Clinicians' communication with patients receiving a MCI diagnosis: The ABIDE project.

BACKGROUND: We aimed to explore clinicians' communication, including the discussion of diagnosis, cause, prognosis and care planning, in routine post-diagnostic testing consultations with patients with Mild Cognitive Impairment (MCI).
METHODS: Thematic content analysis was used to analyze audiotaped consultations in which 10 clinicians (eight neurologists and two geriatricians) from 7 memory clinics, disclosed diagnostic information to 13 MCI patients and their care partners. We assessed clinician-patient communication regarding diagnostic label, cause, prognosis and care planning to identify core findings.
RESULTS: Core findings were: clinicians 1) differed in how they informed about the MCI label; 2) tentatively addressed cause of symptoms; 3) (implicitly) steered against further biomarker testing; 4) rarely informed about the patient's risk of developing dementia; 5) often informed about the expected course of symptoms emphasizing potential symptom stabilization and/or improvement, and; 6) did not engage in a conversation on long-term (care) planning. DISCUSSION: Clinicians' information provision about the underlying cause, prognosis and implications for long-term (care) planning in MCI could be more specific. Since most patients and care partners have a strong need to understand the patient's symptoms, and for information on the prognosis and implications for the future, clinicians' current approach may not match with those needs.

Introduction

In recent decades, improvement in the diagnostic work-up and understanding of the slow development of Alzheimer’s disease (AD) caused a shift to an earlier diagnosis [1]. The diagnostic label of mild cognitive impairment (MCI) has been introduced to describe a syndrome in which cognitive impairment is objectified, but there is no dementia, and independence in functional abilities is intact [2]. The recently published practice guideline by the American Academy of Neurology (AAN) states that an accurate diagnosis of MCI is important for patients and care partners to understand the cause of their complaints, to discuss the prognostic possibilities (i.e. development of symptoms and/or risk of dementia) and to arrange long-term (care) planning accordingly [3].

Nonetheless, the MCI label remains challenging for clinicians. First, due to much heterogeneity, this diagnostic label does not imply one cause. MCI may be the first cognitive expression of AD, but may also be secondary to other neurological or psychiatric disorders [4]. Diagnostic tests, e.g. CSF AD biomarkers, can help to determine the origin of MCI [5]. However, this raises a second challenge; as MCI patients do not (yet) have dementia, the label of MCI entails a prognosis rather than a diagnosis. In general, prognostic information is difficult for clinicians to convey, and for patients and care partners to understand [6, 7]. It may even be more difficult in dementia, as no fixed events define its onset, which makes it particularly hard to identify transition points for individual patients [4].

Recently, recommendations were published on how to deliver AD biomarker results and subsequent prognostic information in MCI patients [8]. They state that a positive result, indicative for underlying AD pathology, should elicit further monitoring and conversations about future (care) planning. In the case of a negative result, they advise to communicate the uncertainty that is inherent to a negative result, i.e. that patients remain at increased risk for dementia. Not all clinicians make use of AD biomarkers however, resulting in practice variation in the diagnostic work-up [9, 10].

Despite aforementioned challenges, accurate communication about the MCI label, underlying cause of symptoms, prognosis and long-term (care) planning are recommended [3]. Currently, empirical observational data on whether and how clinicians deal with these challenges when informing patients and their care partners about these issues in clinical practice are lacking. In this qualitative study, our objective was to explore clinicians’ diagnostic communication, including the discussion of cause, prognosis and care planning, in post-diagnostic testing consultations with MCI patients.

Methods

Design and study context

We used a qualitative approach to analyze audiotaped clinician-patient consultations in which clinicians disclosed the diagnostic label MCI to patients and their accompanying care partners. This study was part of the Alzheimer’s Biomarkers in Daily Practice (ABIDE) project [11]. ABIDE encompasses an observational study of audiotaped consultations during the routine diagnostic work-up at Dutch memory clinics [12]. Written informed consent was obtained from clinicians, patients and care partners. The board of the Medical Ethics Committee of the Amsterdam UMC reviewed and approved this study.

Sample, procedure and questionnaires

All clinicians involved in diagnostic consultations at seven Dutch memory clinics were eligible and invited to participate (self-selection). The newly referred patients of consenting clinicians (in the Netherlands patients are referred to the memory clinic by another medical doctor, i.e. no self-referral) and accompanying care partners were invited to participate. Next, audio-recordings were made of the clinician-patient consultations when patients were seen as part of normal clinical care at the memory clinic. Since our aim was to observe and describe naturalistic clinician-patient communication during the routine diagnostic work up for dementia in a broad range of memory clinics, no efforts were made to intervene or standardize the diagnostic work up (for a detailed description see Visser et al., [12], and the results section for details on diagnostic tests that patients underwent). For the current study we selected fully audio-recorded post-diagnostic testing consultations with MCI patients (N = 13), as based on their retrospectively obtained medical records (see Fig 1). These were patients labeled explicitly with MCI (N = 11), prodromal AD (N = 1) or objective cognitive disorder(s) without meeting the criteria for dementia (N = 1). We also retrieved patient characteristics, MMSE score and which diagnostic tests were performed from medical records. Clinicians (N = 10) involved in these consultations completed a questionnaire assessing their age, gender, specialty, and level of experience.

Fig 1

Flow diagram of patient inclusion.

Note. Of the total sample of patients participating in the ABIDE observational study [12], 21 (15%) were diagnosed with mild cognitive impairment (MCI), based on their retrospectively obtained medical record data. Post-diagnostic testing audio recordings were available from 13/21 MCI patients, who were seen by ten clinicians in seven Dutch memory clinics.

Analysis

We used descriptive statistics to report characteristics of clinicians, patients and care partners. Two of the authors (LNCV and ISvM, with a background in psychology) listened to all consultations twice and transcribed clinician-patient interactions related to communication about the MCI label, cause, prognostic information and future (care) planning. Qualitative content analysis was used to systematically organize the data [13]. First, an initial coding scheme was developed, based on previous research [14, 15], literature/guidelines on MCI [3, 5, 8], and literature on risk communication/provision of prognostic information [7], to deductively categorize the observed communication. The study-specific coding scheme (S1 Appendix) assessed clinician-patient communication regarding:

diagnostic label, i.e., whether clinicians used the term MCI or Mild Cognitive Impairment, and addressed each of three MCI criteria [2, 5]; 1) observed cognitive impairment, 2) no interference with daily living, and 3) not demented;

cause, i.e. whether and which diagnostic test results were communicated, and whether and which (potential) etiology was discussed by the clinician;

prognosis, i.e., whether and how potential symptom development and the risk of developing dementia were discussed;

next steps in (care) planning, i.e., whether and which next steps were addressed by the clinician.

All transcripts were independently coded by LNCV and ISvM. Codings were then compared and discussed until consensus was reached. The final codes combined with the relevant transcribed segments of the audio-recordings were then inductively categorized further by LNCV, ISvM and EMAS, identifying core findings emerging from the data. Selected representative quotes were translated from Dutch to English by the VU university translating office.

Results

Table 1 presents characteristics of the 13 MCI patients. They were all accompanied by a care partner, and seen by eight neurologists and two geriatricians. Mean age of patients was 73±9, 6/13 were female and their mini mental state examination (MMSE) score was 26±4 (M±SD). Mean age of clinicians was 49±9 (M±SD), 4/11 were female, they had 10±8 (M±SD) years of experience working in a memory clinic and saw on average 18±8 (M±SD) new patients per month. Clinicians were recruited from a broad range of memory clinics, located at one academic hospital, five non-academic teaching hospitals and one non-academic, non-teaching hospital. The mean duration of the post-diagnostic testing consultation was 19±8 minutes (M±SD).

Table 1

Sample characteristics.

