A significant challenge facing societies worldwide is the increasing demand for medical and daily care services among the elderly population, particularly as demographies shift toward older age groups. A recent study conducted by researchers draws attention to this pressing issue by constructing predictive models for the demand of these services among the elderly, utilizing advanced machine learning techniques.
This study, published on March 11, 2025, systematically examines the needs of elderly individuals aged 60 years and older across three major cities in China: Guangzhou, Suzhou, and Qingdao. By distributing 1,380 questionnaires, with 1,291 returned (yielding a 93.6% response rate), the researchers aimed to identify how various factors influence the demand for medical and daily care services.
The results revealed concerning statistics: 87.5% of the participants indicated they required medical services, and 70.8% pointed to needing daily care services. This situation is alarming, considering the skyrocketing number of elderly individuals, now accounting for approximately 264 million—or 18.7%—of China's total population, as stated by the nation's seventh census.
The study employed three algorithms—Random Forest (RF), Gradient Boosting Decision Tree (GBDT), and Light Gradient Boosting Machine (LightGBM)—to develop predictive models. These models were evaluated using accuracy, recall, precision, F1-score, and area under the receiver operating characteristic curve (AUC) metrics. Notably, the LightGBM model emerged as the superior prediction tool, achieving remarkable performance: AUC values of 0.910 for medical service demand prediction and 0.906 for daily care services demand.
Key factors influencing the demand for these healthcare services included the number of chronic diseases, the education level of the respondents, and their financial sources. Chronic diseases play a pivotal role, adversely impacting the quality of life and increasing the necessity for medical and daily care support. The study emphasizes the importance of tailoring healthcare policies to address the unique demands presented by the elderly.
The aging population is not a unique challenge for China alone; it resonates globally. There is significant concern about the adequacy of healthcare systems to meet the rising needs of older populations. Based on the survey results, the researchers advocate for establishing comprehensive eldercare policies and service structures to alleviate the challenges faced by healthcare providers and improve the well-being of the elderly.
Future efforts must focus on creating adaptive frameworks capable of meeting the needs of vulnerable and financially disadvantaged elderly individuals. The integration of predictive modeling through machine learning could offer substantial advancements for optimizing resource allocation, ensuring timely and effective medical and daily care service delivery. An unprecedented surge in demand necessitates demands responsiveness from policymakers and healthcare systems to support aging populations effectively.
Drawing insights from the study, it is pivotal for governments to combine medical insurance with care services to streamline access for the elderly. By breaking down barriers between healthcare and supportive services, meaningful advancements can be made, facilitating the elderly’s ability to receive comprehensive medical and daily care without unnecessary burdens.
While this study offers invaluable insights, limitations were acknowledged, including its cross-sectional nature and the absence of clinical medical indicators, necessitating future longitudinal research to validate these findings and broaden the predictive model's applicability.
Through these efforts, society could forge pathways toward more effective healthcare delivery models, ensuring inclusivity for the elderly as the global demographic trends continue to evolve.