Description Of Work

The project aimed at bringing together existing and planned epidemiological research in a high-quality prospective cohort study of residents near the Semipalatinsk nuclear test site (SNTS). To achieve this, we evaluated issues related to both epidemiologic outcomes and to dosimetry, which underpins all radiation-effect studies.

The two cohorts of interest comprise inhabitants of communities surrounding the former SNTS. The historical cohort includes residents of 10 highly exposed settlements and six unexposed, control settlements. In total, there are 9 850 exposed and 9 604 unexposed inhabitants born before 1 June 1961. Control inhabitants were frequency matched to exposed inhabitants based on age and sex. The new cohort includes inhabitants from 14 exposed villages and 6 unexposed control villages. In total, the new cohort includes 18 204 inhabitants, mostly exposed to low doses of radiation (<0.6 Sv), and some to high doses (2 Sv). Follow-up for the historical cohort ended in 1999, while the new cohort covers the period 2001-2009.

Feasibility of unifying two cohorts:

  • We tested the possibilities of record linkage between the two cohorts, i.e. the available information for each individual was checked. The new and historical cohorts have different structures, however a complete linkage of both databases is possible, but it would be very labour intensive. It is estimated that one person-year would be needed for a complete linkage.

  • During a site visit to Semey (formerly Semipalatinsk) in September 2013, we learnt that one of the SEMI-NUC partners, the Scientific Research Centre of Radiation Medicine (NIIRME) possess a registry of exposed population around the SNTS that contains information for about 180 000 individuals and is larger than both the historical and new cohorts combined. All project participants agreed that the NIIRME registry is the best to serve as basis for the future prospective cohort. The intellectual property rights (IPR) and accessibility to the data in the NIIRME registry are important for future studies, therefore we negotiated with the Minister of Health of Kazakhstan to obtain formal agreements for the use of Registry data.

Identifying outcomes of interest:

  • The project identified the health outcomes that could by studied in the future using the NIIRME registry data. The recommended outcomes include cancer incidence in relation to early life exposure; cardiovascular disease incidence following exposure in childhood and a study of the natural history of thyroid nodules detected at screening examination.

Testing follow-up mechanisms:

  • We identified and reviewed possible follow-up mechanisms for mortality and outcomes of interest, ascertainment of cause of death and vital status among historical, new cohort and NIIRME registry members. The selection criteria, methods of testing, and results, are summarized in the detailed report submitted to EC. The results of comparison suggest that the NIIRME registry of exposed population is a comprehensive source of information that could be used for a future prospective study. Description and comparison of follow-up mechanisms used in the studies of historical and new cohorts are helpful for considering feasibility of a future full-scale epidemiology study.


Identification of key pathways and mechanisms for estimation of cumulative doses:

  • We identified key pathways and mechanisms for formation of cumulative doses in the villages around the Semipalatinsk NTS; identified gaps and discrepancies in existing dose reconstruction methods based on reviewing available reports and publications from the previous studies and surveys conducted in the Semipalatinsk area; outlined the future work for filling the existing gaps in the assessment of external and internal doses to the population residing in the SNTS area and selected the best approach for filling in the existing gaps. A protocol was developed for testing the feasibility of the proposed approach in a selected pilot village.

Comparison of different methods of dose assessment in the two cohorts:

  • We compared the different methods used for dose assessment in the previous studies of exposed populations. We tested and validated a selected approach for calculation of average settlement doses in a selected pilot village and developed an approach for integrated assessment of individual cumulative doses to the population from all potential sources of exposure. Individual doses using the two approaches developed in the two cohorts were calculated for a sample of individuals, and where discrepancies were identified, the reasons for these discrepancies was analyzed. The calculations were validated using biodosimetry (i.e. chromosomal aberration techniques and electron paramagnetic resonance [EPR] or electron spin resonance [ESR] dosimetry) on available samples of biological material.

  • We calculated individual doses from external exposure for those subjects for whom results of tooth enamel EPR measurements and information on residence history and dietary habits from personal interview are available. Individual dose estimates from external exposure based on the proposed methodology were compared with corresponding individual dose estimates based on EPR measurements results, and results of this inter-comparison were analyzed.


A detailed report, summarizing the results of the feasibility assessment for an integrated cohort study in the future, was prepared and evaluated by the External Advisory Board (EAB) before submission to the European Commission.


  • Final meeting, 14 - 15 March 2016, NRPA, Oslo, Norway
  • Dissemination workshop, 9 November 2015, Helmholtz Centre, Munich, Germany
  • External Advisory Board meeting, 11-12 May 2015, IARC, Lyon, France
  • Second site visit, 30 March - 2 April 2015, Semey, Kazakhstan
  • Mid-term meeting, 9-10 April 2014, Munich, Germany
  • First site visit, 17-20 September 2013, Semey, Kazakhstan
  • Feasibility study meeting, 7-8 May 2013, IARC, Lyon, France
  • Kick off meeting, 6 May 2013, IARC, Lyon, France