The diagnosis of neurodegenerative and psychiatric disorders (NPDs) in primary care can suffer from inefficiencies resulting in misdiagnoses and delayed diagnosis, limiting effective treatment options. The development of speech and language-based profiling biomarkers could aid in achieving earlier motor diagnosis for PD for instance, or more accurate diagnosis of clinically similar or late presenting NPDs.

RHAPSODY aims to investigate the feasibility of the remote administration of a battery of speech tasks in eliciting continuous narrative speech across a range of NPDs. The project also aims to determine the feasibility of using acoustic and linguistic biomarkers from speech data to support the clinical assessment and disambiguation of common NPDs

All participants (n=250) will take part in a single virtual telemedicine video conference with a researcher in which they are screened and complete a battery of speech tasks, in addition to cohort-specific screening measures. Over the following month, participants will be asked to complete a series of short, self-administered speech assessments via a smartphone application.

The speech tasks will be audio-recorded and analysed on Novoic’s technology platform. Objectives will be analysed using measures including average length of speech elicitation for speech tasks, intra- and inter-subject variance, differences in linguistic patterns, and response rates to speech assessments.

The analyses could help to identify and validate speech-derived clinical biomarkers to support clinicians in detecting and disambiguating between NPDs with heterogeneous presentations. This should further support earlier intervention, improved treatment options and improved quality of life.

In terms of significant financial interest and relationships, it is emphasised that the private organisation Novoic, who aim to develop speech algorithms for diagnostic use, is the study’s sponsor and employees or former employees of this company comprise

Psychiatric disorders may be one of considerable contributing factors for dry eye symptoms and the severity of subjective symptoms of dry eye are often related to psychological factors.

The aim of this research was to determine which of two chosen questionnaires for subjective symptoms of dry eye (Ocular Surface Disease Index and Schein questionnaire) is more reliable in the assessment of dry eye in patients with schizophrenia.

Our research included 80 patients of both sexes with schizophrenia ranging between the age of 25 and 55 who have been taking antipsychotics (clozapin, olanzapin, quetiapin) for five years and were in remission. All participants were required to satisfy all included and excluded criteria. They all filled out the Schein and OSDI questionnaires for assessment of subjective symptoms. Tear break-up time test (TBUT) for objective evaluation of tear film stability was also performed. In order to determine the correlation between two subjective and objective tests we calculated Spearmans correlation coefficients.

A analysis shows that there are no statistically significant differences between the correlations. Both subjective questionnaires are statistically significantly and negatively related to the TBUT test, showing that an increase in the results on the OSDI and Schein’s questionnaires led to the decreases in the results on the TBUT test.

In patients with schizophrenia the OSDI and Schein questionnaires are equally reliable in the assessment of subjective symptoms of Dry eye disease. Considering that, OSDI is more common in clinical practice, it is recommended for use in patients with schizophrenia.

No significant relationships.

Cushing’s syndrome is a hormonal disorder caused by chronic exposure to excess glucocorticoids, either exogenous or endogenous. The bronchial carcinoid tumor is an extremely rare origin, described in less than 1% of cases. The most frequent psychiatric symptoms are depression and anxiety, with manic and psychotic symptoms being less frequent. Psychotic symptoms are difficult to manage, as they are usually resistant to antipsychotic treatment, which is why it is considered an indication for medical treatment of Cushing’s syndrome.

To give visibility to this type of psychotic disorders of organic origin to deepen their study as well as raise awareness among professionals dedicated to clinical care with the intention of improving their prevention.

A description of a clinical case is made, accompanied by a bibliographic review on psychosis of endogenous corticosteroid origin.

We describe the case of a 44-year-old woman who was admitted to the charge of Internal Medicine due to Cushing’s syndrome. During her admission, she presented a debut of positive psychotic symptoms, so the liaison psychiatry team followed her up. She was diagnosed with an ACTH-secreting lung carcinoid tumor.

