A cross-sectional study assessing the reliability and quality of educational videos on hydrocephalus available on TikTok and Bilibili.

In recent decades, increasing concern about the health of agricultural workers has elevated this issue to a public health priority. Farming populations exhibit distinct health profiles, particularly a higher prevalence of certain neurodegenerative diseases such as Parkinson's disease (PD). PD is the most common movement disorder globally. It is a progressive neurodegenerative condition that affects millions of individuals worldwide, substantially impairing quality of life. Its etiology is multifactorial, involving a complex interplay between genetic, behavioral, and environmental factors. Notably, only about 5–10% of PD cases are monogenic, caused by pathogenic variants in single genes. Overall heritability is commonly estimated to be around 30%, while currently identified genetic variants explain approximately 16–36% of this heritable risk. Aggregate polygenic liability likely interacts with environmental influences to increase disease risk among individuals without monogenic forms. Several of environmental factors, such as smoking and alcohol consumption, have been identified as potential contributors to PD risk. Understanding these associations is critical for informing prevention strategies. However, the precise role of many environmental factors remains unclear, particularly regarding whether they are causative, contributory, or protective. This ambiguity may stem from several methodological challenges, including the long prodromal phase of PD, which complicates the differentiation between early symptoms and independent risk factors. Additional obstacles include small sample sizes, limited follow-up durations, and geographical heterogeneity across study populations, as well as the restricted range of risk factors typically examined in individual studies. Consequently, certain environmental exposures, especially pesticides, remain central to ongoing scientific and public health debates regarding PD etiology. A recent review emphasized the urgent need for large-scale cohort studies to advance the understanding of PD risk factors. Although traditional epidemiological designs remain essential, they are often constrained by limited sample sizes, restricted geographic coverage, and resource-intensive data collection. In this context, administrative health data provide a valuable complementary approach. These population-based datasets encompass large cohorts over extended time periods and enable retrospective analyses without requiring new data collection efforts. Recent progress in artificial intelligence, coupled with growing access to real-world data, has further expanded opportunities in public health research, particularly through the application of machine learning to identify risk factors. Applying artificial intelligence to administrative health data may therefore yield important insights into public health challenges such as identifying potential PD predictors for hypothesis generation. Overall, the integration of artificial intelligence with administrative health data represents a promising avenue for uncovering potential contributors to PD at the population scale.