This paper is in the following e-collection/theme issue:

Formative Evaluation of Digital Health Interventions (2491) Dementia and Cognitive Decline (502) Artificial Intelligence (1421) Decision Support for Health Professionals (1210) Clinical Information and Decision Making (1434) Diagnostic Tools in Mental Health (248) Knowledge and Attitudes of Mental Health Professionals towards Digital Interventions (26) Electronic/Mobile Data Capture, Internet-based Survey & Research Methodology (847) Memory (45)

Published on in Vol 5, No 12 (2021): December

Preprints (earlier versions) of this paper are available at https://preprints.jmir.org/preprint/31053, first published .
Assessing the Views of Professionals, Patients, and Care Partners Concerning the Use of Computer Tools in Memory Clinics: International Survey Study

Assessing the Views of Professionals, Patients, and Care Partners Concerning the Use of Computer Tools in Memory Clinics: International Survey Study

Assessing the Views of Professionals, Patients, and Care Partners Concerning the Use of Computer Tools in Memory Clinics: International Survey Study

Authors of this article:

Aniek M van Gils1 Author Orcid Image ;   Leonie NC Visser1, 2 Author Orcid Image ;   Heleen MA Hendriksen1 Author Orcid Image ;   Jean Georges3 Author Orcid Image ;   Majon Muller4 Author Orcid Image ;   Femke H Bouwman1 Author Orcid Image ;   Wiesje M van der Flier1, 5 Author Orcid Image ;   Hanneke FM Rhodius-Meester1, 4 Author Orcid Image
Article Authors Cited by (9) Tweetations (6) Metrics

    Original Paper

    1Department of Neurology, Alzheimer Center Amsterdam, Amsterdam Neuroscience, Amsterdam UMC, Location VUmc, Amsterdam, Netherlands

    2Department of Neurobiology, Care Sciences and Society, Division of Clinical Geriatrics, Center for Alzheimer Research, Karolinska Institutet, Stockholm, Sweden

    3Alzheimer Europe, Luxembourg, Luxembourg

    4Department of Internal Medicine, Geriatric Medicine Section, Amsterdam Cardiovascular Sciences Institute, Amsterdam UMC, Location VUmc, Amsterdam, Netherlands

    5Department of Epidemiology and Data Science, Amsterdam UMC, Location VUmc, Amsterdam, Netherlands

    Corresponding Author:

    Aniek M van Gils, MSc, MD

    Department of Neurology, Alzheimer Center Amsterdam, Amsterdam Neuroscience

    Amsterdam UMC, Location VUmc

    De Boelelaan 1118

    Amsterdam, 1081 HZ

    Netherlands

    Phone: 31 204440685

    Email: a.vangils@amsterdamumc.nl


    Background: Computer tools based on artificial intelligence could aid clinicians in memory clinics in several ways, such as by supporting diagnostic decision-making, web-based cognitive testing, and the communication of diagnosis and prognosis.

    Objective: This study aims to identify the preferences as well as the main barriers and facilitators related to using computer tools in memory clinics for all end users, that is, clinicians, patients, and care partners.

    Methods: Between July and October 2020, we sent out invitations to a web-based survey to clinicians using the European Alzheimer’s Disease Centers network and the Dutch Memory Clinic network, and 109 clinicians participated (mean age 45 years, SD 10; 53/109, 48.6% female). A second survey was created for patients and care partners. They were invited via Alzheimer Europe, Alzheimer’s Society United Kingdom, Amsterdam Dementia Cohort, and Amsterdam Aging Cohort. A total of 50 patients with subjective cognitive decline, mild cognitive impairment, or dementia (mean age 73 years, SD 8; 17/34, 34% female) and 46 care partners (mean age 65 years, SD 12; 25/54, 54% female) participated in this survey.

    Results: Most clinicians reported a willingness to use diagnostic (88/109, 80.7%) and prognostic (83/109, 76.1%) computer tools. User-friendliness (71/109, 65.1%); Likert scale mean 4.5, SD 0.7), and increasing diagnostic accuracy (76/109, 69.7%; mean 4.3, SD 0.7) were reported as the main factors stimulating the adoption of a tool. Tools should also save time and provide clear information on reliability and validity. Inadequate integration with electronic patient records (46/109, 42.2%; mean 3.8, SD 1.0) and fear of losing important clinical information (48/109, 44%; mean 3.7, SD 1.2) were most frequently indicated as barriers. Patients and care partners were equally positive about the use of computer tools by clinicians, both for diagnosis (69/96, 72%) and prognosis (73/96, 76%). In addition, most of them thought favorably regarding the possibility of using the tools themselves.

    Conclusions: This study showed that computer tools in memory clinics are positively valued by most end users. For further development and implementation, it is essential to overcome the technical and practical barriers of a tool while paying utmost attention to its reliability and validity.

    JMIR Form Res 2021;5(12):e31053

    doi:10.2196/31053

    Keywords

    artificial intelligence; clinical decision support systems; dementia; diagnostic testing; diagnosis; prognosis; communication 


    Background

    Dementia is a major health problem worldwide, with its prevalence expected to rise to 75 million patients in 2050 [1]. A timely and accurate diagnosis is essential for providing adequate care and appropriate treatment [2]. Diagnosing Alzheimer disease dementia or another type of dementia can be challenging, as clinical presentations overlap and multiple pathologies often co-occur [3,4]. Furthermore, increasing biomarker availability creates the possibility of diagnosing the early stages of Alzheimer disease before the onset of dementia and paves the way for individual dementia risk estimation [5,6].

    With the availability of many different diagnostic tests, clinicians have the difficult task of combining and interpreting all the test results to come to an accurate diagnosis and prognosis [7-10] and communicating these results to patients and care partners [5,11,12]. Despite the increasing number of diagnostic tools, uncertainty in the diagnosis and prognosis of dementia remains, complicating the process of clearly explaining the test results [13]. Recent research has shown that most patients and care partners greatly value precise information on diagnosis and prognosis [12,14]. However, clinicians are often reluctant to address these topics during consultations, leaving these informational needs unmet [12,15].

