Patient participation in decision-making about cardiovascular preventive drugs – resistance as agency

Abstract

Objective

The aim of the study was to describe and explore patient agency through resistance in decision-making about cardiovascular preventive drugs in primary care.

Design

Six general practitioners from the southeast of Sweden audiorecorded 80 consultations. From these, 28 consultations with proposals from GPs for cardiovascular preventive drug treatments were chosen for theme-oriented discourse analysis.

Results

The study shows how patients participate in decision-making about cardiovascular preventive drug treatments through resistance in response to treatment proposals. Passive modes of resistance were withheld responses and minimal unmarked acknowledgements. Active modes were to ask questions, contest the address of an inclusive we, present an identity as a non-drugtaker, disclose non-adherence to drug treatments, and to present counterproposals. The active forms were also found in anticipation to treatment proposals from the GPs. Patients and GPs sometimes displayed mutual renouncement of responsibility for decision-making. The decision-making process appeared to expand both beyond a particular phase in the consultations and beyond the single consultation.

Conclusions

The recognition of active and passive resistance from patients as one way of exerting agency may prove valuable when working for patient participation in clinical practice, education and research about patient–doctor communication about cardiovascular preventive medication. We propose particular attentiveness to patient agency through anticipatory resistance, patients’ disclosures of non-adherence and presentations of themselves as non-drugtakers. The expansion of the decision-making process beyond single encounters points to the importance of continuity of care.

KEY POINTS

• Guidelines recommend shared decision-making about cardiovascular preventive treatment. We need an understanding of how this is accomplished in actual consultations.This paper describes how patient agency in decision-making is displayed through different forms of resistance to treatment proposals.

• •The decision-making process expands beyond particular phases in consultations and beyond single encounters, implying the importance of continuity of care.

• •Attentiveness to patient participation through resistance in treatment negotiations is warranted in clinical practice, research and education about prescribing communication.

Introduction

Parallell to an increasing demand for patient participation in health care is an increasing use of preventive drugs. Cardiovascular prevention has become a major task for primary care. Guidelines state at what levels of risk preventive drugs are recommended, but also advocate risk assessments to be the basis of shared decisions [1,2]. Although often advocated for the sake of outcome, patient participation in decision-making has an ethical value in its own right.

Swedish law mandates patient participation in health care through information and sharing of decisions between patients and physicians. The communication of risk in health care is considered to be problematic as risk statistics are often misunderstood, by physicians as well as laymen [3]. To obtain shared decisions, patients and physicians need to mutually engage in meaningful conversations and have access to individualised evidence in formats they can understand [4]. They also need to decide who decides, and to be aware of on what grounds physicians and patients base their respective decisional authority [5,6].

Tools to evaluate physicians’ communication skills with regard to shared decision-making assess their competences to involve patients in decisions [7]. The preoccupation with the physician’s skills solely, infers that for patient participation to occur, the patient needs to be invited by the physician.

Contrary to such assumptions, conversation analytical studies have shown that patients do participate actively and uninvited in decision-making [8–10]. Unlike interview-based research, conversational analysis and other interactional analytical methods are used to study naturally occurring social interaction such as clinical encounters.

Here, the analytical focus is neither on the GPs’ nor the patients’ descriptions, but on their actions as participants in communication: their display of their sense making of what goes on, through their use of language in interaction [11]. Theme-oriented discourse analysis takes into account both the micro level of detailed features of talk, and a wider context, including the level of whole encounters. It combines the analytical themes from studies of naturally occuring human interaction, such as the display of agency in decision-making, with the focal themes of concern for the profession, such as shared decision-making.

Acting as a decisional agent has been described as “to have choices and the competencies to act on them” [12]. Patients exert agency when they influence or make decisions about their health care.

