Deciding if lifestyle is a problem: GP risk assessments or patient evaluations? A conversation analytic study of preventive consultations in general practice

Abstract

Objective. The aim of this study is to analyse the interaction between patients and GPs in preventive consultations with an emphasis on how patients answer GPs’ questions about lifestyle, and the conditions these answers impose on the process of establishing agreement about lifestyle as a problem or not. Design. Six general practitioners (GPs) video-recorded 15 annual preventive consultations. From these, 32 excerpts of discussions about lifestyle were analysed using conversation analysis (CA). Results. GPs used an interview format to assess risk in patients’ lifestyles. In some cases patients adhered to this format and answered the GPs’ questions, but in many cases patients gave what we have termed “anticipatory answers”. These answers indicate that the patients anticipate a response from their GPs that would highlight problems with their lifestyle. Typically, in an anticipatory answer, patients bypass the interview format to give their own evaluation of their lifestyle and GPs accept this evaluation. In cases of “no-problem” answers from patients, GPs usually encouraged patients by adding support for current habits. Conclusion. Patients anticipated that GPs might assess their lifestyles as problematic and they incorporated this possibility into their responses. They thereby controlled the definition of their lifestyle as a problem or not. GPs generally did not use the information provided in these answers as a resource for further discussion, but rather relied on standard interview procedures. Staying within the patients’ frame of reference and using the patients’ anticipatory answers might provide GPs with a better point of departure for discussion regarding lifestyle.

Key Words::

• Communication
• conversation analysis
• Denmark
• general practice
• lifestyle
• qualitative research
• risk assessment

• Danish GPs are required to risk assess chronically ill patients and this includes a discussion concerning lifestyles.

• In a substantial minority of cases patients made evaluations of their lifestyle ahead of the GP's lifestyle interview (anticipatory answers).

• By way of conversation analysis the article investigates how GPs proceed with the lifestyle discussion after anticipatory answers.

• To develop a fruitful discussion about lifestyle after anticipatory answers it is recommended that GPs focus more on the patient's frame of reference and less on risk.

Introduction

Consultations in general practice are traditionally initiated by patients attending with health problems [1–5]. However, in preventive work establishing agreement on the existence of a problem is often the first task. Several studies have indicated that such an agreement cannot always be reached. For example Sorjonen et al. found that most patients in acute and follow-up visits in Finnish general practice thought of their lifestyles as non-problematic [6]. The process of collaborative construction of a problem has been described in studies of general practitioners (GPs) and nurses in general practice [7]. In this process the health professionals depend on the patient's response to establish a mutual understanding of lifestyle characteristics as a problem [7,8]. It is preferred that the problem is not defined by health professionals alone [7–9]. When the problem is collaboratively constructed the risk that later advice could be redundant or undesired is controlled [9].

The recommended technique for lifestyle discussions in general practice is motivational interviewing (MI). MI represents the ideal collaborative conversation and aims to let patients themselves make the argument for change [10]. It is a core idea in MI not to persuade patients to change their lifestyle as this will cause resistance to change [11]. While the MI method explains in detail how to negotiate change with patients [8,10], it does not, to the same extent, describe how to negotiate whether or not certain behaviours should be considered a problem.

In Denmark, as in many other countries, a proactive approach to patient lifestyle is developing. Among such initiatives are annual check-ups with preventive objectives for patients with chronic disease, and for patients with a high risk of developing chronic disease. The GP invites a patient to attend a fixed-appointment preventive consultation (preventive consultation), which has a pre-agreed agenda [12]. In these consultations a risk assessment is mandatory. Lifestyles are aspects of risk and the guideline for preventive control of diabetes, for example, specifies that assessments and discussions about smoking, diet, and physical activity should be included in these consultations [13]. The guideline also states that the evidence for the effect of lifestyle interventions on complications and mortality is scarce but that it is included anyway since it was part of the studies testing the pharmacological treatments recommended [13].

Similar recommendations are outlined in guidelines for other chronic diseases such as chronic obstructive pulmonary disease.

Preventive consultations provide a particular frame for discussions of lifestyle, as a lifestyle assessment is a mandatory part of the consultation.

The aim of this study was to analyse the interaction between patients and GPs in preventive consultations with an emphasis on how patients answered GPs’ questions about lifestyle, and the conditions these answers imposed on the process of establishing agreement about lifestyle as a problem or not.