Patient ID	Age	Gender	MMSE	Duration of consultation (min)	Care partner (Female/Male)	Clinician
1	70	Male	28	17	Spouse/partner (F)	Neurologist 1
2	65	Female	26	17	Spouse/partner (M)	Neurologist 2
3	81	Female	21	19	Spouse/partner (M)	Neurologist 3
4	63	Female	22	16	Son (in law) (M)	Neurologist 4
5	88	Female	17	28	Daughter (in law) (F)	Neurologist 4
6	91	Female	27	14	Daughter (in law) (F)	Geriatrician 1
7	73	Male	MOCA 22/30	16	Spouse/partner (F)	Neurologist 5
8	73	Male	28	14	Spouse/partner (F)	Neurologist 5
9	73	Male	29	9	Spouse/partner (F)	Neurologist 6
10	73	Male	29	16	Spouse/partner (F)	Neurologist 7
11	64	Male	29	38	Spouse/partner (F)	Geriatrician 2
12	79	Female	29	13	Son/Daughter (in law) (F)	Geriatrician 2
13	61	Male	26	27	Spouse/partner (F)	Neurologist 8

Notes. Each row represents a patient and the patient ID represents the same patient across all tables. MMSE = Mini-mental State Examination, MOCA = Montreal Cognitive Assessment.

Table 2 provides an overview of the diagnostic tests that patients underwent and whether individual test results were communicated to be abnormal, normal, or unclear by the clinician. Categorized communication behavior is displayed in Table 3 (concerning the MCI label, cause of symptoms, and prognostic information), and Table 4 (next steps in (care) planning). We allocated a unique ID number to each patient, and included those ID numbers in all tables in such a way that each row represents a patient.

Table 2

Clinicians’ communication concerning results of diagnostic tests that patients underwent.

Patient ID	Neuropsychological assessment	Imaging (CT/MRI)	CSF by lumbar puncture	Amyloid PET	Visual aid used?
1	Abnormal	Unclear	Not performed	Results not yet available	CT/MRI images
2	Unclear	Normal	Abnormal	Not performed	No
3	Abnormal	Abnormal	Not performed	Not performed	No
4	Unclear	Normal	Not performed	Not performed	CT/MRI images
5	Abnormal	Abnormal	Not performed	Not performed	CT/MRI images
6	Not performed	Not performed	Not performed	Not performed	A drawing
7	Abnormal	Unclear	Not performed	Not performed	CT/MRI images
8	Unclear	Normal	Not performed	Not performed	CT/MRI images
9	Unclear	Abnormal	Normal	Not performed	CT/MRI images
10	Abnormal	Normal	Normal	Not performed	No
11	Normal	Abnormal	Abnormal	Not performed	CT/MRI images
12	Unclear	Normal	Not performed	Not performed	CT/MRI images
13	Abnormal	Normal	Not performed	Not performed	CT/MRI images

Notes. This table indicates: i) whether a patient underwent a diagnostic test (if not, then indicated as ‘not performed’), ii) whether test results were communicated to be abnormal, normal, or unclear by the clinician. The ‘unclear’ category was used when clinicians communicated that test results were unclear/ambiguous, or when clinicians’ communication about the test results was unclear. In addition, the last column indicates whether the clinicians used visual aids to support the conversation with the patient and care partner. CSF = Cerebrospinal Fluid, CT = Computer Tomography, MRI = Magnetic Resonance Imaging, PET = Positron Emission Tomography.

Table 3

Clinicians’ communication regarding MCI label, cause, symptoms development, and risk of dementia.

	A) Clinicians' communicationregarding the MCI label				B) Clinicians' communication about the cause of symptoms	C) Clinicians' communication about the expected course of symptoms			D) Clinicians' communication about the risk of developing dementia
Patient ID	MCI term used?	1. Cognitive impairment	2. No interference daily living	3. Not demented	Underlying cause addressed?	Progression addressed?	Improvement addressed?	Stabilization addressed?	Risk addressed?	How? Formats used
1	Yes	Yes	No	No	Yes; possible AD	No	No	No	Yes	Verbal label, percentage, fifty-fifty, with time frame
2	Yes	Yes	No	Yes	Yes; probable AD	Yes	No	Yes	No	NA
3	No	No	No	No	Yes; possible psych distress	No	No	Yes	No	NA
4	Yes	Yes	Yes	Yes	Yes; possible AD	Yes	No	Yes	Yes	Verbal label only
5	No	Yes	Yes	Yes	Yes; aging	No	Yes	No	No	NA
6	Yes	Yes	No	Yes	Yes; possible AD	Yes	No	Yes	Yes	Verbal label, percentage, framing, with time frame
7	Yes	Yes	Yes	Yes	No	Yes	No	Yes	Yes	Verbal label only
8	No	No	No	Yes	Yes; psych distress	Yes	Yes	No	No	NA
9	No	Yes	No	Yes	Yes; potential vascular	Yes	No	Yes	No	NA
10	Yes	Yes	Yes	Yes	Yes; inexplicit	No	No	No	No	NA
11	No	No	No	No	Yes; possible AD	Yes	No	No	No	NA
12	No	Yes	No	No	No	Yes	Yes	Yes	No	NA
13	Yes	Yes	Yes	Yes	Yes; possible ADHD	No	No	No	Yes	Verbal label, fifty-fifty

Notes. In the medical record, the patient with ID 10 was labeled as an individual with ‘objective cognitive disorder(s) without meeting the criteria for dementia’ and the patient with ID 11 as an individual with ‘prodromal AD’. All other patients were labeled in the medical record as individuals with MCI. AD = Alzheimer’s Disease, NA = not applicable, MCI = Mild Cognitive Impairment.

Table 4

Overview of next steps in healthcare and disease management that were addressed by the clinician.

Patient ID	Further diagnostic testing	Medication / Medical treatment	Follow-up at memory clinic	Lifestyle adjustment/ advice	Study/trial participation	Driving ability testing	Follow-up by general practitioner	Referral to other healthcare professionals
1	No	Yes, no medical treatment available	Yes	No	Yes	No	No	No
2	No	Yes, no medical treatment available	Yes	Yes	Yes	No	Yes	Yes, psychologist
3	No	Yes, medication for mood disorder	Yes	Yes	No	No	Yes	Yes, psychiatrist
4	Yes, not pursued	Yes, no medical treatment available	Yes	Yes	No	No	Yes	No
5	No	Yes, adjusting medication for thyroid disorder	No	No	No	No	Yes	Yes, case manager
6	Yes, not pursued	No	Yes	No	No	No	Yes	No
7	No	Yes, no medical treatment available, vitamin B1 supplementation	Yes	Yes	No	No	No	Yes, neuropsychologist
8	No	No	Yes	No	No	No	No	Yes, psychologist
9	No	Yes, treatment for hypertension and hypercholesterolemia	Yes	Yes	No	Yes	No	No
10	No	No	Yes	No	No	No	No	No
11	No	Yes, no medical treatment available	Yes	Yes	Yes	No	No	No
12	No	No	Yes	Yes	No	No	No	No
13	Yes, pursued	Yes, cholinesterase inhibitors	Yes	No	No	No	No	Yes, occupational therapist

Overall, qualitative analysis of the audiotaped post-diagnostic testing consultations revealed variation in clinicians’ communication regarding the MCI label, the underlying cause of symptoms, prognostic information, and (care) planning. Based on further inspection of these results and relevant transcripts we formulated six core findings that comprise the essence of our results.

Clinicians differed in the way they informed about the MCI label

Clinicians used the term MCI in approximately half of the consultations (Table 3 section A, and see also representative quotes 1.1 and 1.2 below). They also differed in whether they addressed the three MCI criteria to explain the patient’s situation, i.e., clinicians often, yet not always, mentioned that the patient showed cognitive impairment and emphasized that the patient was not (yet) demented. In some consultations clinicians mentioned that there was no interference with daily living. The following two quotes illustrate how patients were provided with different amounts of information about the MCI label. In the first quote (1.1), the clinician thoroughly explains the patient’s diagnosis by mentioning MCI and addressing all three MCI criteria.

Quote 1.1 (Patient ID 7): “We do not think you have dementia. If it really were dementia, then you would expect to see more disorders, more tasks that you could not perform well during the tests. And you would also expect it to cause many more problems with daily activities. What it comes down to: Slight memory problems. […] This is called mild cognitive impairment in English, or MCI.”

