Edited by: Ying Xu, University at Buffalo, USA
Reviewed by: Gianfranco Spalletta, Fondazione Santa Lucia (IRCCS), Italy; Michele Fornaro, New York State Psychiatric Institute, Columbia University, USA
*Correspondence: Auriane Gros
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution and reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Information and communication technologies (ICT) currently represent pervasive assistive tools for our daily life, and their use is fast growing also in the health domain, where they can improve diagnosis and stimulation strategies in many medical fields. In neuropsychiatry, the development of ICT is changing the use of conventional assessment methods, by progressively integrating computer technologies in their methods of administration, in order to obtain a more reliable and reproducible assessment (Epstein and Klinkenberg,
For example ICT includes:
Computerized cognitive tests: they are more and more suited to tactile tablet format and can be used in nursing homes and also at home (Brown et al.,
Serious games: they allow the management and training of different cognitive deficits (Robert et al.,
Virtual reality: it is used particularly in the training of visuospatial abilities, memory functions, attention, speed monitoring (Legault et al.,
Wearable sensors and environmental sensors: they can be used for the evaluation of walking abilities (Moufawad El Achkar et al.,
Despite several aforementioned limitations, scientific literature on ICT is reporting an increasing number of studies using ICT dedicated to different health-related conditions. In the field of aging, recommendations for the use of ICT and Serious Games have been recently proposed in order to provide new insights and marks in this emerging field for the diagnosis and management of frailty and dementia (Robert et al.,
ICTs may also constitute an added value for the assessment and follow-up of several clinical parameters for which evaluation biases are commonly reported. This is particularly true for neuropsychiatric symptoms, for which there is: (1) a lack of objectivity of current assessment questionnaires and interviews; (2) subjectivity linked to the patient’s self-reports; (3) subjectivity of the rater; (4) cultural differences in symptoms expressivity; (5) agnosia due to cognitive impairment (Orfei et al.,
For affective disorders, several authors have already reported promising results to improve the quality of assessment methods with the integration of physiological parameters of emotional states: voice analysis showing specific patterns in depressed individuals (Mundt et al.,
Despite previous results, affective disorders may have specific characteristics in aging populations that need to be taken into account to provide dedicated ICT-based solutions.
In elderly populations, affective disorders may have specific clinical symptoms compared to younger adults (Valiengo et al.,
Depression is frequently over diagnosed in elderly populations due to the clinical misdiagnosis with apathy. Apathy is a core symptom of depression, associated to sadness, and is frequently observed in aging. Persons with persistent apathy were less likely to recover from depression than those who remitted from apathy and persistent apathy was associated with worse outcome of depression (Groeneweg-Koolhoven et al.,
Several age-related biological changes that usually interfere with stress and affective phenomenology, such as the hypothalamo-pituitary axis (likely to be less downregulated in aging) with changes in chronic cortisol secretions, may also partially explain the variability in affective expressivity across ages (Lenze and Wetherell,
The presence of cognitive disturbances and eventually anosognosia often leads to assessment issues when interviewing elderly individuals with affective symptoms. Current assessment methods for affective disorders, mainly relying on interviews and questionnaires with the patient and the family caregivers, may lack objectivity especially when patient’s and/or caregiver’s insight is impaired, despite this they have usually been validated and are currently used in daily clinical practice. Also higher levels of burden related to care were associated with higher levels of perceived stress and cognitive difficulties experienced which shows that stress and burden of caregivers makes judgment difficult (Luchesi et al.,
However, “pencil-paper” tests can encompass differences in instructions depending on the evaluators, consecutive errors in manual data entry, and errors in the transfer of such data to statistical software. To reduce these biases and to obtain a more reliable assessment of a given construct, ICTs have enabled the development of computer-based testing (Epstein and Klinkenberg,
Other measurements from the ICT and assessing the physiological component of emotions grow more and more, as they avoid the drawbacks and the subjective nature of self-report questionnaires. These measures include skin conductance response (SCR) and HR using wearable sensors placed on the skin. It has been shown a deceleration of HR during negative emotions (Palomba et al.,
The IA workshop 2015, organized by the Cognition-Behavior-Technology (CoBTek) research team of the University of Nice-Sophia Antipolis, took place in Nice on November 12th 2015.
