Introduction
In the last decade, there has been a shift to a patient-centred approach when evaluating the outcomes of healthcare interventions. This reflects an attempt to understand patients’ perception of clinical effectiveness, which can differ from clinical-based outcomes. A plethora of quality of life questionnaires have been described in the ENT literature; these have been used to assess the value of an intervention from the patients’ perspective and to measure satisfaction.Reference Koenraads, Aarts, van der Veen, Grolman and Stegeman1, Reference Naunheim, Wittenberg and Shrime2 Quality-adjusted life-years questionnaires and stated preference methods have been introduced more recently in an attempt to not only identify perceived improvement after an intervention, but also to quantify the outcome based on patients’ preferences.Reference Naunheim, Wittenberg and Shrime2, Reference Cheraghi-Sohi, Hole, Mead, McDonald, Whalley and Bower3 The latter focuses on both health and non-health outcomes to understand patients’ choices and satisfaction.Reference Naunheim, Rathi, Naunheim, Alkire, Lam and Song4
The majority of research in this field focuses on subjective patient-centred outcomes after the completion of a given clinical intervention.Reference Chen and Boss5 However, management is generally based on the clinician's presumptions of what is wrong with the patient and what their concerns are. These can be surprisingly different from what patients want and are expecting to receive.Reference Kenten, Bowling, Lambert, Howe and Rowe6, Reference Kamper, Haanstra, Simmons, Kay, Ingram and Byrne7 This suggests that patient-centred care models could be improved further by considering patients’ concerns from the outset, rather than solely measuring satisfaction levels post-management.
Understanding patients’ expectations is particularly important in otolaryngology, a specialty where a significant part of the clinical workload is focused on providing quality of life interventions.Reference Koenraads, Aarts, van der Veen, Grolman and Stegeman1 Some attempts have been made to explore this aspect, but the focus has turned more towards clinical factors, such as the seniority of doctors in the out-patient clinic or operating theatre, waiting times, and timing of consent, rather than understanding what clinical questions the patient wants answered from the clinical encounter.Reference Virk, Awad, Singh and Khali8, Reference Awad, Virk and Singh9
Whilst this has not yet been explored in detail in the ENT literature, research from other medical fields has highlighted areas of mismatch between patients’ expectations and current models of care. In the field of geriatrics, for example, disclosure of dementia is not routine in general practice and varies significantly in specialist clinics. Yet a study of patients’ expectations prior to attending a dementia clinic showed that most patients and carers want to be informed of their diagnosis in order to plan for the future, receive treatment and learn coping strategies. More than half of the patients were concerned that they may have signs of early dementia prior to being seen in the clinic.Reference Mastwyk, Dow, Ellis and Ames10
Routinely, ENT clinicians use the general practitioner's referral letter when determining the patient's reason for clinic attendance and to prepare the setting prior to assessing the patient. Nevertheless, studies have shown that general practitioners’ reasons for referral and provisional diagnosis can contradict patients’ expectations.Reference Crosbie, McKendrick, Corson, Lowit and MacKenzie11, Reference Rosendal, Carlsen and Rask12
In order to address the above points, and better understand patients’ thoughts and concerns, we devised an open-structured questionnaire. This led to 14 leading questions that patients wanted answered during a laryngology clinic consultation. These questions were further tested for internal reliability. Removal of non-informative questions produced the eight-item Assessing and Caring for Patients’ Expectations in Laryngology (‘ACaPELa’) questionnaire.Reference Crosbie, McKendrick, Corson, Lowit and MacKenzie11
Since its initial development two years ago, it has become common practice to ask all patients attending our dedicated voice clinic to complete the Assessing and Caring for Patients’ Expectations in Laryngology questionnaire prior to their consultation. This involves the selection and rank ordering, in order of importance, of the questions they want answered. The responses are then reviewed by the clinicians at the beginning of the consultation to guide the discussion with patients, following completion of a focused ENT history-taking and examination.