The second quote (1.2) illustrates how clinicians sometimes explained the MCI label less extensively. Here, the clinician only mentions that the patient shows cognitive impairment, without using the term MCI.

Quote 1.2 (Patient ID 12): “We have more or less said that, and this is not a real diagnosis but a description of the clinical picture, that the test results show a condition that is characterized by slight comprehension problems, apparently …, and some trouble remembering new information. Yes, there is a term used to describe your symptoms, but I won't mention it because you'll just forget and it's not really that important.”

Clinicians often addressed the cause of patients’ symptoms in a tentative manner

In the majority of consultations, we observed communication regarding the underlying cause(s) of the patient’s symptoms, as shown in section B of Table 3. This varied from a short statement by the clinician to a more elaborate conversation. If cause was addressed, the clinician often expressed himself/herself in a tentative manner, i.e., by using words such as possible, probable, potential, and may(be). This was independent of whether or not biomarker testing had been performed and independent of the biomarker result (normal/abnormal). This tendency to keep all options open, is illustrated by the next two quotes (2.1 and 2.2).

Quote 2.1 (Patient ID 2 –abnormal biomarker status): “But now that we see that the proteins in the cerebrospinal fluid are abnormal, we think this might suggest Alzheimer's disease after all.”

Quote 2.2 (Patient ID 10 –normal biomarker status): “The lumbar puncture shows no signs of Alzheimer's disease. […] Does this also mean that we can rule out an underlying disease? No, unfortunately we cannot really say that. The future will sort of reveal this.”

In case of AD as the potential underlying cause of the patient’s symptoms, only one clinician explained the difference between Alzheimer’s disease and dementia as illustrated below (quote 2.3).

Quote 2.3 (Patient ID 6): “I can then prove, for example, that you have Alzheimer's disease, but that this has not yet developed into Alzheimer's dementia. There is a difference. You can have Alzheimer's disease for 20 years before you get Alzheimer’s dementia.”

Clinicians (implicitly) steered against further biomarker testing by an unbalanced presentation and phrasing of arguments

If biomarker testing was not yet performed, clinicians sometimes introduced the possibility of additional/further biomarker testing (Table 4, and see representative quotes below). Clinicians mentioned as the main reason for biomarker testing to demonstrate the presence of AD pathology as the underlying cause of the patient’s symptoms. However, they minimized the potential benefits/impact of biomarker testing by emphasizing that an abnormal biomarker result would not have (major) implications (as illustrated by quote 3.1 below). Of note, that a normal biomarker result would increase the likelihood of another (potentially treatable) cause, such as psychiatric problems, was never presented as an argument for biomarker testing. In addition, clinicians implicitly discouraged further biomarker testing by using diminutives and referring to biomarker tests as ‘tricks’, as illustrated by quote 3.2.

Quote 3.1 (Patient ID 4): “But with further tests we will be able to determine whether or not you have Alzheimer's disease. With a spinal tap, a lumbar puncture. However, the question is whether that is advisable, whether it is of any use to you, whether it would result in treatment. This is not the case. It would be purely for diagnostic purposes as there is no treatment.”

Quote 3.2 (Patient ID 6): “We can play all kinds of tricks, including a PET scan or examining the cerebrospinal fluid. Those can then help me confirm, for example, that you have Alzheimer's disease.”

Few clinicians informed about the patient’s (personalized) risk of developing dementia

Clinicians communicated about the patient’s risk of developing dementia in some consultations, as shown in Table 3 section D. If risk was discussed, this was most often discussed in a verbal format (see quote 4.1 below) without the use of percentages, natural frequencies, reference class or time frame.

Quote 4.1 (Patient ID 4): “Of course, this could be an early warning sign of Alzheimer's disease with a dementia syndrome.”

Still, the following quote (4.2) illustrates how some clinicians explicitly and elaborately explained a probability. Of note, the communicated risk was always in general terms and never personalized, e.g., taking advantage of knowledge on demographic or clinical characteristics, or (ab)normal biomarker result.

Quote 4.2 (Patient ID 6): “Half of these people will remain stable, and the symptoms of about half of them will worsen and progress to dementia [verbal format]. […] Let's say that the likelihood that you develop dementia within 5 years [time frame] is 50% [percentage; negative framing]. But the likelihood that you remain stable is also 50% [percentage; positive framing].”

Clinicians often emphasized potential symptom stabilization or improvement

As displayed in Table 3 section C, clinicians often provided prognostic information by addressing the expected course of patient’s symptoms. If addressed, the likelihood of symptom progression was often counterbalanced by addressing the unpredictability of symptom development, emphasizing that symptoms could also improve or stabilize. The next two quotes illustrate this tendency to foster hope.

Quote 5.1 (Patient ID 9): “Yes, it is uncertain how it will progress, whether it will worsen or remain stable. This is something the tests can’t tell us. […] Anyway, let’s hope your condition remains stable.”

Quote 5.2 (Patient ID 2): “Alzheimer's disease is, of course, a condition that is progressive, so it gets worse. But I can’t predict how quickly this will happen. Your condition could remain stable for a long time and then deteriorate slowly, or it could progress more quickly. I can't predict that. […] Yes, it may well remain stable for quite some time. Yes, that is quite possible.”

Clinicians informed about current next steps, but did not discuss long-term (care) planning

Table 4 shows that all clinicians discussed one or more next steps in care or symptom management, mostly medication. Regarding long-term (care) planning, clinicians only addressed follow-up at the memory clinic to ‘keep an eye on things’, but in none of the consultations clinicians engaged in a conversation on future care or other long-term planning, e.g. concerning legal or financial issues.

Many clinicians provided general advise on how to manage/cope with the current cognitive symptoms, emphasizing what patients and their care partners could do themselves, e.g., concerning life style (quote 6.1 below) or the interaction between patient and care partner (quote 6.2).

Quote 6.1 (Patient ID 7): In general, it has been shown that a healthy lifestyle is really very important for the whole body, including the brain. In terms of food, i.e. alcohol and smoking, and in terms of exercise. Memory training exercises have also been looked into, but there is not a specific test or exercise that helps.”

Quote 6.2 (Patient ID 8): “In general, I think it is important that the memory problems do not lead to too much irritation and frustration. But that is easier said than done, of course. So you must really do your best if you [patient] think that she is saying something important; try to really concentrate and remember it. And for you [partner], make sure that you choose the right moment to say something, just the two of you in a quiet room, with no TV or radio on, and clearly say: I really want you to remember what I'm telling you now. And repeat what you have to say a few times. These things actually always help.”

In addition, some clinicians informed on how to treat or cope with current psychological/psychiatric problems, by means of medication or referral to a mental health professional, as illustrated by the next quote (6.3).

Quote 6.3 (Patient ID 2): “As far as counseling and guidance is concerned, I can't offer you much help from here. However, if you feel like, this message has come as a real shock, I could do with some support, then we do have a medical psychologist here who could help you and give you some extra guidance. If you are interested, just let us know. I can then arrange this for you. It is always difficult to gauge how you will react to the news. It is possible that once you are home you have a lot of difficulty coping with the situation.”

Discussion

In this observational study we aimed to explore memory clinic clinicians’ communication during post-diagnostic testing consultations with MCI patients and their care partners, resulting in six core findings. First, we showed that approximately half of the clinicians used the term MCI, and they did not consistently use the three MCI criteria [5] to explain MCI. Second, clinicians often addressed the cause of patients’ symptoms in a tentative manner, leaving the options open as it were, even when biomarker information was available and independent of biomarker status. Third, if biomarker testing was not yet performed, clinicians seemed to steer towards the decision against further testing. Fourth and fifth, few clinicians provided specific or personalized information on the risk to develop dementia, but many emphasized potential symptom stabilization or improvement. Finally, all clinicians advised on one or more short-term next steps in symptom management, but they rarely addressed long-term (care) planning topics.