An acceptance to a proposal is the normative preferred response, in everyday social interaction [13], as well as in health care settings [8–10,14]. When a treatment proposal is not readily followed by uptake from the patient, the interaction takes on the form of a negotiation with collaborative efforts from the participants to reach a mutually agreed decision. In spite of the negative ring to the word, resistance from patients to treatment proposals is one way to accomplish their legitimate agency in decision-making, which is a prerequisite for shared decision-making and patient participation.

Dispreferred responses are potentially face threatening in all interaction, and not specific for patients responding to physicians’ treatment proposals [15]. Signals of potential disagreements in general conversations tend to be hedged and subtle, guided by politeness strategies, and include silences, hesitations, requests for clarification and weakly stated agreements [13,16].

Previous studies from health care settings (general internal medicine, oncology and pediatrics) have shown the same mechanisms at play. Patient resistance here was shown to be displayed through withheld responses and weak agreements (passive resistance), counterproposals and questions (active resistance) [8–10].

Shared decision-making is called for in guidelines for cardiovascular prevention. Although patient participation in decision-making through various forms of resistance to proposals has been described in other settings, there is a paucity of studies of actual clinical encounters where decisions about cardiovascular preventive drugs are being made.

Aim

The aim was to describe and explore patient agency through resistance in decision-making about cardiovascular preventive drugs.

Materials and methods

The material was collected within a larger project with the overarching aim to explore how cardiovascular risk and recommendations for medication to prevent cardiovascular disease is contextualised in actual clinical encounters. Data were selected from 80 patient–GP–encounters in primary care in the southeast of Sweden, audiorecorded 2008–2010.

GPs were recruited through a brief e-mail invitation to all GPs in the area. Those who responded with an interest to participate were given further information by author JH. Six GPs participated after written consent, see Table 1 for characteristics.

Table 1.

GP characteristics.

Age	41–64
Years in practice	12–36
Male/female	3/3
Rural/urban	3/3

They were equipped with a digital recording device, and administred inclusion of patients and recordings. The authors were not present during data collection. Adult patients, able to communicate in Swedish without an interpreter, were included after verbal and written consent.

The GPs were instructed to ask all patients with scheduled appointments during 1 or 2 days, with no prior selection of specific complaints or expected reasons for the patient’s visit. This was to ensure the capture of all talk about cardiovascular preventive drugs, also that occurring en passant in visits for other reasons.

The recordings were transcribed verbatim by a secretary, and refined by author J.H., according to conversational analytical convention [11]. For transcription symbols, see Table 2.

Table 2.

Transcription conventions.

UPPERCASE	loud talk
emphatic	with emphatic stress
she said [that	overlapping talk starting at [[right
(xxx)	undecipherable talk
=well but	no pause between turns, latching
(.)	micropause (less than 1/4 s)
(2.0)	timed pause (here: 2.0 s)
ye:s	lengthening of a sound
tra-	speaker interupts herself in word
in case–	speaker leaves utterance incomplete
°now°	speech in low volume (“sotto voce”)
*here*	laughter in speaker’s voice
((phone rings))	comments
h	exhalation
.ah	spoken on inhalation
pt, mt	smacking, clicking sound

J.H. performed the initial sorting of the data. Sequences involving treatment proposals from GPs concerning cardiovascular preventive drugs were selected for further analysis. They included suggestions for new prescriptions, changes of dose, stopping medication, and the proposal for no change. Within these sequences, all instances where patients did not promptly accept the GP’s proposal were selected.

Different ways for patients to resist proposals were related to previous findings about patient resistance as agency [8,9], and other interactional phenomena concerning agency in decision-making [10,16–18]. As previously described, physicians treat patient acceptance of treatment as necessary to proceed to the next activity [8,9,19]. All other responses from patients to GPs’ proposals than prompt acceptance were met with interactional work from the GPs in pursuit of acceptance from the patients. Such reactions from the GPs were regarded as indications of patient resistance. Responses from the GPs included intensifications and modifications of their initial proposal. Intensification was accomplished through provision of more information, rhetorical reinforcement, such as change of framing, addition of arguments, and the invocation of external authorities such as hospital doctors, guidelines or the medical expert community in general. Modifications included postponement of decision and adjustments according to objections brought forth by patients, including their counterproposals.

