Abstract
Objective
To investigate patients' agendas before consultation and to assess which aspects of agendas are voiced in the consultation and the effects of unvoiced agendas on outcomes.
Design
Qualitative study.
Setting
20 general practices in south east England and the West Midlands.
Participants
35 patients consulting 20 general practitioners in appointment and emergency surgeries.
Results
Patients' agendas are complex and multifarious. Only four of 35 patients voiced all their agendas in consultation. Agenda items most commonly voiced were symptoms and requests for diagnoses and prescriptions. The most common unvoiced agenda items were: worries about possible diagnosis and what the future holds; patients' ideas about what is wrong; side effects; not wanting a prescription; and information relating to social context. Agenda items that were not raised in the consultation often led to specific problem outcomes (for example, major misunderstandings), unwanted prescriptions, non-use of prescriptions, and non-adherence to treatment. In all of the 14 consultations with problem outcomes at least one of the problems was related to an unvoiced agenda item.
Conclusion
Patients have many needs and when these are not voiced they can not be addressed. Some of the poor outcomes in the case studies were related to unvoiced agenda items. This suggests that when patients and their needs are more fully articulated in the consultation better health care may be effected. Steps should be taken in both daily clinical practice and research to encourage the voicing of patients' agendas.
Introduction
Research into communication in general practice has focused on either the consultation or interviews with doctors and patients.1–4 Researching the consultation in isolation tends to neglect those aspects of communication that remain unspoken. Conducting interviews in isolation focuses too much on participants' generalised views at the expense of their specific communication behaviours in the medical interaction. To determine what is unspoken in the consultation requires both doctors and patients to be interviewed outside the consultation and a recording of the interaction to be made. We examined the absent discourse that emerges when this approach is taken and its effect on outcomes.
In developing their model of patient centred medicine Levenstein et al introduced the concept of agendas as the key to understanding patients.5 They found that doctors failed to elicit 54% of patients' reasons for consulting and 45% of their worries.6 Campion et al showed that social and emotional agendas are the most likely issues to be underrepresented in the consultation.7 The concept of patients' total agendas is preferable to the narrower and yet more difficult to define concept of patients' expectations. It includes all the reasons for encounter and encompasses patients' ideas, concerns, and expectations. Expectations include specific behaviour that patients would like to occur in the consultation and more general aspects concerning the relationship and interaction with the doctor.4
What doctors both believe and do influences the expression of patients' agendas. Doctors may overestimate the extent to which patients are primarily concerned with medical treatment rather than with gaining information and support. Unless patients are overtly distressed doctors may have trouble in recognising those who are seeking support.8
Participants and methods
We describe the first phase of a two part study, “improving doctor-patient communication about drugs.” We aimed to describe current communication practice among general practitioners through a qualitative approach (phase 1) and from this to develop and test an educational intervention to improve communication about drugs (phase 2). We conducted phase 1 in 20 practices in the West Midlands and south east England. Ethical approval was obtained from 11 local ethics committees. The methods have been reported in detail elsewhere.9
Sampling
To represent a diversity of doctors' sex, practice size, location (urban, suburban, rural), and fundholding status, we purposively sampled 20 of 101 (16%) general practitioners who responded positively to a letter outlining the research.10 The letter was sent to 645 general practitioners in 11 health authorities across the West Midlands and south east England.
We recruited patients over the age of 18, or the parents of patients under 18, from the participating practices in one of two ways. In 13 practices receptionists recruited 44 patients when they booked appointments, and the researcher contacted those who agreed. As few patients met our initial recruitment criteria of consulting for a new problem for which a prescribing decision was likely, we also recruited patients who wanted to discuss a previously prescribed drug. As patients with acute problems were underrepresented, in the final seven practices we recruited 18 patients attending emergency surgeries.
Data collection
We interviewed patients with appointments in their home before attending the doctor, and we interviewed those without appointments in the surgery before they consulted the doctor. The interviews were conducted by a psychologist (CAB) and a sociologist (FAS). To conceal the identity of the study patients from the doctors, we audiotaped the consultations of all patients attending the selected surgeries, who were agreeable.
