Abstract
Soliciting a patient’s agenda (the reason for their visit, concerns and expectations) is fundamental to health care but if not done effectively outcomes can be adversely affected. Forms to help patients consider important issues prior to a consultation have been tested with mixed results. We hypothesized that using an agenda form would impact the extent to which patients felt their doctor discussed the issues that were important to them. Patients were randomized to receive an agenda form to complete whilst waiting or usual care. The primary outcome measure was the proportion of patients agreeing with the statement ‘My doctor discussed the issues that were important to me’ rated on a four-point scale. Secondary outcomes included other experience and satisfaction measures, consultation duration and patient confidence. There was no significant effect of agenda form use on primary or secondary outcomes. Post hoc exploratory analyses suggested possible differential effects for new compared to follow-up patients. There was no overall benefit from the form and a risk of detrimental impact on patient experience for some patients. There is a need for greater understanding of what works for whom in supporting patients to get the most from their consultation.
Introduction
When a patient visits the doctor, he or she usually has a ‘total agenda’ encompassing his or her reason for visiting, ideas, concerns and expectations. Exploring this is viewed by some as a fundamental aspect of health care, garnering vital information and improving patient satisfaction 1 –3 and can reduce the frequency of ‘by-the-way’ syndrome whereby the patient raises a new problem at the closure of a consultation. 4
Soliciting the patient’s agenda need takes little time but is not always done effectively. 1,3,5,6 While symptoms, requests for diagnosis and prescriptions are commonly raised, worries are most often left unvoiced. 1,7 This may cause misunderstandings, unwanted prescriptions and nonadherence. 1,8,9
‘Prompt sheets’ or ‘agenda forms’, to help patients consider important issues and questions before a consultation, have shown mixed results. Outcomes are reported for consultation process (e.g. patient question-asking behaviour), consultation outcomes (e.g. patient health, care and satisfaction) and service outcomes (e.g. consultation length). 10 Reviews report a lack of consistent findings and variable success. In some cases there is increased question-asking behaviour and more discussion of difficult areas like prognosis and pain, but there is inconsistent evidence for satisfaction, anxiety and depression and insufficient evidence for effect on consultation duration. 10 –17 There is no clear benefit of clinician training in such interventions on patient satisfaction or consultation duration 10 nor any reporting of effects of specialist or primary care settings. 10,14,18
To our knowledge there are no studies of the impact of agenda forms in respiratory settings with a high proportion of patients living with chronic conditions such as chronic obstructive pulmonary disease or asthma. Vocalizing one’s agenda is a key self-management skill and good self-management can contribute to outcomes such as fewer acute exacerbations. 19,20 To support patients in this we conducted a non-randomized pilot of an intervention in a respiratory outpatient clinic: a patient agenda form with no clinician preparation. Whilst integrated interventions are recommended for greatest benefit, 21 providing a form just for patients is simple and of low-cost. If it were of some, albeit lesser, value then low implementation cost could still make it cost-effective. Compared to those receiving usual care, patients receiving the form reported higher mean satisfaction and confidence and had shorter consultation times, though these were not statistically significant. We therefore tested the form in a randomized, powered study. We hypothesized that use of the agenda form would impact the extent to which patients felt their doctor discussed the issues that were important to them.
Methods
Development of materials
The agenda form, based on similar forms tested in self-management programmes in the local setting, was developed with clinicians and patients using the cycle plan-do-study-act: plan the change, carry it out, study the results, plan the next iteration. 22 Feedback on usefulness, ease of comprehension and suggested improvements contributed to the final design.
Participants
Approval was sought from 11 consultants in a respiratory outpatient clinic to invite into the trial patients aged over 18 years on designated clinic lists between March 2014 and May 2014. Study information was mailed to patients 1 week before their appointment. Patients were eligible if literate and able to give informed consent.
Design and procedures
Patients were allocated to receive the agenda form or usual care by blocked randomization (block size 6), stratified by consultant.
Sample size was calculated from the pilot study. Seventy-nine patients were required in each group to provide 80% power to detect the difference between a control group proportion of 0.400 and an agenda form group proportion of 0.640 (odds ratio of 2.667), where the proportion represents the patients who strongly agree with the statement ‘My doctor discussed the issues that were important to me’.