This entity should be taken into account in cases of atypical psychosis in patients with compatible phenotypic characteristics.

No significant relationships.

Psychosis of epilepsy has intrigued many neurologists and psychiatrists. We attempt to summarize the phenomenon, suggested diagnostic criteria and distinguishing features between different clinical entities linked with epilepsy.This case series is unique and rare as we include the case that meets full criteria of forced normalization.

1) To understand the concept and diagnostic criteria of Forced Normalization 2) To differentiate different psychiatric manifestations of epilepsy

A total of 13 studies were reviewed using the key words from 1999 –2021 using different search engines- Google scholar, Pub-med, Elsevier, Dynamed.

Patients with epilepsy have an eightfold increased risk of psychosis (6). Forced Normalization has been described as the onset of psychotic or mood symptoms after the resolution or remission of >50% of seizures, evidenced by normal EEG. It was first described in 1950’s and has been extensively studied since 19thcentury. The age of onset can be 8 years to 71 years of age (mean - 28.3). The exact mechanism is still unknown. Different factors have been linked to this phenomenon like kindling, neurotransmitters etc.

It is interesting to understand the antagonistic relationship between epilepsy and psychosis. Forced normalization is a rare entity because it is hard to diagnose due to possible overlap with other clinical entities like post-ictal or side effects of AED. The prognosis seems to be favorable depending on the trigger for the symptoms with better prognosis if the resolution of seizures was achieved AED. Mood disorders had worse prognosis than dissociation and psychosis.

No significant relationships.

In recent years, more and more attention has been paid to the risks of using SSRIs. This group of antidepressants may be associated with an increased risk of gastrointestinal bleeding. This risk would be even further increased with concomitant use of NSAIDs. A number of studies have described this interaction, however they reported conflicting results.

Our objective was to investigate the risk of gastrointestinal bleeding with SSRIs, with or without NSAID use.

We performed a literature search, using Pubmed, EMBASE, and Cochrane library, in order to investigate controlled trials, cohort, case-control and cross-sectional studies that reported the incidence of gastrointestinal bleeding s on SSRIs with or without concurrent NSAID use, compared to placebo or no treatment.

15 case-control studies and 4 cohort studies were included in the analysis. There was an increased risk of gastrointestinal bleeding with SSRIs in the cohort studies and case-control studies. The risk of gastrointestinal bleeding was even further increased with the combined use of both SSRIs and NSAIDs.

SSRIs are associated with a modest increase of gastrointestinal bleeding. However, this risk is significantly increased when SSRIs are used in combination with NSAIDs. Psychiatrists should be aware of the hazards in prescribing these medications together.

No significant relationships.

Neuropsychiatric disorders including Generalized Anxiety Disorder (GAD), Obsessive-Compulsive Disorder (OCD), Major Depressive Disorder (MDD), Bipolar Disorder (BD), and Schizophrenia (SZ) have been considered distinct categories of diseases despite their overlapping characteristics and symptomatology.

We aimed to provide an in-depth review elucidating the role of glutamate/Glx and white matter (WM) abnormalities from a transdiagnostic perspective.

The PubMed online database was searched for studies published between 2010 and 2021. After careful screening, 399 studies were included.

The findings point to decreased levels of glutamate in the Anterior Cingulate Cortex in both SZ and BD, whereas Glx is elevated in the Hippocampus in SZ and MDD. With regard to WM abnormalities, the Corpus Callosum and superior Longitudinal Fascicle were the most consistently identified brain regions showing decreased fractional anisotropy (FA) across all the reviewed disorders, except GAD. Additionally, the Uncinate Fasciculus was found to be affected in all the reviewed disorders, except OCD. Decreased FA was also found in the inferior Longitudinal Fasciculus, inferior Fronto-Occipital Fasciculus, Thalamic Radiation, and Corona Radiata in SZ, BD, and MDD. Decreased FA in the Fornix and Corticospinal Tract were found in BD and SZ patients. The Cingulum and Anterior Limb of Internal Capsule exhibited decreased FA in MDD and SZ patients.