The doctor was interviewed the next day, and study patients were interviewed for a second time in their homes a week later to investigate subsequent medicine taking and other outcomes. Ten doctors attended one of two feedback sessions to respond to the findings. Summaries were also sent to the participating patients. Interviews and consultations were recorded and transcribed. The resulting dataset comprises a set of case studies of linked data.
Analysis
All five authors, representing four disciplines (general practice, pharmacy, psychology, and sociology), were involved in the analysis.11 CAB and FAS conducted a preliminary analysis of patient's agendas with NUDIST software. The other three authors acted as second coders for 10% of patients. Given the volume of data, a subset of 35 patients was chosen for detailed analysis from the 62 complete cases. These patients were selected to represent all 20 doctors and a range of patient characteristics. The analysis for this paper was conducted by CAB and second coded by FAS.
Patients' agendas were determined from the interview data. The transcripts from the consultation were used to determine voiced agenda items.
The short term outcomes of these consultations were analysed and compared with unvoiced agendas for each patient to look for possible links. The outcomes were: whether patients achieved wanted and unwanted actions, including prescriptions, examinations, tests and referrals, and information and reassurance; whether major misunderstandings were present or absent12; whether prescriptions were presented to a pharmacist, adherence was self reported, and problems with drugs were reported; and whether patient and doctor satisfaction was expressed in the semistructured interviews.
CAB and FAS coded all cases according to these outcomes and independently produced a composite rating on a scale of 1 (good outcome) to 4 (problem outcome). When the ratings did not agree discussions were held to reach a consensus between the two raters.
Results
Patients' agendas
We have treated as patients' agendas their ideas, concerns, and expectations according to their response to interview questions (box B1). Agendas were classed as symptoms, diagnosis theories, illness fears, wanted and unwanted actions, self treatment, and emotional and social issues.
Other research focusing on the consultation alone or on patients answering structured questionnaires in the waiting room defines agenda more narrowly.13 For example, one study included complex and psychosocial agendas in the categorisation of 210 consultations, and nearly half of the consultations were categorised as straightforward (44%).14 Fifteen of our 35 consultations (42%) would have fulfilled their criteria as straightforward. In the light of further data from the interviews, however, only three of the patients' agendas would now be categorised as straightforward. The other 12 patients had complex agenda items such as social issues or not wanting a prescription, but these aspects did not get voiced in the consultation.
Of the 35 patients in this study no one had only one agenda item and most had five or more. To illustrate the diversity of patients' agendas, we present two case studies, one concerning a focused agenda and one a more complex agenda. The first patient had many unvoiced agenda items whereas most of the agenda items were voiced by the second patient. These cases are broadly typical of other patients in the sample.
Case studies
A focused patient agenda, mostly unvoiced
Victoria Morton was worried about her three year old daughter Charlotte's cough, which had not responded to three weeks of homoeopathic treatment prescribed by her private homoeopath. Dr Parker diagnosed an ear infection. Despite the commonplace nature of Charlotte's problem there were nine items on her mother's agenda, of which only three were voiced in the consultation (box B2).
A complex patient agenda, mostly voiced
Tony Byron, a 42 year old lorry driver, had stomach problems and more broad stress related problems and health worries (box B3). Although Tony's agenda was mostly voiced there was one problematic unvoiced item: he did not want to be prescribed antidepressants.
Doctors' response to initial data
These detailed agendas produced quite strong reactions in doctors. For example, during the second feedback session doctor number 19 said:
“I got so depressed when you described this man whose list of expectations went on to about 18 (sic) points . . . There's no way that I am ever going to be able to address even three of these, let alone 18 expectations.”
At a later academic presentation a non-participant doctor labelled Tony as a “heartsink” patient. These emotional responses suggest two interpretations: a mismatch between patients' actual agendas and doctors' views of them and the possibility that doctors prefer patients with simpler agendas. The doctors in another study reported greater satisfaction with consultations involving simple agendas.14
Agenda items voiced in, or left out of, consultations
Only four of the 35 patients voiced their full agendas. These items tended to represent biomedical issues, mainly symptoms (table 1). Only two patients did not have symptoms to report and of those 33 who did, 24 managed to relay all their symptoms to the doctor. Eight, however, only managed to impart some of their symptoms and one did not mention his symptoms at all. Unvoiced agendas tended to represent psychosocial issues and reflected patient's autonomy.
