Artificial Intelligence: An Interprofessional Perspective on Implications for Geriatric Mental Health Research and Care

Artificial intelligence (AI) in healthcare aims to learn patterns in large multimodal datasets within and across individuals. These patterns may either improve understanding of current clinical status or predict a future outcome. AI holds the potential to revolutionize geriatric mental health care and research by supporting diagnosis, treatment, and clinical decision-making. However, much of this momentum is driven by data and computer scientists and engineers and runs the risk of being disconnected from pragmatic issues in clinical practice. This interprofessional perspective bridges the experiences of clinical scientists and data science. We provide a brief overview of AI with the main focus on possible applications and challenges of using AI-based approaches for research and clinical care in geriatric mental health. We suggest future AI applications in geriatric mental health consider pragmatic considerations of clinical practice, methodological differences between data and clinical science, and address issues of ethics, privacy, and trust.

Download Full-text

1636. Variation in Clinical Practice Patterns Among Infectious Diseases Faculty at a Large Academic Institution

Open Forum Infectious Diseases ◽

10.1093/ofid/ofy209.106 ◽

2018 ◽

Vol 5 (suppl_1) ◽

pp. S44-S44

Author(s):

Varun K Phadke ◽

Jennifer O Spicer

Keyword(s):

Clinical Practice ◽

Clinical Decision Making ◽

Practice Patterns ◽

Clinical Care ◽

Clinical Decision ◽

Educational Interventions ◽

Fixed Dose ◽

Published Data ◽

Routine Practice ◽

Clinical Practice Patterns

Abstract Background Clinical practice patterns vary between providers, but few studies have examined this variation among infectious disease (ID) physicians. Characterizing these differences in practice can help identify areas where targeted educational interventions or further research are needed to improve clinical decision-making. We describe a faculty survey conducted at our institution designed to identify clinical practice variation within a large academic ID division. Methods In January 2017, an electronic survey was distributed to all clinical ID faculty at our institution. The survey collected baseline demographic information as well as responses to 28 common clinical dilemmas encountered in routine practice. Descriptive statistics were performed. Results Twenty-four (44%) of 54 active clinical ID faculty (12 assistant professors, 6 associate professors, and 6 professors) completed the survey. Examples of clinical dilemmas with >80% agreement among faculty included: (1) S. aureus bacteremia should be a mandatory ID consult (88%) and (2) lumbar puncture should be performed for all patients with suspected ocular syphilis (88%). The majority of clinical dilemmas had less than 80% agreement, and these spanned the range of routine ID practice. Examples included: (1) use of ceftriaxone for outpatient antibiotic therapy for nonbacteremic invasive methicillin-susceptible S. aureus infections (58% agree), (2) length of treatment for guideline-defined uncomplicated S. aureus bacteremia (50% 2 weeks, 50% 4 weeks), (3) use of fixed-dose dolutegravir/abacavir/lamivudine as a single-drug regimen for an HIV-infected patient with an M184V mutation (42% agree), and (4) benefit of routine anal Pap smears among HIV-infected men who have sex with men (50% agree). Conclusion Practice patterns vary between ID physicians within our institution, particularly for clinical dilemmas for which there is insufficient or conflicting published data. Further studies to examine practice pattern variation among ID physicians across institutions and geographic regions could identify areas where further research or educational interventions are needed to enhance clinical care. Disclosures All authors: No reported disclosures.

Download Full-text

Artificial Intelligence, Heuristic Biases, and the Optimization of Health Outcomes: Cautionary Optimism

Journal of Clinical Medicine ◽

10.3390/jcm10225284 ◽

2021 ◽

Vol 10 (22) ◽

pp. 5284

Author(s):

Michael Feehan ◽

Leah A. Owen ◽

Ian M. McKinnon ◽

Margaret M. DeAngelis

Keyword(s):

Artificial Intelligence ◽

Health Outcomes ◽

Clinical Decision Making ◽

Clinical Care ◽

Clinical Decision ◽

Health Inequity ◽

Great Promise ◽

Patient Health ◽

Improve Patient ◽

Regulatory Landscapes

The use of artificial intelligence (AI) and machine learning (ML) in clinical care offers great promise to improve patient health outcomes and reduce health inequity across patient populations. However, inherent biases in these applications, and the subsequent potential risk of harm can limit current use. Multi-modal workflows designed to minimize these limitations in the development, implementation, and evaluation of ML systems in real-world settings are needed to improve efficacy while reducing bias and the risk of potential harms. Comprehensive consideration of rapidly evolving AI technologies and the inherent risks of bias, the expanding volume and nature of data sources, and the evolving regulatory landscapes, can contribute meaningfully to the development of AI-enhanced clinical decision making and the reduction in health inequity.

