Skip to main content
The BMJ logoLink to The BMJ
. 2000 Oct 7;321(7265):867–871. doi: 10.1136/bmj.321.7265.867

Do patients wish to be involved in decision making in the consultation? A cross sectional survey with video vignettes

Brian McKinstry 1
PMCID: PMC27496  PMID: 11021866

Abstract

Objective

To determine patients' preferences for a shared or directed style of consultation in the decision making part of the general practice consultation.

Design

Structured interview, with video vignettes of acted consultations.

Setting

5 practices in Lothian, Scotland.

Participants

410 patients (adults and adults accompanying children) attending surgery appointments.

Main outcome measures

Preference for shared or directed form of video vignette for five different presenting conditions.

Results

Patients varied in their preference for involvement in decision making in the consultation. Under multiple regression analysis, patients' preference was found to be independently predicted by the problem viewed (patients presented with physical problems preferred a directed approach), patients' age (patients aged 61 or older were more likely to prefer the directed approach), social class (social classes I and II were more likely to prefer the shared approach), and smoking status (smokers more likely to prefer the shared approach). Those patients who were able to answer (or who thought their doctor's style similar to those in the vignettes) were more likely to describe their own doctor's style as similar to their preferred style. No major association in preference was found with sex, frequency of attendance, or perceived chronic ill health.

Conclusion

Patients may vary in their desire for involvement in decision making in consultations. Although this variation seems to depend on the presenting problem, age, social class, and smoking status, these associations are not absolute, with large minorities in each group. Doctors need the skills, knowledge of their patients, and the time to determine on which occasions, with which illnesses, and at which level their patients wish to be involved in decision making.

Introduction

The evidence that patients are more satisfied and more likely to comply with treatment when doctors allow them to express their concerns and ideas in the consultation is powerful.15 Some authors have suggested that patients should routinely be involved in decision making in consultations,3,6,7 and this concept has been accepted by those involved in the training of general practitioners.8 There is, however, little evidence that patients find shared decision making acceptable.9 I aimed to determine patients' preferences for participation in decision making in consultations for different types of medical problems.

Participants and methods

Video vignettes

Pairs of video vignettes of five common scenarios in consultations were made, one in a style that involved the patient in deciding on management (shared approach), and one in a style where the doctor largely decided management (directed approach). The video covered only the decision making part of the consultation, the history having been described in a brief introduction. The vignettes represented five presenting problems: serious acute (bleeding mole), minor acute (sprained calf), chronic (unresponsive rheumatoid arthritis), mental health (depression), and lifestyle advice (smoking).

The actors followed a script that contained exactly the same information in both approaches (see examples in boxes). To avoid the possibility that perceived differences in the versions might be due to the personalities of the actors, two sets of two actors each made two versions, one depicting the shared approach and the other the directed approach. Twenty vignettes were thus available for use. The scenarios were shown with the approaches in varying order.

The patient is a 30 year old woman. She has come to see her doctor after a bout of bronchitis to get a certificate to go back to work. She is fully recovered now and is expecting a brief consultation. Her doctor is concerned that she is continuing to smoke despite having had three bouts of bronchitis in the past year. Her doctor decides to use the consultation to talk about her smoking.

Shared approach

Doctor: Are you still smoking?

Patient: I'm afraid so doctor.

Doctor: This last infection didn't put you off then. Do you think the smoking is connected to these chest infections you've had?

Patient: I dare say. I wouldn't mind stopping, but it's not easy to give up.

Doctor: I know it's very difficult. Quite a lot of my patients say that. Have you ever tried to give up?

Patient: Yes, a few years ago I gave up for four months.

Doctor: Well that was good. What made you start again?

Patient: It was stupid really. I was at a wedding, had a few drinks, and thought one drag wouldn't hurt, and that was it.

Doctor: Was it hard to stop?

Patient: That was the odd thing, then I didn't really find it that hard.

Doctor: I heard recently that it takes an average of three tries to stop smoking. It's worth trying again, because the smoking definitely appears to be catching up with you. Do you think you will give it another go?

Patient: Well, maybe.

Doctor: The other thing that might be worth considering is cutting down. There's good evidence to show that the fewer cigarettes you smoke the less the risk. Would that be easier?

Patient: No. If I was going to stop I would stop completely.

