Using large language model (LLM) to identify high-burden informal caregivers in long-term care.

BACKGROUND: The rising global elderly population increases the demand for caregiving, yet traditional methods may not fully assess the challenges faced by vital informal caregivers. OBJECTIVE: To investigate the efficacy of Large Language Model (LLM) in detecting overburdened informal caregivers, benchmarking against rule-based and machine learning methods. METHODS: 1,791 eligible informal caregivers from Southern Taiwan and utilized their textual case summary reports for the LLM. We also employed structured questionnaire results for machine learning models. Furthermore, we leveraged the visualization of the LLM's attention mechanisms to enhance our understanding of the model's interpretative capabilities. RESULTS: The LLM achieved an Area Under the Receiver Operating Characteristic (AUROC) curve of 0.84 and an Area Under the Precision-Recall Curve (AUPRC) of 0.70, marking an 8% and 14% improvement over traditional methods. The visualization of the attention mechanism accurately reflected the evaluations of human experts, concentrating on descriptions of high-burden descriptions and the relationships between caregivers and recipients. CONCLUSION: This research demonstrates the notable capability of LLM to accurately identify high-burden caregivers in Long-term Care (LTC) settings. Compared to traditional approaches, LLM offers an opportunity for the future of LTC research and policymaking.

Understanding and alleviating informal caregiver burden through the development and validation of a caregiver strain index-based model in Taiwan.

BACKGROUND: Quantifying the informal caregiver burden is important for understanding the risk factors associated with caregiver overload and for evaluating the effectiveness of services provided in Long-term Care (LTC). OBJECTIVE: This study aimed to develop and validate a Caregiver Strain Index (CSI)-based score for quantifying the informal caregiver burden, while the original dataset did not fully cover evaluation items commonly included in international assessments. Subsequently, we utilized the CSI-based score to pinpoint key caregiver burden risk factors, examine the initial timing of LTC services adoption, and assess the impact of LTC services on reducing caregiver burden. METHODS: The study analyzed over 28,000 LTC cases in Southern Taiwan from August 2019 to December 2022. Through multiple regression analysis, we identified significant risk factors associated with caregiver burden and examined changes in this burden after utilizing various services. Survival analysis was employed to explore the relationship between adopting the first LTC services and varying levels of caregiver burden. RESULTS: We identified 126 significant risk factors for caregiver burden. The most critical factors included caregiving for other disabled family members or children under the age of three (ß = 0.74, p < 0.001), the employment status of the caregiver (ß = 0.30-0.53, p < 0.001), the frailty of the care recipient (ß = 0.28-0.31, p < 0.001), and the behavioral symptoms of dementia in care recipients (ß = 0.28-2.60, p < 0.05). Generally, caregivers facing higher burdens sought LTC services earlier, and providing home care services alleviated the caregiver's burden. CONCLUSION: This comprehensive study suggests policy refinements to recognize high-risk caregivers better early and provide timely support to improve the overall well-being of both informal caregivers and care recipients.

Predicting Long-Term Care Service Demands for Cancer Patients: A Machine Learning Approach.

BACKGROUND: Long-term care (LTC) service demands among cancer patients are significantly understudied, leading to gaps in healthcare resource allocation and policymaking. OBJECTIVE: This study aimed to predict LTC service demands for cancer patients and identify the crucial factors. METHODS: 3333 cases of cancers were included. We further developed two specialized prediction models: a Unified Prediction Model (UPM) and a Category-Specific Prediction Model (CSPM). The UPM offered generalized forecasts by treating all services as identical, while the CSPM built individual predictive models for each specific service type. Sensitivity analysis was also conducted to find optimal usage cutoff points for determining the usage and non-usage cases. RESULTS: Service usage differences in lung, liver, brain, and pancreatic cancers were significant. For the UPM, the top 20 performance model cutoff points were adopted, such as through Logistic Regression (LR), Quadratic Discriminant Analysis (QDA), and XGBoost (XGB), achieving an AUROC range of 0.707 to 0.728. The CSPM demonstrated performance with an AUROC ranging from 0.777 to 0.837 for the top five most frequently used services. The most critical predictive factors were the types of cancer, patients' age and female caregivers, and specific health needs. CONCLUSION: The results of our study provide valuable information for healthcare decisions, resource allocation optimization, and personalized long-term care usage for cancer patients.

Prevalence and predictive modeling of undiagnosed diabetes and impaired fasting glucose in Taiwan: a Taiwan Biobank study.