The results suggest a gradual increase in severity from GAD to SZ defined by the number of brain regions with WM abnormality which may be partially caused by abnormal glutamate levels. WM damage could thus be considered a potential marker of some of the main neuropsychiatric disorders.

No significant relationships.

Attention Deficit / Hyperactivity Disorder (ADHD) is a highly prevalent neurodevelopmental condition characterized by inattention, motor hyperactivity and impulsivity. ADHD cognitive and behavioral presentation is characterized by a high heterogeneity (APA, 2013). Indeed, a complex diagnostic process, that considers several validated tools, is, to date, necessary.

The main aim is to develop supervised machine learning (ML) algorithms that could be used to support the diagnostic process for ADHD, by identifying the most relevant features in discriminating between the presence or absence of the ADHD diagnosis in children.

We analyzed data from 342 children (Mean age: 8y 8m ± 1y; 61 F) referred for possible ADHD symptomatology. Assessments were performed by an expert clinician and through questionnaires: Social Responsiveness Scale (SRS), Child Behavior Checklist (CBCL), Conners Rating Scale for Parents (CPRS) and for Teachers (CTRS). Data were analyzed using a decision tree classifier and random forest algorithms.

The decision tree model performed an accuracy of 0.71. The random forest model that was identified as the best tested, performed an accuracy of 0.77 (Figure 1) and it identified as most informative parent- and teacher-rated DSM-oriented ADHD symptoms (Figure 2).

Figure 1: Random forest confusion matrix and statistics.

Figure 2: Ranking of variables importance.

A random forest classifier could represent an effective algorithm to support the identification of ADHD children and to simplify the diagnostic process as an initial step. The use of supervised machine learning algorithms could be useful in helping the diagnostic process, highlighting the importance of a personalized medicine approach.

No significant relationships.

Telepsychiatry is the best-documented e-Mental Health application. It refers to the use of videoconferencing in the provision of mental health services. During the COVID19 pandemic, in response to physical distancing, mental health services worldwide have turned to online consultations. For the vast majority of clinicians, it was the first time they use telepsychiatry, and very few have received training in how to do it.

- to present the main objectives and messages of the WPA Global Guidelines for Telepsychiatry related to competencies & skills, educational & legislative needs, and international collaboration.

A structured review of the main challenges, innovations, and settings in the first Global Telepsychiatry Guidelines, published by WPA in February 2021.

The benefits of increased access to telehealth services are apparent for telepsychiatry, but benefits can only be realized if the tools are used by clinicians who have the appropriate training and guidance. With proper preparation and thoughtful risk management, telepsychiatry can be an invaluable tool for allowing greater access to care. However, certain prerequisites must be fulfilled to achieve the desired goals. These prerequisites are e.g. choice of the technology, settings, patient/provider preferences as well as competencies and skills described in this document.

The need for training among health care professionals is the highest priority. The urgent need for clinical training and skills building around e-mental health inclusive telepsychiatry, will determine the influence that psychiatry can have in addressing the mental health sequelae of the COVID19 pandemic via competent practice and increased international collaboration.

I am the main author of “WPA Telepsychiatry Global Guidelines”

Despite the high prevalence of substance use disorders, the majority of affected individuals do not seek any medical help or receive treatment targetting mainly symptoms of intoxication, withdrawal or general medical conditions due to chronic use of psychoactive substances. Patients with substance use disorders are more likely to remain undiagnosed regarding other psychiatric illnesses. Electronic patient-reported outcomes (ePRO) provide an easy-to-use instrument for detailed assessment at low economic cost.

To assess patients’ attitude towards self-reporting of symptoms related to substance use, mood, anxiety, quality of sleep, medication intake, social performance, and psychotic symptoms.

Mobile application consisting of seven questionnaires (Mood, Anxiety, Substance Use, Sleep, Medication, Social Activity and Various symptoms) was offered for use to patients with substance use disorder. Enrolled subjects were encouraged to use the app to report their actual condition in accordance with their own willingness and lifestyle.

