Quantifying the Level of Knowledge of Health Risks for New Tobacco Products: Development of a Measurement Certainty Index

      Kallischnigg, G.; Weitkunat, R.; Browne, H.; Smith, M.; Sanders, E.; Dempsey, R.; H-Urban, H-J. Urban.
      Conference date
      Dec 7, 2008
      Conference name
      Society for Risk Analysis (SRA) 2008

      Epidemiological risk assessment of new, potentially reduced-risk tobacco products (PRRTP) relates use of these products to morbidity and mortality risks. It quantifies the probability of developing a disease under specific exposure conditions. Although an epidemiological approach is the benchmark for health-risk assessment of tobacco products, it is often not applicable for risk assessment of new and innovative products where it is necessary to provide reliable predictive estimates of comparative risk prior to launch on a market. This is especially true when there is a long latency period between exposure and health outcome. The aim of this paper is to introduce the measurement certainty index (MCI) as a method based on objective rules to determine the level of certainty provided by a given dataset in a risk assessment context. The level of certainty depends on methodological aspects as well as on the underlying data, and the two core dimensions of the mci are reliability and validity. While reliability of measurement is primarily a technical issue, validity is more complex. The MCI is a quantitative measure of empirical evidence weighting different criteria of reliability and validity with respect to the level of certainty of a risk assessment. Application of such an index can help with both the retrospective assessment of existing data sets and the prospective design of assessment strategies to determine the probability of reduced risk prior to launch (and subsequent collection of epidemiological data).