Improving cost‐effectiveness of epidemiological studies via designed missingness strategies

摘要: Modern epidemiological studies face opportunities and challenges posed by an ever-expanding capacity to measure a wide range of environmental exposures, along with sophisticated biomarkers exposure response at the individual level. The challenge deciding what is further complicated for longitudinal studies, where logistical cost constraints preclude collection all possible measurements on participants every follow-up time. This true National Children’s Study (NCS), large scale study that will enroll children starting in early pregnancy gather information their development environment through adulthood. success NCS significantly depend accurate, yet cost-effective, characterization exposures thought be related health outcomes interest. purpose this paper explore use saving, valid adequately powered statistical approaches gathering within cohort studies. proposed approach involves detailed assessment specially selected subset population, less-costly, presumably less-detailed less-burdensome, surrogate measures across entire cohort. We show efficiency costs burden may achieved without making substantive sacrifices ability draw reliable inferences concerning relationship between outcome. Several scenarios are provided document how targeted sub-sampling design strategy can benefit such as NCS, well other more focused epidemiologic