Material and methods

The study was carried out with six GPs in Danish general practice. It was based on 32 samples from 15 video-recorded consultations, which were subsequently transcribed. The samples represented the moments when doctors commenced a new lifestyle discussion, that is, when they initiated inquiries about subjects such as smoking, physical activity, diet, alcohol, and weight. Our analyses concentrated on how patients responded to doctors’ initial inquires, and which conversational consequences applied to the different types of responses.

Cases were drawn from fixed-appointment preventive consultations, most of them annual preventive controls of chronic disease. In these consultations, the GP contract encourages the assessment of patients’ health risks and discussions about lifestyle. They are also recognized by GPs themselves as the type of consultation where lifestyle discussions often take place. GPs who had the highest number of preventive consultations during 2010, according to Danish national registries [14], were invited to participate in our study. The GPs were from two of the five regions in Denmark, covering both urban and rural communities. The patients were 10 women and five men, aged between 43 and 80. The GPs were three women and three men, with varying practice tenure, aged 42–64 years. Four of the GPs were working in partnership practices and two in cooperative practice. GPs who participated received financial compensation equalling the time spent on preparation for the project.

We investigated the data using conversation analysis (CA), which is multidisciplinary and crosses both sociology and linguistics. It is used to study how talk-in-interaction is organized, and how members employ its structures to reach common understandings [e.g. 15–17]. Several of the important contributions to CA have focused on doctor–patient interaction [e.g. 18–22]. There is a large body of research on the ways in which general practice consultations predominantly consist of question-and-answer-driven interactions, that is, doctors ask questions, to which patients provide answers [22–24]. In fact, this conversational organization occasionally makes it troublesome for patients to contribute information to the dialogue that exceeds merely answering questions [25,26].

The CA method implies a commitment to consider the possible significance of the smallest paralinguistic details of conversational contributions. This commitment has led to a rigorous notation standard, among other things, which seeks to depict not only what participants say, but also how they say it [27]. Thus, CA researchers transcribe their data rigorously by means of the Jeffersonian principles of notation [28]. A list of symbols used for this study (Supplementary Appendix 1 available online at http://informahealthcare.com/doi/abs/10.3109/02813432.2015.1078564) and transcripts with CA symbols (Supplementary Appendix 2 available online at http://informahealthcare.com/doi/abs/10.3109/02813432.2015.1078564) may be found at the end of the paper. Excerpts are identified by a letter and a number identifying the GP (A–F) and the patient, e.g. A3. Original Danish data are provided in Supplementary Appendix 3 available online at http://informahealthcare.com/doi/abs/10.3109/02813432.2015.1078564.

Results

Lifestyle interview adherence

Lifestyle discussions were organized in conversational trajectories where the parties addressed different lifestyle issues. Most of these trajectories took the form of questions and answers, and they all started with an interview, led by the doctor, assessing the patient's risk as in excerpt (1):

(Excerpt 1) (C2)

01. DOCTOR (DO): And you don't smoke do you, as far as I recall?

02. PATIENT (PA): No I've never smoked a cigarette.

A similar trajectory commencement is found in excerpt (2) in which a patient corrects her answer to provide a more accurate one:

(Excerpt 2) (B2)

01. DO: What about, erhm, tobacco?

02. (short pause)

03. DO: Do you smoke?

04. (short pause)

05. PA: No I never have.

06. DO: No.

07. PA: I have tried it though.

Patients’ responses regarding tobacco consumption in excerpts (1) and (2) illustrate what we term “lifestyle interview adherence”. This behaviour entails that patients provide answers that conform with doctors’ questions. Also, patients accept and await the trajectories, which are governed by the doctors’ questions.

This is also the case for the answer concerning tobacco in excerpt (3):

(Excerpt 3) (B1)

01. DO: And tobacco, you don't smoke?

02. PA: No.

03. DO: Excellent and your weight, what does it say?

04. PA: Well it's sort of on the rise.

05. DO: Hmm how much do you weigh now?

The way in which the patient in excerpt (3) designs her response concerning weight, however, does not meet the criteria for lifestyle interview adherence. The response, rather than providing information on “what the weight says”, hints at a problematic development (“sort of on the rise”), which anticipates a discussion about the need for lifestyle changes.

In cases of lifestyle adherence it is usually the GP who makes an evaluation of the patient's lifestyle as problematic or not. In cases where patients respond with an anticipatory answer, such evaluations are, rather, made by the patients. We investigated further this latter type of trajectory as it poses a challenge to the GPs’ ability to make a risk assessment and eventually establish patients’ lifestyles as a problem.