Throughout the study a total of 1077 completed questionnaires were submitted, of which 193 (17.9%) were on mood, 188 (17.5%) - on substance use, 187 (17.4%) – on sleep, 155 (14.4%) – on anxiety, 139 (12.9%) – on medication intake, 111 (10.3%) – on psychotic symptoms, and 104 (9.7%) – on social performance.

Our research revealed that patients with substance use disorder are likely to share concerns regarding variety of psychiatric symptoms besides these attributed to their primary diagnosis. Implementation of ePROs can be a valuable tool for in-depth assessment and subsequent meeting the needs of such patients.

No significant relationships.

The increasing global burden of mental disorders has led to rising demand for mental healthcare services. Effective resource management is essential to ensure safe and timely access to care. Electronic health records (EHRs) provide a real-time source of data on clinical presentation and prognostic factors that could be harnessed to provide clinicians with actionable insights to prioritise mental healthcare delivery. We describe the development and evaluation of MaST, an EHR data visualisation tool that provides information to clinicians on risk of mental health crisis defined as an admission to a psychiatric hospital or acceptance into a community crisis service.

(i) To develop an EHR-data driven risk prediction tool for risk of crisis. (ii) To evaluate predictive performance in a real-world clinical setting.

The risk of crisis algorithm was developed and evaluated with EHR data from six UK NHS mental health providers using Ordered Predictor List propensity scores grouped into 5 quintiles. The predictor variables were clinical and sociodemographic factors including previous mental health service contacts.

Data from 2,620 patients contributed to algorithm development which was subsequently tested on data from 107,879 patients. The risk of crisis algorithm performed well with an overall accuracy for predicting the greatest risk of crisis (top quintile) ranging from 64% to 80%.

The MaST algorithm accurately predicted risk of mental health crisis in UK community mental health services. EHR data visualisation tools can provide actionable insights to clinicians to prioritise mental healthcare delivery in real-world clinical practice.

This study was funded in full by Holmusk.

Italy was the first European country to face up with COVID-19 pandemic, which posed challenges to National Health System (NHS), including the need to adapt mental health services/infrastructures and implement digitalization.

Despite telepsychiatry (ie., delivery of psychiatric care remotely through IT), is extensively used in non-European countries, only during the COVID-19 pandemic, became a convincing alternative to face-to-face modality for many psychiatrists in their clinical practice. Our aim was investigating Italian psychiatrists’ opinion about telepsychiatry.

A questionnaire, disseminated during the third Italian phase, constituted by three sections (socio-demographic, opinions and personal experience about/with telepsychiatry) was build by adapting the 42-item questionnaire by Schubert (2019) and CAMH’s Client Experience Survey from the psychiatrist’s perspective.

90 questionnaires were collected from a sample of 54 women (60%) with an average age of 43(SD=11.4). Mostly were psychiatrists (85.6%) working in NHS (66.7%) with an average working years of 13.7(SD=11.5) and a previous experience in telepsychiatry (71.1%). Overall, participants do not believe that telepsychiatry is comparable with face-to-face modality. A significant positive opinion was reported among younger psychiatrists compared to those more experienced, regarding efficacy, feasibility and mental health access (p<0.05). No significant differences were reported in psychiatrists’ opinion, according to the level of telepsychiatry use in their clinical practice.

Overall, sufficient digital skills and knowledge of technological tools are evident among younger psychiatrists who also appeared to be more prone to implement telepsychiatry in their clinical practice.

No significant relationships.

In many countries, people face problems regarding access to care, 24/7 support and evidence-based support. Digital interventions and services, such as chatbots, can be one option to tackle these challenges. There is a lack knowledge regarding how mental health chatbots are developed and how to ensure that there is collaboration between mental health and digital technology experts and users.

This presentation describes the phases of the development for the ChatPal mental health and wellbeing chatbot.