In our initial publication on the Assessing and Caring for Patients’ Expectations in Laryngology questionnaire,Reference Crosbie, McKendrick, Corson, Lowit and MacKenzie11 we reported on the process of arriving at the eight-item survey from the original open-structured questionnaire. No clinical validation of the questionnaire had been performed at that stage, and there are currently no other reports in the literature on the usability of the questionnaire. However, informal feedback received from our patients indicated that the questionnaire was proving difficult to rank order, with patients finding some of the options very similar in content. We therefore felt that a formal audit of clinical applicability of the questionnaire was warranted. In addition, it seemed appropriate to ask if the patients’ ranked questions had been addressed or not, and so these questions were added to the revised questionnaire.
This study aimed to: (1) audit the clinical applicability of the Assessing and Caring for Patients’ Expectations in Laryngology tool; (2) revise the tool if necessary; (3) introduce an outcome set of questions; and (4) validate the revised version using a new cohort of patients.
Materials and methods
This study involved a two-cycle audit conducted at the ENT Department of the Glasgow Royal Infirmary. In the first cycle, we retrospectively audited patients’ responses to the original eight-question version of the Assessing and Caring for Patients’ Expectations in Laryngology tool, to investigate what kinds of changes might be necessary. All new patients seen in the dedicated laryngology clinic between September 2016 and September 2017 were included in this analysis. All patients had completed the original questionnaire prior to their clinic consultation.
Subsequently, the questionnaire was revised by reducing the question set. The new tool was then validated by prospectively collecting patients’ responses between February and July 2018. During this second cycle, patients were also asked to comment on whether their questions were answered during the consultation, by replying yes or no next to their ranking of the revised Assessing and Caring for Patients’ Expectations in Laryngology survey questions. In order to ensure unbiased responses, this was performed after the patient had left the consulting room. Patients’ answers were checked by the clinic nurse and fed back to the consultant.
All questionnaire replies and satisfaction responses were transferred to an Excel® spreadsheet software file and stored on a National Health Service computer. Caldicott approval had been obtained to run the study as a clinical effectiveness audit. The responses were anonymised; therefore, ethics committee approval was not required.
The chi-square or Fisher's exact test and analysis of variance test were performed for comparison of categorical data, and comparison of continuous with categorical data, respectively. The Friedman test with post-hoc pairwise comparisons was used to identify redundant questions in the Assessing and Caring for Patients’ Expectations in Laryngology survey, to enable the merging and collapse of questions covering broader themes. The level of significance was set at 0.05. Statistical software IBM SPSS® version 20.0 was used for data analysis.
Results
A total of 242 patients completed the questionnaire during the first cycle. The population was female-predominant, with a female-to-male ratio of 1.7:1 (62.4 per cent females). The patients’ mean age was 52.8 years (standard deviation (SD) = 17.8). Patients’ baseline characteristics were comparable with the study population used to develop the Assessing and Caring for Patients’ Expectations in Laryngology questionnaire (Table 1). Thirty-seven patients (15.3 per cent) did not complete the ranking as instructed, giving the same ranking to more than one question. This indicated that there were similarities between questions in the survey. These responses were not included in the subsequent pairwise rank ordering comparison analysis.
Table 1. Patients’ baseline characteristics from the current and previous voice clinic cohort
*Data of this cohort were used to develop the Assessing and Caring for Patients’ Expectations in Laryngology questionnaire. †Number and percentage not recorded. ACaPELa = Assessing and Caring for Patients’ Expectations in Laryngology questionnaire; SD = standard deviation
The questionnaire results are shown in Table 2. Question 1 (‘I want to know what is wrong with my voice’) was most commonly ranked first, followed by question 2 (‘I want to know what is wrong with my throat’) and question 3 (‘I want to know what is wrong/diagnosis’). These, along with question 6 (‘Tell me what I can do to make my throat problem better’), were the most common expectations, with the latter being ranked first less frequently. Despite the fact that only eight patients (3.3 per cent) were referred to the clinic urgently with a suspicion of cancer, 65.3 per cent of patients (n = 158) wanted to know if they had cancer, with 41 (16.9 per cent) ranking this expectation first. There was no correlation between general practitioners’ reasons for referral and patients’ ranked expectations (p > 0.05). Order of ranking was not associated with patients’ age or gender (p > 0.05 for each of the eight questions).
