Artificial intelligence and increasing misinformation.

With the recent advances in artificial intelligence (AI), patients are increasingly exposed to misleading medical information. Generative AI models, including large language models such as ChatGPT, create and modify text, images, audio and video information based on training data. Commercial use of generative AI is expanding rapidly and the public will routinely receive messages created by generative AI. However, generative AI models may be unreliable, routinely make errors and widely spread misinformation. Misinformation created by generative AI about mental illness may include factual errors, nonsense, fabricated sources and dangerous advice. Psychiatrists need to recognise that patients may receive misinformation online, including about medicine and psychiatry.

Artificial intelligence and cybercrime: implications for individuals and the healthcare sector.

The malicious use of artificial intelligence is growing rapidly, creating major security threats for individuals and the healthcare sector. Individuals with mental illness may be especially vulnerable. Healthcare provider data are a prime target for cybercriminals. There is a need to improve cybersecurity to detect and prevent cyberattacks against individuals and the healthcare sector, including the use of artificial intelligence predictive tools.

Implications of Online Self-Diagnosis in Psychiatry.

Online self-diagnosis of psychiatric disorders by the general public is increasing. The reasons for the increase include the expansion of Internet technologies and the use of social media, the rapid growth of direct-to-consumer e-commerce in healthcare, and the increased emphasis on patient involvement in decision making. The publicity given to artificial intelligence (AI) has also contributed to the increased use of online screening tools by the general public. This paper aims to review factors contributing to the expansion of online self-diagnosis by the general public, and discuss both the risks and benefits of online self-diagnosis of psychiatric disorders. A narrative review was performed with examples obtained from the scientific literature and commercial articles written for the general public. Online self-diagnosis of psychiatric disorders is growing rapidly. Some people with a positive result on a screening tool will seek professional help. However, there are many potential risks for patients who self-diagnose, including an incorrect or dangerous diagnosis, increased patient anxiety about the diagnosis, obtaining unfiltered advice on social media, using the self-diagnosis to self-treat, including online purchase of medications without a prescription, and technical issues including the loss of privacy. Physicians need to be aware of the increase in self-diagnosis by the general public and the potential risks, both medical and technical. Psychiatrists must recognize that the general public is often unaware of the challenging medical and technical issues involved in the diagnosis of a mental disorder, and be ready to treat patients who have already obtained an online self-diagnosis.

Telepsychiatry in an Era of Digital Mental Health Startups.

PURPOSE OF REVIEW: Telepsychiatry practiced by psychiatrists is evidence-based, regulated, private, and effective in diverse settings. The use of telemedicine has grown since the COVID-19 pandemic as people routinely obtain more healthcare services online. At the same time, there has been a rapid increase in the number of digital mental health startups that offer various services including online therapy and access to prescription medications. These digital mental health firms advertise directly to the consumer primarily through digital advertising. The purpose of this narrative review is to contrast traditional telepsychiatry and the digital mental health market related to online therapy. RECENT FINDINGS: In contrast to standard telepsychiatry, most of the digital mental health startups are unregulated, have unproven efficacy, and raise concerns related to self-diagnosis, self-medicating, and inappropriate prescribing. The role of digital mental health firms for people with serious mental illness has not been determined. There are inadequate privacy controls for the digital mental health firms, including for online therapy. We live in an age where there is widespread admiration for technology entrepreneurs and increasing emphasis on the role of the patient as a consumer. Yet, the business practices of digital mental health startups may compromise patient safety for profits. There is a need to address issues with the digital mental health startups and to educate patients about the differences between standard medical care and digital mental health products.

Challenges and Ethical Considerations to Successfully Implement Artificial Intelligence in Clinical Medicine and Neuroscience: a Narrative Review.

This narrative review discusses how the safe and effective use of clinical artificial intelligence (AI) prediction tools requires recognition of the importance of human intelligence. Human intelligence, creativity, situational awareness, and professional knowledge, are required for successful implementation. The implementation of clinical AI prediction tools may change the workflow in medical practice resulting in new challenges and safety implications. Human understanding of how a clinical AI prediction tool performs in routine and exceptional situations is fundamental to successful implementation. Physicians must be involved in all aspects of the selection, implementation, and ongoing product monitoring of clinical AI prediction tools.

Longitudinal Digital Mood Charting in Bipolar Disorder: Experiences with ChronoRecord Over 20 Years.