Using a Delphi method, an expert panel involving at least 25 professionals was chosen to respond to the following aim: “to propose recommendations for the use of ICT in elderly populations with affective disorders”. Four analysts (two psychiatrists and psychologists, two ICT engineers) were involved in the selection of the experts, in the development and analysis of a questionnaire built to drive discussions toward recommendations, and in the editing process of the final report. The analysts and the majority of the experts have been involved in the previous recommendations published during IA workshops 2013 and 2014 (Robert et al.,
Participants were health care professionals (psychiatrists (
Experts were then divided into four-person subgroups (with at least one psychiatrist and one psychologist in each subgroup), and asked to answer questions related to the use of ICT for affective disorders as well as to perform a Strengths, Weaknesses, Opportunities, Threats (SWOT) analysis on the same topic.
Many of the questions asked during this face-to-face session were derived from the results of our previous workshops (for which population and severity stages? what type of ICT should be preferred? for what purpose? which affective disorders should be more specifically targeted? which frequency of use? where? with whom?).
Based on the workshop results, a web survey was designed and sent to local general practitioners (GP) involved in the diagnosis and management of both affective disorders and Alzheimer’s disease and related disorders in elderly populations, and to all interns in psychiatry in France (in order to capture the overall feeling of current and future physicians regarding the use of ICTs in their daily practice). GPs are not specialists for the diagnosis and management of affective disorders but they often constitute first-line practitioners and their feedback thus appears to be valuable. The internship in psychiatry in France has a 4-year duration. All interns from 1st, 2nd, 3rd and 4th year received a web invitation to participle in the survey. All GPs working in the city of Nice also received the web invitation to participate in the survey. Participants were presented with the list of questions and items employed during the Workshop and asked to rate each item on a 0–2 scale (0 = not adapted at all; 1 = neutral; 2 = very adapted).
One-hundred and forty-one interns in psychiatry and 34 GPs have responded to the survey.
The following diagnostic criteria related to specific medical conditions with cognitive decline were considered in the writing process of the present recommendations for the use of ICT: frailty syndrome (Fried et al.,
In each four-participant subgroup there was at least one psychiatrist and one psychologist. Other participants included physicians (GPs, neurologists, geriatricians), ICT engineers or social workers. Participants were asked to discuss a set of several questions regarding the use of ICT for the diagnosis, follow-up and management of affective disorders in elderly populations, with and without associated cognitive disturbances (from Mild Cognitive Impairment to Dementia). All responses were then synthesized and summarized as follows.
Subgroups reported an overall interest in using ICTs for individuals with Mild Cognitive Impairment, frailty syndrome (Fried’s criteria), or mild to moderate dementia syndrome (MMSE between 26 and 16), but did not recommend using ICT for severe stages of dementia (MMSE < 15). The severity of the cognitive decline is expected to decrease the ICT acceptability (lack of understanding of ICTs utility, feeling of intrusive devices) and to induce anxiety or delusion. Also, patients with severe cognitive decline may be unable to give an opinion about the device’s functioning, and have usage difficulties, which make the device not appropriate to their needs (Faucounau et al.,
Experts reported that all types of devices and/or solutions could be considered, but should take into account the subjects’ sensory impairments, cultural aspects and associated cognitive disturbances.
For example, subjects with hearing impairment should not be assessed using speech analysis solutions or ICTs with verbal instructions. Subjects with hallucinations or delusions should not be evaluated using skin sensors or environmental sensors.
Results from recent studies have shown that several stimulation-oriented ICT could likely improve affective affects, such as using serious games on tablet showing interest for apathetic individuals (Manera et al.,
Regarding affective disorders more specifically, VR solutions have shown interesting results for the management of anxiety-related disorders in adults (Parsons and Rizzo,
ICT-based solutions could be used for the assessment in elderly populations with affective disorders through more objective assessments, including devices or solutions which are able to capture physiological parameters related to emotional states.
ICTs could be used for the assessment of individuals with limited mobility or living far from specialized centers.
ICTs could be used for in-home monitoring in order to have continuous measures (24/7-monitoring), as well as for the regular follow-up of outpatients consulting in specialized settings.