    Currently, artificial intelligence solutions are rapidly being developed and can aid clinicians in addressing these challenges in several ways. Artificial intelligence–based computer tools for dementia diagnosis and prognosis have demonstrated diagnostic accuracy equal to that of clinicians’ performance. These tools support individual risk estimation and increase clinicians’ confidence in diagnosis and prognosis [16-19]. Web-based cognitive test tools have shown promising results, enabling cost-effective testing [20-22]. From other medical fields, such as oncology, we know that computer-based tools can also support the communication process, for example, by engaging patients and their families more actively in the diagnostic decisions or by supporting clinicians in the clear communication of results [23,24]. To date, the actual implementation of such computer tools in memory clinic practice has been limited [25,26].

    Barriers and Facilitators

    Several barriers to the acceptance and implementation of tools have been identified in different health care areas. The main concerns regarding computer tools are related to the physician–patient relationship: the fear of interfering with this relationship when using a tool and affecting patient communication. Furthermore, clinicians fear the disturbance of clinical work and the loss of clinical autonomy when using a tool. In addition, a time-consuming tool, a tool that does not fit into the workflow, complexity of a tool, and computer literacy have been frequently mentioned as barriers. On the other hand, good training before the use of a tool, user-friendliness, relevancy, transparency, and reliability are stimulating factors in the use of a computer tool in clinical practice [27-33].

    It is not known if the same barriers and facilitators apply to computer tools in memory clinics. The nature of the patient population—older adults with cognitive decline—and the vast number of diagnostic tests involved in the diagnostic process might lead to a different set of relevant barriers and facilitators. In addition, patients’ and care partners’ opinions regarding the use of computer tools by their clinicians might be a potential barrier to or facilitator of clinicians’ use of a tool.

    Objective

    Therefore, this study aims to understand preferences and identify the main barriers to and facilitators of using computer tools in the dementia workup from the perspectives of clinicians, patients, and care partners, that is, the end users.


    Design

    We conducted 2 surveys, 1 for clinicians and 1 directed at patients and care partners, both in the fall of 2020. In addition, to aid the interpretation of the survey results, we conducted an interactive panel session with clinicians. The study was approved by the medical ethical committee of the Amsterdam University Medical Center, Vrije Universiteit Medical Center, Amsterdam. All participants provided digitally informed consent.

    Survey for Clinicians

    Participants

    Between July and October 2020, we invited clinicians from memory clinics in Europe via the European Alzheimer’s Disease Consortium and the Dutch Memory Clinic network (Nederlands Geheugenpoli Netwerk [NGN]) through emails that contained a link to participate in the web-based survey. Furthermore, we invited all participants during the annual NGN conference to share their thoughts in our web-based interactive panel session.

    Survey

    The survey was created in the web-based survey tool Survalyzer (Survalyzer AG) [34] and translated into Dutch and English. The survey was adaptive; that is, certain questions were only conditionally displayed based on responses to other items. Furthermore, participants could scroll through the survey to edit their answers. The survey comprised 3 parts. In the first part, we collected background information (eg, age, gender, profession, and specialization). In the second part, we used a funneled method to examine the current opinions on computer tools and identify the barriers and facilitators. First, we asked if the clinician would be willing to use computer tools in general, after which we asked them to explain their opinions in an open-ended question. Subsequently, we provided them with a list of barriers and facilitators compiled from barriers and facilitators known from the existing literature [27,28,32], and we asked them which factors would stimulate or discourage them from using a tool. Clinicians could complement this list with their own perspectives. We then asked them to rate the importance of these factors using the Likert scale (1=very unimportant, 2=unimportant, 3=neutral, 4=important, and 5=very important). Finally, we asked the participants how likely they were to use diagnostic and prognostic tools. In the third part, we explored clinicians’ opinions on additional computer tools, that is, web-based cognitive testing, communication support, and communication skills training.

    Interactive Panel Session

    During the NGN annual conference (held on the web on November 10, 2020), 2 authors (HFMRM and LNCV) presented the preliminary results of the survey. To help interpret these results, they asked all conference participants several in-depth questions using Mentimeter [35]. These questions were related to the importance of several factors that stimulate their trust in a tool; factors that would convince them of the usability, reliability, and validity of a tool; and the primary outcome measures of a tool.

    Survey for Patients and Care Partners

    Participants

    Between July and October 2020, we invited a mixed memory clinic population comprising patients with subjective memory complaints (subjective cognitive decline [SCD]), mild cognitive impairment, and dementia and care partners to participate in the web-based survey. To be included, patients had to be able to understand the questionnaire in Dutch or English. In this study, care partner refers either to an informal caregiver or a close relative or friend of the patient who provides either or both emotional and practical support. A general link to the survey was sent via a newsletter and social media to the members (patients and care partners) of Alzheimer Europe and directly to the members of the Alzheimer’s Society United Kingdom. Next, we invited both patients and care partners from the Amsterdam Dementia Cohort of the Alzheimer Center Amsterdam [36,37] and the Amsterdam Aging Cohort of the outpatient geriatric clinic of the Amsterdam University Medical Center (Amsterdam University Medical Centers) [38]. Patients and care partners were approached and informed by phone or email, and when they confirmed their participation, they were sent a personalized link to the web-based survey. The survey was adaptive to reduce the number of questions. Furthermore, participants could scroll through the survey to edit their answers.

    Survey

    A total of 2 versions of the survey were created, 1 directed at patients and 1 at care partners, both comprising 3 parts. In part 1, we collected background information regarding the participants (eg, age, gender, and diagnosis). In part 2, we asked for their opinion on clinicians’ use of (1) a computer tool that analyzes the results of the diagnostic tests (diagnostic tool), (2) a computer tool to help predict the course of their symptoms (prognostic tool), and (3) a tool to help communicate the test results in day-to-day language with the patient (communication tool). We adjusted the predefined list of barriers to and facilitators of using computer tools to the patient and care partner perspectives. We provided participants with this list and asked the extent (1=strongly disagree, 2=disagree, 3=neutral, 4=agree, and 5=strongly agree) to which the different items applied to them. In the last part, we asked their opinion on computer tools directed at patients and care partners, that is, web-based cognitive testing and tools that could support and empower them in their communication with the clinicians. The survey was piloted in a test panel of 3 patients (2 with SCD and 1 with dementia) and 1 care partner of a person with dementia.