INTRODUCTION: Longitudinal study is an essential methodology for understanding disease trajectories, treatment effects, symptom changes, and long-term outcomes of affective disorders. Daily self-charting of mood and other illness-related variables is a commonly recommended intervention. With the widespread acceptance of home computers in the early 2000s, automated tools were developed for patient mood charting, such as ChronoRecord, a software validated by patients with bipolar disorder. The purpose of this study was to summarize the daily mood, sleep, and medication data collected with ChronoRecord, and highlight some of the key research findings. Lessons learned from implementing a computerized tool for patient self-reporting are also discussed. METHODS: After a brief training session, ChronoRecord software for daily mood charting was installed on a home computer and used by 609 patients with affective disorders. RESULTS: The mean age of the patients was 40.3±11.8 years, a mean age of onset was 22±11.2 years, and 71.4% were female. Patients were euthymic for 70.8% of days, 15.1% had mild depression, 6.6% had severe depression, 6.6% had hypomania, and 0.8% had mania. Among all mood groups, 22.4% took 1-2 medications, 37.2% took 3-4 medications, 25.7 took 5-6 medications, 11.6% took 7-8 medications, and 3.1% took >8 medications. CONCLUSION: The daily mood charting tool is a useful tool for increasing patient involvement in their care, providing detailed patient data to the physician, and increasing understanding of the course of illness. Longitudinal data from patient mood charting was helpful in both clinical and research settings.

Expectations for Artificial Intelligence (AI) in Psychiatry.

PURPOSE OF REVIEW: Artificial intelligence (AI) is often presented as a transformative technology for clinical medicine even though the current technology maturity of AI is low. The purpose of this narrative review is to describe the complex reasons for the low technology maturity and set realistic expectations for the safe, routine use of AI in clinical medicine. RECENT FINDINGS: For AI to be productive in clinical medicine, many diverse factors that contribute to the low maturity level need to be addressed. These include technical problems such as data quality, dataset shift, black-box opacity, validation and regulatory challenges, and human factors such as a lack of education in AI, workflow changes, automation bias, and deskilling. There will also be new and unanticipated safety risks with the introduction of AI. The solutions to these issues are complex and will take time to discover, develop, validate, and implement. However, addressing the many problems in a methodical manner will expedite the safe and beneficial use of AI to augment medical decision making in psychiatry.

Commercial Use of Emotion Artificial Intelligence (AI): Implications for Psychiatry.

PURPOSE OF REVIEW: Emotion artificial intelligence (AI) is technology for emotion detection and recognition. Emotion AI is expanding rapidly in commercial and government settings outside of medicine, and will increasingly become a routine part of daily life. The goal of this narrative review is to increase awareness both of the widespread use of emotion AI, and of the concerns with commercial use of emotion AI in relation to people with mental illness. RECENT FINDINGS: This paper discusses emotion AI fundamentals, a general overview of commercial emotion AI outside of medicine, and examples of the use of emotion AI in employee hiring and workplace monitoring. The successful re-integration of patients with mental illness into society must recognize the increasing commercial use of emotion AI. There are concerns that commercial use of emotion AI will increase stigma and discrimination, and have negative consequences in daily life for people with mental illness. Commercial emotion AI algorithm predictions about mental illness should not be treated as medical fact.

Increasing Cybercrime Since the Pandemic: Concerns for Psychiatry.

PURPOSE OF REVIEW: Since the pandemic, the daily activities of many people occur at home. People connect to the Internet for work, school, shopping, entertainment, and doctor visits, including psychiatrists. Concurrently, cybercrime has surged worldwide. This narrative review examines the changing use of technology, societal impacts of the pandemic, how cybercrime is evolving, individual vulnerabilities to cybercrime, and special concerns for those with mental illness. RECENT FINDINGS: Human factors are a central component of cybersecurity as individual behaviors, personality traits, online activities, and attitudes to technology impact vulnerability. Mental illness may increase vulnerability to cybercrime. The risks of cybercrime should be recognized as victims experience long-term psychological and financial consequences. Patients with mental illness may not be aware of the dangers of cybercrime, of risky online behaviors, or the measures to mitigate risk. Technology provides powerful tools for psychiatry but technology must be used with the appropriate safety measures. Psychiatrists should be aware of the potential aftermath of cybercrime on mental health, and the increased patient risk since the pandemic, including from online mental health services. As a first step to increase patient awareness of cybercrime, psychiatrists should provide a recommended list of trusted sources that educate consumers on cybersecurity.

Comparison of potential psychiatric drug interactions in six drug interaction database programs: A replication study after 2 years of updates.