Anticipating issues of problematic lifestyle and possible recommendations

Patients’ responses commonly anticipated talk about the problematic nature of their lifestyles, even in cases, such as in excerpt (3), where doctors designed their questions as relatively neutral inquires. Consider the trajectory commencement of excerpt (4):

(Excerpt 4) (A2)

01. DO: Then I just need to check on smoking;

02. how are things going with that?

03. PA: Well it's going a lot better but I haven't quit entirely yet.

04. DO: Yes.

05. DO: No. What – how many do you smoke a day?

The patient's response (line 03) is markedly different from what was observed in the previous examples of lifestyle interview adherence. It answers more than the question. The answer consists of two parts: the first part (“well … better”) emphasizes an improvement; the second part (“but … yet”) admits that the patient still smokes. Together, not least because of the tying conjunction “but”, the answer conveys awareness of smoking being a bad habit and a hope to overcome it in the future.

Most of the examples so far have, for the sake of comparison, concerned smoking. But the distinction between replies that meet lifestyle interview adherence, versus replies that anticipate recommendations for lifestyle changes, apply equally well to other lifestyle issues. For instance, in excerpt (5) the discussion is about exercise:

(Excerpt 5) (A1)

01. DO: Then, erhm, we usually also talk about exercise what … what …

02: PA: I walk and bicycle a lot.

03. DO: Yes.

04. PA: But I don't visit the gym or anything like that.

05. DO: No, no.

06. PA: But I consider myself very active.

07. DO: But that's also … the main thing is that you move your body and bicycling will also get your heart rate up and …

08. PA: Yes.

09. DO: Are you active on a daily basis?

Before the doctor has even completed her initial question this patient provides a series of responses, which anticipate either an evaluation of lifestyle as problematic, or advice to change lifestyle. The responses emphasize the patient's active lifestyle (line 02 and 06), but also seek to moderate its extent (line 04).

Anticipatory answers take different shapes. Most consist of patients’ own evaluations of their lifestyle, as in excerpts (3–5) above. Some evaluate the lifestyle issue positively, as in (excerpt 5, line 06) “but I consider myself very active”; others evaluate it as problematic, as in (excerpt 3, line 04) “well it's sort of on the rise”. Still other responses describe lifestyle issues as matters that are already taken care of. This is illustrated in excerpt (6):

(Excerpt 6) (B3)

01. DO: How about your weight; is it somewhat stable?

02.PA: Actually I'm in the process of losing weight.

03.DO: Excellent!

By answering that she is losing weight (line 02), the patient indicates both that she considers her weight a problem and also that the problem is already being taken care of.

Proceeding after anticipatory answers

GPs continued the lifestyle interview assessing patients’ risks after the anticipatory answers. Examples are seen in excerpt 3, line 05 where the GP asks how much the patient weighs now; in excerpt 4, line 05 where the GP asks how many cigarettes a day the patient smokes; and in excerpt 5, line 09 where the GP asks about frequency of physical activity.

How the GPs proceeded with the lifestyle interview after anticipatory answers depended on whether or not the patients’ anticipatory answers evaluated their lifestyle as problematic.

In cases where patients evaluated their lifestyle as problematic, the GPs accepted this evaluation and continued the interview, probing the patient for possible change. Prior to excerpt (7), the patient has explained that he enjoys gardening on a regular basis:

(Excerpt 7) (C4)

01. DO:What, what d… what other kinds of exercise do you do?

02. PA:Naturally we bicycle.

03. DO:Yes.

04. PA:But I must con… I must confess that we probably haven't kept it up this XX summer.

05. DO:Oh.

06. PA:Among others because of the weather.

07. (short pause)

08. PA:It has been windy as hell. Heh heh.

09. DO:Yes.

10. PA:But of course now we are about to go for walks.

11. (short pause)

12. DO:XX

13. PA:Then it will be better. We walk in the wintertime.

14. DO:Yes.

15. PA:Bicycle in the summer.

16. DO:Every day or what?

17. PA:Two, three times a week.

18. DO:Yes for how long?

19. (short pause)

20. PA:About an hour.

21. DO:Yes.

22. (short pause)

23. PA:Approximately five kilometres.

24. DO: What prevents you from doing this every day?

25. PA:Nothing does.

26. (short pause)