Doctor: I have some information here which you might find useful. It tells you about some of the aids we have now to stop smoking, such as nicotine patches and gum, along with other common sense stuff. If I can do anything to help you with this, or can give you advice, please let me know.

Patient: Thanks doctor I'll think about it. Smiling.

Direct approach

Doctor: Well you seem to have shaken off another of these infections, but that's the third this year. It can't go on like this. You really have to stop smoking.

Patient: It's not easy doctor!

Doctor: I know it's difficult, all my smoking patients tell me this, but if you keep trying you will be successful. I read somewhere that on average people have to try three times before they eventually stop smoking. I'm sure someone like you can do that.

Brief gap.

Even if you can't stop you should cut down. The less you smoke the less the risk. If I can be of any help to you I will. So give it a go. I have some information here which you might find useful. It tells you about some of the aids we have now to stop smoking, such as nicotine patches and gum, along with other common sense stuff. If I can do anything to help you with this, or can give you advice, please let me know.

Patient: Thanks doctor. I'll think about it.

Doctor: Don't think about it, do it!

Patient: OK. Smiling.

The videos were subjected to discussion by groups selected largely from the primary care team (see website). The groups were asked to score both versions of the consultation against 50 different variables about the doctor, patient, and consultation by using Lickert scales. The main differences between the directed and shared versions were in the variables of power, authority, directing, sharing, cooperation, negotiation, and one sidedness. The groups also noted differences in consultation length, which was inevitably longer in the shared version.

Power calculation

Ethical approval was obtained for the study. I chose a sample size of 400 to enable detection of significant differences in the order of 15% in preferences between dichotomous groupings of the patients on the basis of other factors. Adult patients and adults accompanying children attending five Lothian general practices of varying demographies and list size were invited to take part. They were shown one of 10 video “couplets,” comprising an introduction followed by two different versions, shared and directed, by one or other of the sets of actors. Patients who agreed to take part were shown the next in sequence from one of two tapes (one showing scenarios in reverse order). The tapes were viewed by up to four patients at any one time. Immediately after viewing the interviewer asked patients which version (shared or directed) they thought was best, which was most like their own doctor's style, and what they thought was the biggest difference between the versions. The patients' age, sex, history of chronic ill health, smoking status, frequency of attendance at surgery, age of leaving full time education, and social class were recorded.

Multiple logistic regression was used to test for those factors significantly and independently associated with preference for a shared or directed approach for each scenario. Confidence intervals were calculated where appropriate.

Results

Overall, 410 of 631 patients (65%) who were approached took part. Patients often apologised for not being able to take part, citing pressure of time. No data were available on those patients who did not take part. Table 1 lists the characteristics of the sample.

Table 1.

Preference for shared or directed consultations related to age, social class, smoking status, and scenario. Values are numbers (percentages) unless stated otherwise

 Type of consultation
Odds ratio (95% CI)*
Shared Directed
Age
 15-60 140 (42.9) 186 (57.1) 1.00
⩾61  23 (27.4) 61 (72.6) 2.03 (1.14 to 3.63)
Social class
 I  13 (52.0) 12 (48.0) 0.35 (0.13 to 0.94)
II  55 (50.7) 55 (50.3) 0.32 (0.17 to 0.63)
III  40 (35.1) 74 (64.9) 1.00
IV  25 (33.8) 49 (66.2) 0.86 (0.43 to 1.71)
V  30 (34.5) 57 (65.5) 1.05 (0.54 to 2.02)
Scenario
 Bleeding mole  35 (44.3) 44 (55.7) 1.62 (0.81 to 3.21)
Injured leg  13 (14.4) 77 (85.6) 7.56 (3.4 to 16.7)
Rheumatoid arthritis  30 (36.1) 53 (63.9) 1.97 (0.99 to 3.88)
Depression  49 (58.3) 35 (41.7) 0.69 (0.35 to 1.36)
Smoking advice  36 (48.6) 38 (51.4) 1.00
Smoker
 Yes  60 (54.1) 51 (45.9) 0.29 (0.17 to 0.49)
No 103 (34.6) 195 (65.4) 1.00
Total 163 (39.8) 247 (60.2)
*

Odds ratios and 95% confidence intervals for a preference for directing are shown relative to reference category of each factor, adjusted for other independently significant factors in multiple logistic regression. 