Our coding of the observed communication was partially based on the recently updated American Academy of Neurology (AAN) practice guideline [3]. Despite the challenges regarding the use of the MCI label, this practice guideline explicitly recommends to discuss the MCI diagnosis with patients and their care partners, to carefully educate them about the cause of symptoms and prognosis, and to encourage participation in long-term planning including living wills, finances and advance directives [3]. Discussing MCI with patients and their care partners might cause relieve, enabling them to label the patient’s condition [16, 17]. Furthermore, an accurate MCI diagnosis might enable patients and care partners to be more engaged in healthcare and disease management and to prepare for the future [18]. However, the MCI label should then be communicated to them. Our results showed that clinicians do not consistently inform about MCI in practice. This is in line with previous research, where clinicians indicated to find it difficult to present MCI as a diagnosis, as its implications for the individual are quite uncertain [15]. Nevertheless, variation between clinicians with regard to disclosure strategies may not be unique to MCI [19], and is also reflected in their diagnostic communication with patients with dementia and their care partners [20].

Knowledge about the underlying cause of patients’ symptoms could offer clinicians the opportunity to provide specific, more personalized, information about an individual patient’s prognosis and, consequently, future (care) planning. Most patients and care partners also have a strong need to understand the patient’s symptoms, and for information on the prognosis and implications [14]. Grill and colleagues state in their practice guideline that amyloid testing may offer such information [8]. Their point of view is more progressive then what is recommended in the AAN [3] or the Dutch MCI practice guidelines [21]. The Dutch guideline states that there are no accepted biomarkers available at this time to predict progression to dementia in patients with MCI, but stimulates clinicians to use diagnostic testing as a means to confirm alternative causes of the patient’s cognitive decline that could be treatable, such as depression. Nonetheless, biomarker evidence could establish support for the underlying etiology (AD or non-AD, e.g., psychiatric problems) [5, 22]. A negative biomarker status may be especially informative, since it increases the likelihood of a, potentially treatable, etiology other than AD [23]. Still, in our sample, amyloid testing was infrequently used and did not have major consequences for clinicians’ communication. That is, both negative and positive biomarkers results gave rise to a rather cautious discussion about the underlying cause of symptoms, and neither positive nor negative results were used to personalize the provision of information.

In fact, the patient’s prognosis was only discussed in general terms, whereby the clinicians emphasized that symptoms may well remain stable or improve over time. Clinicians thus seemed to use this uncertainty to foster hope, by especially highlighting the possibility that positive outcomes could occur [24, 25]. This observation may also illustrate how clinicians try to balance the seemingly contrasting needs that could exist among patients, i.e., the needs for honest information versus optimism [26, 27]. However, since many patients and care partners prefer to reduce uncertainty about the diagnosis and prognosis to allow for preparation and future (care) planning, and report unmet information needs [12, 14, 28], clinicians’ current approach may not match with patients’ and care partners’ needs. Information on prognosis could be personalized by using demographic (e.g. age) or clinical information (e.g. cognitive testing). In addition, ancillary investigations such as MRI, PET or CSF biomarkers could be used. Thus, even though biomarker testing does not necessarily decrease uncertainty, it still could provide patients and care partners with desired information about the future, even if the future is uncertain [29]. To attune to the individual’s needs and preferences, clinicians, patients and care partners could discuss beforehand what potential test results might yield, and if biomarker testing is the preferred action at this point in time. Thus, the decision for or against testing, and which tests to use, should preferably be a shared decision [8, 30]. Our results however suggest that clinicians who addressed the possibility of further biomarker testing, did not present all arguments for biomarkers testing. Clinicians did communicate that a positive biomarker result would not impact any clinical decisions, as treatment is not available. From literature however, it becomes increasingly clear that biomarkers have a particular good negative predictive value and may in turn increase or decrease the likelihood of another etiology, such as psychiatric problems (with potential treatment options) [3, 23, 31]. However, in our study, clinicians never mentioned this as an argument for additional biomarker testing.

Although the available practice guidelines recommend on what topics should be discussed when communicating about MCI, those guidelines do not provide clinicians with practical advise on how to optimally communicate these. To this background, clinicians have increasing amounts of (uncertain) information at their disposal that they could or should share, but they may not know how to meaningfully convey this information to their MCI patients. In addition, some clinicians might not be aware of the clinical value of the latest recommendations, since some of the recently published international frameworks are intended for research purposes [22]. Moreover, interpretation of biomarker results to gain estimates of the individual risk to develop dementia is not straightforward. Prediction models and computer-assisted clinical decision support systems may aid the clinician in interpreting clinical data and biomarker values and providing evidence-based individually-tailored prognostic information [23, 32, 33]. Nonetheless, even with a precise risk estimation in hand, the communication of this risk and its (uncertain) implications for the patient’s future remains challenging [34], particularly so in cognitively impaired individuals. Therefore, communication-skills training teaching strategies and skills crucial for uncertainty communication and/or shared decision-making could also support clinicians to enhance their communication in clinical practice and, consequently, better attune to the individual needs of patients and their care partners.

Among the strengths of this study is its multicenter design, resulting in a heterogeneous group of clinicians and patients, contributing to the generalizability of the results. Moreover, we followed good practice in qualitative research [13], including double-coding of the data. However, some limitations deserve mentioning. First, the sample size of the current study was small. This study included all MCI patients (15%) from a larger observational study which included consecutive newly-referred memory clinic patients [12]. Compared to other (inter)national memory clinics [35, 36] and looking at the general population [37], this percentage of MCI patients appears very similar, increasing the likelihood that our sample is representative of the population. Still, sample bias might have occurred among participating clinicians (recruited based on self-selection), i.e., clinicians who are relatively comfortable with communication research and perhaps more skilled. In addition, our sample comprised more MCI patients seen by neurologists than geriatricians. Perhaps MCI patients are typically younger and therefore more often seen by a neurologist, instead of a geriatrician. In addition there might also be a clinician-induced bias, i.e., geriatricians may less often diagnose patients with MCI, for example because it is more difficult to determine objective cognitive decline in older patients. Since this was an explorative study, future research should investigate in a larger sample: 1) what factors could explain the variation in/are associated with clinicians’ communication, e.g., consult duration, patient characteristics (for example the suspected underlying aetiology of the MCI), and/or clinician characteristics (such as clinicians’ medical specialty), and; 2) the impact/consequences of different communication strategies on patients and their care partners, in terms of quality of life, emotional well-being, and (their engagement in) health and disease management. Based on that evidence, we could (further) develop recommendations on how to optimally disclose an MCI diagnosis and prognosis, and interventions to support clinicians in their adherence to MCI guidelines. In addition, investigating the recall and/or understanding of patients and care partners of the information provided during consultations would also be an interesting avenue for future investigation.

Conclusion

The term MCI was used in half of the post-diagnostic testing consultations. Although variation existed, patients and care partners were often not provided with specific information about the underlying cause, risk to develop dementia and implications for long-term (care) planning. Since most patients and care partners have a strong need to understand the patient’s symptoms, and for information on the prognosis and implications for the future, clinicians’ current approach may not match with those needs.

The study-specific coding scheme.

(DOCX)

Click here for additional data file.

27 Sep 2019

PONE-D-19-25712

Clinicians’ communication with patients receiving a MCI diagnosis: the ABIDE project.

PLOS ONE

Dear Dr. Visser,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration by 2 Reviewers and an Academic Editor, all of the critiques of both Reviewers must be addressed in detail in a revision to determine publication status. If you are prepared to undertake the work required, I would be pleased to reconsider my decision, but revision of the original submission without directly addressing the critiques of the two Reviewers does not guarantee acceptance for publication in PLOS ONE. If the authors do not feel that the queries can be addressed, please consider submitting to another publication medium. A revised submission will be sent out for re-review. The authors are urged to have the manuscript given a hard copyedit for syntax and grammar.

==============================

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Overall this is a potentially very useful study for researchers and practitioners in the area of MCI research. I'm not aware of any studies to date that have examined clinical consultations for applications of recommended best practice, so this could be a valuable addition to the literature. It was appropriate in this case not to make the data fully available as per PLOS policy, given the nature of the data (clinical consultations). However I also think there are some important ways the manuscript could be improved.

