises issues that need to be verified in future research. The low response rate was investigated informally amongst GP and nurse participants, and all mentioned—and were surprised by—the relative lack of patients presenting with clinical symptoms meeting the inclusion criteria. In retrospect, it would have been appropriate to investigate the process of antibiotic supply decision making and recruitment using qualitative methods prior to beginning the questionnaire study. Initially, this low recruitment rate during the summer months was not unexpected. However, as time progressed and recruitment remained low, various actions were taken. For example, each GP and walk-in centre was contacted by telephone and letter on a 3–4 weekly basis. In addition, the data collection period was extended by 3 months beyond the original study completion date of September 2001. To ensure that the low recruitment rate was not due to forgetfulness or study apathy, one walk-in centre offered to interrogate its database of patient attenders. The number of patients recruited was found to match the number of presenting patients who fulfilled the inclusion criteria.
Nonetheless, the method assumes that the health care professional included all eligible patients and distributed all questionnaires at the end of the consultation. Difficulties in either of these processes, due to practice or workload reasons, may have accounted for the low numbers returned. The health care professional may have decided on treatment (an antibiotic) first and then decided on a diagnosis (non-respiratory tract infection) to justify this treatment, thereby missing the study inclusion criteria. In addition, as there were a relatively small number of walk-in centres that had been supplying antibiotics for acute respiratory tract infections for >3 months, the sample size of health care professionals available for the study was small. PGDs in walk-in centres with high levels of antibiotic supply, which were selected for this study, might be working differently from PGDs in walk-in centres with low levels of supply. The walk-in centres and the participating GPs may not be typical of their wider populations, thus it is not possible to extrapolate these findings more generally. For these reasons, these findings need to be seen in the context within which they were generated, suggestive of issues which need to be verified in future research.
Interestingly, this study found that GPs prescribed in 74% of those consultations where patients presented with cough or sore throat and met the inclusion criteria. This is comparable with the findings of Macfarlane15 in their study of patients presenting with cough where an antibiotic was prescribed in 75% of patients. In respect of patient expectation for a prescription, other authors have also found a significant association between this and the GP's perception of this expectation.10 In Cockburn and Pit's study, when a patient did expect a medication, the GP's judgement agreed in 65% of cases, analogous to the 62% found in the current study.
One possible reason why walk-in centre nurses may be more certain that an antibiotic is clinically indicated is because their PGDs provide them with guidance on appropriate antibiotic supply. However, this appears unlikely given that the information on the PGD tends to be very brief and descriptive. Previous research has shown that PGDs can vary greatly in terms of the quantity and format of information they provide.16 The guidance for PGDs states that only fully competent, qualified and trained professionals operate within directions and that these professionals should act within their Code of Professional Conduct.2 This means that the nurse should feel competent within the clinical area and undertake the necessary qualifications, training and experience considered relevant for the medicines in the protocol. Within these broad principles, the training undertaken by the nurses, which was not investigated in this study, is likely to have been specific to the needs of individual nurses and resources in the local area. The extent to which any training included a discussion of the influence of patient expectation on prescriber decision making is not known. Any training that was provided is likely to have instilled an ethos to adhere to the PGD, which would be complemented by decision support software using clinical algorithms to guide decision making. While such information is helpful, the empirical use of antibiotics, based on clinical signs, symptoms and history, is difficult. Prescribing an antibiotic for sore throat remains controversial.17,18 For cough, findings from history and physical examination alone are unable to rule in a diagnosis of pneumonia19 where the use of an antibiotic would be deemed more appropriate. Indeed, previous evidence suggests that prescribing decisions in acute respiratory tract infections are likely to be influenced by social factors such as patient expectations.20,21
Previous research has found a comparable prescribing rate between doctors and nurses,22 while this study found a (non-significant) trend for nurses to supply an antibiotic less frequently than GPs. This may be related to the length of consultation, since the main evaluation of NHS walk-in centres demonstrated that nurses had an average consultation length of 14 min7 compared with a typical GP consultation length of 8 min.23 Nurses may have used this time to explain the lack of efficacy of antibiotics in viral infections and been more selective in the cases where they chose to supply antibiotics. As less antibiotic prescribing is perceived from a public health perspective as more appropriate prescribing, then nurses may be ‘better prescribers’. Unfortunately, the gold standard or ‘true’ level of clinically appropriate antibiotic prescribing is not known for any of these consultations. Nurses more frequently asserted that their antibiotic supply was clinically appropriate than GPs. However, whether this was due to a genuine, more appropriate use of antibiotics or is a method of compensating for their lack of security in their new role, by frequently asserting that their antibiotic supply is clinically appropriate, is unknown.