The descriptions of resistance were refined, discussed by the authors and brought back to the empirical examples for further definition. Finally we went back to the whole encounters to get an overview of the findings in their context.

The analysis deals with the intersection between two themes: the general interactional concepts of resistance and agency, and the clinically relevant focal theme of decision-making about prescribing cardiovascular preventive drugs [11].

Results

35 out of the 80 encounters concerned cardiovascular prevention. 28 of the 35 encounters contained a treatment proposal, and of these 18 contained instances with resistance from patients to treatment proposals from GPs. We describe seven different types of resistance, illustrated by transcribed excerpts translated from Swedish into English. See Supplementary Appendix online for original transcripts. The encounters are referred to with GP coded with letters A–F and the patients numbered in consecutive order.

Withheld response

Silence, even very brief, from the patient following a treatment proposal from the GP is interactionally handled as withholdment of acceptance.

Excerpt 1: F05

01 D:pt cause now you’ve taken it for a month

02right

03 P:a yes h a month or five weeks

04or what it was

05 D:yes that’s right

06(2.9)

07 D:an’ then we oughta have seen effect (.) fully

08from it

09(2.4)

10 D:pt ha let’s see

11(1.0)

12 D:pt the alternative is to add a

13e-medicine

14(1.7)

15 D:but (.) I think it’s better

16to raise this

17 P:*yes please*

This excerpt was preceeded by a proposal from the GP to raise the dose of an antihypertensive drug, which the patient initially did not provide acceptance to. Her immediate response in the turn after the proposal to raise the dose is to ask “to one and half tablets then”, treated by the GP as a non-acceptance of the proposal, displayed by her continued argumentation that they should have seen the full effect by now. The patient continues to withhold acceptance, with long pauses in lines 6 and 9. In line 12, the GP presents the alternative to add another medicine, framing it as a choice between that and a raised dose, again met with silence from the patient. When the GP, within the new framing, proposes a raised dose in line 15, the patient accepts and they go on to the practicalities of prescription and follow-up arrangements.

Minimal unmarked acknowledgement, identity as non-drugtaker, contest address

The following excerpt contains examples of different forms of resistance. It is also illustrative of a typical cardiovascular preventive treatment discussion in our data, with its orientation towards measurable values, such as blood pressure, rather than numerical risk estimates, or other explicit assessments, of cardiovascular risk. There were no examples in our data of the use of SCORE charts or other decision aids with a quantification of the risk.

Excerpt 2: B05

01 D:when I see this value that is a little

02elevated then I’d like considering that

03you have a diabetes and have heart problems

04and so forth and that it runs in the family

05too put you on (.) eh lipid lowering medication

06 P:mhm

07 D:that one takes at night then

08 P:mm

09 D:and it is like an extra protection then

10 P:mm

The treatment proposal is backed up with arguments enhancing acceptance as the preferred response. The patient gives minimal acknowledgements without delay. It is not treated as acceptance by the GP who adds information in line 7 and motivates it in line 9, still with minimal response from the patient. The GP modifies the framing with acknowledgements about the patient’s good habits building the argument that it is not enough, before repeating the proposal with emphasis in lines 18–21. The patient continues to respond minimally until the GP stresses the elevated value in line 21, to which she responds with a question.

18 D:ehm but then one sh-shall add some

19medication too when one sees that it

20 P:mm

21 D:after all is=

22 P:=what what’s that value then

The GP states the patient’s cholesterol value and that it is too high, reinforcing the proposal that is repeated in lines 33–34. The patient continues to respond minimally to the GPs’ arguments, and the decision-making process is not proceeding. In line 39, the patient repeats her minimal response ending with “hm”. Here the GP changes footing, leaves her line of argumentation and asks what the patient thinks about it, which turns their communication in a new direction.