1. Diagnostic process and categories: Could the authors explain a little more about the diagnostic pathway for MCI in the Netherlands? It looks as though all patients were recruited at the point when the MCI diagnosis was first to be delivered to them, is this correct? Would all the patients have had the same battery of tests to confirm AD/ dementia/ MCI status? Additionally, two of the patients were assigned different clinical labels - prodromal AD and objective cognitive disorders without dementia. I would expect this to affect the sorts of discussions the clinicians had with patients (especially since prodromal AD suggests a more known prognosis than MCI, which may not develop into AD)

2. Recruitment/ sampling: The sample of consultations is quite small at N=13. Quality over quantity in qualitative research, fair enough. But how was the sample size decided on, and what recruitment approach was used?

3. Data presentation: Generally I found the Tables in the manuscript fairly helpful. The one exception is Table 5. Personally, I'd prefer these 'raw' qualitative data included in the flow of the text to illustrate the findings as one reads along. My other point here is that seeing some of the contrasts in the data would help. Eg. Quote 1.1. - the clinician explicitly uses MCI, but in half the cases they did not - what happened in such cases?

4. Data analysis - an a priori coding framework was used, based on current guidelines. This seems like a sensible approach. Was there provision for 'bottom-up' coding of any relevant data that didn't fit the categories? Why/ why not?

5. Clinical implications: In the discussion, there is some implicit criticism of current clinical practice which I think may be a little unfair. I have in mind especially the suggestion to 'clearly communicate' the prognosis (p. 15) - given the uncertainty of prognosis, as you then go on to acknowledge, I think this is a real challenge for clinicians to provide good information. Should risk be positively or negatively framed? Given as absolute or relative risks? etc? This is a really complex issue and clearly, further work is needed to understand what risk information needs giving to which patients, in what format.

I also think there is some nuance to add to the points about fostering hope on p. 16. Here I think clinicians are negotiating some tricky, contrasting needs - on the one hand, patients want clinicians to communicate honestly, but often also desire optimism from clinicians (this is quite an old paper now, but illustrates the point nicely: Kutner, J. S., Steiner, J. F., Corbett, K. K., Jahnigen, D. W., & Barton, P. L. (1999). Information needs in terminal illness. Social science & medicine, 48(10), 1341-1352.). Maybe the point here is to improve communication about lifestyle modifications that may have a chance of improving prognosis.

6. Finally, the focus on clinicians' communication was helpful, but I wonder what the patients made of the communication? Were they passive in accepting the clinicians' words? Did they ask further questions if/ when needed? Did they seem to understand their prognosis? Where information was provided on causes, was this proactively given by the clinician or sought by the patient? Etc. I think there is some really useful info you'll have here that could considerably add to the interest and importance of the article. To take just one example, quote 1.2 in Table 5 is a genuinely bad example of clinical communication - what did the patient do in response?

Reviewer #2: This is a nice paper on an important topic that has received surprisingly little attention—diagnostic disclosure of a Mild Cognitive Impairment (MCI) diagnosis. The authors present data from a subset of participants in a large qualitative study, providing an in depth assessment of the delivery of MCI diagnosis in 13 patient interactions (by 7 different clinicians). They observe heterogeneity in diagnostic delivery, including some surprising things that are frequently missing from clinical interactions like discussion of long-term prognosis (risk for dementia) and long-term planning. Yet, the sample is small. And it is quite possible that important trends (e.g., 4 out of 5 exchanges where the presumed etiology was Alzheimer’s disease did include discussion of potential worsening over time [presumably to dementia]) were not fully fleshed out. There is also a question of selection bias that must be addressed. Altogether, this paper is an important contribution and should sound an alarm for clinicians and professional organizations alike, to use formal or informal checklists of the essential topics that should be addressed when delivering the MCI diagnosis.

Abstract

• Should include the physician specialties

• The first conclusion in the abstract is that information provided to patients and families about prognosis could be more personalized. But could it? While some risk estimates exist, are these “ready for primetime?” At best, they face tremendously limitations in generalizability.

• Alternatively, perhaps the point above about 4 out of 5 exchanges with patients for whom the presumed etiology was Alzheimer’s disease including discussion of risk for progression indicates that more personalized assessments are being performed for a subset of patients.

Introduction

• Well written and pithy. It may make sense (here or elsewhere) to give additional context for the study. This might briefly include important information about national practice guidelines (e.g., approved therapies, reimbursement for biomarkers, etc.)?

Methods

• There is no discussion of saturation or how the number of participants or interviews were determined

• Similarly, there is imbalance in the number of interviews from neurologists vs. geriatricians

• Presumably these are a subset of patients from a larger study of diagnostic exchange (Visser et al. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring. 2019). Yet, the selection does not appear to be as simple as all MCI diagnoses from the previous study (n=21). How then were these cases selected and what risk of bias comes with the selection? The paper would be strengthened if all MCI cases from the previous study were used.

Results

• The heading for section 3.2 is unclear

Discussion

• The duration of the consult (Table 1) seems worth comment.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Joshua D. Grill

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

==============================

We would appreciate receiving your revised manuscript by February, 2020. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Stephen D. Ginsberg, Ph.D.

Section Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

7 Dec 2019

Review Comments to the Author

Reviewer #1: Overall this is a potentially very useful study for researchers and practitioners in the area of MCI research. I'm not aware of any studies to date that have examined clinical consultations for applications of recommended best practice, so this could be a valuable addition to the literature. It was appropriate in this case not to make the data fully available as per PLOS policy, given the nature of the data (clinical consultations). However I also think there are some important ways the manuscript could be improved.

1. Diagnostic process and categories:

- Could the authors explain a little more about the diagnostic pathway for MCI in the Netherlands? It looks as though all patients were recruited at the point when the MCI diagnosis was first to be delivered to them, is this correct?

We selected MCI patients from the larger ABIDE observational study, where audio-recordings were made of the routine clinician-patient consultations with 136 newly referred patients during their diagnostic work up (for a detailed description, see Visser et al. (1)). In the Netherlands, patients are referred to the memory clinic by another medical doctor, mostly their general practitioner, i.e. no self-referral. Of the total sample of patients participating in the larger ABIDE study, 21 (15%) were diagnosed with mild cognitive impairment (MCI), based on retrospectively obtained medical record data. Fully audio-recorded post-diagnostic testing consultations were available for 13/21 MCI patients. These audio-recorded consultations comprise the data used for the current manuscript.

We agree with the reviewer that we could have explained this more clearly, and now elaborated a bit further in the method section 2.2 of our manuscript on page 6:

“All clinicians involved in diagnostic consultations at seven Dutch memory clinics were eligible and invited to participate (self-selection). The newly referred patients of consenting clinicians (in the Netherlands patients are referred to the memory clinic by another medical doctor, i.e. no self-referral) and accompanying care partners were invited to participate. Next, audio-recordings were made of the clinician-patient consultations when patients were seen as part of normal clinical care at the memory clinic. For the current study we selected fully audio-recorded post-diagnostic testing consultations with MCI patients (N=13), as based on their retrospectively obtained medical records.”

In addition, we added Figure 1, a flow diagram, on page 7 with the following note:

“Note. Of the total sample of patients participating in the ABIDE observational study (1), 21 (15%) were diagnosed with mild cognitive impairment (MCI), based on their retrospectively obtained medical record data. Post-diagnostic testing audio recordings were available from 13/21 MCI patients, who were seen by ten clinicians in seven Dutch memory clinics.”

- Would all the patients have had the same battery of tests to confirm AD/dementia/MCI status?

No efforts were made to standardize the diagnostic work up between memory clinics or clinicians, since our aim was to observe and describe naturalistic clinician-patient communication during routine diagnostic clinical consultations in a wide range of memory clinics. This resulted in variation between patients in diagnostic tests used; some memory clinics had a fixed diagnostic care path, i.e., they used a standard diagnostic protocol in which the decisions regarding diagnostic testing were already made on a memory clinic level (what tests to use and in which order), in other memory clinics, decisions about testing were made at a patient level (2). We added the diagnostic tests that each patient underwent, and the communicated test results in Table 2. In addition, we now elaborated on this in the method section on page 6:

“Since our aim was to observe and describe naturalistic clinician-patient communication during the routine diagnostic work up for dementia in a broad range of memory clinics, no efforts were made to intervene or standardize the diagnostic work up (for a detailed description see Visser et al., (1) and Table 2 for details on diagnostic tests that patients underwent).”