ICT could be used for the non-pharmacological management of affective disorders.
For example, VR has been used extensively in the treatment of phobias such as claustrophobia (Bruce and Regenbrecht,
Experts reported that anxiety disorders should be the main target of ICT use, but also depressive symptoms and apathy may represent interesting targets.
However experts believe that psychotic disorders are not prime targets for ICT, as well as disorders with significant cognitive deficits.
Experts reported that ICT could be used on demand for individuals at home, and likely with a more organized schedule for individuals in hospitals or long-term care facilities due to staff constraints.
The experts suggested that in some cases the frequency of use should be indicated by the doctor or be defined in the context of a medical prescription.
However, home video sensors for monitoring must be installed for sufficient time to be able to identify regular events.
Participants proposed several locations for the use of ICT: at home, in hospital, in long-term care facilities, as well as waiting room of GP’s office.
The choice of ICTs will depend on the location:
At home, environmental sensors can be recommended to monitor several behavioral changes (nighttime behaviors, wandering) that could indirectly reflect emotional changes.
At the hospital, wearable sensors are the most recommended in order to provide objective and controlled assessments.
In GP’s offices, computerized tests are more recommended because of the possibility to control the understanding of instructions.
Several experts reported the need to use ICT after medical prescription to give a frame in the use of these new types of devices. The importance of the GP has been highlighted in the use of ICT. Several reasons were advanced: (1) GPs are first-line physicians and could contribute to improve prevention and early diagnosis of affective disorders using objective methods involving ICTs, despite their lack of training in the screening of affective disturbances; and (2) GPs’ consultations are usually too short to fully capture patients’ global medical status. ICT could be used, for example, in the waiting room, to start recording several relevant medical parameters that could be provided to the GP during the consultation. Also, involving GPs, or caregivers, will allow them to explain to the patients how they can use ICTs, at what times and remind them the instructions required. This would enable to customize ICT to each patient. In care facilities, use of ICT should be done under staff supervision.
We also asked the panel to conduct a SWOT analysis for the use of ICT in affective disorders among elderly individuals. A summary of the SWOT analysis is presented in Table
Strenghts | Weaknesses |
---|---|
Objective evaluation | Devices too sophisticated and complicated to use |
Interface adapted and tailored to the user, | Expensive equipments |
Possibility to use at home | Addiction. |
Real-time feedback delivery for the user and the professionals | Risk of overdiagnosis |
Improve screening and early diagnosis, improvement of mass screening | Risk to induce new symptoms for advanced dementia stage (anxiety, delirium, persecution) |
Possibility to provide prolonged or continuous evaluation | Risk of decreasing social, familial and outdoors activities |
Geographic equity | Ethical challenges |
Adapted to the new generations | Perception that ICT will replace clinicians, disappearance of human relations |
Provide homogeneous therapeutic actions in care structures, provide therapeutic actions on patients’ living place | Negative representation (intrusive devices and privation violation) |
Possibility to provide professional training | Risk of standardization, separate affective disorders from temper, personal history, traumatic life events |
Embedded physiological measure of affective symptoms (respiratory and heartbeat, sudation, …) | Lack of consensus on specific markers for affective disorders |
It is recognized that ICT allows a better homogenization in the awarding and issuing instructions during evaluations (Barak,
Finally, ICT allow more objective evaluations because they limit the consecutive errors deriving from a manual data entry, and allow an easier transfer of data to statistical software (Chuah et al.,
In the context of ICT it is possible to create interfaces adapted to the users. One of the advantages of ICT is that it is possible to individualize and personalize the evaluation. Adaptation is important for elderly people because they can have sensorial problems (auditory, visual) and memory problems, which require repeating instructional cues.
The home-based and independent practice would contribute to reduce the burden to public healthcare system. ICT can promote activity automation, and generalization of the learned skills to every-day behavior. The possibility to carry out an autonomous activity can also lead to mood improvements and self-esteem enhancement. Finally, the possibilities to use ICT at home represent an opportunity for a safe testing and training environment, minimizing risks due to travel.