    Analysis

    Completion of the survey was enforced using mandatory questions. Proceeding with the survey was not possible when a question was unanswered. Furthermore, it was not possible to complete the survey more than once, as Survalyzer solely allowed unique visitors. Only completed surveys were analyzed. We analyzed participant characteristics and survey outcomes using descriptive statistics. Chi-square tests were used to compare answers between patients and care partners. For clinicians, we compared answers between groups based on age, sex, profession, and specialization. When using the 5-point Likert scale, the mean Likert scale scores were calculated per item. Frequencies were calculated for all the barriers and facilitators from the predefined list. We combined the frequencies with mean Likert scale scores to define the most important barriers and facilitators (eg, the item with the highest frequency combined with the highest mean Likert scale score was regarded as the most important). SPSS, version 22.0 (IBM Corporation) was used to analyze the quantitative data. P values <.05 were considered significant.

    The answers to the open-ended questions were analyzed in MAXQDA software (VERBI Software) [39] using a process of deductive thematic content analysis [40,41]. A total of 2 authors, AMVG (physician) and HMAH (neuropsychologist), independently generated the initial thematic codes based on the existing literature and data. Subsequently, 1 author (AMVG) generated a thematic framework and used this framework to code all the given answers. The codes were then sorted into broad categories.


    Demographics

    Sample descriptions have been presented in Table 1 for clinicians and Table 2 for patients and care partners. Clinicians were, on average, aged 45 (SD 11) years and had 16 (SD 13) years of experience. Most participating clinicians were medical specialists working in neurology (60/109, 55%) or internal or clinical geriatric medicine (33/109, 30.3%). Patients were in general older (mean age 73 years, SD 8) than care partners (mean age 65 years, SD 12), who were mostly a partner or spouse (33/46, 72%) or a granddaughter, daughter, grandson, and son (in-law; 12/46, 26%). Participating patients were most often diagnosed with SCD (21/50, 42%), whereas participating care partners were mostly those of patients with dementia (36/46, 78%).

    Table 1. Sample demographics of clinicians participating in the web-based survey and interactive panel session (N=294).
    CharacteristicsWeb-based survey (n=109)Interactive panel sessiona (n=184)
    Age (years), mean (SD)45 (11)43 (11)
    Sex (female), n (%)53 (48.6)98 (85.9)
    Cohort, n (%)

    		European Alzheimer’s Disease Consortium53 (48.6)N/Ab

    		Dutch Memory Clinic network56 (51.4)N/A
    Profession, n (%)

    		MDc, specialist87 (79.8)60 (54.5)

    		MD, specialist training or not in specialist training12 (10.9)1 (0.9)

    		Physician assistant, nurse specialist, or specialized nurse3 (2.8)23 (20.9)

    		Neuropsychologist or psychologist6 (5.5)16 (114.5)

    		Other1 (0.9)10 (9.1)
    Experienced (years), mean (SD)16 (13)N/A
    Specializatione, n (%)

    		Neurology60 (50.4)N/A

    		Clinical geriatric or internal medicine33 (30.3)N/A

    		Nursing home physician or general practitioner2 (1.8)N/A

    		Psychiatry9 (8.3)N/A

    		Other9 (8.3)N/A
    Institutionf, n (%)

    		Academic or university hospital68 (62.4)N/A

    		Nonacademic teaching hospital32 (29.4)N/A

    		Nonteaching hospital8 (7.3)N/A

    		Mental health service2 (1.8)N/A

    		Other3 (2.8)N/A

    aOwing to the hybrid conference setting, not every description was available for all participants. For sex, n=114 participants replied to the question. For profession, n=110 participants replied to the question.

    bN/A: not applicable.

    cMD: medical doctor.

    dOnly applicable for medical specialists.

    eSome clinicians had ≥1 specialization.

    fSome clinicians worked in ≥1 institution.

    Table 2. Sample demographics of patients and care partners participating in the web-based survey (N=96).
    CharacteristicsPatientsa (n=50)Care partners (n=46)
    Age (years), mean (SD)73 (8)65 (12)
    Sex (female), n (%)17 (34)25 (54)
    Cohort, n (%)

    		Alzheimer Europe or Alzheimer’s Society United Kingdom2 (4)14 (30)

    		Amsterdam dementia cohort25 (50)27 (52)

    		Amsterdam aging cohort23 (46)5 (18)
    Diagnosisb, n (%)

    		SCDc21 (42)2 (4)

    		MCId16 (32)8 (17)

    		Dementia13 (26)36 (78)
    Educatione, n (%)

    		Low1 (2)1 (2)

    		Middle22 (45)16 (36)

    		High26 (53)27 (61)

    aOf these 50 patients, 20 (40%) completed the survey together with their care partner.

    bFor the care partners, the numbers represent the diagnosis of their loved ones.

    cSCD: subjective cognitive decline.

    dMCI: mild cognitive impairment.

    eAccording to the Dutch Verhage scale (low 1-3; middle 5; high 6-7).

    Survey of Clinicians

    Opinions on the Use of Computer Tools

    In response to whether they would be willing to use computer tools in their daily clinical practice, 51.4% (56/109) of clinicians said they would probably use a diagnostic tool, and 29.4% (32/109) said they would certainly use a diagnostic tool. Furthermore, 7.3% (8/109) said they would be unlikely to use a tool, 11.9% (13/109) answered neutrally, and none of the clinicians reported that they did not want to use a tool. The results were similar for prognostic tools; of the 109 clinicians, 53 (48.6%) said they would probably use a prognostic tool, and 30 (27.6%) said they would certainly use a prognostic tool. Furthermore, 2.8% (3/109) were unlikely to use a prognostic tool, and 0.9% (1/109) would certainly not use a prognostic tool. The remaining participants responded as neutral (22/109, 20.2%). We found no differences in both diagnostic and prognostic tools based on age (P=.20 and P=.49, respectively), sex (P=.14 and P=.73, respectively), or profession (P=.61 and P=.98, respectively). We found that neurologists indicated more willingness to use prognostic tools (P=.04) than clinicians from other specializations.

    Content analysis of clinicians’ explanations of their opinion on the use of computer tools resulted in 6 main topics: support, clinical expertise, efficiency, accuracy, clinician–patient relationship, and care of the future. Each of the topics has been described in Table 3, and illustrative quotes have been provided.