OBJECTIVE: Drug interaction database programs are a fundamental clinical tool. In 2018, we compared the category of potential drug-drug interaction (DDI) provided by six drug interaction database programs for 100 drug interaction pairs including psychiatric drugs, and found the category often differed. This study replicated the comparison in 2020 after 2 years of updates to all six drug interaction database programs. METHODS: The 100 drug pairs included 94 different drugs: 67 pairs with a psychiatric and non-psychiatric drug, and 33 pairs with two psychiatric drugs. The assigned category of potential DDI for the drug pairs was compared using percent agreement and Fleiss kappa statistic of interrater reliability. RESULTS: Despite 67 updates involving 46 of the 100 drug pairs, differences remained. The overall percent agreement among the six drug interaction database programs for the category of potential DDI was 67%. The interrater agreement results did not change. The Fleiss kappa overall interrater agreement was fair. The kappa agreement for a drug pair with any severe category rating was substantial, and the kappa agreement for a drug pair with any major category rating was fair. CONCLUSIONS: Physicians should be aware of the inconsistency among drug interaction database programs in the category of potential DDI for drug pairs including psychiatric drugs. Additionally, the category of potential DDI for a drug pair may change over time. This study highlights the importance of ongoing international efforts to standardize methods used to define and classify potential DDI.

Association between Adherence with an Atypical Antipsychotic and with Other Psychiatric Drugs in Patients with Bipolar Disorder.

BACKGROUND: Using U.S. pharmacy and medical claims, medication adherence patterns of patients with serious mental illness suggest that adherence to atypical antipsychotics may be related to adherence to other prescription drugs. This study investigated whether adherence to an atypical antipsychotic was related to adherence to other prescribed psychiatric drugs using self-reported data from patients with bipolar disorder. METHODS: Daily self-reported medication data were available from 123 patients with a diagnosis of bipolar disorder receiving treatment as usual who took at least 1 atypical antipsychotic over a 12-week period. Patients took a mean of 4.0±1.7 psychiatric drugs including the antipsychotic. The adherence rate for the atypical antipsychotic was compared to that for other psychiatric drugs to determine if the adherence rate for the atypical antipsychotic differed from that of the other psychiatric drug by at least ±10%. RESULTS: Of the 123 patients, 58 (47.2%) had an adherence rate for the atypical antipsychotic that differed from the adherence rate for at least 1 other psychiatric drug by at least±10%, and 65 (52.8%) patients had no difference in adherence rates. The patients with a difference took a larger total number of psychiatric drugs (p<0.001), had a larger daily pill burden (p=0.020) and a lower adherence rate with the atypical antipsychotic (p=0.007), and were more likely to take an antianxiety drug (p<0.001). CONCLUSION: Adherence with an atypical antipsychotic was not useful for estimating adherence to other psychiatric drugs in about half of the patients with bipolar disorder.

Potential Drug interactions with Drugs used for Bipolar Disorder: A Comparison of 6 Drug Interaction Database Programs.

BACKGROUND: Patients with bipolar disorder frequently experience polypharmacy, putting them at risk for clinically significant drug-drug interactions (DDI). Online drug interaction database programs are used to alert physicians, but there are no internationally recognized standards to define DDI. This study compared the category of potential DDI returned by 6 commercial drug interaction database programs for drug interaction pairs involving drugs commonly prescribed for bipolar disorder. METHODS: The category of potential DDI provided by 6 drug interaction database programs (3 subscription, 3 open access) was obtained for 125 drug interaction pairs. The pairs involved 103 drugs (38 psychiatric, 65 nonpsychiatric); 88 pairs included a psychiatric and nonpsychiatric drug; 37 pairs included 2 psychiatric drugs. Every pair contained at least 1 mood stabilizer or antidepressant. The category provided by 6 drug interaction database programs was compared using percent agreement and Fleiss kappa statistic of interrater reliability. RESULTS: For the 125 drug pairs, the overall percent agreement among the 6 drug interaction database programs was 60%; the Fleiss kappa agreement was slight. For drug interaction pairs with any category rating of severe (contraindicated), the kappa agreement was moderate. For drug interaction pairs with any category rating of major, the kappa agreement was slight. CONCLUSION: There is poor agreement among drug interaction database programs for the category of potential DDI involving psychiatric drugs. Drug interaction database programs provide valuable information, but the lack of consistency should be recognized as a limitation. When assistance is needed, physicians should check more than 1 drug interaction database program.

International multi-site survey on the use of online support groups in bipolar disorder.

