Navigation and service

ProZES: Calculation of the probability of causation of cancer by radiation exposure

The software ProZES calculates the probability that a previous occupational radiation exposure of a specific person in a given exposure situation has resulted in cancer. The Institute of Radiation Protection at the Helmholtz Zentrum München has developed the software by order of the Federal Environment Ministry and the Federal Office for Radiation Protection (BfS). The software can be downloaded free of charge.

What is ProZES doing?

Based on individual information on disease, radiation type and doses, ProZES calculates the probability that the occurrence of cancer in a specific person has been caused by previous occupational radiation exposure.

Some experts suggest that in compensation claims not only the probability that radiation exposure has caused cancer should be considered, but also the probability that radiation exposure has accelerated cancer development. At present ProZES does not consider this aspect explicitly. Practically, in compensation claims in Germany only the probability of causation is relevant at present.

ProZES uses state-of-the-art models of radiation risk as well as demographic and incidence data specifically related to Germany.

One particular strength of ProZES is that it not only calculates one single estimated value for the assigned share but also its range of uncertainty.

An exemplified calculation using ProZES ProZES: ExampleProZES User interface with an exemplified calculation

Who is ProZES aimed at?

ProZES is designed for professionals dealing with probabilities of occurrence of cancer after occupational radiation exposure. Application of ProZES requires expert knowledge on radiation risk and calculation of organ doses/exposure data. It is important to familiarize with the program with aid of the help file and the tutorial before application. (Help file and tutorial are included in the software.) For experts and staff of workers' compensation boards ProZES can serve as a tool.

Information on latest versions

If you are interested in news about ProZES, we will be happy to inform you about current developments. Please send us a short e-mail stating your name and the desired form of address to ProZES@bfs.de.
You can of course revoke your consent to the use of your contact details at any time. To do so, send an e-mail to ProZES@bfs.de and your data will be deleted immediately.

Further information on data protection, in particular on your rights under the GDPR, can be found at the BfS Privacy Policy.

Legal Notes

Terms of License for the German Federal Office for Radiation Protection (BfS)'s ProZES software are included in the software and have to be accepted for setup.

More information

Detailed information on ProZES can be found in the scientific reports for the research projects during which ProZES was developed (part 1, part 2 and part 3). They are available in DORIS - the Digital Online Repository and Information System of the BfS.

Comments and feedback regarding the software are appreciated and may be sent via email to ProZES@bfs.de.

Questions and answers on how to use ProZES

What is the difference between ProZES and IREP?Show / Hide

Like ProZES, the US programme IREP also calculates the probabilities of causation for selected diseases. Both programmes are relatively similar in terms of methodology. One crucial difference is that unlike IREP, ProZES is designed for the population of Germany. German incidence data are thus used in order to determine the underlying risks of disease. An overview of the methodological differences can be found in Ulanowski et al. (2020). Analyses for selected endpoints are provided in the departmental research report (in German) Quantitative Abschätzung des Strahlenrisikos unter Beachtung individueller Expositionsszenarien.

Underground work in uranium mining usually led to exposure to not only radon but also gamma radiation and/or long-lived radionuclides. How can such exposures be taken into account in ProZES?Show / Hide

In order input a radon exposure in mining, you have to select the tab “Radon Mines” and then enter the exposures in working level month (WLM) for the respective years of exposure. In ProZES, information on exposure (e.g. to gamma radiation or long-lived radionuclides) can be provided under the “General” tab. For this purpose, the equivalent dose from the gamma radiation or the long-lived radionuclides must have been determined in advance. When calculating the probability of causation, ProZES then includes the data on exposures for both “Radon Mines” and “General”. You can find more information in the help and tutorial.

A person has been diagnosed with squamous cell carcinoma. Is using the diagnosis “Non-melanoma skin cancer” appropriate in order to be able to determine the probability of causation for this disease?Show / Hide

Not particularly. The diagnosis “Non-melanoma skin cancer” includes different subtypes of non-melanoma skin cancer such as basal cell carcinoma and squamous cell carcinoma. An overestimation of the probability of causation is to be expected because the association between basal cell carcinoma and radiation exposure is assumed to be stronger than between squamous cell carcinoma and radiation exposure. You can find more information in the help and tutorial.

Which smoking behaviour is assumed when selecting “unknown” for smoking behaviour?Show / Hide

Two cases must be distinguished here:

  • If it is specified that it is unknown whether the person is a smoker, it is decided randomly in each iteration of the ProZES algorithm whether they are a smoker. A male is assumed to have a 35 % probability of being a non-smoker, whereas a female is assumed to have a 53 % probability. If the person is assumed to be a smoker in a particular iteration, a typical German smoking behaviour is determined for them based on sex, birth cohort, and age (for details, see departmental research report, p. 49/50). The values are based on reports by the RKI. The users cannot change this assumed smoking behaviour.
  • If it is stated that the person is a smoker but the details are unknown, a typical German smoking behaviour is assumed for them based on sex, birth cohort, and age in analogy to the procedure described above. If the assumed smoking behaviour seems inappropriate for the respective person, it is possible to switch to “smoking habits known”, and plausible statements can be made.

A person has developed lung cancer after exposure to radiation (not radon). How can it be that the probability of causation Z (the probability that the lung cancer was caused by radiation) is higher for smokers than for non-smokers?Show / Hide

On one hand, radiation exposure and smoking are competing risk factors with regard to lung cancer: The higher the probability that the lung cancer was caused by smoking, the lower the probability that it occurred because of radiation exposure. This would suggest that non-smokers would have a higher Z for the same radiation exposure than smokers because smoking is ruled out as a possible cause. On the other hand, studies suggest an interaction effect between radiation exposure and smoking in relation to the development of lung cancer. This means that radon and smoking mutually reinforce each other with respect to their effect on lung cancer. This would suggest that smokers have a higher Z than non-smokers. Which of these two opposing effects predominates depends on the magnitude of the interaction effect. In the case of a strong interaction effect between radiation exposure and smoking, it is plausible that a higher Z would result for smokers than for non-smokers. You can find more information in the help and tutorial.

A person has developed lung cancer after exposure to radiation (not radon). What should you choose if you do not want to take into account the interaction between smoking and radiation exposure, which can lead to higher probabilities of causation for smokers than for non-smokers?Show / Hide

If this type of interaction is not to be taken into account in the calculation of the probability of causation, the option “non-smoker” should be selected regardless of smoking behaviour. You can find more information in the help and tutorial.

State of 2024.01.15

How do you rate this article?

Site information and functions

© Bundesamt für Strahlenschutz