    Table 3. Clinicians’ opinions on the use of computer tools in memory clinics.
    TopicsDescriptionQuotes

    		Facilitating factorsHindering factors
    SupportSupport to the diagnostic process from screening or prescreening to follow up, support data storage, support research purposesNot applicable for specific patient populations
    • “[...] I would welcome a tool that would be implemented with the available clinical data and help reach a diagnosis (ie, considering the neurocognitive and neuroimaging data, in patients with such profile, expert diagnosis would be...with a probability of...%—which could be increased by the use of...biomarker).” [Male, 42 years, MDb, physician working in neurology]
      • ‎
    Clinical expertiseComplementary to clinical expertise (eg, an aid for complex cases or with interpretation of test results) and contributory to evidence-based medicine [42]Tools should not be a replacement for clinical expertise
    • “Computer tools and AI might be a way to have an evidence-based standard procedure in addition to my own long time clinical experience.” [Female, 59 years, MD, geriatrician];
      • ‎
    • “[...] I consider the clinical view as most important. A computer tool cannot (partly) replace this.” [Female, 38 years, MD, geriatrician]
      • ‎
    Efficiency The ability to standardize the diagnostic process, if easy to use, if connecting with electronic patient file, and time-efficiencyA tool not connected to the electronic patient file, information technology issues
    • “A quick and useful way to get practical answers on the workplace.” [Male, 62 years, MD, neurologist];
      • ‎
    • “For tools that are not implemented in the electronic patient file I foresee barriers in the implementation.” [Female, 45 years, MD, geriatrician]
      • ‎
    AccuracyComputer tools could help in making a more accurate diagnosis, providing additional objective information, and overcoming human errorsTools might generate results of no use and fear of loss of important clinical information
    • “Sometimes we can be influenced by the patient we have in front of us. We can diagnose them too easily or consider them as (sub)normal because their general behavior makes us think so. A computer could be more objective than we are in some cases.” [Male, 26 years, MD, neurology resident];
      • ‎
    • “[...] I am afraid that there will be an outcome that is of no use for me, such as 64% chance of Alzheimer’s disease.” [Female, 38 years, MD, geriatrician]
      • ‎
    Clinician–patient relationshipImproving patient communicationA tool might have a negative impact on the relationship between clinicians and patients
    • “[...] It facilitates the communication to the patient.” [Female, 49 years, MD, neurologist];
      • ‎
    • “Patients also come for attention and care, which they get less if we look at the screen more often.” [Male, 32 years, MD, physician working in neurology]
      • ‎
    Care of the futureThe use of tools is considered part of the care of the futureN/Aa
    • “AI and big data are the future, they make the invisible visible [...].” [Male, 33 years, MD, internal (geriatric) medicine resident]
      • ‎

    aN/A: not applicable.

    bMD: medical doctor.

    Barriers and Facilitators

    Clinicians’ selection of factors that would hinder the use of computer tools has been displayed in Figure 1A. None of the barriers were rated using a mean Likert scale score ≥4. Figure 1B shows which factors would stimulate the use of computer tools; items marked with an asterisk were rated with a mean Likert scale score of ≥4. The most important barriers were (1) a tool not being well connected with the electronic patient file (47/109, 43.2% of clinicians), with a mean Likert scale score of 3.8 (SD 1.0), and (2) important information, such as observations, cannot be added to a tool (49/109, 44.9% of clinicians), with a mean Likert scale score of 3.7 (SD 1.2). The most important facilitators were (1) user-friendliness (73/109, 66.9% of clinicians; mean 4.5, SD 0.7), and (2) increasing diagnostic accuracy (78/109, 71.6% of clinicians; mean 4.3, SD 0.7).

    ‎
    Figure 1. Frequencies of barriers to (A) and facilitators of (B) the use of computer tools in daily practice according to Dutch and European clinicians. EPF: electronic patient file. *Items rated with a mean Likert scale score of ≥4.
    View this figure
    Additional Tools for Memory Clinics

    Many clinicians (48/109, 44%) reported a willingness to use both web-based communication tools and skills training. Over half of the clinicians (62/109, 56.9%) indicated that they would like to test the patient’s cognition via the computer at home before the appointment. Frequently selected reasons for web-based cognitive testing were triage (62/109, 56.9%) and shortening of the test battery at the clinic (37/109, 33.4%). Not being able to observe the patient during testing was mentioned by 89% (42/47) of the clinicians who would not want to test the patient’s cognition on the web.

    Interactive Panel Session

    Both transparencies about the objectives of the tool provider (Likert scale score mean 4.3, SD 1.1) and honesty about the possibilities and limitations of a tool (Likert scale scores mean 4.4, SD 0.9) were considered important to strengthen clinicians’ trust in a tool. The most important aspects of convincing clinicians of usability, reliability, and validity of a tool were the explicit provision of information regarding the tool (mean 4.3, SD 0.9), scientific publications on underlying models (mean 4.3, SD 0.9), obtaining hands-on experience with the tool (mean 4.2, SD 0.6), and a randomized controlled trial (RCT) to test the effectiveness of the tool in clinical practice (mean 4.0, SD 0.8). Diagnostic accuracy (31/100, 31%) and patient-related outcome measures (38/100, 38%), such as quality of life, were most frequently selected as the ideal primary outcome measures of such an RCT.

    Survey of Patients and Care Partners

    Tools Used by Clinicians

    The results of patients’ and care partners’ opinions on diagnostic, prognostic, and communication tools have been presented in Table 4. Most patients and care partners were positive regarding their clinician using these tools. No differences were found between patients and care partners (diagnostic tools, P=.36; prognostic tools, P=.36; communication tools, P=.63) or different syndrome diagnoses (diagnostic tools, P=.64; prognostic tools, P=.69; communication tools, P=.92).

    Figure 2 shows an overview of the responses of patients and care partners to several statements regarding the use of computer tools by clinicians. Items marked with an asterisk are items rated with a mean Likert scale score ≥4. We found no differences between the responses of patients and those of the care partners (P values in order of appearance of topics from top to down in the figure: P=.75, P=.62, P=.70, P=.55, P=.78, P=.21, and P=.60).