Nurses did supply an antibiotic for 191 of the 291 patients coming to see them. Patient expectations were comparable between the two sites yet, compared with GPs, nurses were less likely to report a patient expecting an antibiotic. In addition, GPs' perceptions of patient expectation were likely to accord with the patient's view, whereas a much weaker relationship existed between nurses' perceptions of patient expectation for an antibiotic and the patient's view (Table 5). In this context, the trend towards lower prescribing by nurses would have been more reassuring if it had reflected a shared agreement with the patient about the need for an antibiotic.
It may be that nurses are less willing to admit to patient expectation as an influence on their supply of antibiotics because they are less used to making diagnoses. Nurses may not recognize that such an admission might be beneficial to their development as a prescriber by giving them greater understanding and insight into the consultation process. The influence of non-clinical factors on prescriber decision making has been described as part of a rational human decision-making process, although clearly irrational from a health policy perspective which takes a narrower view, emphasizing the scientific basis of prescribing.24 Given the overwhelming evidence that non-clinical factors do affect prescribing decisions, nurses may be better off realizing that patient expectations and other non-clinical factors can similarly affect them. The acknowledgement of this influence releases nurses to engage in an explicit discussion of the patient's expectations for the consultation where both the patients' and the prescribers' agendas are considered. This should enhance the nurse–patient relationship and could guide the development of specific consultation strategies to promote more appropriate use of antibiotics in primary care.25
Conclusion
This study investigated different health professionals' views of patient expectation for an antibiotic. Nurses reported a lower level of patient expectation for an antibiotic than GPs, although actual patient expectations for a prescription were comparable between the two sites. Some evidence presented suggests that patients' views about whether they wanted a prescription or considered an antibiotic to be beneficial were assessed accurately by nurses less frequently than by GPs. Nurses were also less likely to report being influenced by patient expectations for an antibiotic than GPs. The findings suggest that training for new prescribers should reflect an acknowledgement of the influence of non-clinical factors, such as patient expectation, on prescribers' decision making. This includes engaging with the notion that prescribing an antibiotic as a result of patient expectation is a rational—in the widest sense of the word—human response. The recognition that decision making is influenced by non-clinical factors, and is a rational human response, are necessary precursors to understanding the consultation process, where such expectations are recognized as being worthy of explicit discussion in the consultation itself. For new prescribers continually to assert the lack of influence of non-clinical factors on prescribing is to deny the existence of an influence recognized as part of normal human interactions, and forestalls the development of appropriate methods to deal with this process.
|
Declaration |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1 QuestionnairesReferences |
|---|
Funding: Department of Health.
Ethical Approval: South & West Multi-Centre Research Ethics Committee.
Conflicts of interest: None.
|
Appendix 1 Questionnaires |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1 QuestionnairesReferences |
|---|
Health professional questionnaire
- To what extent did you feel the patient expected an antibiotic? (extremely, quite a bit, a little, not at all)
- Did you prescribe an antibiotic? (yes/no)
- How certain were you about using an antibiotic? (definitely indicated, probably indicated, probably not indicated, definitely not indicated)
- To what extent did patient expectation for an antibiotic influence your decision to prescribe? (extremely, quite a bit, a little, not at all)
- What other relevant social factors influenced your decision to prescribe an antibiotic? (forthcoming weekend; social factors for the patient, i.e. holiday, exam, etc.; your time or work pressures; other—please describe).
Patient questionnaire
- 1. Can you give brief details of the symptoms or problem that prompted your visit to the walk-in centre or GP: (open text)
- 2. Have you treated the symptoms with anything before coming to see the doctor or nurse? (yes, go to question 2a; no, go to question 3)
- 2a. If so, what have you tried: (open text)
- 3. Do you think your symptoms are caused by an infection? (yes, no)
- 4. Did you expect to receive a prescription for these symptoms? (yes, go to question 4a; no, go to question 5)
- 4a. If so, was there anything in particular you wanted prescribed:
- 5. Did you feel you would benefit from an antibiotic? (yes, go to question 5a; no, go to question 6)
- 5a. Why did you feel you needed an antibiotic?
- 6. To what extent, if at all, do you think antibiotics might help your symptoms? (absolutely necessary, probably necessary, don't know, probably not necessary, not necessary).