Download Full-text

Psychiatric ethics

New Oxford Textbook of Psychiatry ◽

10.1093/med/9780199696758.003.0006 ◽

2012 ◽

pp. 28-32

Author(s):

Sidney Bloch ◽

Stephen Green

Keyword(s):

Mental Health ◽

Decision Making ◽

Clinical Practice ◽

Clinical Decision Making ◽

Ethical Dilemmas ◽

Theoretical Perspective ◽

Codes Of Ethics ◽

Clinical Decision ◽

Mental Health Practice ◽

Ethical Problems

A myriad of ethical problems pervade clinical practice and research in psychiatry. Yet with few exceptions, psychiatric ethics has generally been regarded as an addendum to mainstream bioethics. An assumption has been made that ‘tools’ developed to deal with issues like assisted reproduction or transplant surgery can be used essentially unmodified in psychiatry. These tools certainly help the psychiatrist but the hand-me-down approach has meant that salient features of psychiatric ethics have been prone to misunderstanding. Psychiatric ethics is concerned with the application of moral rules to situations and relationships specific to the field of mental health practice. We will focus on ethical aspects of diagnosis and treatment that challenge psychiatrists, and on codes of ethics. Resolution of ethical dilemmas requires deliberation grounded in a moral theoretical framework that serves clinical decision-making, and we conclude with our preferred theoretical perspective.

Download Full-text

A Mobile Sensing App to Monitor Youth Mental Health: A Feasibility Study (Preprint)

10.2196/preprints.20638 ◽

2020 ◽

Author(s):

Banuchitra Suruliraj ◽

Dominik Gall ◽

Lucy MacLeod ◽

Kitti Bessenyei ◽

Alexa Bagnell ◽

...

Keyword(s):

Mental Health ◽

Clinical Decision Making ◽

Clinical Care ◽

Active Role ◽

Clinical Decision ◽

Mobile Sensing ◽

Clinical Monitoring ◽

Health States ◽

Clinical Symptomatology ◽

Prediction Tools

BACKGROUND Current methods of predicting mental health crises usually rely on subjective symptom ratings obtained at discrete time points during routine clinical care. But clinical decision-making based on such subjective information is challenging, as changes in symptoms might be sub-threshold, context dependent, or variable over time. Therefore, novel prediction tools need to be developed meeting the highest standards of reliability, feasibility, scalability, and affordability. Smartphones might configure such prediction tools, as they are ubiquitous and afford a wide variety of types of behavioural data that can be automatically recorded by their built-in sensors. OBJECTIVE To facilitate the collection of high-quality, passive mobile sensing data, we built the Predicting Risk and Outcomes of Social InTeractions (PROSIT) tool, a mobile sensing app that runs on both Android and iOS operating systems. In addition we aimed to ensure the acceptability and usability of this tool in youth. METHODS The PROSIT tool captures multiple indices of a youth’s daily life behaviors via their naturalistic use of a smartphone. These indices include physical activity, geolocation, sleep, phone use, typed text, music choice, and acoustic vocal quality. Importantly, the PROSIT tool records most of these data passively with only minimal burden to youth. All the time-intensive, detail-rich data streams that the tool captures to make inferences about youth’s mental health states are encrypted and uploaded to a secure server at our clinic. Although other mobile data collection tools exist, the PROSIT tool places a unique emphasis on the designing the tool for youth. RESULTS In a pilot study (N=61), participants tolerated the PROSIT tool well, reporting only minimal burden. Over 85% of youth were using the tool for the whole study period, although they were suffering from severe clinical symptomatology. But not only youth accepted the PROSIT tool well, for youth under the age of 15 we requested consent of parents, which 80% of parents provided. CONCLUSIONS The PROSIT tool offers novel way for clinical monitoring in youth with mental disorder. The high acceptability rates indicate that mobile sensing technologies can successfully be used even in youth with severe clinical symptomatology. We built the PROSIT tool to assist in clinical monitoring, with the ultimate goal of leveraging individual big data to empower youth patients to take on a more active role in the management of their clinical symptomatology.

Download Full-text

Use of Machine Learning and Artificial Intelligence Methods in Geriatric Mental Health Research Involving Electronic Health Record or Administrative Claims Data: A Systematic Review

Frontiers in Psychiatry ◽

10.3389/fpsyt.2021.738466 ◽

2021 ◽

Vol 12 ◽

Author(s):

Mohammad Chowdhury ◽

Eddie Gasca Cervantes ◽

Wai-Yip Chan ◽

Dallas P. Seitz

Keyword(s):