INTRODUCTION: We investigated the prevalence of undiagnosed diabetes and impaired fasting glucose (IFG) in individuals without known diabetes in Taiwan and developed a risk prediction model for identifying undiagnosed diabetes and IFG. RESEARCH DESIGN AND METHODS: Using data from a large population-based Taiwan Biobank study linked with the National Health Insurance Research Database, we estimated the standardized prevalence of undiagnosed diabetes and IFG between 2012 and 2020. We used the forward continuation ratio model with the Lasso penalty, modeling undiagnosed diabetes, IFG, and healthy reference group (individuals without diabetes or IFG) as three ordinal outcomes, to identify the risk factors and construct the prediction model. Two models were created: Model 1 predicts undiagnosed diabetes, IFG_110 (ie, fasting glucose between 110 mg/dL and 125 mg/dL), and the healthy reference group, while Model 2 predicts undiagnosed diabetes, IFG_100 (ie, fasting glucose between 100 mg/dL and 125 mg/dL), and the healthy reference group. RESULTS: The standardized prevalence of undiagnosed diabetes for 2012-2014, 2015-2016, 2017-2018, and 2019-2020 was 1.11%, 0.99%, 1.16%, and 0.99%, respectively. For these periods, the standardized prevalence of IFG_110 and IFG_100 was 4.49%, 3.73%, 4.30%, and 4.66% and 21.0%, 18.26%, 20.16%, and 21.08%, respectively. Significant risk prediction factors were age, body mass index, waist to hip ratio, education level, personal monthly income, betel nut chewing, self-reported hypertension, and family history of diabetes. The area under the curve (AUC) for predicting undiagnosed diabetes in Models 1 and 2 was 80.39% and 77.87%, respectively. The AUC for predicting undiagnosed diabetes or IFG in Models 1 and 2 was 78.25% and 74.39%, respectively. CONCLUSIONS: Our results showed the changes in the prevalence of undiagnosed diabetes and IFG. The identified risk factors and the prediction models could be helpful in identifying individuals with undiagnosed diabetes or individuals with a high risk of developing diabetes in Taiwan.

Cost-Effectiveness Analysis of Different Genetic Testing Strategies for Lynch Syndrome in Taiwan.

Patients with Lynch syndrome (LS) have a significantly increased risk of developing colorectal cancer (CRC) and other cancers. Genetic screening for LS among patients with newly diagnosed CRC aims to identify mutations in the disease-causing genes (i.e., the DNA mismatch repair genes) in the patients, to offer genetic testing for relatives of the patients with the mutations, and then to provide early prevention for the relatives with the mutations. Several genetic tests are available for LS, such as DNA sequencing for MMR genes and tumor testing using microsatellite instability and immunohistochemical analyses. Cost-effectiveness analyses of different genetic testing strategies for LS have been performed in several studies from different countries such as the US and Germany. However, a cost-effectiveness analysis for the testing has not yet been performed in Taiwan. In this study, we evaluated the cost-effectiveness of four genetic testing strategies for LS described in previous studies, while population-specific parameters, such as the mutation rates of the DNA mismatch repair genes and treatment costs for CRC in Taiwan, were used. The incremental cost-effectiveness ratios based on discounted life years gained due to genetic screening were calculated for the strategies relative to no screening and to the previous strategy. Using the World Health Organization standard, which was defined based on Taiwan's Gross Domestic Product per capita, the strategy based on immunohistochemistry as a genetic test followed by BRAF mutation testing was considered to be highly cost-effective relative to no screening. Our probabilistic sensitivity analysis results also suggest that the strategy has a probability of 0.939 of being cost-effective relative to no screening based on the commonly used threshold of $50,000 to determine cost-effectiveness. To the best of our knowledge, this is the first cost-effectiveness analysis for evaluating different genetic testing strategies for LS in Taiwan. The results will be informative for the government when considering offering screening for LS in patients newly diagnosed with CRC.

Policy Design of Multi-Year Crop Insurance Contracts with Partial Payments.

Current crop insurance is designed to mitigate monetary fluctuations resulting from yield losses for a specific year. However, yield realization tendency can vary from year to year and may depend on the correlation of yield realizations across years. When the current single-year Yield Protection (YP) and Area Risk Protection Insurance (ARPI) contracts are extended to multiple periods, actuarially fair premium rate is expected to decrease as poor yield realizations in a year can be offset by another year's better yield realizations. In this study, we first use simulations to demonstrate how significant premium savings are possible when coverage is based on the sum of yields across years rather than on a year-by-year basis. We then describe the design of a multi-year framework of crop insurance and model the insurance using a copula approach. Insurance terms are extended to more than a year and the premium, liability, and indemnity are determined by a multi-year term. Moreover, partial payment is provided at the end of each term to offset the possibility of significant loss in a single term. County-level data obtained from the U.S. Department of Agriculture are used to demonstrate the implementations of the proposed multi-year crop insurance. The proposed multi-year plan would benefit farmers by offering insurance guarantees across years for significantly lower costs.

A two-stage random forest-based pathway analysis method.

Pathway analysis provides a powerful approach for identifying the joint effect of genes grouped into biologically-based pathways on disease. Pathway analysis is also an attractive approach for a secondary analysis of genome-wide association study (GWAS) data that may still yield new results from these valuable datasets. Most of the current pathway analysis methods focused on testing the cumulative main effects of genes in a pathway. However, for complex diseases, gene-gene interactions are expected to play a critical role in disease etiology. We extended a random forest-based method for pathway analysis by incorporating a two-stage design. We used simulations to verify that the proposed method has the correct type I error rates. We also used simulations to show that the method is more powerful than the original random forest-based pathway approach and the set-based test implemented in PLINK in the presence of gene-gene interactions. Finally, we applied the method to a breast cancer GWAS dataset and a lung cancer GWAS dataset and interesting pathways were identified that have implications for breast and lung cancers.