27. DO:No.

28. PA:Nothing.

29. DO: Couldn't you do this every day?

30. PA:We could easily do that.

This patient replies to the doctor's question about his exercise habits (line 01) with an elaborate anticipatory answer (lines 02–16). In this answer, the patient anticipates that the doctor will assess his exercise habits as insufficient by “confessing” how relatively little he bicycles (lines 04–05); by excusing himself with reference to bad weather (lines 07–09); and, unsolicited, by adding that he and his wife are about to engage in their wintertime walks. The doctor continues the lifestyle interview asking the patient specifically how often (line 17) and for how long (line 19) he walks. After this inquiry the doctor asks a confrontational question about why the patient does not walk every day (line 25). When the patient admits that nothing prevents him (lines 26–29), the doctor, in turn, is able to pose a final question, which comes very close to explicit advice regarding walking every day (line 30).

In cases where patients evaluated their lifestyle as unproblematic, on the other hand, the GPs did not probe for change, as in excerpt (7), but rather offered support for the current habit. The statement by the GP in excerpt 5 (lines 07–08) is typical of these trajectories. The GP supports the patient's view that her level of physical activity is sufficient, while at the same time incorporating the supportive information that what counts in physical activity is to get the heart rate up. Another example of support for a healthy habit is seen in the continuation of excerpt (6):

(Excerpt 6, continued) (B3)

04. PA:Excellent.

05. DO:That's really great.

06. (short pause)

07. PA:So I've lost four kilos.

08. (short pause)

09. DO:During how long?

10. PA:Six months I think.

11. DO:Six months? Well that's marvellous!

12. PA:So erh-

13. (short pause)

14. DO:That's something you can sense yourself, right?

15. PA:Yeah.

16. DO:That … it's … it's really good, right?

17. PA:Sure, yes.

18. (short pause)

19. PA:I can sense it both on my clothing …

20. DO:Also when you consider hips and these things.

21. PA:Yes.

22. DO: You know it's really good for …

23. PA:Sure it is.

24. DO:… joints and all these things.

After the patient's anticipatory answer that she is in the process of losing weight (line 02), the GP demonstrates appreciation in several steps telling the patient that this is great (line 03 and 05) and even marvellous (line 11). The GP also adds support by stressing the medical benefits of weight loss (lines 22 and 24). This kind of support was given in almost all cases of a “no-problem” answer from patients. The GPs did not challenge patients’ own evaluation of their lifestyle as unproblematic. In one case, however, the question was reintroduced after poor test results indicated that there may, after all, be a problem.

Discussion

Principal findings

GPs conducted lifestyle interviews to establish whether patients’ lifestyles posed a health risk. The interview questions were usually answered by the patients, but in a substantial minority of cases patients’ self-evaluations of lifestyle were added in anticipation of advice or recommendations from the GP. In some cases, such self-evaluations were given instead of answers conforming to the GP's question. In cases where the lifestyle issue was considered problematic by the patient, the GP probed for possible change; and in cases where the lifestyle issue was considered unproblematic by the patient, the GP supported current habits. GPs usually did not challenge the patients’ own evaluation of their lifestyle as unproblematic even though they generally asked further questions about quantity and frequency of habits.

Strengths and weaknesses of the study

Recruiting among GPs with the most activity in prevention gave us rich material with an abundance of lifestyle discussions. GPs with a low level of activity in lifestyle discussions were not included in the study and their preventive consultations may differ from those represented here.

It is possible that GPs may have conducted more lifestyle discussions than usual to satisfy the researcher and enrich the recordings. However, there is documented research indicating that recording has very little effect on the content of consultations [29], suggesting that this would probably not be an important factor.

The preventive annual controls and discussions about lifestyle we investigated shared some aspects of structure and organization. The aspects shared by the practices in our study are expected to apply to more practices due to common institutional goals and shared competences of interaction. There were also many differences and variations. By focusing on the aspects that the consultations had in common, we did not address all the variations in style expressed by the GPs who participated.

Findings in relation to other studies

By putting their evaluations first, the patients’ answers determined whether or not their lifestyle was considered a problem, and also the ongoing trajectory of the lifestyle interview. It is clear from previous research that establishing a problem in preventive work is often done in a stepwise fashion in collaboration between the patient and the professional [7, 9]. It seems, however, that when patients put their evaluations first, they challenge this stepwise process of recognizing that a problem may exist. Furthermore, patients often demonstrated knowledge of lifestyle issues through their anticipatory answers. Previous studies in CA have described how people pitch what they say to meet the knowledge they believe the people they are talking to already have [30]. This has become established as a norm: that is, people do not tell others what they believe they already know [31]. Given this norm, the patients who demonstrated their knowledge of lifestyle issues did not invite GPs to provide more information about lifestyle.