    Table 4. Opinion of patients and care partners on clinicians’ use of diagnostic, prognostic, and communication tools, illustrated with quotes (N=96).
    OpinionPatients (n=50), n (%)Care partners (n=46), n (%)Quotes
    Diagnostic tool

    		I think that is a good thing38 (76)31 (67)“The more information, the better. As long as the computer program is in addition to the doctor’s expertise and not a replacement, I think it would be a good idea.” [Female 60 years, care partner]

    		I would not want that4 (8)4 (9)“I think face to face contact between the doctor and the patient is essential.” [Female 74 years, care partner]

    		I do not know or no opinion8 (16)11 (24)“Depends on how good the program is.” [Male 76 years, patient, dementia]
    Prognostic tool

    		I think that is a good thing41 (82)32 (70)“There is nothing against the use of a computer in predicting the disease process. It remains an aid to the physician. [...] He/she should remain leading.” [Male 78 years, patient, SCDa]

    		I would not want that4 (8)6 (13)“I want to know so I can plan ahead. However, with the variation in the progression rate, I don’t see how this could be sufficiently accurate. If not accurate, I would not want it.” [Female, 61 years, patient, dementia]

    		I do not know or no opinion5 (10)8 (17)“My husband approves [the use of a prognostic tool], me as his wife, do not know if I would like it. What if the prediction is somber! We would instantly be depressed.” [Female (age unknown), care partner]
    Communication tool

    		I think that is a good thing39 (78)38 (83)—b

    		I would not want that3 (6)1 (2)—

    		I do not know or no opinion8 (16)7 (15)—

    aSCD: subjective cognitive decline.

    bNo quotes available.

    ‎
    Figure 2. Agreement of patients and care partners on several statements regarding computer tools on a 5-point Likert scale. * Items rated with a mean Likert scale score of ≥4.
    View this figure
    Tools Directed at Patients and Care Partners

    Most patients (35/50, 70%) and care partners (26/46, 57%) expressed a preference for a list of example questions to select questions they wanted to ask their clinician. Furthermore, 42% (21/50) of patients and 50% (23/46) of care partners said they would prepare their visit to the memory clinic by watching informational videos. A smaller but considerable proportion of care partners (11/46, 24%) indicated that they would use a web-based communication tool to practice their communication skills, and some (5/50, 10%) patients reported wanting to do the same. A small proportion of patients (4/50, 8%) and care partners (8/46, 17%) did not want to use any web-based communication tools.

    Most patients (35/50, 70%) and care partners (28/46, 61%) were positive regarding web-based cognitive testing at home. Reasons for not wanting to perform cognitive testing at home differed between patients and care partners (P=.01). The most frequently selected reason for patients was their preference for personal contact with the clinician (18/26; 69%). The most selected reason (5/12, 42%) for care partners was that web-based cognitive testing was too difficult to perform for their loved ones.


    Principal Findings

    This study showed that most clinicians, patients, and care partners were supportive of the use of computer tools in memory clinics. This holds true for diagnostic and prognostic tools, tools that support communication, and web-based cognitive testing. Despite acknowledging their potential barriers, the general attitude of clinicians toward these tools was positive. The facilitating factors were mainly practical (tools should be user-friendly) and technical (connection with electronic patient files) and that tools should increase diagnostic accuracy. The identified barriers mainly focused on doubts regarding reliability and validity, preservation of clinical autonomy, and fear of losing important clinical information. Furthermore, the tools should be considered in addition to the current working methods and not as a replacement.

    We hypothesized that the barriers to and facilitators of tools in memory clinics might differ from those identified in other health care areas because of the nature of the patient population (older adults with cognitive decline) and the large number of diagnostic tests used in the clinical workup of dementia. However, the barriers and facilitators we found in this survey study largely corresponded to the existing literature on barriers to and facilitators of the use of computer tools in other medical fields [27-33]. Furthermore, we found that most of the given answers to the open-ended question were in accordance with our predefined list, which was based on studies on applying computer tools in other health care areas [27,28,32]. In addition, it is conceivable that patients’ and their care partners’ (negative) opinions regarding the tools were a possible barrier to clinicians using a tool. In this study, we showed that patients, despite their age and (potential) cognitive decline, are mainly positive regarding the use of computer tools. Most of them embrace the possibility of using a tool themselves, and their care partners share this opinion. A computer tool must support rather than replace clinicians, who would then view it as an aid appropriate to modern times.

    Increase Acceptance of Tools

    We found the overall attitude among clinicians toward tools to be highly positive. Nevertheless, none of the available tools are regularly used in daily practice, and it seems there is a major information gap and educational need to make clinicians understand the possibilities of such computer tools [43]. The results from our survey provide direction for the way to increase the acceptance of computer tools in memory clinic practice. First, clinicians indicated that an RCT on the efficiency in clinical practice would boost their confidence in the reliability of computer tools. RCTs are considered important and robust methods for assessing the impact of a tool [44,45]. However, no RCT on the application of computer tools has yet been performed in memory clinics. Thus, our findings suggest a need for RCTs with diagnostic accuracy and patient-related outcome measures as primary outcome measures.

    Second, computer tools could contribute to evidence-based medicine (EBM) [42]. EBM concerns medical practice based on the best available evidence, clinical experience, and patient preferences [42]. Within the concept of EBM, clinical experience is highly essential and should not be replaced by a tool. Computer tools could support EBM by making the best available evidence more accessible to clinicians, and computer tools could clarify patient preferences. Acknowledging computer tools as a part of EBM might lead to clinicians viewing these tools as an aid complementary to their own clinical experience rather than a threat to their clinical autonomy [27,29,32,46]. In addition, there must be scientific publications regarding the underlying models and the transparent provision of information regarding a tool’s reliability to increase clinicians’ trust in the tools [28,32,47].

    Then, clinicians’ confidence in the tools might be strengthened if the tools are under the jurisdiction of a regulatory body to authorize and supervise the quality. To date, there are no formal regulatory standards for tools used to support clinicians in decision-making [28]. When developers claim that their software has a medical purpose, it becomes a medical device, and manufacturers themselves have to proclaim that their device meets the safety and performance requirements. The new European Medical Devices Regulation implemented in May 2021 makes manufacturers adhere to more strict guidelines for ensuring the safety of their products, including assessment of their device by a notified body. Involving an independent notified body in the approval of a tool might be the first step toward improving its transparency and acceptance. Eventually, as a next step, the use of software should be included in the guidelines of professional associations [26].

