Optimizing the diagnosis and management of dementia within primary care: a systematic review of systematic reviews

Development of an AI-Based Predictive Algorithm for Early Diagnosis of High-Risk Dementia Groups among the Elderly: Utilizing Health Lifelog Data

BACKGROUND/OBJECTIVES

This study aimed to develop a predictive algorithm for the early diagnosis of dementia in the high-risk group of older adults using artificial intelligence technologies. The objective is to create an accessible diagnostic method that does not rely on traditional medical equipment, thereby improving the early detection and management of dementia.

METHODS

Lifelog data from wearable devices targeting this high-risk group were collected from the AI Hub platform. Various indicators from these data were analyzed to develop a dementia diagnostic model. Machine learning techniques such as Logistic Regression, Random Forest, LightGBM, and Support Vector Machine were employed. Data augmentation techniques were applied to address data imbalance, thereby enhancing the model performance.

RESULTS

Data augmentation significantly improved the model's accuracy in classifying dementia cases. Specifically, in gait data, the SVM model performed with an accuracy of 0.879. In sleep data, a Logistic Regression was performed, yielding an accuracy of 0.818. This indicates that the lifelog data can effectively contribute to the early diagnosis of dementia, providing a practical solution that can be easily integrated into healthcare systems.

CONCLUSIONS

This study demonstrates that lifelog data, which are easily collected in daily life, can significantly enhance the accessibility and efficiency of dementia diagnosis, aiding in the effective use of medical resources and potentially delaying disease progression.

Increasing Identification and Follow-Up of Cognitive Impairment in Primary Care: A Quality Improvement Pilot Study

INTRODUCTION

Beyond our population growing older and living longer, there is an increased risk of developing a cognitive disorder. Standardized screening during a routine visit in primary care may be ideal for early detection of mild cognitive impairment (MCI) and follow-up for cognitive changes.

AIM

This quality improvement (QI) project aimed to determine the impact of implementing the Mini-Cog© quick screening for early dementia detection to identify and follow up on the cognitive impairment of older adults in a primary care clinic setting.

METHODS

Implementation started in February 2024 in a primary care clinic in Southern California. Data was collected for this project over a total of 16 weeks. This QI project implemented a routine cognitive screening using the Mini-Cog©. Cognitive impairment was identified, and if indicated by the Mini-Cog© scores, follow-up for a cognitive assessment and care plan services were initiated. Data were obtained from the project site's electronic medical record on a total sample size of 471 participants (n = 382 in the pre-implementation group and n = 89 in the post-implementation group).

RESULTS

Pearson's chi-square test indicated a statistically significant improvement in the identification rate of cognitive impairment, increasing from 11.8% (n = 45 out of 382) at pre-implementation to 34.8% (n = 31 out of 89) at post-implementation, and specifically, mild cognitive impairment increased from zero identified in pre-implementation to 12.4% (n = 11 out of 89) post-implementation. Lastly, follow-up rates improved from 91.1% (n = 41 out of 45) to 100% (n = 31 out of 31) in post-implementation, and clinical significance was evident based on the phi-coefficient (φ = 0.196), indicating a small effect size and a 100% follow-up rate.

CONCLUSIONS

The findings of this project suggest older adults should receive cognitive screenings to help identify early cognitive impairment and increase follow-up for further evaluation, treatment, and advanced care planning.

Understanding barriers to and facilitators of clinician-patient conversations about brain health and cognitive concerns in primary care: a systematic review and practical considerations for the clinician

BACKGROUND

Primary care clinicians (PCCs) are typically the first practitioners to detect cognitive impairment in their patients, including those with Alzheimer's disease or related dementias (ADRD). However, conversations around cognitive changes can be challenging for patients, family members, and clinicians to initiate, with all groups reporting barriers to open dialogue. With the expanding array of evidence-based interventions for ADRD, from multidomain care management to novel biotherapeutics for early-stage AD, incorporating conversations about brain health into routine healthcare should become a standard of care. We conducted a systematic review to identify barriers to and facilitators of brain health conversations in primary care settings.