- Additionally, two of the patients were assigned different clinical labels - prodromal AD and objective cognitive disorders without dementia. I would expect this to affect the sorts of discussions the clinicians had with patients (especially since prodromal AD suggests a more known prognosis than MCI, which may not develop into AD)

The reviewer addresses an interesting point. Patient ID 11 was labeled with prodromal AD. The coded findings for this patients (Table 3) show that the clinician addressed progression of symptoms, and that this was the only patients where this statement was not counterbalanced with a statement on possible stabilization of symptoms. For this patient, one might expect the clinician to communicate in more details about this patients’ risk of developing a dementia syndrome in the future. However, such a discussion on risk of developing dementia was not observed in this patient. Patient ID 10 was labeled with objective cognitive disorder without dementia in the medical record. But in the clinical consultation, the diagnosis for this patient was clearly communicated as an MCI label (Table 3). Our sample is however too small to make sensible conclusions based on these observations. For the reader to appreciate possible differences in communication, we now included a note underneath Table 3, stating which patient had the prodromal AD label (patient ID 11) and which patient was labeled with objective cognitive disorder without dementia (patient ID 10).

2. Recruitment/ sampling: The sample of consultations is quite small at N=13. Quality over quantity in qualitative research, fair enough. But how was the sample size decided on, and what recruitment approach was used?

The MCI patients were selected from a larger study that investigated naturalistic clinician-patient communication during the diagnostic work-up for dementia, in the context of routine clinical care. In this larger study, all newly referred patients and accompanying care partners were invited to participate. Of the total sample of patients participating in the larger observational study, 21 (15%) were diagnosed with mild cognitive impairment (MCI), as based on retrospectively collected medical record data. Fully audio-recorded post-diagnostic consultations were available for 13/21 MCI patients. We have now added this to the method section (2.2 Sample, procedure and questionnaires, page 6.) and included a flow diagram to describe this patient flow (Figure 1, page 7).

3. Data presentation:

- Generally I found the Tables in the manuscript fairly helpful. The one exception is Table 5. Personally, I'd prefer these 'raw' qualitative data included in the flow of the text to illustrate the findings as one reads along.

We have now included the quotes in the main text as suggested, and excluded Table 5.

- My other point here is that seeing some of the contrasts in the data would help. Eg. Quote 1.1. - the clinician explicitly uses MCI, but in half the cases they did not - what happened in such cases?

Throughout the manuscript we have aimed to show contrasts by means of representative quotes. An example of what happened in the cases where a clinician did not explicitly use MCI, is shown in Quote 1.2. Here, the clinician only mentions that the patient shows cognitive impairment, without using the term MCI.

Quote 1.2 (Patient ID 12): “So the point is that I find it hard to explain how we can clarify this. The tests are also not conclusive, so yes... [...]. We have more or less said that, and this is not a real diagnosis but a description of the clinical picture, that the test results show a condition that is characterized by slight comprehension problems, apparently.... and some trouble remembering new information. Yes, there is a term used to describe your symptoms, but I won't mention it because you'll just forget and it's not really that important.”

Another example of contrasts in the data are for example shown in section 3.4. The first quote illustrates that if risk was discussed, this was most often discussed in a verbal format without the use of percentages, natural frequencies, reference class or time frame.

Quote 4.1 (Patient ID 4): “Of course, this could be an early warning sign of Alzheimer's disease with a dementia syndrome.”

On the other hand, some clinicians explicitly and elaborately explained a probability, shown by the following quote:

Quote 4.2 (Patient ID 6): “Half of these people will remain stable, and the symptoms of about half of them will worsen and progress to dementia [verbal format]. [...] Let's say that the likelihood that you develop dementia within 5 years [time frame] is 50% [percentage; negative framing]. But the likelihood that you remain stable is also 50% [percentage; positive framing].”

That we aimed to show contrasts in the data, might have been unclear as all quotes were previously summarized in table 5. We have now included the quotes in the main text, and we now better refer to those representative quotes, which hopefully sufficiently illustrates the contradictions in the qualitative data.

4. Data analysis - an a priori coding framework was used, based on current guidelines. This seems like a sensible approach. Was there provision for 'bottom-up' coding of any relevant data that didn't fit the categories? Why/ why not?

First, we indeed used deductive coding based on codebook inspired by the current guidelines and previous research. In that coding scheme (see S1 Appendix), there was some room for bottom-up coding, e.g., by means of the category ‘other’ with regard to next steps in care / planning.

Second, we took an inductive/bottom-up approach, by using thematic content coding looking at all coded categories together with relevant transcripts/quotes, identifying six core findings/themes that emerged from the data. We now better clarified this in the method section (2.3) on page 8:

“The final codes combined with the relevant transcribed segments of the audio-recordings were then inductively categorized further by LNCV, ISvM and EMAS, identifying core findings emerging from the data.”

5. Clinical implications:

- In the discussion, there is some implicit criticism of current clinical practice which I think may be a little unfair. I have in mind especially the suggestion to 'clearly communicate' the prognosis (p. 15) - given the uncertainty of prognosis, as you then go on to acknowledge, I think this is a real challenge for clinicians to provide good information. Should risk be positively or negatively framed? Given as absolute or relative risks? etc? This is a really complex issue and clearly, further work is needed to understand what risk information needs giving to which patients, in what format.

We agree with the reviewer that prognostic information is often uncertain and that risks are difficult/challenging to communicate clearly, because of various reasons. That is, in fact the point we try to make in the paragraph on page 20. We thank the reviewer for pointing this out, since it is absolutely not our intention to criticize clinicians on this matter. We now changed our suggestion on page 15 and excluded prognosis from that statement, which now reads:

“However, the MCI label should then be communicated to them.”

- I also think there is some nuance to add to the points about fostering hope on p. 16. Here I think clinicians are negotiating some tricky, contrasting needs - on the one hand, patients want clinicians to communicate honestly, but often also desire optimism from clinicians (this is quite an old paper now, but illustrates the point nicely: Kutner, J. S., Steiner, J. F., Corbett, K. K., Jahnigen, D. W., & Barton, P. L. (1999). Information needs in terminal illness. Social science & medicine, 48(10), 1341-1352.). Maybe the point here is to improve communication about lifestyle modifications that may have a chance of improving prognosis.

This point is well taken, and we would like to thank the reviewer for this concrete suggestion about an alternative explanation of our observations. We now added the following sentence to our discussion on page 19:

“This observation may also illustrate how clinicians try to balance the seemingly contrasting needs that could exist among patients, i.e., the needs for honest information versus optimism (3, 4). However, since many patients and care partners prefer to reduce uncertainty about the diagnosis and prognosis to allow for preparation and future (care) planning (1, 5), clinicians’ current approach may not match with patients’ and care partners’ needs.”

We also tried to add some nuance to this discussion by emphasizing that many patients and care partners prefer information, and that clinicians’ approach may not match with their needs.

6. Finally, the focus on clinicians' communication was helpful, but I wonder what the patients made of the communication? Were they passive in accepting the clinicians' words? Did they ask further questions if/ when needed? Did they seem to understand their prognosis? Where information was provided on causes, was this proactively given by the clinician or sought by the patient? Etc. I think there is some really useful info you'll have here that could considerably add to the interest and importance of the article. To take just one example, quote 1.2 in Table 5 is a genuinely bad example of clinical communication - what did the patient do in response?