ICT offer the therapists and caregivers the possibility to record and visualize the activity immediately, while the classical evaluation of patient performances typically involves a
The frequency and duration of medical consultations are limited, which may bias the assessment results because the patient, for instance, may be tired during the consultation. Motivation is important not only for a successful care but also for the scores obtained during an evaluation. A continuous and repeated analysis can avoid the bias due to the moment when the evaluation took place.
The ICTs for the larger dissemination of evaluations may allow evaluating a larger portion of the population compared to conventional ratings, which require the presence of a therapist. Similarly, ICTs, by finer assessments, are more sensitive methods.
Many elderly people are not used to interacting with high-tech interfaces, and the initial approach with ICT may cause them a high cognitive load. This high cognitive load can represent a bias for the assessment because it can make the task more difficult than what it actually is. Also to employ ICT solutions it is often necessary to connect cables and modify setups, which may be too difficult for non-expert users, especially for the elderly. Finally it may be also difficult for the therapists to use the applications to visualize performance results and analyze data.
Equipment for ICT is a stuff that we already have at home such as tablets, television, PC and laptops. However, some ICT require the purchase of less common and more expensive materials such as CAVE or immersive screen. Equipment costs may risk limiting the use of ICT to people who have the most money, thus creating discrimination.
Some ICT, especially for assessment, require the installation of video-cameras or sensors in patients’ homes. These sensors can be perceived as intrusive in the patient’s privacy, in particular when they must be installed in the bedroom to evaluate sleep disorders.
As with video games, new technologies can create addictions for patients, especially if their use is not governed by rules and indeed, currently, the use of new technologies in patients’ homes cannot be controlled, which makes it dangerous.
It appears that the use of ultra-sensitive diagnostic techniques is most often accompanied by a significant increase in positive diagnosis, seeming to be more of overdiagnosis. ICT, by finer assessments could thus result in increased diagnosis of affective disorders and therefore an overdiagnosis.
Several types of ICTs can lead to other symptoms in a patient. For example, the installation of cameras at a patient’s home can cause delusions of persecution, the use of VR can lead to a phobia or anxiety, use of sensors on the skin can generate hypochondria. ICTs are intrusive sensors and are therefore particularly at risk to generate a secondary pathology.
The use of ICT is often individual and therefore can generate a feeling of not needing the others, thereby reducing external activities and social relations in the patient. Also, very few ICT are used outdoors and therefore it does not encourage the patient to go outside.
Patients who live far from the cities go less often to the doctor or the therapist and therefore have less regular care than people living in the city. ICTs can reduce this inequality by enabling patients living far from the cities to access care without having to move and therefore to be able to have an evaluation and a more regular care.
The elderly in the future will be the youth of today who are more accustomed to new technologies than to the “paper and pencil” tasks. Indeed the new generations are faster with ICT. New technologies are intuitive to them and are therefore better for the assessment and the training.
It is difficult to assess and take care of a patient in an environment different from where he/she lives. Indeed, the patient can succeed in the classical “paper and pencil” assessments, but no have autonomy in real activities of daily living, or be autonomous in daily life and still fail in “paper and pencil” assessments. ICT, analyzing the behavior at home, can allow a more accurate assessment of patient skills. Furthermore, the use of ICT in nursing homes can enable taking care of the patients in an automated and therefore more regular manner, overcoming the problem of the limited care giver availability.
ICTs, with wide distribution, can afford to train a large number of professionals at the same time, where they want and when they want. Thus, these technologies make easier access to assessment methods and specific support that were previously reserved for specialists.
Physiological measurements allow an objective assessment of affective disorders and are not dependent on factors such as social status or culture. In addition, these signals, such as the heartbeat, or the skin conductance, allow for a more accurate and sensitive measurement.
The use of ICT poses a number of ethical questions such as:
The respect of privacy and private life, the processing of the images (using video sensors).
The data processing and storage.
These ethical questions are even more compelling for older and frail people who cannot always accurately understand the functioning of ICT.
Some uses of ICT can promote short and little personalized exchanges between patient and therapist. In those situations where therapists have little time for more informal exchanges, the patient can develop a sense of isolation, an erosion of the sense of caring and more overall a formalization of exchanges.