Development process was conducted in five and with four different languages. First, using an electronic survey for mental health professionals (n =190) we screened how familiar they are with chatbots and how they evaluated their potential. Second, university students and staff, mental health professionals and service users (n=78) participated in workshops to design the chatbot content. Finally, the content and scripts of chatbot were written in multi-professional and multi-national collaboration.

ChatPal is based on the PERMAH model of positive psychology and on the idea that we all have mental health which needs boosting and support from time to time. ChatPal includes relevant mental health information, exercises, mood diaries and simple monitoring and self-care tools. Based on preliminary evaluations, the ChatPal chatbot offers an option to offer support in areas where other mental health services are lacking or are insufficient.

ChatPal is already freely available in application stores and first scientific trials are have started. Preliminary results of 4-week and subsequent 12-week in-the-wild trials will be in place at the time of EPA 2022 conference.

No significant relationships.

Therapists’ forced transition to provide psychotherapy remotely during the COVID-19 pandemic offers a unique opportunity to examine therapists’ views and challenges with teletherapy.

We aimed to develop predictive models of three aspects of psychotherapists’ acceptance of teletherapy during the COVID-19 pandemic; attitudes towards teletherapy, concerns about using teletherapy, and intention to use it in the future.

In an international survey, therapists (N = 795) completed a survey about their experiences during the pandemic, including quality of therapeutic relationship, professional self-doubt, vicarious trauma, and telepsychotherapy acceptance. Regression decision trees machine learning analyses were used to build prediction models for each aspects of telepsychotherapy acceptance.

Attitudes toward telepsychotherapy were most positive for therapists who reported neutral or strong online working alliance, especially if they experienced little professional self-doubt and were younger than 40 years old. Therapists who were most concerned about telepsychotherapy, were those who reported higher levels of professional self-doubt, particularly if they also reported vicarious trauma experiences. Therapists who reported low working alliance were the least likely to use telepsychotherapy in the future.

Therapists’ professional self-doubt and the quality of their working alliance with their telepsychotherapy patients appear to be the most pertinent factors associated with therapists’ acceptance of telepsychotherapy during COVID-19, and should be addressed in future training and research.

No significant relationships.

During the early months of the COVID-19 pandemic, patients receiving individual psychotherapy needed to transition to telepsychotherapy (TP). Since telemental health appears to be here to stay after the pandemic ends, it is crucial to understand factors that determine whether telemental health is a good fit for patients.

The aim of the present study was to (1) explore patients’ perception of the therapeutic relationship and attitudes towards TP, and (2) identify predictors of patients’ TP acceptance.

We used a longitudinal design, where patients (N = 719) receiving individual TP during the pandemic participated in an online survey, in which they responded to demographic questions and completed measures of symptom severity, Covid-related distress, attachment style (avoidant/anxious), perceived quality of the therapeutic relationship (working alliance and real relationship), and TP acceptance.

We found that (1) patients perceived the quality of the therapeutic relationship as reasonably good, and patients’ TP acceptance was moderately high. (2) patients’ TP acceptance was predicted by their attachment avoidance and their perception of the real relationship, whereas attachment anxiety, working alliance, as well as demographic variables, symptom severity, and Covid-related distress were unrelated to TP acceptance. The final model showed that perceived strength of the real relationship mediated the relationship between attachment avoidance and TP acceptance.

Both general (attachment) and situational (therapeutic relationship) relational variables are important predictors of patient’s acceptance of TP, and should be considered during decision making about suitability of TP to patients.

No significant relationships.

Adult attention deficit hyperactivity disorder (ADHD) is associated with reduced work performance. Online interventions increase accessibility of services to clients by removing barriers such as physical distance, which may prevent care.

This study aimed to assess the efficacy of an innovative functional metacognitive intervention for work-performance improvement of adults with ADHD.