Table 1.
Base | Agenda items
|
No of patients with item | |
---|---|---|---|
Voiced items | Unvoiced items | ||
Symptoms | 32 | 9 | 33 |
Prescription request | 17 | 9 | 22 |
Previous self treatment | 13 | 6 | 18 |
Request for diagnosis | 13 | 4 | 16 |
Theories about diagnosis | 12 | 12 | 21 |
Reporting of, or discussion about, side effects | 11 | 8 | 15 |
Worries about diagnosis or prognosis | 11 | 14 | 20 |
Not wanting a prescription | 3 | 6 | 9 |
Social context | 3 | 5 | 8 |
Total | 115 | 73 | 35 |
Effects of unvoiced agenda items
It is recognised that the consultation is a dynamic process and that in theory something important to a patient beforehand may seem less so as the consultation proceeds. This may explain why some patients' agendas were unvoiced. Agenda items that were not raised, however, often seemed to be associated with specific problem outcomes, such as major misunderstandings.12
Many of the unvoiced agenda items in our study that caused problem outcomes were related to treatment: not revealing that a prescription was not wanted and not reporting side effect problems with drugs or self treatment before consultation. The well known and sizeable problems of adherence15 may well be avoidable if such issues are aired in the consultation.
In all of the 14 consultations with problem outcomes, at least one of the problems was associated with an unvoiced agenda item. In addition to the two case studies discussed, table 2 gives two further examples. The patients may not have thought it a problem not voicing all of their agenda. We did not directly ask them this and neither did they mention it spontaneously. Irrespective of whether or not the patients thought it was important to voice their agendas, there were often ensuing problems.
Table 2.
Case details | Unvoiced agenda item | Related problem outcome |
---|---|---|
Doctor No 2, patient No 7 | Patient has underlying worry that menstrual problems might be recurrence of cancer (in remission) | Doctor doesn't pick up on patient's worries about this problem being cancer related, so therefore cannot allay these fears |
Friend and family think problems might be related to “the change” (patient is 40 years old) | Doctor doesn't think she is in menopause yet so this possibility is not explored, even though patient makes oblique reference to mother's and sister's menopause. Patient is left not knowing whether she is experiencing menopausal symptoms | |
How does doctor know what tablets to give if he hasn't tested her blood for hormone imbalance? | Doctor thinks she is happy about treatment but patient doesn't use prescription and asks her cancer specialist for second opinion. She doesn't trust doctor's diagnosis in absence of blood test | |
Confusion about whether prescription is hormone replacement therapy or hormone drugs | Patient is left not trusting doctor and waits to ask her cancer specialist for advice. He advises against hormone treatment | |
Does not want hysterectomy | Doctor thinks drugs have 50% chance of working and if not she will have to have surgery but doesn't tell her this | |
Doctor No 9, patient No 28 | Patient is worried about nose surgery (broken in fight leading to breathing problems) as friends say they have never been right since similar operations. Also worried about losing sense of smell | Patient does not get information about either his nose or risks and benefits of surgery |
Patient mentions cold but is not troubled by it; he is worried about long term breathing problems | Doctor thinks patient is there to get antibiotics for his cold and patient gets two unwanted prescriptions: antibiotics and linctus | |
Wants to know if antibiotics have side effects and whether he can drink with them | If doctor had warned him about side effects he wouldn't have used prescription for antibiotics. Patient doesn't finish course of antibiotics when he gets side effects |
Victoria, Charlotte's mother, had several problem outcomes related to her five unvoiced agenda items (see box B2). Dr Parker assumed that Victoria had come for antibiotics and prescribed amoxycillin. In the interview afterwards he reported that he had realised from Victoria's body language that he had misunderstood her. On rethinking the consultation he realised that he need not have prescribed. Victoria, however, presented the prescription straight away and gave Charlotte a whole course of unnecessary antibiotics, believing that if antibiotics were prescribed her daughter must have a serious infection. Had Victoria's unspoken agendas been voiced the doctor would have been much clearer about her reasons for attending. He would have known of her use of homoeopathy and her antipathy to antibiotics. This would have helped him to avoid the unnecessary prescription and put her mind at rest about complications.