    The attitude toward tools is one of the key characteristics of eventual acceptance [30,32]. Several user acceptance models have been proposed to further encourage the acceptance of tools in medical practices [32]. In this context, we would like to address the user acceptance and system adaptation design model [32]. This model aims to include end users as the central point in the design process of a computer tool. In this model, user expectations and needs need to be thoroughly understood before developing a tool. The development of a tool is an iterative process, and end users should be continuously involved throughout the development process. In congruence with this user acceptance model, we took the first step toward accepting computer tools by identifying the barriers to and facilitators of computer tools according to the end users in memory clinics. On the basis of the user acceptance and system adaptation design model, the next step in the iterative process would be the evaluation of usability. Then, a pilot study would be needed to evaluate user acceptance. The results of each step should direct improvements in the tool and be evaluated in the next step. However, it should be mentioned that in this study, we asked participants about the barriers to and facilitators of computer tools in general. Therefore, the barriers and facilitators found in this study should be considered a starting point. When implementing specific tools, further exploration of user expectations and needs for those specific tools might be necessary.

    Strengths and Limitations

    One of the strengths of this study is that we involved all the end users of computer tools in memory clinics, that is, clinicians, patients, and care partners. Including patients and care partners originating from Europe and a heterogeneous population of clinicians contributed to the generalizability of the results. Furthermore, we used a funneled method in which we started the survey with open-ended questions and worked toward closed questions. By doing so, we actively asked and stimulated clinicians’ own input.

    Nevertheless, our study has several limitations. First, we distributed the survey via a web-based link, which might have caused selection bias by only involving people with sufficient digital skills who might have had a more positive approach toward computer tools. We tried to minimalize this risk for patients and care partners by including participants from both geriatric and neurology departments with different cognitive impairment stages. Furthermore, participants’ ages ranged from relatively young to older patients, who might have had less experience with digital tools. Second, we have no data available on the origin of the European participants, which might have led to an uneven distribution of participants among countries. Nonetheless, the international character of this study contributes to the generalizability of the results. Third, there might be a risk of response bias; people who are less inclined to use computer tools may not have responded. We could not estimate the response rates as the link to the survey was spread among an unknown number of people. Nonetheless, we gained insight into the important barriers and facilitators based on inquiry among large samples of the most important stakeholders.

    Conclusions

    In conclusion, this study shows broad support for the use of computer tools in memory clinic practices by clinicians, patients, and care partners. To stimulate the implementation of tools in daily memory practice, the tools should overcome several technical and practical barriers. Moreover, clinicians have to be convinced regarding the reliability and validity of the tool. By identifying the potential barriers and facilitators, we have paved the way for further development and implementation of the tools. Our results provide an important step in the iterative process of developing computer tools for memory clinics in cocreation with end users.

    Acknowledgments

    Research of the Alzheimer’s Center Amsterdam is part of the neurodegeneration research program of Amsterdam Neuroscience. The Vrije Universiteit Medical Center Alzheimer Center is supported by the Stichting Alzheimer Nederland and Stichting Vrije Universiteit Medical Center Fonds. The clinical database structure was developed with funding from the Stichting Dioraphte. The collaboration project Dementia diagnostics using Artificial Intelligence (light weighted computer tool for local memory clinics; project number LSHM19123-HSGF) is cofunded by the public–private partnership allowance made available by Health-Holland, Top Sector Life Sciences and Health, to stimulate public–private partnerships. The chair of WMVDF is supported by the Pasman Stichting. WMVDF and LNCV are recipients of ABOARD (A personalized medicine approach for Alzheimer disease), which is a public–private partnership receiving funding from ZonMW (number 73305095007) and Health-Holland, Top Sector Life Sciences and Health (public–private partnership allowance; number LSHM20106). LNCV is supported by a fellowship grant received from Alzheimer Nederland (WE.15-2019-05). WF, LNV and JG are recipients of the EU Joint Programme - Neurodegenerative Disease Research (JPND) project EURO-FINGERS (ZonMW-Memorabel #733051102) which is supported through the following funding organizations under the aegis of Joint Programme-Neurodegenerative Disease: Finland, Academy of Finland; Germany, Federal Ministry of Education and Research; Spain, National Institute of Health Carlos III; Luxemburg, National Research Fund; Hungary, National Research, Development and Innovation Office; The Netherlands, Netherlands Organisation for Health Research and Development; Sweden, Swedish Research Council (grant agreement INTER/JPND/19/BM/14012609). These funding sources were not involved in the study design, collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.

    Conflicts of Interest

    HFMRM performs contract research for Combinostics; all funding is paid to her institution. WMVDF performs contract research for Biogen. Research programs of WMVDF have been funded by ZorgOnderzoek Nederland, area of Medical Sciences, Dutch Research Council, European Union Framework Programmes, Alzheimer Nederland, Cardiovascular Onderzoek Nederland, Gieskes-Strijbis fonds, Pasman Stichting, Boehringer Ingelheim, Piramal Neuroimaging, Combinostics, Roche BV, AVID. She has been an invited speaker at Boehringer Ingelheim and Biogen. All funding is paid to her institution. FHB performs contract research for Optina Dx and Optos, she has been an invited speaker at Roche and has been invited for expert testimony at Biogen. All funding is paid to her institution. All other coauthors report no conflicts of interest.

