Large long-term epidemiological studies
Large-scale and long-term studies generate the highest-quality epidemiological evidence on the relationships between common exposures and health conditions. They also present special challenges, for example in ensuring that data provided by study participants is accurate and suitable for research on health outcomes over a long period of time.
Validity of data is a major focus of our methodological work. Information on characteristics such as height and weight, blood pressure and use of prescription medications is reported by participants at various time points throughout our studies. By studying the reliability and repeatability of these data, we can estimate how reporting or measurement errors may affect the results of epidemiological analyses.
Equally important is the validity of information on health conditions obtained through linkage to other data sources (see Data linkage/e-health data). We compare diagnoses and medical procedure data from multiple sources to determine which conditions can be reliably identified from routine health data, including hospital records, cancer registries, and primary care records.