31 D:and the desirable level is below zero point six

32 P:ah ye:s

33 D:.hhm so that’s why I think we ought to

34add (.) another tablet then

35 P:mm

36 D:to lower the blood lipids

37 P:mm

38 D:further then

39 P:mm mm .hhm

40 D:but what do you say about that

41 P:eh well I’m not all that

42fond of *medicines*

43 D:no

44 P:but of course if you judge it

45 P:like that then °I° hav–

46 D:yes today one has eh eh one considers

47that one has some extra protection

48 P:mm

49 D:when one has diabetes

50 P:mm

51 D:and heart problems

52 P:mm (.)

53 D:then one has some to gain from it

54 P:but one can try it and see

55if i:t i- make- gives an effect

56 D:absolutely

57we follow it up

58 P:yes

The patient’s response in lines 41–42: “I’m not all that fond of medicines” is mitigated by a tone of laughter in her voice, signalling that she is delivering a dispreferred response. It constitutes continued resistance. She presents herself as someone – an identity marker – not prone to take medicines.

The address, a tentative inclusive “we”, in the final proposal in lines 33–34 is contested by the patient in lines 44–45: “if you … I hav– … ”. She opens for an acceptance of the treatment proposal, but renounces the responsibility for the decision and puts it with the GP, contesting the inclusive we from the GP.

The GPs’ “one considers” in line 46 in response to the patients placement of responsibility on her, detaches her from personal responsibility and passes it on to an unspecified “one”; the medical expert community in general.

The agent responsible for the decision is under negotiation here, as well as the treatment decision. The decision to treat is reached when the patient accepts both the proposal modified according to her conditions (to have it followed up), and the partial responsibility for the decision in the negotiated decisional agent “we”.

Ask questions

Patients’ questions in immediate response to treatment proposals were seen in Excerpt 1: (raise the dose) “to one and a half, then”, and in Excerpt 2: “what’s that value then”. To ask questions was the most common form of resistance in our data, with instances from all GPs.

Excerpt 3: E21

01 D:(:) I’d like us to actually add

02one more one blood pressure tablet and that

03may sound like a lot cause you already have three

04 P:yeah what model

05 D:yes it is another

06model [than those yes another variant

07 P:[yet a model

08 P:ah yes

09 D:cause it eh it’s no use increasing

10the dose of those you already have

Here the patient responds with a question about what “model” of drug the GP proposes. By doing so, he withholds acceptance. He also makes himself accountable as a decision-maker through claiming information on which to base a possible final decision. The GP not only answers the question, but also provides more information, reinforcing the line of argumentation for his proposal. The patient’s question is not an explicit questioning of the treatment proposal but it still challenges it. By asking a clarifying neutral question the patient withholds acceptance, sets the agenda, and claims knowledge relevant for an eventual decision, thereby displaying agency in the decision-making.

Counterproposals

Counterproposals were seen as a response from the patient in the turn after a treatment proposal, but also like in this example, within the context of evaluating the treatment in a follow-up visit, before an explicit treatment proposal from the GP. Counterproposals included suggestions to change doses, to stop medication, or to postpone the decision, typically with the proposal for diet or exercise as an alternative to the medication proposed by the GP. The counterproposals were sometimes backed up with reference to a non-present authority such as advice or assessments from other health care professionals, and sometimes with the expression of an identity as someone who doesn’t take medication.

In Excerpt 4, the GP introduces the topic of treating the blood pressure, as a means to prevent stroke and dementia, and the patient responds with talk about physical exercise.