    1. The epidemiology and impact of dementia: current state and future trends. World Health Organization. 2015.   URL: https://www.who.int/mental_health/neurology/dementia/dementia_thematicbrief_epidemiology.pdf [accessed 2020-11-16]
    2. Dubois B, Padovani A, Scheltens P, Rossi A, Dell'Agnello G. Timely diagnosis for Alzheimer's disease: a literature review on benefits and challenges. J Alzheimers Dis 2016;49(3):617-631 [FREE Full text] [CrossRef] [Medline]
    3. Prince M, Ferri C. World Alzheimer Report 2011: the benefits of early diagnosis and intervention. Alzheimer’s Disease International (ADI) and Institute of Psychiatry, King's College London. 2011.   URL: https://www.alzint.org/u/WorldAlzheimerReport2011.pdf [accessed 2021-10-29]
    4. Robinson L, Tang E, Taylor J. Dementia: timely diagnosis and early intervention. Br Med J 2015 Jun 16;350:h3029 [FREE Full text] [CrossRef] [Medline]
    5. 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 Nov;18(11):1034-1044. [CrossRef]
    6. Scheltens P, De Strooper B, Kivipelto M, Holstege H, Chételat G, Teunissen CE, et al. Alzheimer's disease. Lancet 2021 Apr;397(10284):1577-1590. [CrossRef]
    7. McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack CR, Kawas CH, et al. The diagnosis of dementia due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 2011 May;7(3):263-269 [FREE Full text] [CrossRef] [Medline]
    8. Román GC, Tatemichi TK, Erkinjuntti T, Cummings JL, Masdeu JC, Garcia JH, et al. Vascular dementia: diagnostic criteria for research studies. Report of the NINDS-AIREN International Workshop. Neurology 1993 Mar;43(2):250-260. [CrossRef] [Medline]
    9. Rascovsky K, Hodges JR, Knopman D, Mendez MF, Kramer JH, Neuhaus J, et al. Sensitivity of revised diagnostic criteria for the behavioural variant of frontotemporal dementia. Brain 2011 Sep;134(Pt 9):2456-2477 [FREE Full text] [CrossRef] [Medline]
    10. McKeith IG, Boeve BF, Dickson DW, Halliday G, Taylor J, Weintraub D, et al. Diagnosis and management of dementia with Lewy bodies: fourth consensus report of the DLB Consortium. Neurology 2017 Jul 04;89(1):88-100 [FREE Full text] [CrossRef] [Medline]
    11. Vos SJ, Verhey F, Frölich L, Kornhuber J, Wiltfang J, Maier W, Alzheimer’s Disease Neuroimaging Initiative, et al. Prevalence and prognosis of Alzheimer's disease at the mild cognitive impairment stage. Brain 2015 May 17;138(Pt 5):1327-1338 [FREE Full text] [CrossRef] [Medline]
    12. Fruijtier AD, Visser LN, Bouwman FH, Lutz R, Schoonenboom N, Kalisvaart K, et al. What patients want to know, and what we actually tell them: the ABIDE project. Alzheimers Dement (N Y) 2020;6(1):e12113 [FREE Full text] [CrossRef] [Medline]
    13. Visser LN, Pelt SA, 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 2020 Feb;23(1):52-62 [FREE Full text] [CrossRef] [Medline]
    14. Kunneman M, Pel-Littel R, Bouwman FH, Gillissen F, Schoonenboom NS, 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 Sep;3(3):314-322 [FREE Full text] [CrossRef] [Medline]
    15. Visser LN, Kunneman M, Murugesu L, van Maurik I, Zwan M, Bouwman FH, et al. Clinician-patient communication during the diagnostic workup: the ABIDE project. Alzheimers Dement (Amst) 2019 Dec;11:520-528 [FREE Full text] [CrossRef] [Medline]
    16. Bruun M, Frederiksen KS, Rhodius-Meester HF, Baroni M, Gjerum L, Koikkalainen J, et al. Impact of a clinical decision support tool on dementia diagnostics in memory clinics: the PredictND validation study. Curr Alzheimer Res 2019;16(2):91-101. [CrossRef] [Medline]
    17. van Maurik IS, Visser LN, Pel-Littel RE, van Buchem MM, Zwan MD, Kunneman M, et al. Development and usability of adappt: web-based tool to support clinicians, patients, and caregivers in the diagnosis of mild cognitive impairment and Alzheimer disease. JMIR Form Res 2019 Jul 08;3(3):e13417 [FREE Full text] [CrossRef] [Medline]
    18. Tolonen A, Rhodius-Meester HF, Bruun M, Koikkalainen J, Barkhof F, Lemstra AW, et al. Data-driven differential diagnosis of dementia using multiclass disease state index classifier. Front Aging Neurosci 2018 Apr 25;10:111 [FREE Full text] [CrossRef] [Medline]
    19. Rhodius-Meester HF, van Maurik IS, Koikkalainen J, Tolonen A, Frederiksen KS, Hasselbalch SG, et al. Selection of memory clinic patients for CSF biomarker assessment can be restricted to a quarter of cases by using computerized decision support, without compromising diagnostic accuracy. PLoS One 2020;15(1):e0226784 [FREE Full text] [CrossRef] [Medline]
    20. Zygouris S, Tsolaki M. Computerized cognitive testing for older adults: a review. Am J Alzheimers Dis Other Demen 2015 Feb;30(1):13-28. [CrossRef] [Medline]
    21. Rhodius-Meester HF, Paajanen T, Koikkalainen J, Mahdiani S, Bruun M, Baroni M, et al. cCOG: a web-based cognitive test tool for detecting neurodegenerative disorders. Alzheimers Dement (Amst) 2020;12(1):e12083 [FREE Full text] [CrossRef] [Medline]
    22. Hansen TI, Haferstrom EC, Brunner JF, Lehn H, Håberg AK. Initial validation of a web-based self-administered neuropsychological test battery for older adults and seniors. J Clin Exp Neuropsychol 2015;37(6):581-594 [FREE Full text] [CrossRef] [Medline]
    23. Licqurish SM, Cook OY, Pattuwage LP, Saunders C, Jefford M, Koczwara B, et al. Tools to facilitate communication during physician-patient consultations in cancer care: an overview of systematic reviews. CA Cancer J Clin 2019 Nov;69(6):497-520 [FREE Full text] [CrossRef] [Medline]
    24. Stacey D, Légaré F, Lewis K, Barry MJ, Bennett CL, Eden KB, et al. Decision aids for people facing health treatment or screening decisions. Cochrane Database Syst Rev 2017 Dec 12;4:CD001431. [CrossRef] [Medline]