This study used a wait-list control group design, with a study and a comparison group (total 46 adults, mean age of 35.65 years). All participants had been diagnosed with ADHD, worked at least 3 months at the same place, and were willing to improve their work performance. Intervention sessions were provided mostly online and focused on the adults’ occupational goals in a workplace context. The intervention’s efficacy was evaluated with a focus on participants’ work performance (Canadian Occupational Performance Measure) executive functions (Behavior Rating Inventory of Executive Function-Adult), organisation in time (Time Organisation and Participation Scale), and quality of life (Adult ADHD Quality of Life Questionnaire).

Participants’ work performance, executive functions, organisation in time and quality of life significantly improved following the intervention. Their achievements were maintained through to the 3-month follow-up.

The online metacognitive functional intervention for work-performance improvement of adults with ADHD was found to be efficient and suitable for clinical use among this population. Future studies with larger samples and additional objective measures are needed to further validate these findings.

No significant relationships.

There is a rapid increase in the use of mHealth apps to provide care and support to people with deminetia (pwd) and their caregivers.

Functionality as well as scientific evidence of mHealth apps were analysed from a Health Technology Assessment (HTA) perspective

mHealth apps for pwd and their caregivers were identified in the app stores and assessed for functionality and methodological quality of the underlaying scientific evidence. Functionality was assessed with the Mobile App Rating Scale-German (MARS- G), methodological quality of studies using the Critical Appraisal Skills Program (CASP) checklists.

A total of 20 mHealth apps for were identified via systematic search in the Apple App Store and the Google Play Store. The overall quality of the apps can be rated medium with a mean score of 3.38. Studies have been published for only 30% of the apps (n = 6). 13 studies were included in the assessment of methodological quality, but the app itself was the object of study in only 2 publications. In summary methodological weaknesses such as small group sizes, a short study duration and insufficient comparative therapy were frequent.

Our results demonstrate that the majority of apps is lacking reliable scientific evidence. To support pwd and caregivers as end users in their choice standardised HTA criteria has to be defined and applied for mHealth apps.

The Digital Dementia Registry Bavaria (digiDEM Bayern) is funded by the Bavarian Ministry of Health (No.: G42d-G8300-2017/1606-381).

PROSPERO Registration Number: CRD42021260309.

No significant relationships.

Telemedicine has been at the heart of healthcare system’s strategic response to the COVID-19 pandemic. Within psychiatry, there has been a surge of research and guidelines into the use of video-teleconferencing to replace face to face consultations across clinical settings. Clinical ward rounds are central to inpatient psychiatric care yet little guidance is available on how best to integrate telemedicine into the multidisciplinary work of inpatient psychiatry.

We report on the introduction of video teleconferencing for psychiatric ward rounds on our acute inner-London psychiatric unit during the outbreak of COVID-19.

In undertaking the rapid transition to tele-ward rounds, we had to reconcile the multiple functions of psychiatric ward rounds with the technological resources available to us.

Tele-ward rounds helped simplify care delivery, facilitate multidisciplinary collaboration and improve accessibility for patients and relatives in a time of crisis. The transition to tele-ward rounds also brought about technical, operational and communication issues that may impact on the patient experience and quality of care including governance challenges, contextual dissonance and technological limitations.

The routine use of newer technology in psychiatry ward rounds is unlikely to succeed on the basis of improvisation, particularly given the stream of technical innovations in telemedicine, and the multifarious quality of social interactions in our clinical setting. Staff training and the development of an adapted etiquette and code of communication are both essential. Patient participation in future developments will also help ensure tele-ward rounds continue to meet the standards of high quality inpatient psychiatric care beyond the COVID-19 pandemic.

No significant relationships.

Electronic health (eHealth) interventions integrate different elements of care in treating and preventing mental ill-health in patients with somatic illnesses. Identifying different sociodemographic characteristics that might be associated with higher perceived usability can help in improving the usability of these e-health interventions.