For economic reasons and/or for easier and faster usability, ICTs might be used in a standardized way that could minimize the interest of ICTs to propose tailored assessments and managements.
Several ICT use the video sensors to evaluate affective disorders. These sensors analyze behavioral disorders but the specific behaviors reflecting emotional problems still require a better specification and precision. Similarly, data on voice markers, are so far not numerous enough to be certain of their reliability.
One-hundred and forty-one interns in psychiatry and 34 GPs responded to the survey.
The questionnaire results showed that, in the general population, the pathologies considered the most difficult to identify for interns in psychiatry were bipolar disorder (28.6%), followed by emotional lability (25%) and generalized anxiety disorder (14.3%). GPs identified Obsessive Compulsive Disorder (38.2%), phobic disorders (38.2%) and depressive episode (23.5%) as difficult diagnoses.
Around 25% of participants (both in interns and GPs subgroups) found ICTs relevant for the detection of affective disorders (with higher rates of responses considering ICT inappropriate for affective disorders). Among positive responders, the use of computerized tools was considered more relevant compared to wearable sensors. ICTs were considered as promising tools for the patients’ follow-up, after the diagnosis has been made, as well as for non-pharmacologic strategies (see Table
Considering elderly populations and subjects living with dementia (Table
The difference of opinion of interns and GPs was also significant in the relevance of using ICTs as treatment (
Among the most difficult disorders to identify, GPs found that ICTs are irrelevant, whether for obsessive-compulsive disorder (54.5%), phobic disorders (44.1%) or depressive episodes (63.6%).
We noted a significant difference with interns who found that the use of ICT can help in screening depressive episodes (
Finally, the survey revealed that interns and GPs find the use of new technologies more adapted for healthy elderly people than for AD patients (Figure
Several major conclusions can be drawn after the general questions, SWOT analysis and web survey.
First, ICT could contribute to provide better recognition and earlier diagnosis of affective disorders in elderly populations, because ICT could be designed to specifically recognize affective symptoms using specific and dedicated algorithms. However, in order to become clinically accepted, therapists should undergo ICT trainings to learn how to employ them correctly and in a standardized way.
Second, first-line physicians, mainly GPs, could benefit from ICTs: (1) as they are usually “reference physician” (and officially declared to the health regulatory authorities for health coverage), receive more patients compared to specialists, and often consult in patients’ home; and (2) as they usually lack specific training to recognize affective symptoms despite being first-line practitioners. In this line, use of ICT aimed at recognizing affective symptoms could help GPs in their daily practice but could also contribute to large population screening. The limited amount of responses gathered from GPs as well as several comments noticed after our web survey tend to indicate an overall lack of knowledge on ICT capabilities for health management. In this line, increased diffusion of ICT is required among physicians.
Third, ICTs for affective disorders should embed measures of physiological parameters linked to affects and emotions such as respiratory frequency, heartbeat, skin conductance, and so on. However, more specific markers of affective disorders are likely needed to be recognized to better participate in the assessment and follow-up process of affective symptoms in addition to current clinical markers. Participants reported the use of ICT at home as a strength for affective disorders. Developers and clinicians should thus also discuss the use of devices with ambulatory affective markers that could be sensitive to at-home changes. Participants in the web survey did not consider the use of wearable sensors as adapted, especially among interns in psychiatry.
Fourth, ICTs could be helpful in prevention, diagnosis and non-pharmacologic treatment for several affective symptoms (mainly anxiety disorders and depressive symptoms). Experts pointed out that ICTs could potentially have a negative impact for individuals with advanced dementia stages (increased risk of delirium avec persecution, increased consecutive anxiety) whereas they appear to be adapted for preclinical, frail and early to moderate stage demented individuals.
We found important differences between the reports of the expert panel and GPs and interns interviewed in the web survey. Indeed, the experts recommended the use of ICT for patients with AD, while the majority of GPs and interns suggested that ICTs are not adapted for AD patients.
Similarly, while experts recommend the use of ICTs for evaluation, monitoring and treatment, interns found ICTs adapted only for the monitoring and treatment and GPs, more generally, found ICT irrelevant.