In the second case, Tony revealed most of his agenda owing to Dr O'Neill's patient centred behaviour, but he did not voice his strong antipathy towards antidepressants (see box B3). He was worried about drowsiness because he was a lorry driver and about having to explain to his family he was on antidepressants. His preference was for non-drug solutions: counselling and making life changes. Dr O'Neill was not party to this information and as well as endorsing Tony's suggestion of using the gym and referring him for counselling, he gave Tony a prescription for sleeping tablets. Tony mistook them for antidepressants, did not present the prescription, and was feeling awkward about how to admit this to the doctor at his next appointment. This may have been a factor in his not returning to the doctor as requested.
Discussion
The pattern of the main voiced and unvoiced agendas reveals systematic differences between how patients present in consultations with how they present in research interviews. In consultations patients seem only partially present, with only limited autonomy—that is, to make requests but not to suggest solutions. Outside consultations patients are more fully present: as socially and contextually situated, thinking, feeling people, with their own ideas on their medical condition and opinions and possible criticisms of medical treatments. Applying Habermas' work on the sociology of communication to the medical context, Mishler described the two presences as two voices: the voice of medicine, in which the consultation is conducted, and the voice of the lifeworlds (reports of contextually grounded experience of events and problems expressed in everyday language), which is largely left outside the consultation.16 This suggests that in the consultation the patient is most commonly construed as a purely “biomedical” entity—that is, a person with disconnected bodily symptoms, wanting a label for what is wrong and a prescription to put it right. Even under this guise the patient still sometimes fails to report their full biomedical agenda. Not all symptoms were reported (nine patients) and not all desires for a prescription were voiced (nine).
Lazarus showed that although patients' interactions with their doctors coincided with their versions of the biomedical model, they did not coincide with their expectations of health care and how it should be delivered. Maybe patients are behaving as they believe they are expected to rather than as they would like.4,17
Some of the poor outcomes in our study were associated with unvoiced agenda items. Patients have many needs and when these needs are not voiced they can not be addressed. When patients and their needs are more fully present in the consultation better healthcare can be conducted. Some of the work in patient centred medicine supports this.6
A more complete view of the patient's agenda was only possible through a methodology that asked patients to present their full selves. When research methods are structured closer to the lifeworld—qualitative, loosely structured, open ended, people centred—a fuller more complex situated view of people and their agendas is gained. Can lessons for consultation behaviours be learned from these research methods to assist both doctors and patients to encourage the patient to be more fully present?
There are some indications that neither doctors nor patients are open to the presentation of fuller agendas, the doctors perhaps lacking confidence to deal with complex agendas and seeing them as overly time consuming, the patients worried about what is deemed appropriate to communicate and about wasting doctor's time. Yet this partial voicing and facilitating of agendas can produce less effective consultations. Even apparently simple presentations, for example a child's chesty cough, can mask more complex agendas. When left unvoiced these can affect outcomes. Our research suggests that some doctors can facilitate patients to reveal fuller agendas, as shown in the case study of Tony Byron.
Both doctors and patients need to change their behaviour to improve outcomes. We believe that by changing doctors' views and behaviours, patients can also be facilitated to change.
What is already known on this topic
Most research on patients' agendas has focused either on the consultation or on interviews with either doctors or patients
Such studies have shown that doctors fail to elicit all of patients' reasons for attending and that emotional and social agendas are likely to be underrepresented in the consultation
Direct comparisons between patients' agendas outside and inside the consultation have rarely been conducted, and previous research has mainly categorised agendas into broad quantitative categories such as “social”
What this study adds
The case study approach allows a more detailed look at what patients' agendas comprise, which can relate specific unvoiced agendas to problem outcomes
Interviews with patients and doctors and transcripts of consultations showed the complexity of patients' agendas and that more of the agendas are unvoiced than was thought
There is a pattern to what is not said and there may be implications for outcomes of consultations in general practice
Acknowledgments
We thank the Department of Health for funding the research and the participating receptionists, doctors, and patients.