Data missing for one patient. 

This survey contained a higher ratio of women than men (2.6:1) than has been found in other studies of attendees at general practices.10 This is probably because women usually accompanied children. For analysis, patients were divided into those who had attended more than six times and those who had attended less than six times in the past year, as the average number of attendances stated by patients was 5.8 per year. Analysis showed that neither the order of presentation of the scenarios nor the actor affected the results.

Patients' preferences for shared or directed versions of scenarios were significantly associated with the patients' age, smoking status, and social class, the scenario, and their perception of their own doctor as being one who shared or directed. Multiple logistic regression showed that these variables were also independent predictors of preference (table 1). Although patients who left full time education aged less than 17 were significantly less likely to prefer the shared scenarios (78/225 (35%) versus 85/185 (46%), 95% confidence interval for odds ratio for sharing 0.41 to 0.95), this was not found to be an independent predictor. Although highly significant, the variable of “own doctor's style” was not included in the regression model because it was poorly answered. Patients who saw multiple doctors could not attribute a style to their own doctor. Generally, patients described their own doctor as having the same style as their preferred style (table 2). No significant associations were found with sex, frequency of attendance, or stated chronic ill health. Patients clearly preferred the directed approach for all the scenarios except those for depression and smoking advice. There was no significant evidence that the strength of age, social class, or smoking effects differed between the five scenarios. Smoking status was initially included because it was thought it might be a factor in the smoking advice scenario (smokers were more likely to prefer the shared version (15/19 (79.0%) versus 21/55 (38.2%), 95% confidence for odds ratio 1.65 to 28.64)); however, it was found to be a significant factor in all the scenarios.

The patient is a 32 year old man who has quite severe rheumatoid arthritis. He has tried a variety of treatments, which have not been very successful. He is in constant pain, but has chosen to keep on working as long as he can. He is married and has two children aged 10 and 14. He is currently receiving gold injections for his arthritis. They have not helped. He is disappointed as he had been told this treatment is usually successful, and he had started it with high hopes. He has found the injections and blood tests a real nuisance, and he wants to stop the treatment. His doctor has phoned the specialist who has recommended a higher dose of the drug. His doctor can think of no other course of action at the moment.

Shared approach

Doctor: Well, how are things?

Patient: Not great.

Doctor: Has there been any improvement since we last spoke?

Patient: I'd love to say yes, but there hasn't.

Doctor: What do you feel about the treatment then?

Patient: I think I've given it a good trial. It hasn't worked. I'd like to stop.

Doctor: I spoke to the specialist, she's very keen to try a higher dose.

Patient: Look of exasperation and disbelief. You're not serious!

Doctor: I know, I don't blame you, that would probably be my reaction too.

Patient: Do you really . . . I want you to be honest . . . think it will work?

Doctor: Honestly . . . I don't know for certain. It might. I have seen a higher dose work before. I think if it were me, and I know it is very hard to know how you are feeling at the moment, I would probably give it a go, but I would be going in with my eyes open not expecting too much.

Patient: I'm just fed up with being disappointed.

Doctor: I know.

Patient: OK lets do it, what have we got to lose.

Directed approach

Doctor: Well, how are things?

Patient: Not great.

Doctor: Well we'll have to do something about that then. I know you've been disappointed by the gold so far, so I've been on to the specialist. She says that much better results are obtained from higher doses. She recommends that we double the dose. I think that that's what we should do.

Patient: Look of exasperation and disbelief. You can't be serious!

Doctor: I know you're fed up, but I really hope this will make a difference.

Patient: It's just that I've heard this before.

Doctor: Believe me, we've got very few other options. I think this represents our best hope.

Patient: So I have to keep going with these darn injections and blood tests. I wish to goodness I thought they would work.

Doctor: Dr Johnson is an expert in this. She wouldn't recommend it if she didn't think it would work. I think you should try it.

Patient: Well, I don't suppose I have a choice. Let's do it.

Table 2.

Preference for shared or directed consultation related to perception of own doctor as one who shares or directs consultations

Own doctor's style Type of consultation preferred
Prefer sharing scenario Prefer directing scenario
No reply (n=60)  18 (30.0)  42 (70.0)
Shares (n=137) 106 (77.4)  31 (22.6)
Directs (n=213)  39 (18.3) 174 (81.7)
Total (n=410) 163 (39.8) 247 (60.2)

Odds ratio 15.3 (95% confidence interval 8.7 to 26.91) for doctor's style agreeing with patient's preference. 