Mental Health ◽

Artificial Intelligence ◽

Machine Learning ◽

Systematic Review ◽

Health Research ◽

Mental Health Research ◽

Risk Of Bias ◽

Geriatric Mental Health ◽

Electronic Health

Introduction: Electronic health records (EHR) and administrative healthcare data (AHD) are frequently used in geriatric mental health research to answer various health research questions. However, there is an increasing amount and complexity of data available that may lend itself to alternative analytic approaches using machine learning (ML) or artificial intelligence (AI) methods. We performed a systematic review of the current application of ML or AI approaches to the analysis of EHR and AHD in geriatric mental health.Methods: We searched MEDLINE, Embase, and PsycINFO to identify potential studies. We included all articles that used ML or AI methods on topics related to geriatric mental health utilizing EHR or AHD data. We assessed study quality either by Prediction model Risk OF Bias ASsessment Tool (PROBAST) or Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) checklist.Results: We initially identified 391 articles through an electronic database and reference search, and 21 articles met inclusion criteria. Among the selected studies, EHR was the most used data type, and the datasets were mainly structured. A variety of ML and AI methods were used, with prediction or classification being the main application of ML or AI with the random forest as the most common ML technique. Dementia was the most common mental health condition observed. The relative advantages of ML or AI techniques compared to biostatistical methods were generally not assessed. Only in three studies, low risk of bias (ROB) was observed according to all the PROBAST domains but in none according to QUADAS-2 domains. The quality of study reporting could be further improved.Conclusion: There are currently relatively few studies using ML and AI in geriatric mental health research using EHR and AHD methods, although this field is expanding. Aside from dementia, there are few studies of other geriatric mental health conditions. The lack of consistent information in the selected studies precludes precise comparisons between them. Improving the quality of reporting of ML and AI work in the future would help improve research in the field. Other courses of improvement include using common data models to collect/organize data, and common datasets for ML model validation.

Download Full-text

The Medical Model in Mental Health

10.1093/med/9780198807254.001.0001 ◽

2019 ◽

Cited By ~ 7

Author(s):

Ahmed Samei Huda

Keyword(s):

Mental Health ◽

Health Research ◽

Teaching And Learning ◽

Clinical Decision Making ◽

Mental Health Problems ◽

Medical Model ◽

General Medicine ◽

Clinical Information ◽

Clinical Decision ◽

Mental Health Research

The medical model is a biopsychosocial model assessing a patient’s problems and matching them to the diagnostic construct using pattern recognition of clinical features. Diagnostic constructs allow for researching, communicating, teaching, and learning useful clinical information to influence clinical decision-making. They also have social and administrative functions such as access to benefits. They may also help explain why problems occur. Diagnostic constructs are used to describe diseases/syndromes and also other types of conditions such as spectrums of conditions. Treatments in medicine and psychiatry have several treatment objectives including cure or reducing distress and a variety of mechanisms of action apart from reversing disease/cure. Causation of conditions in medicine and psychiatry are often complex. The medical model allows doctors to assess and offer effective treatments to large numbers of patients and provide emergency cover. Diagnostic constructs in psychiatry and general medicine overlap for attributes such as clinical utility (e.g. predicting likely outcomes) and validity (e.g. lack of boundaries between different diagnostic constructs) and importance of social factors. There is an overlap in effectiveness between psychiatric and general medicine treatments and many general medicine medications do not reverse disease processes. Different mental health classifications have particular strengths and weaknesses for clinical, research, and social functions. Mental health research into understanding causes and mechanisms may need other classifications than diagnosis. As doctors in all specialties encounter mental health problems, there will always be psychiatric diagnostic constructs compatible with their training. Mental health research and service provision will always need to address psychosocial issues.

Download Full-text

Transitioning towards official clinical guidelines for the diagnostics and treatment of severe behavioural problems in children across Europe: current state and future directions.

10.31234/osf.io/8hsmy ◽

2017 ◽

Author(s):

Alexandra-Raluca Gatej ◽

Audri Lamers ◽

Robert Vermeiren ◽

Lieke van Domburgh

Keyword(s):

Mental Health ◽

Clinical Guidelines ◽

Clinical Decision Making ◽

Daily Practice ◽

Clinical Decision ◽

European Guidelines ◽

Current State ◽

Assessment And Treatment ◽

Evidence Based Treatments ◽

Multifactorial Approach

Severe behaviour problems (SBPs) in early childhood include oppositional and aggressive behaviours and predict negative mental health outcomes later in life. Although effective treatments for this group are available and numerous clinical practice guidelines have been developed to facilitate the incorporation of evidence-based treatments in clinical decision-making (NICE, 2013), many children with SBPs remain unresponsive to treatment (Lahey & Waldman, 2012). At present, it is unknown how many countries in Europe possess official clinical guidelines for SBPs diagnosis and treatment and what is their perceived utility. The aim was to create an inventory of clinical guidelines (and associated critical needs) for the diagnostics and treatment of SBPs in youth mental health across Europe according to academic experts and mental health clinicians’ opinions. To investigate the aim, two separate online semi-structured questionnaires were used, one directed at academics (N=28 academic experts; 23 countries), and the other at clinicians (N=124 clinicians; 24 countries). Three key results were highlighted. First, guidelines for SBPs are perceived as beneficial by both experts and clinicians. However, their implementation needs to be reinforced and content better adapted to daily practice. Improvements may include taking a multifactorial approach to assessment and treatment, involving the systems around the child, and multidisciplinary collaboration. Second, academic experts and clinicians support the need for further developing national / European guidelines. Finally, future guidelines should address current challenges identified by clinicians to be more applicable to daily practice.