Table 2. ACaPELa patients’ questionnaire responses
Data represent numbers (and percentages). *Thirty-seven patients (15.3 per cent) gave the same ranking to more than one question. ACaPELa = Assessing and Caring for Patients’ Expectations in Laryngology questionnaire
Pairwise comparisons of the ranked questions across our sample identified redundant questions within the original Assessing and Caring for Patients’ Expectations in Laryngology survey. These were the questions with no detectable difference in patients’ rank ordering (post-hoc Friedman test p > 0.05). Question 3 (‘I want to know what is wrong/diagnosis’) was consistently given a similar ranking to questions 1 (‘I want to know what is wrong with my voice’) and 2 (‘I want to know what is wrong with my throat’) (p = 0.637 and p = 0.895 on pairwise comparisons respectively), and hence could be incorporated within the first two questions. The same applied to questions 4 (‘I want to know if I have cancer’) and 5 (‘I want to know if I have something seriously wrong’) (p = 0.615). Similarly, question 8 (‘I want to know what the ENT team can do to make me better’) had a close ranking to questions 6 (‘Tell me what I can do to make my throat problem better’) or 7 (‘Tell me how I can improve my voice/I want my voice to improve’) (p = 0.179 and p = 0.983 respectively). On this basis, questions with non-detectable differences on ranking were collapsed to cover broader themes. This resulted in the Assessing and Caring for Patients’ Expectations in Laryngology – Revised (‘ACaPELa-R’) questionnaire. Table 3 shows the modified question set.
Table 3. Revised ACaPELa (‘ACaPELa-R’) questionnaire
ACaPELa = Assessing and Caring for Patients’ Expectations in Laryngology questionnaire
In order to validate and evaluate the usability of the refined questionnaire, we audited a further 115 patients’ responses to the revised questionnaire. Of these patients, 68.7 per cent were female (n = 79), and the mean age was 50.4 years (SD = 18.8). The results for the revised questionnaire are shown in Table 4.
Table 4. ACaPELa-R patients’ questionnaire responses
Data represent numbers (and percentages). *Five patients (4.4 per cent) gave the same ranking to more than one question. ACaPELa-R = Assessing and Caring for Patients’ Expectations in Laryngology – Revised questionnaire
There was a statistically significant drop in the number of patients who completed the questionnaire incorrectly by giving the same ranking to more than one question, with a decrease from 15.3 per cent (37 out of 242 patients) to 4.4 per cent (5 out of 115 patients) (p = 0.003).
In response to the additional follow-up question regarding whether their queries or concerns had been addressed, 95.7 per cent of patients (n = 110) indicated that this was the case for the full set of questions ranked. Five patients (4.4 per cent) noted in their responses that between one and all of their questions had not been answered satisfactorily. In three of these cases, this result was inevitable. In the first case, no diagnosis could be given to the patient, as further investigations were required. In the second case, the patient felt he had been provided with insufficient advice on how he could improve his voice, as this aspect formed part of his future management with speech and language therapy services, rather than being something which could be provided during the consultation. The third patient had not understood the follow-up question and therefore indicated incorrectly that his questions had not been answered.
This left two patients who were actually dissatisfied with the outcome (1.7 per cent). Both patients had ranked the question about cancer exclusion in fifth place (i.e. lowest ranking priority), but felt that this issue had not been covered sufficiently during the consultation. This suggests that it is important to address all of a patient's concerns, even if they are not indicated as high priority.
Discussion
This study set out to validate a tool developed in our laryngology clinic to capture patients’ expectations and concerns, and in this process to refine the tool and establish its clinical applicability. During this process, we evaluated the tool both as a means to guide the consultation process, and as a measure of patient satisfaction for quality control purposes.
Our results showed that, in its original form, the Assessing and Caring for Patients’ Expectations in Laryngology questionnaire contained too many options, some of which were similar to each other, creating difficulties for patients with the ranking process. By narrowing down the questions from eight to five in the revised version of the questionnaire, we were able to reduce the number of incorrectly completed questionnaires to an acceptable level (4.4 per cent). It is likely that, for some patients, certain aspects might have equal priority, irrespective of the number of choices offered.