    25. Frisoni GB, Boccardi M, Barkhof F, Blennow K, Cappa S, Chiotis K, et al. Strategic roadmap for an early diagnosis of Alzheimer's disease based on biomarkers. Lancet Neurol 2017 Dec;16(8):661-676. [CrossRef] [Medline]
    26. Turner RS, Stubbs T, Davies DA, Albensi BC. Potential New Approaches for Diagnosis of Alzheimer's Disease and Related Dementias. Front Neurol 2020 Jun 5;11:496 [FREE Full text] [CrossRef] [Medline]
    27. Kilsdonk E, Peute LW, Jaspers MW. Factors influencing implementation success of guideline-based clinical decision support systems: a systematic review and gaps analysis. Int J Med Inform 2017 Dec;98:56-64. [CrossRef] [Medline]
    28. Shortliffe EH, Sepúlveda MJ. Clinical decision support in the era of artificial intelligence. J Am Med Assoc 2018 Dec 04;320(21):2199-2200. [CrossRef] [Medline]
    29. Varonen H, Kortteisto T, Kaila M, EBMeDS Study Group. What may help or hinder the implementation of computerized decision support systems (CDSSs): a focus group study with physicians. Fam Pract 2008 Jun;25(3):162-167. [CrossRef] [Medline]
    30. Devaraj S, Sharma SK, Fausto DJ, Viernes S, Kharrazi H. Barriers and facilitators to clinical decision support systems adoption: a systematic review. J Busi Admin Res 2014 Jul 24;3(2):5217. [CrossRef]
    31. Kortteisto T, Komulainen J, Mäkelä M, Kunnamo I, Kaila M. Clinical decision support must be useful, functional is not enough: a qualitative study of computer-based clinical decision support in primary care. BMC Health Serv Res 2012 Oct 08;12:349 [FREE Full text] [CrossRef] [Medline]
    32. Khairat S, Marc D, Crosby W, Al Sanousi A. Reasons for physicians not adopting clinical decision support systems: critical analysis. JMIR Med Inform 2018 Apr 18;6(2):e24 [FREE Full text] [CrossRef] [Medline]
    33. Sutton RT, Pincock D, Baumgart DC, Sadowski DC, Fedorak RN, Kroeker KI. An overview of clinical decision support systems: benefits, risks, and strategies for success. NPJ Digit Med 2020;3:17 [FREE Full text] [CrossRef] [Medline]
    34. Survalyzer - to survey and analyze. Utrecht.   URL: https://education.survalyzer.com/ [accessed 2020-09-11]
    35. Mentimeter. Mentimeter AB, Stockholm, Sweden.   URL: https://mentimeter.com [accessed 2020-12-14]
    36. van der Flier WM, Pijnenburg YA, Prins N, Lemstra AW, Bouwman FH, Teunissen CE, et al. Optimizing patient care and research: the Amsterdam Dementia Cohort. J Alzheimers Dis 2014;41(1):313-327. [CrossRef] [Medline]
    37. van der Flier WM, Scheltens P. Amsterdam Dementia Cohort: performing research to optimize care. J Alzheimers Dis 2018;62(3):1091-1111 [FREE Full text] [CrossRef] [Medline]
    38. Rhodius-Meester HF, van de Schraaf SA, Peters MJ, Kleipool EE, Trappenburg MC, Muller M. Mortality risk and its association with geriatric domain deficits in older outpatients: the Amsterdam Ageing Cohort. Gerontology 2021 Jan 13;67(2):194-201 [FREE Full text] [CrossRef] [Medline]
    39. MAXQDA Standard 18.2.0 ed.   URL: https://www.maxqda.com/new-maxqda-2018 [accessed 2021-10-29]
    40. Clarke V, Braun V. Thematic analysis. J Posit Psychol 2016 Dec 09;12(3):297-298. [CrossRef]
    41. Hsieh H, Shannon SE. Three approaches to qualitative content analysis. Qual Health Res 2005 Nov;15(9):1277-1288. [CrossRef] [Medline]
    42. Masic I, Miokovic M, Muhamedagic B. Evidence based medicine - new approaches and challenges. Acta Inform Med 2008;16(4):219-225 [FREE Full text] [CrossRef] [Medline]
    43. Sittig DF, Wright A, Osheroff JA, Middleton B, Teich JM, Ash JS, et al. Grand challenges in clinical decision support. J Biomed Inform 2008 Apr;41(2):387-392 [FREE Full text] [CrossRef] [Medline]
    44. Bright TJ, Wong A, Dhurjati R, Bristow E, Bastian L, Coeytaux RR, et al. Effect of clinical decision-support systems: a systematic review. Ann Intern Med 2012 Jul 3;157(1):29-43. [CrossRef] [Medline]
    45. Liu JL, Wyatt JC. The case for randomized controlled trials to assess the impact of clinical information systems. J Am Med Inform Assoc 2011;18(2):173-180 [FREE Full text] [CrossRef] [Medline]
    46. Murray-Torres T, Casarella A, Bollini M, Wallace F, Avidan MS, Politi MC. Anesthesiology Control Tower-Feasibility Assessment to Support Translation (ACTFAST): mixed-methods study of a novel telemedicine-based support system for the operating room. JMIR Hum Factors 2019 Apr 23;6(2):e12155 [FREE Full text] [CrossRef] [Medline]
    47. Kelly CJ, Karthikesalingam A, Suleyman M, Corrado G, King D. Key challenges for delivering clinical impact with artificial intelligence. BMC Med 2019 Oct 29;17(1):195 [FREE Full text] [CrossRef] [Medline]

    ‎
    EBM: evidence-based medicine
    NGN: Nederlands Geheugenpoli Netwerk
    RCT: randomized controlled trial
    SCD: subjective cognitive decline

    Edited by G Eysenbach; submitted 08.06.21; peer-reviewed by S Hasselbalch, D Hu; comments to author 08.08.21; revised version received 07.09.21; accepted 18.09.21; published 03.12.21

    Copyright

    ©Aniek M van Gils, Leonie NC Visser, Heleen MA Hendriksen, Jean Georges, Majon Muller, Femke H Bouwman, Wiesje M van der Flier, Hanneke FM Rhodius-Meester. Originally published in JMIR Formative Research (https://formative.jmir.org), 03.12.2021.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Formative Research, is properly cited. The complete bibliographic information, a link to the original publication on https://formative.jmir.org, as well as this copyright and license information must be included.