Although chronic ill health was not found to be a significant factor, it was confounded by being more frequent in elderly patients, a group that had independently been shown to prefer the directed approach. Further analysis of younger (less than 61 years), chronically ill patients showed that they were more likely than other younger patients to prefer shared consultations, but this fell short of significance (35/68 (51.5%) versus 104/257 (40.5%), 95% confidence interval for odds ratio 0.88 to 2.76). Most patients saw the main difference between the versions of the scenarios as being one of direction or control.

Patients were usually sure about their decision and often vocal in justifying their choice. What was considered “decisive” and “sharing the patient's viewpoint” by some was considered “overbearing” and “shilly-shallying” by others. Only in the depression scenario was it obvious that some people (still a minority) were uncertain about which to choose.

Discussion

Watching videos of consultations is not the same as experiencing them as a patient. Although most patients have experienced an acute injury, fewer experience an illness such as rheumatoid arthritis, a scenario included to portray patients who are “expert in their illness.”7 It was clear from comments made by patients to me about this scenario that they viewed it as a complex medical problem of which they knew little, and they assumed that the patient was equally ignorant and should therefore follow the doctor's advice. Smokers had a notably different view from non-smokers on the smoking scenario, suggesting that personal experience might change a patient's view.

It is difficult to know if those taking part in the survey were different from those who did not. Although there were more women in the sample than men, no relation was found between sex and patients' preferences. The surgeries provided a spread of social classes and general practitioners. Patients with long term illness constituted 26% of the sample, which was comparable to other studies.10 Housebound patients, however, with whom doctors may have a stronger relationship,were excluded, and patients from Lothian may not be representative of the United Kingdom as a whole.

The study showed that a large number of patients preferred directed consultations when viewing the scenarios. This may mean that they still seek some direction from their doctor. Most studies of general practitioners' consultations show a strong degree of professional control.11,12 As in this study, most patients' experience is probably of a doctor who adopts a directed approach in consultations.

The patient is a 30 year old woman. She is a keen runner. Yesterday, while out running, she fell and hurt her right leg, which now has a bad bruise. She knows it is not serious, but the paracetamol she has taken hasn't helped. Her doctor has examined her, and they are now discussing what to do.

Shared approach

Doctor: Well, what do you think you've done there?

Patient: I think it's just a bad bruise . . . don't you?

Doctor: Yes. What did you hope I would do for you?

Patient: Something to ease the pain would be nice.

Doctor: I think that would be OK. What have you tried already?

Patient: Just paracetamol, and that was useless.

Doctor: What exactly did you have in mind?

Patient: I don't know. What do you suggest doctor?

Doctor: I was thinking of ibuprofen. I see you've had it a few times before. That will help the pain and possibly reduce the swelling a bit too. Do you think that would be a reasonable one to try?

Patient: That sounds fine to me.

Doctor: How do you feel about not going running for a couple of weeks?

Patient: Do I have to give it up?

Doctor: It's up to you, but I think it would be much better if you rested that leg.

Patient: OK. Thanks doctor.

Direct approach

Doctor: Well that's a nasty bump, but nothing serious. It must be sore. I take it you want something for it. Have you tried anything yet?

Patient: Just paracetamol, and they haven't helped much.

Doctor: Well, I think we can do better than that. I'll give you a prescription for ibuprofen; you've had it before. It's good for this sort of thing—it reduces the swelling a bit as well as easing the pain. Now I know you're not going to like this, but I think you should give the running a miss for a couple of weeks.

Patient: What?

Doctor: If you want it to get better quickly that's what you have to do.

Patient: OK. Thanks doctor.

