Search

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. [ABSTRACT FROM AUTHOR]

Background: Behavioral and psychological symptoms of dementia (BPSD) are associated with increased stress, burden, and depression among family caregivers of people with dementia. STAR-Caregivers Virtual Training and Follow-up (STAR-VTF) is adapted from an evidence-based, in-person program that trains family caregivers to manage BPSD. We used a human-centered design approach to obtain feedback from family caregivers about STAR-VTF. The program will be evaluated using a pragmatic randomized trial. Objective: The objective of the study was to understand the needs of family caregivers for improving BPSD management and the extent to which caregivers perceived that STAR-VTF could address those needs. Methods: Between July and September 2019, we conducted 15 semistructured interviews with family caregivers of people with dementia who receive care at Kaiser Permanente Washington in the Seattle metropolitan area. We identified participants from electronic health records, primarily based on a prescription for antipsychotic medication for the person with dementia (a proxy for caregivers dealing with BPSD). We showed caregivers low-fidelity prototypes of STAR-VTF online self-directed materials and verbally described potential design elements. We obtained caregiver feedback on these elements, focusing on their needs and preferences and perceived barriers to using STAR-VTF. We used a hybrid approach of inductive and deductive coding and aggregated codes to develop themes. Results: The idea of a virtual training program for learning to manage BPSD appealed to caregivers. They said health care providers did not provide adequate education in the early disease stages about the personality and behavior symptoms that can affect people with dementia. Caregivers found it unexpected and frustrating when the person with dementia began experiencing BPSD, symptoms they felt unprepared to manage. Accordingly, caregivers expressed a strong desire for the health care organization to offer programs such as STAR-VTF much sooner. Caregivers had already put considerable effort into problem solving challenging behaviors. They anticipated deriving less value from STAR-VTF at that point. Nonetheless, many were interested in the virtual aspect of the training due to the convenience of receiving help from home and the perception that help from a virtual program would be timelier than traditional service modalities (eg, face to face). Given caregivers' limited time, they suggested dividing the STAR-VTF content into chunks to review as time permitted. Caregivers were interested in having a STAR-VTF provider for additional support in managing challenging behaviors. Caregivers reported a preference for having the same coach for the program duration. Conclusions: Caregivers we interviewed would likely accept a virtual training program such as STAR-VTF to obtain information about BPSD and receive help managing it. Family caregivers anticipated deriving more value if STAR-VTF was offered earlier in the disease course. JMIR Aging 2021;4(1):e24965 doi:10.2196/24965 [ABSTRACT FROM AUTHOR]

Programs such as the Pediatric Access Line in Washington state have shown decreases in antipsychotic medication use by youth with non-psychotic disorders. Program outcomes have been studied with observational designs. This manuscript describes the protocol for Targeted and Safer Use of Antipsychotics in Youth (SUAY), a randomized controlled trial of psychiatrist review of prescriptions and facilitated access to psychosocial care. The aim of the intervention is to reduce the number of person-days of antipsychotic use among participants. Recruitment occurs at 4 health systems. Targeted enrollment is 800 youth aged 3–17 years. Clinicians are block randomized to intervention versus usual care prior to the study. Youth are nested within the arm of the prescribing clinician. Clinicians in the intervention group receive an EHR-based best practice alert with options to expedite access to psychosocial care and all medication orders are reviewed by a child and adolescent psychiatrist with feedback provided to the prescriber. The primary outcome is person-days of antipsychotic medication use in the 6 months following the initial order. All randomized individuals contribute data regardless of their level of participation (including declining all services). The trial has been approved by the institutional review boards at each of the 4 sites. The intervention has 4 novel design features including automated recruitment using a best practice alert, psychiatrist medication order review and consultation, telephone navigation to psychosocial care, and telemental health visits. Recruitment began in March of 2018 and will be completed in June 2020. Follow-up will be completed December 31, 2020. Clinicaltrials.gov , NCT03448575 [ABSTRACT FROM AUTHOR]

Rates of child maltreatment (CM) obtained from electronic health records are much lower than national child welfare prevalence rates indicate. There is a need to understand how CM is documented to improve reporting and surveillance. To examine whether using natural language processing (NLP) in outpatient chart notes can identify cases of CM not documented by ICD diagnosis code, the overlap between the coding of child maltreatment by ICD and NLP, and any differences by age, gender, or race/ethnicity. Outpatient chart notes of children age 0–18 years old within Kaiser Permanente Washington (KPWA) 2018–2020 were used to examine a selected set of maltreatment-related terms categorized into concept unique identifiers (CUI). Manual review of text snippets for each CUI was completed to flag for validated cases and retrain the NLP algorithm. The NLP results indicated a crude rate of 1.55 % to 2.36 % (2018–2020) of notes with reference to CM. The rate of CM identified by ICD code was 3.32 per 1000 children, whereas the rate identified by NLP was 37.38 per 1000 children. The groups that increased the most in identification of maltreatment from ICD to NLP were adolescents (13–18 yrs. old), females, Native American children, and those on Medicaid. Of note, all subgroups had substantially higher rates of maltreatment when using NLP. Use of NLP substantially increased the estimated number of children who have been impacted by CM. Accurately capturing this population will improve identification of vulnerable youth at high risk for mental health symptoms. [ABSTRACT FROM AUTHOR]