\u8b1b\u00a0 \u984c\uff1aCausal Inference in Real-World Evidence: Applications to Population Health Research<\/p>\n
\u4e3b\u8b1b\u4eba\uff1a\u9ec3\u610f\u5a77 \u52a9\u7406\u6559\u6388(\u4e2d\u592e\u7814\u7a76\u9662\u7d71\u8a08\u79d1\u5b78\u7814\u7a76\u6240)<\/p>\n
\u6642\u00a0 \u9593\uff1a2026\u5e7410\u670815\u65e5\uff08\u661f\u671f\u56db\uff09\u4e0b\u534802\uff1a10 – 04\uff1a00<\/p>\n
\u5730\u00a0 \u9ede\uff1aB302A\uff08\u6de1\u6c34\u6821\u5712\u5546\u7ba1\u5927\u6a13\uff09<\/p>\n
\u8336\u00a0 \u6703\uff1a2026\u5e7410\u670815\u65e5\uff08\u661f\u671f\u56db\uff09\u4e0b\u534801\uff1a30 (\u5546\u7ba1\u5927\u6a13 B1102)<\/p>\n
Real-world data offer unique opportunities to study disease progression and evaluate healthcare interventions at the population level. Yet, the scale and richness of these data do not automatically translate into reliable evidence. Challenges such as confounding, selection, outcome definition, and the timing of exposure and events require careful integration of epidemiologic study design and statistical methodology.<\/p>\n
This presentation highlights two lines of research on transforming large-scale observational health data into interpretable real-world evidence. The first examines the natural history of hepatitis C virus infection using longitudinal data to characterize disease progression and long-term clinical outcomes, with particular attention to time-to-event processes and the challenges of analyzing complex disease trajectories. The second evaluates influenza vaccine effectiveness among adults aged 65 years and older in Taiwan during the 2023\u20132024 and 2024\u20132025 influenza seasons, illustrating how causal inference methods, including propensity score approaches and time-to-event analyses, can be applied to address confounding and time-related biases in observational vaccine studies.<\/p>\n
Together, these studies illustrate how causal inference principles and rigorous statistical analysis can strengthen evidence generated from observational health data. Building on this work, my research also extends to more complex causal questions, including mediation analysis for longitudinal and time-to-event data, with the goal of improving causal interpretation in real-world evidence studies.<\/p>\n