Editorial by Charles
Footnotes
Funding: The study on which this paper is based is funded by the Department of Health as part of the prescribing research initiative. The views expressed in this paper are those of the authors and not the Department of Health. FAS is supported by Sir Siegmund Warburg's voluntary settlement.
Competing interests: None declared.
References
- 1.Frankel R, Beckman H. Evaluating the patient's primary problems. In: Stewart M, Roter D, editors. Communicating with medical patients. 1989. [Google Scholar]
- 2.Henbest RJ, Stewart M. Patient-centredness in the consultation, 1: method for measurement. Fam Pract. 1989;6:249–254. doi: 10.1093/fampra/6.4.249. [DOI] [PubMed] [Google Scholar]
- 3.Bradley CP. Uncomfortable prescribing decisions: a critical incident study. BMJ. 1992;304:294–296. doi: 10.1136/bmj.304.6822.294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Stimson G, Webb B. Going to see the doctor. The consultation process in general practice. London: Routledge Paul; 1975. [Google Scholar]
- 5.Levenstein JH, MeCracken EC, McWhinney IR, Stewart M, Brown JB. The patient-centred clinical method, 1: a model for the doctor-patient interaction in family medicine. Fam Pract. 1986;3:24–30. doi: 10.1093/fampra/3.1.24. [DOI] [PubMed] [Google Scholar]
- 6.Stewart M, McWhinney IR, Buck C. The doctor-patient relationship and its effect on outcome. J R Coll Gen Pract. 1979;29:77–82. [PMC free article] [PubMed] [Google Scholar]
- 7.Campion PD, Butler NM, Cox AD. Principle agendas of doctors and patients in general practice consultations. Fam Pract. 1992;9:181–190. doi: 10.1093/fampra/9.2.181. [DOI] [PubMed] [Google Scholar]
- 8.Salmon P, Sharma N, Valori R, Bellenger N. Patients' intentions in primary care: relationship to physical and psychological symptoms, and their perception by general practitioners. Soc Sci Med. 1994;38:585–592. doi: 10.1016/0277-9536(94)90255-0. [DOI] [PubMed] [Google Scholar]
- 9.Stevenson FA, Barry CA, Britten N, Barber N, Bradley CP. Doctor-patient communication about drugs: the evidence for shared decision-making. Soc Sci Med. 2000;50:829–840. doi: 10.1016/s0277-9536(99)00376-7. [DOI] [PubMed] [Google Scholar]
- 10.Mason J. Qualitative researching. London: Sage; 1998. [Google Scholar]
- 11.Barry CA, Britten N, Barber N, Bradley C, Stevenson F. Using reflexivity to optimise teamwork in qualitative research. Qual Health Res. 1999;9:26–44. doi: 10.1177/104973299129121677. [DOI] [PubMed] [Google Scholar]
- 12.Britten N, Stevenson FA, Barry CA, Barber N, Bradley CP. Misunderstanding in prescribing decisions in general practice: a qualitative study. BMJ. 2000;320:484–488. doi: 10.1136/bmj.320.7233.484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Webb S, Lloyd M. Prescribing and referral in general practice: a study of patients' expectations and doctors' actions. Br J Gen Pract. 1994;44:165–169. [PMC free article] [PubMed] [Google Scholar]
- 14.Winefield HR, Murrell TGC, Clifford JV, Farmer EA. The usefulness of distinguishing different types of general practice consultation, or are needed skills always the same? Fam Pract. 1995;12:402–407. doi: 10.1093/fampra/12.4.402. [DOI] [PubMed] [Google Scholar]
- 15.Royal Pharmaceutical Society. From compliance to concordance. Achieving shared goals in medicine taking. London: Royal Pharmaceutical Society of Great Britain; 1997. [Google Scholar]
- 16.Mishler EG. The discourse of medicine. The dialectics of medical interviews. Norwood, NJ: Ablex; 1984. [Google Scholar]
- 17.Lazarus ES. Theoretical considerations for the study of the doctor-patient relationship: implications of a perinatal study. Med Anthropol Q. 1988;1:34–58. [Google Scholar]