We also identified a mismatch between general practitioners’ letters and patients’ expectations in our cohort. When considering, in particular, the mismatch in expectations about excluding malignancy, this possibly reflects the fact that general practitioners base their referral on the current head and neck suspected cancer referral recommendations for hoarseness, rather than patients’ concerns and expectations.13 Irrespective of the reasons for the mismatch, our results highlight the importance of exploring patients’ concerns before the specialist consultation, in order to address these issues effectively. The Assessing and Caring for Patients’ Expectations in Laryngology questionnaire is one way of bridging this gap.
The Assessing and Caring for Patients’ Expectations in Laryngology questionnaire has now been modified to make it easier for completion by the patients and interpretation by the clinician. In our opinion, it is a useful clinical tool. It can be used in laryngology clinics as an additional source of information, alongside the general practitioner's referral letter and the clinical findings, to inform the clinical encounter.
In its latest revised form, the questionnaire allows post-consultation feedback. Patients can now answer ‘yes’ or ‘no’ next to each of their ranked questions, as an assessment of their satisfaction at the end of the clinic consultation. Despite similar questionnaires being used and discussed in the ENT literature,Reference Virk, Awad, Singh and Khali8, Reference Awad, Virk and Singh9, Reference Schmidt, Meyer, Jahnke, Wollenberg and Schmidt14 the proposed feedback function of the Assessing and Caring for Patients’ Expectations in Laryngology – Revised questionnaire has the uniqueness of assessing how well the patients’ own concerns were addressed. Hence, the revised questionnaire can be used to audit consultation performance as part of a clinical self-audit process. It can be also used as an immediate re-consultation and correction mechanism, whereby patients provide post-consultation feedback, and a clinic nurse then checks the answers and informs the clinician of unsatisfactory responses. The clinician can then briefly see the patient again to address the concerns that remain unanswered. This could further improve patients’ overall clinic satisfaction, resulting in a more patient-centred healthcare system.
We acknowledge that this re-consultation and correction process will incur additional clinic time. The process will therefore not be suitable for all ENT clinics or hospitals. This is particularly so when one takes into account the fact that most ENT clinics across the country are currently overbooked to meet demands, despite the ENT-UK recommendations for the maximum number of patients per clinic.15 The negative feedback rate was low in our cohort (4.4 per cent), which allowed us sufficient time to see patients at the end of the consultation to address any remaining unanswered questions. Nevertheless, this might prove challenging if negative feedback rates are high.
Alternative ways of administering the revised questionnaire could be considered, such as the questionnaire being sent out to patients with their appointment letter, or being handed out at the reception desk, with patients returning the questionnaire directly to the clinician at the start of their consultation. The effectiveness of such alternative approaches would have to be evaluated to determine the best means to ensure that relevant information is available during the consultation.
Whilst the wider use of the Assessing and Caring for Patients’ Expectations in Laryngology – Revised questionnaire in other institutions remains to be explored, we believe that the value this input provides to patients’ healthcare outweighs the costs of implementing this process. We suggest that other ENT subspecialties instigate similar processes that help clinicians to appreciate why patients are attending clinics and allow them to be appropriately prepared to meet their expectations. This could be of particular importance in vertigo, tinnitus and rhinology clinics, where, frequently, there can be many, varied and challenging aspects to patients’ symptoms.
• The Assessing and Caring for Patients’ Expectations in Laryngology questionnaire was developed to understand patients’ expectations and guide consultation
• An audit showed that the questionnaire contained options with similar content, making rank ordering difficult
• The questionnaire was revised to encompass broader themes, integrating a post-consultation feedback function
• The revised questionnaire performed well on validation, with a low rate of same-ranked questions and high post-consultation patient satisfaction
• The questionnaire can be used to understand why patients are attending clinics, and allows clinicians to be appropriately prepared to meet patients’ expectations
Conclusion
This paper proposes a novel way of integrating patient expectations into consultations in the ENT clinic setting, resulting in the provision of more patient-focused healthcare. We have revised and evaluated our questionnaire to the degree where it is easy for patients to complete, and provides sufficient information for the clinician to guide their consultation so as to address patients’ individual needs. In addition, the questionnaire can be used as a performance outcome measure for self-audit purposes.
Competing interests
None declared