BACKGROUND: Peer support is an established component of recovery from bipolar disorder, and online support groups may offer opportunities to expand the use of peer support at the patient's convenience. Prior research in bipolar disorder has reported value from online support groups. AIMS: To understand the use of online support groups by patients with bipolar disorder as part of a larger project about information seeking. METHODS: The results are based on a one-time, paper-based anonymous survey about information seeking by patients with bipolar disorder, which was translated into 12 languages. The survey was completed between March 2014 and January 2016 and included questions on the use of online support groups. All patients were diagnosed by a psychiatrist. Analysis included descriptive statistics and general estimating equations to account for correlated data. RESULTS AND CONCLUSIONS: The survey was completed by 1222 patients in 17 countries. The patients used the Internet at a percentage similar to the general public. Of the Internet users who looked online for information about bipolar disorder, only 21.0% read or participated in support groups, chats, or forums for bipolar disorder (12.8% of the total sample). Given the benefits reported in prior research, clarification of the role of online support groups in bipolar disorder is needed. With only a minority of patients using online support groups, there are analytical challenges for future studies.

Automated Decision-Making and Big Data: Concerns for People With Mental Illness.

Automated decision-making by computer algorithms based on data from our behaviors is fundamental to the digital economy. Automated decisions impact everyone, occurring routinely in education, employment, health care, credit, and government services. Technologies that generate tracking data, including smartphones, credit cards, websites, social media, and sensors, offer unprecedented benefits. However, people are vulnerable to errors and biases in the underlying data and algorithms, especially those with mental illness. Algorithms based on big data from seemingly unrelated sources may create obstacles to community integration. Voluntary online self-disclosure and constant tracking blur traditional concepts of public versus private data, medical versus non-medical data, and human versus automated decision-making. In contrast to sharing sensitive information with a physician in a confidential relationship, there may be numerous readers of information revealed online; data may be sold repeatedly; used in proprietary algorithms; and are effectively permanent. Technological changes challenge traditional norms affecting privacy and decision-making, and continued discussions on new approaches to provide privacy protections are needed.

Gender differences in thyroid system function: relevance to bipolar disorder and its treatment.

OBJECTIVES: Thyroid hormones play a critical role in the functioning of the adult brain, and thyroid diseases impair both mood and cognition. This paper reviews gender differences in thyroid system function that are relevant to the diagnosis and treatment of bipolar disorder. METHODS: The study comprised a comprehensive literature review of gender differences in thyroid disease that are pertinent to mood disorders. RESULTS: The prevalence of thyroid disease was found to be much higher in females than males, and to increase with age. The most commonly detected abnormality was subclinical hypothyroidism, which was found to occur in up to 20% of postmenopausal women. Females also had higher rates of thyroid autoimmunity. Individuals at risk for thyroid disease, such as adult females, may have had less ability to compensate for additional challenges to thyroid metabolism, including lithium treatment. Thyroid abnormalities were associated with a poorer response to standard treatments for mood disorders. Females with treatment-resistant mood disorders may have responded better than males to adjunctive therapy with thyroid hormones. CONCLUSIONS: Disturbances of thyroid system function, which occur commonly in females, may complicate the diagnosis and treatment of mood disorders. In particular, this is clinically relevant during lithium treatment because lithium may impair vital thyroid metabolic pathways secondary to its anti-thyroid activity.

Privacy in the digital world: medical and health data outside of HIPAA protections.

Increasing quantities of medical and health data are being created outside of HIPAA protection, primarily by patients. Data sources are varied, including the use of credit cards for physician visit and medication co-pays, Internet searches, email content, social media, support groups, and mobile health apps. Most medical and health data not covered by HIPAA are controlled by third party data brokers and Internet companies. These companies combine this data with a wide range of personal information about consumer daily activities, transactions, movements, and demographics. The combined data are used for predictive profiling of individual health status, and often sold for advertising and other purposes. The rapid expansion of medical and health data outside of HIPAA protection is encroaching on privacy and the doctor-patient relationship, and is of particular concern for psychiatry. Detailed discussion of the appropriate handling of this medical and health data is needed by individuals with a wide variety of expertise.

New measures of mental state and behavior based on data collected from sensors, smartphones, and the Internet.

With the rapid and ubiquitous acceptance of new technologies, algorithms will be used to estimate new measures of mental state and behavior based on digital data. The algorithms will analyze data collected from sensors in smartphones and wearable technology, and data collected from Internet and smartphone usage and activities. In the future, new medical measures that assist with the screening, diagnosis, and monitoring of psychiatric disorders will be available despite unresolved reliability, usability, and privacy issues. At the same time, similar non-medical commercial measures of mental state are being developed primarily for targeted advertising. There are societal and ethical implications related to the use of these measures of mental state and behavior for both medical and non-medical purposes.

Exploratory study of ultraviolet B (UVB) radiation and age of onset of bipolar disorder.