Patients' preference for directed scenarios depended on the problem presented. This is in keeping with the theories of others.13,14 As with my study, other researchers have shown that for psychological illness or general advice the directed approach was not associated with benefit, but this approach seemed more beneficial for physical problems.15 As in my study, questionnaire surveys in American hospitals found that those patients who were more ill than others preferred to delegate more to doctors.16,17

What is already known on this topic

Patients who are allowed to express ideas and concerns in the consultation are more likely to be satisfied and to comply with treatment

Medical students and general practice registrars are encouraged to routinely involve patients in decision making in the consultation, but there is little evidence that patients want this involvement

What this study adds

Patients may vary in their desire for involvement in decision making in the consultation

Desire for involvement in decision making is associated with the presenting problem, patients' age, social class, educational level, and the style of the doctor usually seen

These associations are far from absolute, however, and doctors need to determine for individual patients how much involvement in decision making they want

That most smokers, despite lower social class, preferred shared consultations was surprising. This may be because they have experienced more authoritarian approaches from doctors and have come to dislike such treatment generally.18 The increased preference by higher social classes for shared decision making is consistent with studies examining time spent by general practitioners and the quality of consultations with this group.1922

I found that older patients preferred a more directed style than younger patients. This has been found by others.16,17 Older patients' experience of doctors in the past, recounted to me at the time of the study, was one of considerably more directiveness than today.

Some authors have suggested that patients tend to prefer what they know and are sceptical about what is new or unfamiliar.23 The finding that patients preferred the style they attributed to their own doctor may be explained in this way but could be equally explained by patients selecting general practitioners with their preferred style.

Analysis of young patients who were chronically ill showed them to be numerically, but not significantly, more likely than other young patients to prefer shared consultations. A larger survey or one directed at chronically ill patients may be necessary to elucidate this.

In my study some patients preferred directiveness in certain circumstances—for example, simple self limiting conditions and serious illness. In these circumstances some patients want the reassurance of certainty or possibly to avoid responsibility for a poor outcome.24 When patients believe they may have more insight into the problem than their doctor, such as for depression or lifestyle, more patients prefer to help decide their management.

Conclusion

Although the case for a listening doctor who is open to the ideas of patients in the history taking part of the consultation is strong, patients may vary in their desire for sharing in the decision making part of the consultation. This variation in desire depends on the presenting problem but is also associated with the age, social class, and educational level of the patient. These associations are not absolute, with large minorities of each group holding opposite views to the majority. For some conditions patients clearly thought that their own views on management must be taken more into account. This seems to be true of mental health and lifestyle problems.

Doctors need both communication skills and time in consultations, along with knowledge of their patients, to determine at which times, with which illnesses, and at which level their patients wish to be involved in decision making.

Supplementary Material

[extra: Videos validation description]

Acknowledgments

This study was part of an MD thesis at the University of Edinburgh. I have had invaluable help from the university's Department of General Practice, in particular Mike Porter my supervisor, but also John Howie, Sally Wyke, Jane Hopton, and Don Thomson. I also thank Graham Buckley, Ruth Liddle, and Denis Pereira Gray for their support and constructive criticism, my actors Bill Patterson, Alison Sinclair, Sophie Pilgrim, and John Liddle, statistician Rob Elton, the Lothian Primary Care Research Network, and all the general practitioners in Lothian who let me interview their patients.

Footnotes

Funding: Lothian Health funded the author's three month sabbatical during which the study was conducted.

Competing interests: None declared.

