Development of a machine learning model to predict mild cognitive impairment using natural language processing in the absence of screening.

BACKGROUND: Patients and their loved ones often report symptoms or complaints of cognitive decline that clinicians note in free clinical text, but no structured screening or diagnostic data are recorded. These symptoms/complaints may be signals that predict who will go on to be diagnosed with mild cognitive impairment (MCI) and ultimately develop Alzheimer's Disease or related dementias. Our objective was to develop a natural language processing system and prediction model for identification of MCI from clinical text in the absence of screening or other structured diagnostic information. METHODS: There were two populations of patients: 1794 participants in the Adult Changes in Thought (ACT) study and 2391 patients in the general population of Kaiser Permanente Washington. All individuals had standardized cognitive assessment scores. We excluded patients with a diagnosis of Alzheimer's Disease, Dementia or use of donepezil. We manually annotated 10,391 clinic notes to train the NLP model. Standard Python code was used to extract phrases from notes and map each phrase to a cognitive functioning concept. Concepts derived from the NLP system were used to predict future MCI. The prediction model was trained on the ACT cohort and 60% of the general population cohort with 40% withheld for validation. We used a least absolute shrinkage and selection operator logistic regression approach (LASSO) to fit a prediction model with MCI as the prediction target. Using the predicted case status from the LASSO model and known MCI from standardized scores, we constructed receiver operating curves to measure model performance. RESULTS: Chart abstraction identified 42 MCI concepts. Prediction model performance in the validation data set was modest with an area under the curve of 0.67. Setting the cutoff for correct classification at 0.60, the classifier yielded sensitivity of 1.7%, specificity of 99.7%, PPV of 70% and NPV of 70.5% in the validation cohort. DISCUSSION AND CONCLUSION: Although the sensitivity of the machine learning model was poor, negative predictive value was high, an important characteristic of models used for population-based screening. While an AUC of 0.67 is generally considered moderate performance, it is also comparable to several tests that are widely used in clinical practice.

Physical Activity Intervention Adaptation: Recommendations from Rural American Indian Older Adults.

Preventive interventions are critical to improving health equity among American Indian (AI) populations, yet interventions that promote physical activity (PA) among AI populations are scarce. This research addresses the research-to-practice gap by informing the adaption and implementation process of evidence-based interventions (EBIs) among rural AI older adults. We used a community-based approach and an Indigenous-focused adaptation theoretical framework. Qualitative, semi-structured interviews elicited detailed information on preferences for PA intervention among rural AI older adults. We applied a collaborative directed content analysis strategy, and established trustworthiness and relevance using an inter-rater reliability process and member checking. We conducted 21 interviews, all participants identified as AI, the mean age was 66 years (SD = 7.6), and 57% were female. Themes characterized contextual and cultural intervention considerations for adapting and implementing evidence-based PA interventions in rural AI older adults. Key findings included an emphasis on social and community interaction, strategies for targeted engagement, preference for group format, pairing PA sessions with shared meals, and inclusiveness in the PA intervention across ability levels and age groups. This study identified opportunities for adaptation of PA-focused EBIs among rural AI older adults. Findings can be applied to support the adaptation and implementation of effective and relevant PA-focused preventive interventions among this population which is at high risk for chronic disease and health disparities.

ACT and Recovery: What We Know About Their Compatibility.

While assertive community treatment (ACT) is a widely implemented evidence-based practice, the extent of its recovery orientation has been debated. A literature search identified 16 empirical articles studying recovery and ACT. These 16 studies were classified as involving stakeholder perceptions, interventions, or fidelity measurement. Stakeholders generally viewed ACT as being recovery oriented; research on both interventions and fidelity measurement showed promising approaches. Overall the literature yielded encouraging findings regarding ACT and recovery, though there remains a dearth of research on the topic. We discuss future directions for research and practice to ensure that ACT programs skillfully support recovery.

Developing home-based telemental health services for youth: Practices from the SUAY Study.

INTRODUCTION: There are no published procedural or safety guidelines for home-based telemental health (TMH) therapy with youth, despite the unique challenges and risks of providing services to this population outside of a traditional clinic setting. We developed clinical, logistical, and safety procedures for home-based TMH with youth in the context of a large clinical trial. METHODS: A Targeted Approach to Safer Use of Antipsychotics in Youth (SUAY) study identifies youth ages 3-17 who are prescribed second-generation antipsychotic medication for non-psychotic disorders within large healthcare systems. Prescribing physicians receive psychopharmacology consultation. Patients receive a "navigator" to coordinate treatments and access to TMH if they do not have a local therapist. We optimized access by allowing TMH sessions to take place in the family's home, while providing guidelines for privacy, safety, and in-session crises. RESULTS: Clinical issues included providing flexibility in the treatment modality and engaging families. Logistical issues included remote consenting for treatment and troubleshooting technological problems. Safety issues included crisis and safety planning with the youth and family before and during treatment. DISCUSSION: The provision of home-based TMH therapy for youth requires adaptations to existing TMH procedural and safety guidelines to optimize clinical care, technology coordination, and safety.Trial registration number and trial register: Clinicaltrials.gov: NCT03448575.

Uplifting the voices of rural American Indian older adults to improve understanding of physical activity behavior.

American Indian (AI) older adults experience pronounced health disparities and demonstrate one of the lowest levels of physical activity (PA) among racial and ethnic groups. Nearly half of AI older adults live in rural areas, indicating distinct challenges to participation in PA. Research to identify factors influencing PA in this population is missing from the literature, yet is critical to informing culturally relevant PA intervention development and implementation. The purpose was to identify barriers to and facilitators of PA among rural AI older adults using the ecological model and qualitative methods. A community-based approach was used to conduct semi-structured interviews with rural AI older adults. Interview questions were based on a multi-level ecological model. Content analysis was performed, using an iterative coding process to identify findings. The mean age of participants (n = 21) was 66 years. Barriers to and facilitators of PA were identified across ecological model levels. Barriers included factors such as caregiving and community responsibilities, lack of acceptable areas for walking, and overall lack of community-level support for older adult health. Facilitators included a personal connection to the land and ancestors through PA, multigenerational participation, and supportive tribal policies. This study addressed a gap in the literature by identifying barriers to and facilitators of PA among rural AI older adults, which can inform PA intervention development. With barriers and facilitators identified by AI older adults themselves, the voices of those directly affected are uplifted to shape efforts toward addressing longstanding health disparities through relevant public health interventions.