GPs supported patients’ habits from a medical perspective if the patients themselves considered their lifestyles unproblematic. In Sorjonen's study [6] GPs also worked to support a “no-problem” evaluation made by patients. Unlike the GPs in Sorjonen’s study, the GPs in our study usually added support or a recommendation for current habits. This difference could be explained by the fact that Sorjonen was investigating acute consultations, whereas we focused on preventive consultations where lifestyle issues are differently framed.

The frame of the preventive consultation might also explain the difference between the findings in our investigation and those in the work by Stivers and Heritage [26]. They described the phenomenon that extended answers, which demonstrate the patient's knowledge of the appropriate course of action, pre-empted the GP from pursuing a lifestyle issue. Their study was based on extended medical interviews, which do not have the emphasis on discussion of lifestyle that the preventive consultations in our study have. We found that GPs maintained the interview format asking “how much”, “how often”, and “how far” even after patients’ anticipatory answers. Determining lifestyle risks and reducing risk are formalized aims of preventive consultations [12]. The aim of reducing risk determines what it is necessary for the doctor to know about the problem and leaves other aspects untouched [32].

From an epidemiological perspective lifestyles are correlated with risks of morbidity and mortality. Such classifications provide doctors with knowledge about illness but also with specific perspectives on people [33]. Quantity and frequency are aspects of lifestyle that are relevant to risk, for example too much food or infrequent physical activity. The general perspective of risk, however, does not include the social context of individual patients. Inquiry that addresses aspects of patients’ answers other than quantity, frequency, or distance might be more fruitful in creating a discussion about lifestyle [11]. A previous investigation of lifestyle counselling in general practice showed that “change talk” was best produced when the nurse stayed within the patient's frame of reference [34].

Anticipatory answers could be seen as a contribution by patients to advancing the activity of lifestyle discussions. Previous studies of conversation describe how answers may, in some cases, not conform to the questions but still contribute to the progressivity of the inferred overall activity [35]. In our study, the overall activity is the process of assessing lifestyle, evaluating it as a problem or not, and discussing problematic lifestyles in terms of changes. By giving their evaluations first, the patients make GPs’ assessments of their lifestyles irrelevant and leave the GPs without the knowledge they need to independently evaluate the lifestyle in question. Treating patients’ anticipatory answers as valid contributions to the progress of the discussion, instead of insisting on interviews about risk, might open new possibilities for discussions of lifestyle.

Our study shows that patients anticipate advice concerning their lifestyle in preventive consultations. The anticipation of advice interferes with the application of MI in general practice consultations. A key aspect of MI is to avoid raising patients’ resistance to change (11). It seems that the orientation of the institution of general practice towards a healthy lifestyle is established to an extent where advice is an expected outcome. The anticipation of advice has also been described in routine consultations [36]. In this respect, the institution itself may act to raise patients’ resistance even in cases when GPs do not give explicit advice.

Conclusion

GPs conducted lifestyle interviews to determine patients’ risk. Patients often anticipated that the GP would consider their lifestyle problematic and provided their own evaluations of whether or not lifestyle was a problem. In cases where the lifestyle issue was considered problematic by the patients, the GPs probed for possible change; and in cases where the lifestyle was considered unproblematic by the patient, the GP supported current habits. GPs usually did not use the substance of patients’ initial answers as a resource for furthering this talk but rather relied on standard interview procedures.

To develop a more fruitful discussion about lifestyle we recommend that GPs explore other strategies than relying on questions about frequency, quantity, and distance. Staying within the patient's frame of reference and developing the conversation on the information the patient provides about relations, everyday life, and experience in anticipatory answers may be one strategy to explore.

Supplementary material available online

Supplementary material, Appendix 1–3 available online at http://dx.doi.org/10.3109/02813432.2015.1078564.

Acknowledgements

The authors would like to thank the patients and the general practitioners who kindly participated in the study. They would also like to thank the Novo Nordisk Foundation and the Committee of Multipractice Studies in General Practice for financial support and Julie Høgsgaard Andersen for valuable comments on the manuscript.

Ethical approval

According to the principles of Danish ethics committees, qualitative studies are not evaluated. Written informed consent was obtained from patients participating in the study.

Declaration of interest

All authors declare having no competing interests. The authors alone are responsible for the content and writing of the paper.