Nevertheless the percentage of neutral responses showed the interest of better informed interns and GPs on ICT interest and opportunities. Indeed we can assume that the persons interviewed found use of ICT not adapted to individuals with AD because, unlike the experts, they do not use ICT’s in their daily practice and they are poorly aware of the available tools.
Opinions are divergent between GPs and interns particularly for the use of computerized tests for diagnosis. We can assume that the interns are part of a generation accustomed to these type of tests, whose use has grown significantly in the clinical practice in recent years.
Using wearable sensors is particularly not adapted to AD patients according to interns (71%). This skepticism may derive from the fact that interns, most likely, use ICT very rarely in their daily practice. But this skepticism may derive also from the fact that wearable sensors can be particularly difficult to use with patients often agitated, presenting aggressive behaviors and which have aberrant wondering behaviors.
In the web survey it is interesting to note that disorders were considered the most difficult to identify by GPs and were those most easily identified by interns, with a significant difference for depressive episode (
Finally, several topics could be discussed on:
–Psychotropic medications:
Several psychotropic medications have shown a deleterious impact on cognitive abilities (e.g., long term exposition to benzodiazepines (Shash et al.,
Thus, all types of psychotropic medications known to have negative impact on vigilance and motivation could potentially reduce the engagement of patients in using ICTs when proposed as long-term monitoring tools and/or non-pharmacologic approaches (i.e., stimulation with serious games).
Additionally, ICT represents an opportunity to provide non-pharmacologic approaches (through serious games for example) in neurodegenerative disorders and affective disturbances (Bruce and Regenbrecht,
–Role of caregivers:
Experts have discussed the role of professional and family caregivers when using ICT. Several aspects could be discussed: what is the influence of caregivers on the willingness of patients to use ICT? To our knowledge, this aspect lacks scientific references due to the recent emergence of ICT. Based on our experience, the use of ICT in long-term care facilities often requires the prompt of the professional caregivers, especially for people with severe dementia. For community-dwelling individuals, the feeling is similar and the use of ICT is often influenced by the presence of a cognitively healthy family caregiver living with the patient. Regarding affective disturbances more specifically, many emotional disturbances (anxiety-related disturbances, post-traumatic stress disorder) are often comorbid with depressive episodes and associated to social isolation, lack of interest and lack of motivation. Lack of motivation is associated with lack of initiation in activities that would therefore reduce patients’ engagement in stimulating activities (such as using serious games, for example, that have to be considered as part of ICT). Thus, the presence of affective disturbances could likely decrease the patients’ willingness to participate in stimulating activities. ICT could potentially offer a way to engage in socially stimulating activities (Coathup et al.,
–Impact of setting:
The point has been discussed by the experts but the literature on ICT is also lacking. According to our experience, the presence of professional caregivers (when trained) in long-term care facilities might increase the patients’ stimulation for activities and potentially the use of ICT designed for stimulation (such as serious games). Additionally, many assistive ICT designed to improve patients’ full-time monitoring in long-term care facilities should offer more availability for professional caregivers by decreasing the time devoted to patients’ surveillance, thus potentially increasing interactions between patients and caregivers for stimulating activities.
AG: drafting the work, conception, interpretation of data. DB: drafting the work, conception, acquisition of data. VM: revising the work, design, interpretation of data. RF: drafting the work, analysis of data. A-MZ-C, ST and MB: drafting the work, acquisition of data. PR: revising the work, acquisition of data. RD: drafting and revising the work, design, interpretation of data.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The Workshop expert panel 2015: MB, Claude Baudu, François Bremond, Jean Claude Broutart, A-MZ-C, Vanessa Cecchetti, Antitza Dantcheva, Guy Darcourt, Nelly Darmon, Renaud David, David Bensamoun, Audrey Deudon, Julie de Stoutz, Roxane Dilcher, Jean Didier Eberhardt, Catherine Felix, Nathalie Fernandez, Cindy Giaume, AG, Sophie Matharan, Marie-Line Menard, Francoise N’Gabira, Marie Pierre Pancrazi, Frederic Prate, Prescillia Pierron, Serge Ricaud, PR, Viviane Roulx-Laty, Florent Salducci, Alexandra Taylor, Wendling Marie.