Agenda form patients received a sealed envelope containing the form (Figure 1) and an invitation to complete it whilst waiting. There were no instructions to share it with the doctor. Control patients received a sealed envelope containing a brief explanation that their appointment would continue as usual. All envelopes contained a request to complete a feedback form that would be provided post-consultation.
Patient agenda form.
Blinding was not possible because patients may have shared the form with their clinician. For maximum equipoise participating clinicians were not involved in designing the study, nor informed of outcome measures or patient allocation.
Measures
Patient-reported outcome measures were collected on the feedback form after the consultation. The primary outcome was the proportion of patients who agreed with the statement ‘My doctor discussed the issues that were important to me’ rated on a four-point scale (strongly disagree/disagree/agree/strongly agree).
Secondary outcomes were the proportion of patients who agreed with the statements (a) ‘I raised with my doctor the issues that were important to me’ and (b) ‘I got the outcome I wanted from my consultation’ rated on the same four-point scale; patient confidence to manage their condition following the consultation on a 0–10 scale where 0 = no confidence and 10 = completely confident; consultation duration timed by stopwatch by a researcher outside the consultation room. Patients were asked to tick on the feedback form items they had discussed with their doctor and to note other issues discussed. The primary outcome measure and secondary outcome (a) were regarded as patient experience as defined by the NHS Patient Experience Framework. 23 Secondary outcome (b) was defined as patient satisfaction.
Agenda form patients were asked to respond to the statement ‘The agenda form helped me to be clear about what I wanted to discuss with my doctor’ on the same four-point rating scale, whether they completed the form (yes/no) and shared it with the doctor (yes/no).
At the end of the trial clinicians completed a feedback form indicating whether they had been trained in negotiated agenda setting, rating their confidence in negotiated agenda setting on a scale of 0–10 and commenting on their experience of the trial.
Statistical methods
The full analysis population consists of all patients who gave written informed consent, were randomized to one of the two groups and provided outcome data.
Categorical data were analysed using Fisher’s exact test to compare the number of responses for all categories within each group. Continuous data were analysed using a two-group test. Analysis was conducted using software package SAS® version 9.2.
Confidence and consultation duration showed positive signals in the pilot data and were continuous variables, amenable to exploratory analysis using analysis of variance, and so we explored for sub-group effects. We looked for interactions between those variables and the following factors: new or follow-up appointment; description of progress/well-being in consultant’s letter (coded 0–3); discharge (yes/no); age; gender; number of co-morbidities; number of symptoms; agenda form completed (yes/no), shared with the doctor (yes/no), how helpful the form was (four-point scale). Factors that influenced confidence or duration were included in a multivariate analysis in which the interactions between group and the factor were included. Estimates of the least squared means for each group were obtained from the model, where the effects of the additional factors are taken into account.
Ethics
Informed written consent was obtained from all patients. The study was approved by North Wales Research Ethics Committee (Central and East) reference 13/WA/0171.
Results
Entry of patients into the trial is shown in Figure 2. Groups were well matched on baseline characteristics of age, gender, respiratory diagnosis, whether new or follow-up appointment and whether discharged (Table 1).
Consort flow diagram.
Characteristics of study participants at baseline.
Primary and secondary outcomes
There was no significant effect of agenda form use on the primary or secondary end-points (Table 2). A majority of patients agreed or strongly agreed to the primary and secondary end-point statements, although a small number did not respond and one strongly disagreed. There was no demonstrable impact of the form on patient confidence to manage their condition or consultation duration. Mean confidence was 8.7 for both groups. Mean duration was 15.2 minutes for the agenda form group and 15.3 minutes for the control group.
Analysis of primary and secondary outcome measures.
SD: strongly disagree; D: disagree; A: agree; SA: strongly agree.
Seventy-one agenda form patients (85.5%) completed the form and 18 (21.7%) shared it with their doctor. Of the 71, 56 agreed/strongly agreed that it helped them be clear about what they wanted to discuss with their doctor, 11 disagreed, three strongly disagreed and one did not respond. One commented that they always prepared notes before appointments.
Issues most commonly discussed were breathlessness (55%), tablet medication (47.2%) and inhalers (46%), with no difference following use of the form.