Excerpt 4: A06

01 D:and the treatment for that is really

02what we’re doing here that is to

03*treat your blood pressure then*

04 P:mm (.) and exercise and so on

The patient accounts for her physical activity and that it makes her feel well. She spontaneously evaluates the medication, saying that she thinks it is working well. Here she displays an interpretation of the situation as an evaluation of the blood pressure treatment in which she includes both medication and exercise. When she describes her present social situation with more spare time, the GP returns to the topic of physical exercise, displaying an acceptance of the suggestion from the patient that the treatment they are following up is the combination of drug treatment, and physical exercise.

185 D:an’ then perhaps one has some more time

186to get in order with the exercise and so on

187 P:yes exactly so °all I want is to

188continue with it now so I’ll see°

189 D:mm

190 P:or continue I’d rather want

191that it would be possible to taper it down

192if one doesn’t need (.)

193but perhaps one doesn’t do (.) that at all

The patient responds with a proposal to stop or at least lower the dose of the medication. It comes before the blood pressure has been measured and anticipates a proposal from the GP to continue or increase the medication. After assessing the blood pressure, the GP proposes to continue with medication without changes, which the patient accepts.

Disclosure of non-adherence

Disclosures from patients that they do not take the medication as prescribed came in response to proposals to continue without change, and in situations where treatments were evaluated. In return visits, like in Excerpt 4, patients often displayed an understanding of the situation as an evaluation of on-going treatment and anticipated a forthcoming proposal for continued medication.

In the beginning of encounter C01 the patient asks to have his cholesterol value measured and says that he is out of tablets. The GP responds with an offer to issue a prescription and asks if there are any problems with the medication. The patient answers that he doesn’t have any side effects, that he used to have problems remembering the tablets, but has made it a habit now, before revealing that he now deliberately has stopped taking them before this appointment. He displays an understanding of cholesterol lowering treatment as possibly temporary with long-lasting effects that he requests a test to evaluate.

Excerpt 5: C01

01 P:but now it’s been over a week

02since I’ve taken any ya know

03 D:aha that was a bit of a pity perhaps

04it’s difficult to take the test then I think

05 P:mm

06 D:perhaps you’d better start over

07an’ come [back

08 P:[I don’t have any

09 D:no I think no but I’ll write you a prescription

The initial testing of waters from the patient about being out of tablets and wanting to test his cholesterol to see if he could do without medication, was not taken up by the GP. When he again says that he is out of tablets, and has not taken any for over a week, the GP makes the connection between the request to have a test and the disclosure about not taking the medication. He turns the request down with a proposal for continued treatment, and a test as an evaluation of it, which the patient accepts. The agreement here is to postpone the definitive decision.

Discussion

Principal findings

We studied decision-making in 28 consultations where GPs proposed cardiovascular preventive drug treatments and found patient resistance to treatment proposals in the forms of withheld response, minimal acknowledgement, questions, contest of ambiguous address, counterproposals, expressions of identity as non-drugtaker and revelations of non-adherence. Through their resistance patients exerted agency with influence on decisions, and in the case of contested inclusive “we” also influence on who became the decision-maker.

Resistance was found in response to treatment proposals. Some forms of resistance also occurred before proposals from the GPs. Such anticipatory resistance was revelations of non-adherence, counterproposals, questions, and the expression of identity as non-drugtaker.

One modification of the treatment proposal from the GP in response to patient resistance was to postpone the prescribing decision.

Strengths and weaknesses

The dataset is large, and rich in relevant content. Yet, our list of different modes of patient resistance should not be regarded as definite or completely exhaustive. There may be others, that are either too infrequent to be found in our data, or that only exist in other settings.

Data were collected 6–8 years ago, but we consider it still useful for the purpose of the study. An approach based on an assessment of cardiovascular risk including recommendations to use risk assessment tools, is in line with the national guidelines at the time, as well as the current European guidelines for primary prevention of cardiovascular disease [1,2].

The study includes both primary and secondary prevention. It is not always legible which from data. We chose not to try to select pure primary prevention for this analysis. It makes the results less specific and transferable to either primary or secondary prevention. On the other hand, this is “the messy clinical reality”, in which patients and GP’s meet, and where demarkations between primary and secondary prevention are not always clear-cut.