This study aimed to identify sociodemographic characteristics that might be associated with the perceived usability of the NEVERMIND e-health system, comprised of a mobile application and a sensorized shirt, developed to reduce co-morbid depressive symptoms in patients with breast or prostate cancer.

The study included 129 patients with a diagnosis of breast or prostate cancer who received the NEVERMIND system. Sociodemographic data were collected at baseline. Usability outcomes included the System Usability Scale (SUS), the Mobile Application Rating Scale: user version (uMARS), and a usage index.

The analysis was based on 108 patients (68 breast cancer and 40 prostate cancer patients) who used the NEVERMIND system. The overall mean SUS score at 12-weeks was 73.4 with no statistical differences among different sociodemographic characteristics. The global uMARS score was 3.8, and females scored the app higher than males (β coefficient= 0.16; p=.03, 95% CI 0.02 - 0.3). Females had significant lower usage (β coefficient= -0.13; p=.04, 95% CI -0.25 to -0.01) after adjusting for other covariates.

There was a higher favourability of the mobile application among females compared to males. However, males had significantly higher usage of the NEVERMIND system. The NEVERMIND system does not suffer from ‘digital divide’ where certain sociodemographic characteristics are more associated with higher usability.

No significant relationships.

Virtual coaches (VCs) can support people with Diabetes Mellitus (DM) by motivating them to better manage their health. Few VCs were aimed at providing psychosocial support. In this regard, motivation is a pivotal construct in diabetes self-management as it allows adults with DM to adhere to the clinical recommendations.

The present study aimed to develop a VC able to motivate adults with DM to adopt and acquire healthier coping strategies, to decrease symptoms of depression, anxiety, perceived stress, and diabetes-related emotional distress, while also improving their well-being.

A total of 12 adults with DM (M=27.91 years; SD=9.82) interacted with a VC, called Motibot using Telegram for an overall duration of 12 sessions. Participants completed a battery of instruments at pre-, post-intervention and follow-up.

highlighted a decrease in anxiety, and depression symptoms between pre-, post-intervention and follow-up, as also showed by the results that emerged through the text mining. Motibot was perceived as motivating and encouraging in the adoption of appropriate coping strategies, such as mindfulness practices. Motibot was also perceived as trustworthy, reflective, and stimulating in its dialogical interaction. Indeed, adults felt involved in the interaction with Motibot, thereby showing an overall perception of a better quality of life, in the absence of diabetes distress.

This study sheds light on the importance of VCs in health care for people with DM for psychosocial support. This is the first experimental study on the matter, and thus, further iterations of the intervention are needed using a larger sample size.

No significant relationships.

There is limited knowledge about the potential role of machine learning (ML) in quality improvement of psychiatric care.

Our case study was to determine whether ML decision trees used on patient databases are suitable for focussing on specific patient population samples of mental healthcare quality audits. Populations were identified by patient and care provider variables, and the time of treatment. Outcomes were defined as hospital mortality, over-long hospitalization (over average +1SD or +2SD); and short hospitalization (under average -1SD; under 3 days).

We conducted a Split Train Test in Python for our outcomes on national mental health inpatient turnover data (2010 through 2018 for training and 2019 for testing). A well-fitting decision tree had the area under the curve (AUC) of the receiver operating characteristic (ROC) >= 0.7, and specificity >= 0.9. Performing qualitative analyses of decision trees, we rejected the ones with little clinical relevance.

Decision trees fit well (AUC = 0.7 to 0.9; specificity = 0.7 to 1.0; sensitivity = 0 to 0.69). For hospital death cases, the decision tree had AUC = 0.86, no difference after controlling for the types of hospital units, and was clinically relevant. Models predicting over-long hospitalization fit well (AUC=0,9); however, controlling for care pathways, good fit and sensitivity both vanished. No valid models emerged for undertime discharges. The decision trees revealed unique combinations of variables.

Our ML decision trees used on healthcare databases proved promising for focussing quality audit efforts. Narrative analysis for the clinical contexts of the decision trees is indispensable.

No significant relationships.