Findings from the larger ABIDE observational study (1) showed that memory clinic patients and care partners only rarely (2% of the consultations) expressed what kind or how much information they preferred, e.g., by asking questions or by a remark. The amount of uncertainty expressions by the clinician in these consultations was the scope of a recently published ABIDE paper, and showed that patients and care partners elicited, respectively, few (15% and 23%) of these uncertainty expressions, through questions (6). Together these findings show that patient and caregiver active involvement in diagnostic consultations at the memory clinic is limited. We do also know, as previously reported based on recent focus group data (5), survey data(1) and a Delphi consensus study(7), that patients and care partners report an unmet need for information, especially about the future and what the test results mean for their daily life.

As the reviewer states, in this current study we focused on clinicians’ communication with MCI patients and did not code MCI patients’ communication in response to the provided diagnosis or prognosis. After the post-diagnostic testing consultation we did aim to survey patients’ and care partners’ understanding of 1) the communicated diagnosis: did you/your relative receive a diagnosis (No or Yes, namely…), and 2) the communicated prognosis: did the clinician disclose the risk of developing Alzheimer’s dementia (No or Yes, namely…)? The majority of patients (8/13) and caregivers (10/13) reported that they had received a diagnosis, however only two patients and two caregivers stated MCI as the received diagnosis. However, the written answers on these questions by patients and care partners were often unclear or their answers to the open-ended questions (namely…?) were lacking completely. We do not know whether this was the result of patients and care partners not understanding/remembering the diagnosis/prognosis, or not understanding the questions or our question format. We were therefore not convinced of the validity/reliability of these data and decided not to incorporate these in the current manuscript. Nevertheless, studying the information recall and/or understanding of patients and care partners of the information provided during consultations would most certainly be an interesting avenue for future investigation.

We now added this future research direction to the discussion on page 21:

“In addition, investigating the recall and/or understanding of patients and care partners of the information provided during consultations would also be an interesting avenue for future investigation.”

Reviewer #2: This is a nice paper on an important topic that has received surprisingly little attention—diagnostic disclosure of a Mild Cognitive Impairment (MCI) diagnosis. The authors present data from a subset of participants in a large qualitative study, providing an in depth assessment of the delivery of MCI diagnosis in 13 patient interactions (by 7 different clinicians). They observe heterogeneity in diagnostic delivery, including some surprising things that are frequently missing from clinical interactions like discussion of long-term prognosis (risk for dementia) and long-term planning. Yet, the sample is small. And it is quite possible that important trends (e.g., 4 out of 5 exchanges where the presumed etiology was Alzheimer’s disease did include discussion of potential worsening over time [presumably to dementia]) were not fully fleshed out. There is also a question of selection bias that must be addressed.

With regard to these issues raised, we provided a point-by-point reply below.

Altogether, this paper is an important contribution and should sound an alarm for clinicians and professional organizations alike, to use formal or informal checklists of the essential topics that should be addressed when delivering the MCI diagnosis.

Abstract

• Should include the physician specialties

We’ve now included the physician specialties (eight neurologists and two geriatricians) in the abstract (page 3).

• The first conclusion in the abstract is that information provided to patients and families about prognosis could be more personalized. But could it? While some risk estimates exist, are these “ready for primetime?” At best, they face tremendously limitations in generalizability.

In the abstract we provide a conclusion based on our six key findings, and stated that “Clinicians’ information provision about the underlying cause, prognosis and implications for long-term (care) planning in MCI could be more specific and personalized”. We did not intend to put emphasis on personalizing prognostic information (only), since we believe it is important to personalize all information provided, for example by adjusting the amount and content of information to the individual patient’s needs and preferences. From our previous publication(1) we know that patients’ and caregivers’ information preferences are however often not discussed / expressed in consultations. We do agree with the reviewer that the data/results included in the current manuscript better fit the conclusion that information provision could be more specific, and provide a less solid base for concluding that this information could be more personalized. The reviewer’s point is thus well-taken. We now deleted ‘personalized’ from the conclusion in the abstract and in the discussion.

• Alternatively, perhaps the point above about 4 out of 5 exchanges with patients for whom the presumed etiology was Alzheimer’s disease including discussion of risk for progression indicates that more personalized assessments are being performed for a subset of patients.

Although recent risk models are promising (8), we agree with the reviewer’s statements above that clinicians might be hesitant to provide precise risk estimates (percentages), and that the existing risk estimation models may not be ready (yet) for primetime in clinical practice. Nevertheless, we do believe that clinicians could use other formats than percentages to inform about the chance of developing dementia (e.g., a verbal label, such as ‘people with MCI have a higher chance of developing dementia’), and that this information is relevant for patients for whom the presumed etiology is Alzheimer’s disease, but also important for patients with other presumed underlying causes. Most patients and caregivers report that such information is very important to discuss during the diagnostic process (7), and many would have liked to receive more information about the future (1). We however observed that in most consultations with our MCI patients (8/13), the chance, likelihood, risk, possibility of developing dementia was not addressed, in fact, dementia was not mentioned at all.

In addition, demographic, clinical and biomarker information could provide guidance to personalize prognostic information for both patients in which AD is the presumed etiology and others. E.g., in two out of the eight consultations (patient ID 2 and ID 11) in which the risk/possibility of developing dementia was not addressed, AD was communicated as a potential/probable cause of MCI, and biomarker results were positive. This information did not gave rise to a conversation about dementia, let alone to personalize information about the chance of dementia. In comparison to other patients with negative biomarker results, these patients have a high(er) chance to develop dementia. And if, for example, neuropsychological testing, anamnestic information, young age or negative biomarker results point in the direction of a psychiatric cause of MCI, developing dementia is less likely. Thus, even though demographic, clinical, or biomarker information could not (yet) result in a precise risk estimate, it still could provide patients and care partners with desired information about the future, even if the future is still uncertain.

We therefore formulated as key finding 4 (section 3.4) that in our sample, information provision about the patient’s risk of developing dementia could be more specific and personalized. Yet, we now added a future research direction in the discussion on page 21 reflecting to reviewer’s hypothesis that clinicians might personalize their assessments and consequently their communication, based on the suspected underlying etiology:

“Since this was an explorative study, future research should investigate in a larger sample: 1) what factors could explain the variation in/are associated with clinicians’ communication, e.g., consult duration, patient characteristics (for example the suspected underlying aetiology of the MCI), and/or clinician characteristics (such as clinicians’ medical specialty)”

Introduction

• Well written and pithy. It may make sense (here or elsewhere) to give additional context for the study. This might briefly include important information about national practice guidelines (e.g., approved therapies, reimbursement for biomarkers, etc.)?

We thank the reviewer for this concrete suggestions. In the discussion on page 19, we mentioned the Dutch (9) MCI-practice guidelines, and we now elaborated on this a bit further:

“The Dutch guideline states that there are no accepted biomarkers available at this time to predict progression to dementia in patients with MCI, but stimulates clinicians to use diagnostic testing as a means to confirm alternative causes of the patient’s cognitive decline that could be treatable, such as depression.”

We now also elaborated a bit more in the methods section on page 6 providing information about the Dutch referral/memory clinic system:

“The newly referred patients of consenting clinicians (in the Netherlands patients are referred to the memory clinic by another medical doctor, i.e. no self-referral) and accompanying care partners were invited to participate. Next, audio-recordings were made of the clinician-patient consultations when patients were seen as part of normal clinical care at the memory clinic. Since our aim was to observe and describe naturalistic clinician-patient communication during the routine diagnostic work up for dementia in a broad range of memory clinics, no efforts were made to intervene or standardize the diagnostic work up (for a detailed description see Visser et al., (1), and Table 2 for details on diagnostic tests that patients underwent).”

Methods

• There is no discussion of saturation or how the number of participants or interviews were determined

We could have been more specific about our sample selection and recruitment. We thank the reviewer for pointing this out. The sample size of this study was not based on saturation. The MCI patients were selected from a larger study that investigated clinician-patient communication during the diagnostic work-up for dementia. In this larger study, all newly referred patients and accompanying care partners were invited to participate (for a detailed description see (1)). Of the total sample of patients participating in the larger ABIDE study, 21 (15%) were diagnosed with mild cognitive impairment (MCI). Fully audio-recorded post-diagnostic consultations were available for 13/21 MCI patients. We have now added this to the method section (2.2 Sample, procedure and questionnaires, page 6.) and included a flow diagram to describe this patient flow (Figure 1, page 7).

