Predictive care: a protocol for a computational ethnographic approach to building fair models of inpatient violence in emergency psychiatry.

INTRODUCTION: Managing violence or aggression is an ongoing challenge in emergency psychiatry. Many patients identified as being at risk do not go on to become violent or aggressive. Efforts to automate the assessment of risk involve training machine learning (ML) models on data from electronic health records (EHRs) to predict these behaviours. However, no studies to date have examined which patient groups may be over-represented in false positive predictions, despite evidence of social and clinical biases that may lead to higher perceptions of risk in patients defined by intersecting features (eg, race, gender). Because risk assessment can impact psychiatric care (eg, via coercive measures, such as restraints), it is unclear which patients might be underserved or harmed by the application of ML. METHODS AND ANALYSIS: We pilot a computational ethnography to study how the integration of ML into risk assessment might impact acute psychiatric care, with a focus on how EHR data is compiled and used to predict a risk of violence or aggression. Our objectives include: (1) evaluating an ML model trained on psychiatric EHRs to predict violent or aggressive incidents for intersectional bias; and (2) completing participant observation and qualitative interviews in an emergency psychiatric setting to explore how social, clinical and structural biases are encoded in the training data. Our overall aim is to study the impact of ML applications in acute psychiatry on marginalised and underserved patient groups. ETHICS AND DISSEMINATION: The project was approved by the research ethics board at The Centre for Addiction and Mental Health (053/2021). Study findings will be presented in peer-reviewed journals, conferences and shared with service users and providers.

Race and Racialization in Mental Health Research and Implications for Developing and Evaluating Machine Learning Models: A Rapid Review.

Machine learning models are often trained on sociodemographic features to predict mental health outcomes. Biases in the collection of race-related data can limit the development of useful and fair models. To assess the current state of this data in mental health research, we conducted a rapid review guided by Critical Race Theory. Findings reveal limitations in the measurement and reporting of race and ethnicity, potentially leading to models that amplify health inequities.

Conceptualising fairness: three pillars for medical algorithms and health equity.

OBJECTIVES: Fairness is a core concept meant to grapple with different forms of discrimination and bias that emerge with advances in Artificial Intelligence (eg, machine learning, ML). Yet, claims to fairness in ML discourses are often vague and contradictory. The response to these issues within the scientific community has been technocratic. Studies either measure (mathematically) competing definitions of fairness, and/or recommend a range of governance tools (eg, fairness checklists or guiding principles). To advance efforts to operationalise fairness in medicine, we synthesised a broad range of literature. METHODS: We conducted an environmental scan of English language literature on fairness from 1960-July 31, 2021. Electronic databases Medline, PubMed and Google Scholar were searched, supplemented by additional hand searches. Data from 213 selected publications were analysed using rapid framework analysis. Search and analysis were completed in two rounds: to explore previously identified issues (a priori), as well as those emerging from the analysis (de novo). RESULTS: Our synthesis identified 'Three Pillars for Fairness': transparency, impartiality and inclusion. We draw on these insights to propose a multidimensional conceptual framework to guide empirical research on the operationalisation of fairness in healthcare. DISCUSSION: We apply the conceptual framework generated by our synthesis to risk assessment in psychiatry as a case study. We argue that any claim to fairness must reflect critical assessment and ongoing social and political deliberation around these three pillars with a range of stakeholders, including patients. CONCLUSION: We conclude by outlining areas for further research that would bolster ongoing commitments to fairness and health equity in healthcare.

"Dirty like a Tenant": Migration and Embodied Dispositions in Malawi.

Much has been written about how structural (e.g. colonialism) and social (e.g. gender) determinants shape embodied health outcomes. However, little attention has been paid to the ways that marginalized populations become complicit in their own oppression. Ethnographic data collected over two years at a rural public hospital in Malawi show that the tobacco political economy produces significant intra-rural inequalities that result in the exclusion of migrant farm workers, called "tenants," from HIV care. Using an analytical framework informed by Bourdieu's concepts of social field and habitus, I illustrate how social inequalities persist unchallenged, even by the most disadvantaged people.