GPs that opted to participate were experienced. They were probably more confident about their consultation skills than the average GP, although the opposite is a theoretical driver for participation, in a quest to learn. Their presumed experience and good communication skills are not necessarily drawbacks, but may serve to enhance and elicit patient agency.

The GPs enrolled the patients. They chose when to participate, with knowledge in advance about which patients were booked. We have no information about patients who declined participation, nor if there were eligible patients not asked by the GPs. Thus, there is a possible selection of patients, whom the GPs felt comfortable to communicate with. Encounters expected by the GPs to be challenging, and therefore possibly of interest regarding treatment negotiations, may be underrepresented.

It can be argued that important information was missed when only audio and not video was recorded. The participants’ orientation towards the electronic health record was audible for example through keyboard clicking, and videouptake may have revealed non-verbal interactional activity during pauses. The recordings were of good quality, and we found the data sufficient for the present analysis. An advantage with audiorecordings is that they inflict minimally on clinical encounters, which facilitated the collection of this relatively large material.

The Swedish primary care setting may limit international transferability of the results due to the local health care culture, such as longer visits than in most other countries, and certain aspects of the Swedish language. Considering the consistency of our findings with previous studies, local traditions don’t seem to have a fundamental impact on the communicative patterns under study. The frequent use of “one” in Swedish, and its importance in the allocation of responsibility and agency in decision-making may not be transferable to languages where pronouns are used differently [20]. On the other hand, although a typical Swedish communicative practice, it accomplishes renouncement of responsibility and agency in decision-making, which is a general phenomenon [5,21].

Discussion of results

We focussed on patient resistance in treatment discussions. It should not be read as our understanding that doctors in general try to persuade patients to take medication, and patients in general ideally should resist it.

Within the analyses, we present an overview of the responses from the GPs to resistance from patients towards treatment proposals. For the purpose of the present study, these responses served as an analytical tool to find sequences that the participants treated as resistance. Further exploration with an analytical focus on physicians’ (re-)actions such as in [19] may render interesting results but is beyond the scope of this paper.

Patients’ communicative resources are likely available as potential tools to exert agency also when they accept treatment proposals. Evaluations merely focusing on physicians’ skills to involve patients, may overlook patients’ potential and actual exertion of agency, and erroneously interpret decisions as not being shared.

Patients’ withholdment of responses, minimal acknowledgements and weak agreements (passive resistance), counterproposals and asking questions (active resistance) following treatment proposals from physicians has been described [8–10]. These findings were based on American data from pediatrics, oncology and general internal medicine, and mainly but not solely from out patient clinics. The presence of the same mechanisms in our data confirms the universality of these forms of resistance as communicative resources for patients to exert agency in treatment decision-making. Cardiovascular preventive drug prescribing does not seem to be an exception in this regard. Neither does it emerge as a particular type of treatment decision-making, although it is depicted so in guidelines for cardiovascular prevention with the advocacy of systematic communication of risk The call for shared decisions based on risk algoritms is stated both in current guidelines, and those applicable during data collection for this study [1,2].

Passive resistance, such as withheld response, appears as a forceful interactional tool in comparison with the active modes with regard to the GPs’ responses. This underlines the importance of recognition and attentiveness to the paradoxically more subtle active resistance in clinical practice and education.

To ask questions is to claim power. It calls for a particular response, restricts the topic and requests information [17]. When posed in response to a treatment proposal, questions constitute resistance, and are a way for patients to exert agency. Patients’ disclosures of “misdeeds” such as changing dose or not taking medication also show their agency in treatment negotiations. “Misdeeds” have been described to be a way to initiate treatment negotiations [18]. We argue that the mechanism at play here is resistance.