BACKGROUND: Sunlight contains ultraviolet B (UVB) radiation that triggers the production of vitamin D by skin. Vitamin D has widespread effects on brain function in both developing and adult brains. However, many people live at latitudes (about > 40 N or S) that do not receive enough UVB in winter to produce vitamin D. This exploratory study investigated the association between the age of onset of bipolar I disorder and the threshold for UVB sufficient for vitamin D production in a large global sample. METHODS: Data for 6972 patients with bipolar I disorder were obtained at 75 collection sites in 41 countries in both hemispheres. The best model to assess the relation between the threshold for UVB sufficient for vitamin D production and age of onset included 1 or more months below the threshold, family history of mood disorders, and birth cohort. All coefficients estimated at P ≤ 0.001. RESULTS: The 6972 patients had an onset in 582 locations in 70 countries, with a mean age of onset of 25.6 years. Of the onset locations, 34.0% had at least 1 month below the threshold for UVB sufficient for vitamin D production. The age of onset at locations with 1 or more months of less than or equal to the threshold for UVB was 1.66 years younger. CONCLUSION: UVB and vitamin D may have an important influence on the development of bipolar disorder. Study limitations included a lack of data on patient vitamin D levels, lifestyles, or supplement use. More study of the impacts of UVB and vitamin D in bipolar disorder is needed to evaluate this supposition.

Impact of sunlight on the age of onset of bipolar disorder.

OBJECTIVE: Although bipolar disorder has high heritability, the onset occurs during several decades of life, suggesting that social and environmental factors may have considerable influence on disease onset. This study examined the association between the age of onset and sunlight at the location of onset. METHOD: Data were obtained from 2414 patients with a diagnosis of bipolar I disorder, according to DSM-IV criteria. Data were collected at 24 sites in 13 countries spanning latitudes 6.3 to 63.4 degrees from the equator, including data from both hemispheres. The age of onset and location of onset were obtained retrospectively, from patient records and/or direct interviews. Solar insolation data, or the amount of electromagnetic energy striking the surface of the earth, were obtained from the NASA Surface Meteorology and Solar Energy (SSE) database for each location of onset. RESULTS: The larger the maximum monthly increase in solar insolation at the location of onset, the younger the age of onset (coefficient= -4.724, 95% CI: -8.124 to -1.323, p=0.006), controlling for each country's median age. The maximum monthly increase in solar insolation occurred in springtime. No relationships were found between the age of onset and latitude, yearly total solar insolation, and the maximum monthly decrease in solar insolation. The largest maximum monthly increases in solar insolation occurred in diverse environments, including Norway, arid areas in California, and Chile. CONCLUSION: The large maximum monthly increase in sunlight in springtime may have an important influence on the onset of bipolar disorder.

Brief depressive symptoms in patients with bipolar disorder: analysis of long-term self-reported data.

OBJECTIVE: Most patients with bipolar disorder experience depressive symptoms outside of an episode of depression as defined by DSM-IV criteria. This study explores the frequency of brief depressive episodes, lasting 1 to 4 days, using daily self-reported mood ratings. METHOD: Mood ratings were obtained from 448 patients (281 bipolar I, 167 bipolar II) using ChronoRecord software (91,786 total days). Episodes of depression and days of depression outside of episodes were determined. The intensity of depressive symptoms (mild versus moderate to severe) was compared. RESULTS: Using the DSM-IV length criteria, 61% of all depressive days occurred outside of a depressed episode. Decreasing the minimum length criterion to 2 days, both the number of patients experiencing a depressed episode (128 to 317) and the mean percent of days spent in a depressed episode by each patient (7.9% to 17.8.%) increased by about 2½ times, and 34.3% of depressed days remained outside of an episode. Depending on the episode length, the proportion of days within an episode with severe symptoms varied from 1/3 to 1/4 for episodes lasting from 14 to 2 days, and 1/4 for single-day episodes. There was no significant difference in the frequency of brief depressive episodes between bipolar I and II disorders. For all episode lengths, patients taking antidepressants spent 4% more days within an episode and 6% more days with depressive symptoms outside of an episode than those not taking antidepressants. CONCLUSION: Brief depressive episodes lasting 1 to 4 days occur frequently in bipolar disorder and do not distinguish between bipolar I and II disorders. Symptoms of moderate to severe intensity occur on 1/4 to 1/3 of the days in brief depressive episodes. This study did not address brief depression in those without bipolar disorder. Patients taking antidepressants experienced more brief depressive episodes. Controlled trials are needed to assess the impact of antidepressants on subsyndromal depressive symptoms.