• Similarly, there is imbalance in the number of interviews from neurologists vs. geriatricians

The MCI patients in this sample are more often seen by neurologists (n=10) than by geriatricians (n=3). Since we did not aim to make a comparison based on specialism, we did not specifically set out to generate a sample of patients seen by a balanced number of neurologists versus geriatricians. But for the larger observational study we did set out to involve a heterogeneous sample of Dutch memory clinics, located at one academic hospital (20 patients, 12 clinicians), six non-academic teaching hospitals (106 patients; 27 clinicians) and one non-academic, non-teaching hospital (10 patients; 2 clinicians). The clinicians were recruited by self-selection, and 41 were willing to participate (42% neurologists, 32% geriatricians, 27% other profession (e.g., psychiatrists)). So, part of the imbalance might be caused by self-selection, whereby neurologists were more often willing to participate.

However, it could also be that there are relatively more neurologists involved at memory clinics in the Netherlands than geriatricians. We don’t have any official numbers to support this hypothesis. Or, alternatively, regarding our selection of MCI patients, it could be that MCI patients are typically younger and are therefore more often referred to a neurologist, instead of a geriatrician. In addition, there might also be a physician induced bias as geriatricians may less often diagnose their older patients with MCI, for example because it is more difficult to determine objective cognitive decline. But also a delay may have occurred, as elderly patients (seen by geriatricians) may more often think that their mild (memory) complaints are a result of normal aging and may therefore less often seek medical help early on (in the stage of MCI). Since this was an explorative study, future research should investigate in a larger sample what factors could explain the variation in clinicians’ communication, e.g., patient or clinician characteristics such as clinicians’ medical specialty. We now elaborated further on this issue, in the discussion section on page 21 (see also our reply to the reviewer’s next point below, the highlighted added text).

• Presumably these are a subset of patients from a larger study of diagnostic exchange (Visser et al. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring. 2019). Yet, the selection does not appear to be as simple as all MCI diagnoses from the previous study (n=21). How then were these cases selected and what risk of bias comes with the selection? The paper would be strengthened if all MCI cases from the previous study were used.

The reviewer is correct that this is a subset of the larger ABIDE observational study. We have now included a flow diagram (see Figure 1, page 7) to clarify the subject selection. From the 21 MCI patients, no fully recorded post-diagnostic consultations were available in n=8. This was for random reasons, e.g., because the clinician forgot to start the audio-recording or the post-diagnostic testing consultation was scheduled beyond our time window for data collection. In the discussion section on page 21 we address our small sample size and potential sampling bias:

“However, some limitations deserve mentioning. First, the sample size of the current study was small. This study included all MCI patients (15%) from a larger observational study which included consecutive newly-referred memory clinic patients (1). Compared to other (inter)national memory clinics (10, 11) and looking at the general population (12), this percentage of MCI patients appears very similar, increasing the likelihood that our sample is representative of the population. Still, sample bias might have occurred among participating clinicians (recruited based on self-selection), i.e., clinicians who are relatively comfortable with communication research and perhaps more skilled. In addition, our sample comprised more MCI patients seen by neurologists than geriatricians. Perhaps MCI patients are typically younger and therefore more often seen by a neurologist, instead of a geriatrician. In addition there might also be a clinician-induced bias, i.e., geriatricians may less often diagnose patients with MCI, for example because it is more difficult to determine objective cognitive decline in older patients. Since this was an explorative study, future research should investigate in a larger sample: 1) what factors could explain the variation in/are associated with clinicians’ communication, e.g., consult duration, patient characteristics (for example the suspected underlying aetiology of the MCI), and/or clinician characteristics (such as clinicians’ medical specialty”

Results

• The heading for section 3.2 is unclear

We have now changed this subheading to “3.2 clinicians often addressed the cause of patients’ symptoms in a tentative manner”, and hope that this adjustment clarifies its meaning (page 13).

Discussion

• The duration of the consult (Table 1) seems worth comment.

We observed some variation in consult duration. However, our sample is too small to analyze differences in communication based on consult duration, or to analyze correlations between consult duration and patient or clinician characteristics. The duration of the post-diagnostic testing consultation in this sample of MCI patients (M= 19, SD= 8 minutes), does not appear to deviate from the larger ABIDE sample, including patients with dementia and without objective cognitive decline (M= 19, SD= 13 minutes). However, we are not sure if the reviewer means that the variation in consult duration is worth comment, or that the duration in itself seems worth a comment (does the reviewer consider the consultation to be long or short?). We now added looking into consult duration as a future research direction, in the discussion section on page 21 (see below). If needed, we are happy to elaborate further on this issue.

“Since this was an explorative study, future research should investigate in a larger sample: 1) what factors could explain the variation in/are associated with clinicians’ communication, e.g., consult duration, …”

References in this revision note.

1. Visser LNC, Kunneman M, Murugesu L, van Maurik I, Zwan M, Bouwman FH, et al. Clinician-patient communication during the diagnostic workup: The ABIDE project. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring. 2019;11:520-8.

2. Kunneman M, Bouwman FH, Smets EMA, van der Flier WM. Diagnostiek van dementie: praktijkvariatie in Nederlandse geheugenpoliklinieken. Neuropraxis. 2018;22(5):137–46.

3. Lakin JR, Jacobsen J. Softening Our Approach to Discussing Prognosis. JAMA Intern Med. 2019;179(1):5-6.

4. Kutner JS, Steiner JF, Corbett KK, Jahnigen DW, Barton PL. Information needs in terminal illness. Soc Sci Med. 1999;48(10):1341-52.

5. Kunneman M, Pel-Littel R, Bouwman FH, Gillissen F, Schoonenboom NSM, Claus JJ, et al. Patients' and caregivers' views on conversations and shared decision making in diagnostic testing for Alzheimer's disease: The ABIDE project. Alzheimers Dement (N Y). 2017;3(3):314-22.

6. Visser LNC, Pelt SAR, Kunneman M, Bouwman FH, Claus JJ, Kalisvaart KJ, et al. Communicating uncertainties when disclosing diagnostic test results for (Alzheimer's) dementia in the memory clinic: The ABIDE project. Health Expect. 2019.

7. Fruijtier AD, Visser LNC, van Maurik IS, Zwan MD, Bouwman FH, van der Flier WM, et al. ABIDE Delphi study: topics to discuss in diagnostic consultations in memory clinics. Alzheimers Res Ther. 2019;11(1):77.

8. van Maurik IS, Vos SJ, Bos I, Bouwman FH, Teunissen CE, Scheltens P, et al. Biomarker-based prognosis for people with mild cognitive impairment (ABIDE): a modelling study. Lancet Neurol. 2019.

9. Richtlijn diagnostiek en behandeling van dementie - Addendum Mild Cognitive Impairment (MCI). Utrecht: Kennisinstituut van de Federatie Medisch Specialisten; 2017.

10. van der Flier WM, Scheltens P. Amsterdam Dementia Cohort: Performing Research to Optimize Care. J Alzheimers Dis. 2018;62(3):1091-111.

11. Roberts R, Knopman DS. Classification and epidemiology of MCI. Clin Geriatr Med. 2013;29(4):753-72.

12. Association As. 2019 Alzheimer’s disease facts and figures. 2019.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

17 Dec 2019

Clinicians’ communication with patients receiving a MCI diagnosis: the ABIDE project.

PONE-D-19-25712R1

Dear Dr. Visser,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Stephen D. Ginsberg, Ph.D.

Section Editor

PLOS ONE

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: The authors have thoroughly addressed the reviewer comments and the manuscript is improved. It is an important addition to the literature.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Tim Gomersall

Reviewer #2: Yes: Joshua Grill

13 Jan 2020

PONE-D-19-25712R1

Clinicians’ communication with patients receiving a MCI diagnosis: the ABIDE project

Dear Dr. Visser:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Stephen D Ginsberg

Section Editor

PLOS ONE