- 7. Did you actually ask the doctor or nurse for an antibiotic? (yes, no)
- 8. Overall, how satisfied were you with the consultation? (very, moderately, slightly, not at all)
- 9. Are you exempt from paying for prescribing? (yes, my prescriptions are free; no, my prescriptions need to be paid for)
- 10. Age (open text)
- 11. Sex (male, female)
- 2. Have you treated the symptoms with anything before coming to see the doctor or nurse? (yes, go to question 2a; no, go to question 3)
Do you have any comments you would like to add?
|
Acknowledgments |
|---|
The authors would like to thank the patients and staff at the walk-in centres and general practices who provided the data on which this study is based. The authors gratefully acknowledge funding from the Department of Health. The research was conducted independently by the University of Bristol. The views expressed in this publication are those of the authors and are not necessarily those of the Department of Health.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationAppendix 1 QuestionnairesReferences |
|---|
1 http://www.doh.gov.uk/supplementaryprescribing/index.htm (accessed 12 May 2003).
2 NHS Executive. Patient Group Directions [England Only], Health Service Circular 2000/026. London: Department of Health.
3 Standing Medical Advisory Committee, Sub-group on antimicrobial resistance. The Path of Least Resistance. London: Department of Health; 1997.
4 Del Mar CB, Glasziou PP, Spinks AB. Antibiotics for sore throat (Cochrane Review). In The Cochrane Library, Issue 2. Oxford: Update Software; 2003.
5 Fahey T, Stocks N, Thomas T. Quantitative systematic review of randomised controlled trials comparing antibiotic with placebo for acute cough in adults. Br Med J 1998; 316: 906–910.
6 NHS Executive. NHS Primary Care Walk-in Centres, Health Service Circular 999/116. Leeds: NHSE; 1999.
7 Salisbury C, Chalder M, Manku-Scott T, Pope C, Moore L. What is the role of walk-in centres in the NHS? Br Med J 2002; 324: 399–402.
8 Bradley C. Uncomfortable prescribing decisions: a critical incident study. Br Med J 1992; 42: 454–458.
9 Schwartz RK, Soumerai SB, Avorn J. Physician motivations for non-scientific drug prescribing. Soc Sci Med 1989; 28: 577–582.[CrossRef][Web of Science][Medline]
10 Cockburn J, Pit S. Prescribing behaviour in clinical practice: patients' expectations and doctors' perceptions of patients expectations—a questionnaire study. Br Med J 1997; 315: 520–523.
11 Britten N, Okoumunne O. The influence of patients' hopes of receiving a prescription on doctors' perceptions and the decision to prescribe: a questionnaire survery. Br Med J 1997; 315: 1506–1510.
12 Little P, Williamson I, Warner G et al. Open randomised trial of prescribing strategies in managing sore throat. Br Med J 1997; 314: 722–727.
13 Macfarlane J, Holmes W, Gard P et al. Prospective study of the incidence, aetiology and outcome of adult lower respiratory tract illness in the community. Thorax 2001; 56: 109–114.
14 StataCorp. Stata Statistical Software: Release 8.0. College Station (TX): Stata Corporation; 2003.
15 Macfarlane J, Lewis SA, Macfarlane R, Holmes W. Contemporary use of antibiotics in 1089 adults presenting with acute lower respiratory tract illness in general practice in the UK: implications for developing management guidelines. Resp Med 1997; 91: 427–434.[CrossRef][Web of Science][Medline]
16 Deave T, Weiss MC, Salisbury C. Antibiotic supply in NHS walk-in centres: how legal is it? Br J Community Nurs 2003; 8: 209–213.[Medline]
17 Little PS, Williamson I. Controversies in management: are antibiotics appropriate for sore throats? Costs outweigh the benefits. Br Med J 1994; 309: 1010–1011.
18 Schwartzman P. Controversies in management. Are antibiotics appropriate for sore throats? Careful prescribing is beneficial. Br Med J 1994; 309: 1011–1012.
19 Metlay JP, Kapoor WN, Fine MJ. Does this patient have community-acquired pneumonia? Diagnosing pneumonia by history and physical examination. J Am Med Assoc 1997; 278: 1440–1445.
20 Virji A, Britten N. A study of the relationship between patients' attitudes and doctors' prescribing. Fam Pract 1991; 8: 314–319.
21 Howie JGR. Clinical judgment and antibiotic use in general practice. Br Med J 1976; 2: 1061–1064.
22 Venning P, Durie A, Roland M, Roberts C, Leese B. Randomised controlled trial comparing cost effectiveness of general practitioners and nurse practitioners in primary care. Br Med J 2000; 320: 1048–1053.
23 Howie JGR, Heaney DJ, Maxwell M et al. Quality at general practice consultations: cross sectional survey. Br Med J 1999, 319: 738–743.
24 Weiss MC, Scott DK. Whose rationality? A qualitative analysis of general practitioners' prescribing. Pharm J 1997; 259: 339–341.
25 Butler CC, Kinnersley P, Prout H, Rollnick S, Edwards A, Elwyn G. Antibiotics and shared decision-making in primary care. J Antimicrobial Chemother 2001; 48: 435–440.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||