The presentation of self is a basic social activity [15]. Patients’ presentation of self as being someone who does not take medication implies that starting with medication would alter identity. In fact, GP consultations may support patients’ change in self-preception necessary to accept disease, and in the extension, to accept long-term medication [22]. From the patient perspective, starting with long-term medication seems to have a strong influence on self-perception. Our data confirm that patients express this in treatment discussions. GPs need to be aware of the effects on patients’ self-perception from the initiation of long-term medication, and responsive to patients who express concern about these effects.

The formulation of treatment proposals often follows the pattern: “I think we… ” [16,23]. This has been suggested to constitute “partnership talk” aiming to obtain consent for the proposal rather than genuine invitations for participation in decision-making [23]. We found contests of an ambiguous address, the tentative inclusive “we”, in treatment proposals. Apparently, patients not only have communicative resources to take on agency in decision-making but also to refrain from it, and the responsibility of being a decisional agent. Switching from “we” to “one” may serve to renounce responsibility in treatment negotiations. It is an example of how patients negotiate who decides, in addition to negotiating the decision.

The impersonal “one” has been described to be used by Swedish phycisians in an ambiguous way, with the potential to mean health care provider/-s including or excluding the patient [20]. GP’s use of one, referring to the medical expert community, is a variant of “health care providers in general excluding the patient” which seems to distance the GP from personal responsibility. In the literature about patient power, the notion that patients and physicians compete to claim power is prevailing. In recommendations for physicians to practice shared decision-making, maximal patient participation is often portrayed as an ideal. Our results, in accordance with recent conversation analytical work, suggest that patients do not always want to decide, or take responsibility for decisions [5,20,24]. This makes GPs’ deferral of responsibility and decisional authority problematic. Although it seemingly allocates power from GPs to patients, it may not empower patients. On the contrary, it may pose a threat to the asymmetry in patient–doctor interaction that “lies at the heart of the medical enterprise: it is founded in what doctors are there for” [25].

Unlike the passive forms of resistance – to withhold response and give weak acknowledgement – active resistance was found both in response to treatment proposals, but also preceeding them. The latter may be regarded as anticipatory resistance, in analogy with the anticipatory answers to life style questions in check-ups for chronic conditions in primary care. It may serve to seize power over the problem definition, and subsequently influence the solutions, thereby constituting patient agency in decision-making [26].

Anticipatory resistance and postponed decisions indicate that the time frame for decision-making extends both the treatment negotiation phase in clinical encounters, and beyond the individual encounter. This has been discussed as problematic in studies of decision-making [27,28]. It is important to be aware of the extended nature of treatment decision-making in research as well as education in communication skills. The extension of decision-making beyond individual encounters underpins the importance of continuity of care for sound decision-making about treatments.

Conclusions

Active and passive resistance to treatment proposals from GPs displays one mode of how patient agency, and the subsequent sharing of decision-making, can be accomplished in clinical encounters.

In clinical practice, education and research about patient–doctor communication about cardiovascular preventive medication we propose particular attentiveness to patient agency through anticipatory resistance, patients’ disclosures of non-adherence and presentations of themselves as non-drugtakers.

The decision-making process about cardiovascular preventive treatment expands beyond single encounters. This decision-making in on-going conversations between patients and GPs underlines the importance of continuity of care, particularly for the use of long-term treatments such as cardiovascular preventive medication.

The sharing of decisions includes a negotiation of who is the decision-maker. Sometimes GPs and patients mutually defer responsibility for decisions through their use of pronouns. Further exploration of how renouncement of agency in decision-making is accomplished may provide clinically useful knowledge.

Ethical approval

Ethical approval was obtained from the Regional Ethical Review Board of Linköping. Reference number M77-08.

Funding Statement

This work was supported by ALF Grants, Region Östergötland, the Research Board of Local Care Eastern Östergötland, the Research and Development Unit of the County Council of Östergötland and the Research Council of South Eastern Sweden (FORSS).

Disclosure statement

The authors declare no conflicts of interest.