Increasing participation in research with therapy dogs: A qualitative study at a large urban mental health and addiction hospital.

The benefits of involving patients as partners in research across diverse medical and psychiatric settings are well established in the literature. However, researchers continue to struggle to access, engage and retain participants from hard-to-reach populations. The main objective of this study was to co-create pet therapy activities with patients admitted for serious and complex mental illness to a large urban mental health and addiction hospital. Informed by the principles of participatory action research methodology, we conducted focus group discussions with 38 inpatients in seven different clinical units. An experienced volunteer handler and a certified therapy dog helped facilitate our discussions. Participating researchers, recreational therapists, volunteer handlers and our participants all reported that the presence of a certified therapy dog at each of our discussions was integral to their success. Certified therapy dogs increased the motivation to participate in our study, helped to build rapport with participants and created connections in our discussions that enriched our data. To our knowledge our study is the first to demonstrate the value of using a therapy dog as a participatory research tool in a healthcare setting. The authors believe that therapy dogs are a low-tech intervention that could be used effectively to engage hard-to-reach populations in research about their treatment and care in a diverse range of medical settings. These findings support the creation of a pilot study to test the value of including therapy dogs in patient-centered research with vulnerable and hard-to-reach populations.

Being there: A scoping review of grief support training in medical education.

INTRODUCTION: Medical education experts argue that grief support training for physicians would improve physician and patient and family wellness, and should therefore be mandatory. However, there is little evidence about the range of curricula interventions or the impact of grief training. The aim of this scoping review was to describe the current landscape of grief training worldwide in medical school, postgraduate residency and continuing professional development in the disciplines of pediatrics, family medicine and psychiatry. METHODS: Using Arksey and O'Malley's scoping review principles, MEDLINE, EMBASE, ERIC, PsychInfo and Web of Science were searched by a librarian. Two levels of screening took place: a title and abstract review for articles that fit a predefined criteria and a full-text review of articles that met those criteria. Three investigators reviewed the articles and extracted data for analysis. To supplement the search, we also scanned the reference lists of included studies for possible inclusion. RESULTS: Thirty-seven articles published between 1979 and 2019 were analyzed. Most articles described short voluntary grief training workshops. At all training levels, the majority of these workshops focused on transmitting knowledge about the ethical and legal dimensions of death, dying and bereavement in medicine. The grief trainings described were characterized by the use of diverse pedagogical tools, including lectures, debriefing sessions, reflective writing exercises and simulation/role-play. DISCUSSION: Grief training was associated with increased self-assessed knowledge and expertise; however, few of the studies analyzed the impact of grief training on physician and patient and family wellness. Our synthesis of the literature indicates key gaps exist, specifically regarding the limited emphasis on improving physicians' communication skills around death and dying and the limited use of interactive and self-reflexive learning tools. Most trainings also had an overly narrow focus on bereavement grief, rather than a more broadly defined definition of loss.

"There was no love there": Intergenerational HIV disclosure, and late presentation for antiretroviral therapy in Northern Malawi.

Despite access to antiretroviral testing and treatment, high rates of mortality among HIV infected infants and young children persist, often because they are diagnosed too late to benefit from treatment. Most research assumes that treatment delays are a proxy indicator for ongoing HIV-related stigma. My argument is different. Instead I argue that secrecy and truth-telling are socially produced; that is I consider how gendered and intergenerational dynamics regulate how and to whom secrets should be shared. In this article I draw on two years of ethnographic fieldwork (2008-2010) in Northern Malawi with 35 HIV positive children, their primary caregivers, as well as multiple interviews with their extended therapy networks [N = 96] and community stakeholders [N = 72] to examine how social hierarchies influenced the timing of an HIV diagnosis for infected infants. My findings indicate that it is neither necessary nor strategically advantageous for some women to disclose their HIV status to their husbands. Rather, grandparents play pivotal roles at facilitating HIV disclosure between intimate partners, which in turn leads to timely HIV diagnoses for infected children. This article contributes to a body of literature that questions the usefulness of the concept of "stigma" for understanding late presentation for ART among infants and children.