Download Full-text

Correction to: Using behavioral insights to design implementation strategies in public mental health settings: a qualitative study of clinical decision-making

Implementation Science Communications ◽

10.1186/s43058-021-00115-y ◽

2021 ◽

Vol 2 (1) ◽

Author(s):

Briana S. Last ◽

Simone H. Schriger ◽

Carter E. Timon ◽

Hannah E. Frank ◽

Alison M. Buttenheim ◽

...

Keyword(s):

Mental Health ◽

Decision Making ◽

Qualitative Study ◽

Clinical Decision Making ◽

Public Mental Health ◽

Implementation Strategies ◽

Clinical Decision ◽

Mental Health Settings

An amendment to this paper has been published and can be accessed via the original article.

Download Full-text

The use of multiple-criteria decision-making theory to measure students’ perceptions of high-fidelity simulation

BMJ Simulation and Technology Enhanced Learning ◽

10.1136/bmjstel-2016-000167 ◽

2017 ◽

Vol 3 (3) ◽

pp. 88-93 ◽

Cited By ~ 1

Author(s):

Maureen Anne Jersby ◽

Paul Van-Schaik ◽

Stephen Green ◽

Lili Nacheva-Skopalik

Keyword(s):

Decision Making ◽

Clinical Practice ◽

Repeated Measures ◽

Clinical Decision Making ◽

Learning Experience ◽

Clinical Decision ◽

High Fidelity ◽

High Fidelity Simulation ◽

Initial Exposure ◽

Repeated Measures Design

BackgroundHigh-Fidelity Simulation (HFS) has great potential to improve decision-making in clinical practice. Previous studies have found HFS promotes self-confidence, but its effectiveness in clinical practice has not been established. The aim of this research is to establish if HFS facilitates learning that informs decision-making skills in clinical practice using MultipleCriteria DecisionMaking Theory (MCDMT).MethodsThe sample was 2nd year undergraduate pre-registration adult nursing students.MCDMT was used to measure the students’ experience of HFS and how it developed their clinical decision-making skills. MCDMT requires characteristic measurements which for the learning experience were based on five factors that underpin successful learning, and for clinical decision-making, an analytical framework was used. The study used a repeated-measures design to take two measurements: the first one after the first simulation experience and the second one after clinical placement. Baseline measurements were obtained from academics. Data were analysed using the MCDMT tool.ResultsAfter their initial exposure to simulation learning, students reported that HFS provides a high-quality learning experience (87%) and supports all aspects of clinical decision-making (85%). Following clinical practice, the level of support for clinical decision-making remained at 85%, suggesting that students believe HFS promotes transferability of knowledge to the practice setting.ConclusionOverall, students report a high level of support for learning and developing clinical decision-making skills from HFS. However, there are no comparative data available from classroom teaching of similar content so it cannot be established if these results are due to HFS alone.

Download Full-text

Clinical Conundrums When Integrating the QbTest into a Standard ADHD Assessment of Children and Young People

Neuropediatrics ◽

10.1055/s-0040-1722674 ◽

2021 ◽

Author(s):

Carsten Vogt

Keyword(s):

Clinical Practice ◽

Clinical Decision Making ◽

Ecological Validity ◽

Neuropsychological Test ◽

Real Life ◽

Clinical Decision ◽

Hyperactivity Disorder ◽

Standard Clinical Practice ◽

Novel Method ◽

The Impact

AbstractThe uptake of the QbTest in clinical practice is increasing and has recently been supported by research evidence proposing its effectiveness in relation to clinical decision-making. However, the exact underlying process leading to this clinical benefit is currently not well established and requires further clarification. For the clinician, certain challenges arise when adding the QbTest as a novel method to standard clinical practice, such as having the skills required to interpret neuropsychological test information and assess for diagnostically relevant neurocognitive domains that are related to attention-deficit hyperactivity disorder (ADHD), or how neurocognitive domains express themselves within the behavioral classifications of ADHD and how the quantitative measurement of activity in a laboratory setting compares with real-life (ecological validity) situations as well as the impact of comorbidity on test results. This article aims to address these clinical conundrums in aid of developing a consistent approach and future guidelines in clinical practice.

Download Full-text