A description of the validation of the videos appears on the BMJ's website

References

  • 1.Stiles WB, Putman SM, Wolf MH, James SA. Interaction exchange structure and patient satisfaction with medical interviews. Med Care. 1979;17:667–681. doi: 10.1097/00005650-197906000-00010. [DOI] [PubMed] [Google Scholar]
  • 2.Roter D, Hall J. Physician interviewing style and medical information obtained from patients. J Gen Intern Med. 1987;2:326. doi: 10.1007/BF02596168. [DOI] [PubMed] [Google Scholar]
  • 3.Heaton PB. Negotiation as an integral part of the physician's clinical reasoning. J Fam Pract. 1981;13:845–848. [PubMed] [Google Scholar]
  • 4.Sackett DL, Haynes RB, Gibson ES, Hackett BC, Taylor DW, Roberts RS, et al. Randomised clinical trial of strategies for improving medication compliance in primary hypertension. Lancet. 1975;i:1205–1207. doi: 10.1016/s0140-6736(75)92192-3. [DOI] [PubMed] [Google Scholar]
  • 5.Stewart M, Brown JB, Weston WW. Patient-centred medicine. Thousand Oaks, CA: Sage; 1995. [Google Scholar]
  • 6.Tuckett D, Boulton M, Olson C, Williams A. Meetings between experts. London: Tavistock; 1985. [Google Scholar]
  • 7.Brody D. The patient's role in clinical decision making. Ann Intern Med. 1980;93:718–722. doi: 10.7326/0003-4819-93-5-718. [DOI] [PubMed] [Google Scholar]
  • 8.Pendleton D, Schofield T, Tate P, Havelock P. The consultation. An approach to learning and teaching. Oxford: Oxford University Press; 1984. pp. 97–103. [Google Scholar]
  • 9.Coulter A. Partnerships with patients: the pros and cons of shared clinical decision-making. J Health Serv Res Policy. 1997;2:112–121. doi: 10.1177/135581969700200209. [DOI] [PubMed] [Google Scholar]
  • 10.Office of Population Censuses and Surveys. General household survey. London: HMSO; 1991. pp. 150–160. [Google Scholar]
  • 11.Boulton M, Tuckett D, Olson C, Williams A. Social class and the general practice consultation. Sociol Health Illness. 1986;8:323–350. [Google Scholar]
  • 12.Byrne PS, Long BEL. Doctors talking to patients. London: HMSO; 1980. [Google Scholar]
  • 13.Szasz TS, Hollender MC. A contribution to the philosophy of medicine: the basic models of the doctor-patient relationship. Arch Intern Med. 1956;97:585–592. doi: 10.1001/archinte.1956.00250230079008. [DOI] [PubMed] [Google Scholar]
  • 14.Botelho R. A negotiation model for the doctor-patient relationship. Fam Pract. 1992;9:210–218. doi: 10.1093/fampra/9.2.210. [DOI] [PubMed] [Google Scholar]
  • 15.Savage R, Armstrong D. Effect of general practitioner's consulting style on patients' satisfaction: a controlled study. BMJ. 1990;301:968–970. doi: 10.1136/bmj.301.6758.968. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Ende J, Kazis L, Ash A, Moskowitz M. Measuring patients' desire for autonomy: decision making and information seeking preferences among medical patients. J Gen Intern Med. 1988;4:23–30. doi: 10.1007/BF02596485. [DOI] [PubMed] [Google Scholar]
  • 17.Cassileth B, Zupkis R, Sutton-Smith K, March V. Information and participation preferences among cancer patients. Ann Intern Med. 1980;92:832–836. doi: 10.7326/0003-4819-92-6-832. [DOI] [PubMed] [Google Scholar]
  • 18.Prochaska JO, Di Clemente CC, Morcross JC. In search of how people change application to addictive behaviour. Am Psychol. 1992;44:1102–1114. doi: 10.1037//0003-066x.47.9.1102. [DOI] [PubMed] [Google Scholar]
  • 19.Buchan I. ‘Time and motion’ in general practice. Practitioner. 1978;221:298–301. [PubMed] [Google Scholar]
  • 20.Bain DJG. Patient knowledge and the content of the consultation in general practice. Med Educ. 1982;11:347–350. doi: 10.1111/j.1365-2923.1977.tb00626.x. [DOI] [PubMed] [Google Scholar]
  • 21.Pendleton DA, Bochner S. The communication of medical information in general practice consultation as a function of patients' social class. Soc Sci Med {Med Psychol Med Sociol} 1980;14a:669–673. doi: 10.1016/0160-7979(80)90072-7. [DOI] [PubMed] [Google Scholar]
  • 22.Cartwright A, O'Brien M. Social class variations in health care and in the nature of general practitioner consultations. Sociol Rev Monogr. 1976;22:77–98. doi: 10.1111/j.1467-954x.1974.tb00019.x. [DOI] [PubMed] [Google Scholar]
  • 23.Porter M, MacIntyre S. What is, must be best: a research note on conservative or deferential responses to antenatal care provision. Soc Sci Med. 1984;11:1197–1200. doi: 10.1016/0277-9536(84)90370-8. [DOI] [PubMed] [Google Scholar]
  • 24.Shapiro J, Shapiro DH. The psychology of responsibility: some second thoughts on holistic medicine. N Engl J Med. 1979;295:939–943. doi: 10.1056/NEJM197907263010410. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

[extra: Videos validation description]

Articles from BMJ : British Medical Journal are provided here courtesy of BMJ Publishing Group

RESOURCES