METHODS

We systematically searched PubMed, Scopus, Web of Science, and the Cochrane Library for qualitative or quantitative studies conducted in the US between January 2000 and October 2022 that evaluated perceptions of cognition and provider-patient brain health conversations prior to formal screening for, or diagnosis of, mild cognitive impairment or ADRD. We assessed the quality of the included studies using the Mixed Methods Appraisal Tool.

RESULTS

In total, 5547 unique abstracts were screened and 22 articles describing 19 studies were included. The studies explored perceptions of cognition among laypersons or clinicians, or provider-patient interactions in the context of a patient's cognitive concerns. We identified 4 main themes: (1) PCCs are hesitant to discuss brain health and cognitive concerns; (2) patients are hesitant to raise cognitive concerns; (3) evidence to guide clinicians in developing treatment plans that address cognitive decline is often poorly communicated; and (4) social and cultural context influence perceptions of brain health and cognition, and therefore affect clinical engagement.

CONCLUSIONS

Early conversations about brain health between PCCs and their patients are rare, and effective tools, processes, and strategies are needed to make these vital conversations routine.

Clinical Uncertainty Influences Antibiotic Prescribing for Upper Respiratory Tract Infections: A Qualitative Study of Township Hospital Physicians and Village Doctors in Rural Shandong Province, China

OBJECTIVE

This study aimed to explore how clinical uncertainty influences antibiotic prescribing practices among township hospital physicians and village doctors in rural Shandong Province, China.

METHODS

Qualitative semi-structured interviews were conducted with 30 township hospital physicians and 6 village doctors from rural Shandong Province, China. A multi-stage random sampling method was used to identify respondents. Conceptual content analysis together with Colaizzi's method were used to generate qualitative codes and identify themes.

RESULTS

Three final thematic categories emerged during the data analysis: (1) Incidence and treatment of Upper Respiratory Tract Infections (URTIs) in township hospitals and village clinics; (2) Antibiotic prescribing practices based on the clinical experience of clinicians; (3) Influence of clinical uncertainty on antibiotic prescribing. Respondents from both township hospitals and village clinics reported that URTIs were the most common reason for antibiotic prescriptions at their facilities and that clinical uncertainty appears to be an important driver for the overuse of antibiotics for URTIs. Clinical uncertainty was primarily due to: (1) Diagnostic uncertainty (establishing a relevant diagnosis is hindered by limited diagnostic resources and capacities, as well as limited willingness of patients to pay for investigations), and (2) Insufficient prognostic evidence. As a consequence of the clinical uncertainty caused by both diagnostic and prognostic uncertainty, respondents stated that antibiotics are frequently prescribed for URTIs to prevent both prolonged courses or recurrence of the disease, as well as clinical worsening, hospital admission, or complications.

CONCLUSION

Our study suggests that clinical uncertainty is a key driver for the overuse and misuse of prescribing antibiotics for URTIs in both rural township hospitals and village clinics in Shandong province, China, and that interventions to reduce clinical uncertainty may help minimize the unnecessary use of antibiotics in these settings. Interventions that use clinical rules to identify patients at low risk of complications or hospitalization may be more feasible in the near-future than laboratory-based interventions aimed at reducing diagnostic uncertainty.

Adapting a Dementia Care Management Intervention for Regional Implementation: A Theory-Based Participatory Barrier Analysis

Dementia is a leading cause of disability and dependency in older people worldwide. As the number of people affected increases, so does the need for innovative care models. Dementia care management (DCM) is an empirically validated approach for improving the care and quality of life for people with dementia (PwD) and caregivers. The aim of this study is to investigate the influencing factors and critical pathways for the implementation of a regionally adapted DCM standard in the existing primary care structures in the German region of Siegen-Wittgenstein (SW). Utilizing participatory research methods, five local health care experts as co-researchers conducted N = 13 semi-structured interviews with 22 local professionals and one caregiver as peer reviewers. Data collection and analysis were based on the Consolidated Framework for Implementation Research (CFIR). Our results show that among the most mentioned influencing factors, three CFIR constructs can be identified as both barriers and facilitators: Patients' needs and resources, Relative advantage, and Cosmopolitanism. The insufficient involvement of relevant stakeholders is the major barrier and the comprehensive consideration of patient needs through dementia care managers is the strongest facilitating factor. The study underlines the vital role of barrier analysis in site-specific DCM implementation.