Analysis of interactions
There was a statistically significant interaction between patient confidence and appointment type and group; the confidence score for new agenda form patients was significantly lower than for patients in the other three groups. This was even more apparent in the least square mean estimated from the model taking the other factors into account (Table 3). Also, new agenda form patients had shorter consultations than new control group patients, whereas there was little difference between groups for follow-up appointments (Table 4).
Interaction between patient confidence, appointment type and group.
aMedical records were not available for two patients, hence they could not be categorized as new or follow-up.
Interaction between consultation duration, appointment type and group.
aMedical records were not available for two patients, hence they could not be categorized as new or follow-up.
Clinician feedback
Six consultants’ feedback. Confidence to support negotiated agenda setting ranged from 7–9 out of 10 (mean = 8). One consultant had attended training in negotiated agenda setting. One commented that most patients with a form checked it at the end of the consultation to ensure they had covered everything (not ‘agenda setting’ per se). Another commented that it was not possible to spot from patients’ behaviour who had received a form.
Discussion
Main findings
The study did not demonstrate that patients receiving an agenda form immediately before their consultation were more likely to agree that their doctor discussed the issues that were important to them. Agenda form users were not more able to raise important issues with their doctor, nor more likely to get the outcome they wanted or to have an impacted consultation duration. Use of the form did not affect the number or types of topics that patients reported discussing with the clinician. Clinicians’ confidence to support patients to negotiate an agenda was high. Exploratory analysis identified that new agenda form patients had lower confidence to self-manage and shorter consultations.
Interpretation of results
Intervention logic proposed that using the form would prompt patients to raise issues at the start of the consultation, which would improve experience, satisfaction and efficient use of time. The negative findings could have resulted from the form having no impact on patient or clinician behaviour. Most patients found it helpful, however, so there may have been impact not captured by the outcome measures.
Targeting the patient with a simple form is a cheap, low time-intensive option compared to clinician training and service re-design. It could be justified because patients do contribute to communication effectiveness 24 and high participation patients can stimulate physicians to adopt a more patient-centred style. 25,26 Perhaps our study, without a specific role for the clinician, was insufficient to impact on patient behaviour. However, interventions in which the clinician played a minor role have been powerful enough to significantly alter patient behaviour in terms of types of questions asked during the consultation. 18 Our study measured outcomes and so perhaps to alter outcomes the clinician does need to address the prompt sheet. 27 Rao et al. 16 argued that leaving out the doctor is like ‘anticipating an elegant waltz to emerge on the ballroom when only one partner has taken dance lessons’. Only 21% of agenda form patients shared the form with their clinician perhaps limiting its power. Some patients may not have been confident to share it without an explicit instruction to do so.
A generic form may not address individual needs and some tailoring may be more beneficial. For example, a trial of ‘TalkToYourDoc’, a pre-consultation, online personalized question prompt to prepare patients for hormone therapy consultations, found that users were more engaged in discussions with their physician, asked more relevant questions, were more satisfied with the discussion and rated the visit as more productive. 28 Other approaches utilizing technology, such as mobile phone apps, would also have potential for personalization.
The form’s impact may have been limited by being single-use. An iterative approach, with the form as part of a process to build confidence and skills, may have had more impact although evidence suggests that complexity is not essential for success and simply providing practitioners with information about patient needs before the consultation can improve some outcomes. 13 Another consideration is that patients in our study may have had insufficient time to establish their agenda. Brandes et al.17 found no indication that the moment of delivery influenced effectiveness and Kinnersley et al. 10 found increased satisfaction only in interventions immediately before the consultation, so immediacy need not be detrimental though patients would have experienced different waiting times during which to complete the form.
