Methods of health risk and impact assessment at industrially contaminated sites: a systematic review.

Background: this paper is based upon work from COST Action ICSHNet. Industrially contaminated sites (ICSs) are a serious problem worldwide and there is growing concern about their impacts on the environment and public health. Health risk assessment methods are used to characterize and quantify the health impacts on nearby populations and to guide public health interventions. However, heterogeneous methods and inconsistent reporting practices compromise comparability risk and impact estimates.

Objectives: to review the literature on assessment of the adverse health effects of ICSs. Specifically, we: - collect published, peer-reviewed literature addressing health assessment of ICSs; - identified and evaluated the methods and tools for the assessment of health impacts related to ICSs; - analysed the methods and tools used in different conditions; - discussed the strengths and weaknesses of the identified approaches; - presented an up-to-date understanding of the available health risk and impact assessment in ICSs. In addition, the terminologies were described and harmonization was proposed.

Methods: we systematically searched PubMed and Web of Science to identify peer-reviewed reviews and original studies from January 1989 to December 2017. We used a qualitative approach for analysing the different elements (type of ICSs, Country of research, active years of working, distance from sources, pollutants, affected population, methods and tools, health outcomes, main founding, method stage, dose-response assessment, risk characterization) of included studies. We divided risk assessment methods used in the papers into four stages: semi-quantitative, quantitative, health impact, and health burden stage.

Results: a total of 92 relevant original papers at ICSs were found and analysed. In current practice, the health risks have been characterized mainly as hazard quotients or hazard indexes (23 studies), and as cancer risk probabilities (60 studies). Only 8 studies estimated the number of cases and one study evaluated years of life lost.

Conclusion: hazard quotients and cancer probabilities are suitable for semi-quantitative and quantitative personal risk estimation, respectively. More comparable risk characterization on public health level requires specificity on the type of outcome and corresponding number of cases. Such data is needed for prioritization of action at low to medium risk sites. We found limited amount of studies that have quantified the health impact at industrially contaminated sites. Most of the studies have used semi-quantitative risk characterization approaches and the adopted methods are mostly of toxicological origin, while epidemiological analysis is almost lacking. There is a need to improve quantitative risk assessment and include health impact and environmental burden of disease assessments at ICSs.