High satisfaction and experience ratings in the control group suggest that there may be little room for improvement in the outcomes measured and this could explain lack of effect. Alternatively, the positive results may be influenced by social desirability; satisfaction ratings with ‘people aspects’ of care may be higher when assessed on site compared to mail. 29 Despite these issues, most agenda form patients agreed that the form helped them be clear about what they wanted to discuss with their doctor, so the study may not have captured the real impact; for example, there is evidence of relationships between agenda forms and recall of information (when combined with provider training), number of questions asked and anxiety at follow-up. 17
Whilst confidence to manage their condition was reasonably high in both groups, new patients using the form reported lower confidence relative to follow-up patients and experienced shorter consultation times than those in the control group. This post hoc analysis is interpreted with caution as the number of new patients was small and consultation durations shorter overall for follow-up patients allowing less scope for difference. However, it highlights the need to consider that some groups may benefit more than others; for example, agenda forms may be useful only for those who find question-asking difficult. 10 New patients may be more anxious overall, 30 but why new agenda form patients might leave feeling less confident than control group patients is unclear. Perhaps new patients attempted an agenda very different to the clinician’s which impacted negatively on the consultation. In any event, high patient confidence to manage one’s condition is not necessarily desirable; in the absence of accompanying knowledge and skills there may be detrimental effects. 31 Shorter consultations may boost service efficiency, but we cannot say if this has a direct impact on patient confidence as our study was not designed to address this question. Reviews of agenda form studies report small effect sizes which may reflect their limited value in a complex communication process, and also the fact that studies included a range of patients and settings in which contextual factors influence the results in differential and unidentified ways.
Implications for practice
Our results do not provide an evidence-based case for routine use of the agenda form, although patients and some clinicians reported that they found it helpful and would like to use it again. This resonates with a debate about tension between evidence-based medicine and individualized clinical autonomy based on a patient-centred approach and clinical experience. 32 Armstrong concludes that it is ‘likely that clinical experience plays a critical role in the process of evaluating “evidence” and incorporating new clinical behaviours for many groups of doctors’.
Strengths and limitations
Strengths of the study include successful randomization and a low attrition rate. Outcomes were recorded for similar numbers in both groups.
High overall ratings for experience and satisfaction may result from selection bias as patients consenting to enter the trial may have been more confident in raising issues. However, we had good uptake from eligible patients. Full blinding was not possible since some patients chose to share their form with their clinician. However, non-production of a form did not necessarily imply allocation to the control group and steps were taken to maintain equipoise as much as possible; participating clinicians were not aware of the outcomes being measured. There may have been potential for contamination by control group patients seeing others using the form and then preparing questions themselves. However researchers were present in the clinic and did not observe this.
The lack of impact on patient-reported outcomes may reflect a ceiling effect. Category measures provided four options which may not have been sensitive enough to detect a difference especially in the case of high pre-existing ratings. A larger range and mid-point ‘unsure’ option would provide greater sensitivity. Furthermore, satisfaction may be limited as an outcome, as patients may be satisfied with less than ideal care. 10 In order that patients could use the agenda form confidentially with their clinician, as would happen in real practice, we did not collect the completed forms. This prevented us from analysing the relationship between content and outcomes which may have helped to understand the results. Also, we did not collect baseline data to compare outcomes over time. An alternative design would be to assess outcomes at baseline and then randomize to agenda form or usual care.
This intervention is naturally limited to a literate population and it was not possible to formally assess literacy in the clinic. Patients’ own confirmation of understanding was taken as a proxy. Patients who could not understand the patient information may have declined to participate or may have consented and then not used the form. Collecting the completed forms would have allowed us to clarify the extent of their actual use.
The data tended to show numerically higher values in the control group than were assumed in the sample size estimate, making an underpowering for effect unlikely for the outcomes measured.
Conclusion
There was no overall benefit from using the agenda form for patient-reported outcomes or consultation duration though the form was regarded as helpful by a majority of patients using it. Exploratory analysis suggested a potential for detrimental impact. These data do not provide an evidence-based case for use of agenda forms within respiratory outpatient clinics. There is a need for greater understanding of what works for whom with regard to supporting patients to get the most from their consultation.
Footnotes
Acknowledgement
We would like to thank Kate Homan for her assistance with the management of the study and with data collection, and Tracy Watts for her assistance with data collection.
Authors’ Contribution
FE: contributed to design and management of the study, data collection, data analysis and writing the manuscript; AJTE: contributed to data collection, data analysis and writing the manuscript; CMO contributed to statistical analysis and writing the manuscript; PLF: revised the manuscript for important intellectual content; JPF: contributed to design of the study, data analysis and writing the manuscript.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This work was supported by the Health Foundation, an independent charity working to improve the quality of health care in the United Kingdom (as part of the Continuing Self-Management Support programme); and by Boehringer Ingelheim (unrestricted educational grant).