¹Krebsregister Mecklenburg-Vorpommern, Institut für Community Medicine, Universitätsmedizin Greifswald, Greifswald, Germany
²Institut für Krebsepidemiologie e. V., Universität zu Lübeck, Lübeck, Germany

Text

Introduction: Standardized incidence ratios (SIRs) and standardized mortality ratios (SMRs) are commonly used to analyze cancer risk and mortality in occupational cohort studies. Such studies can be conducted by matching cohort data with cancer registry data (SIR) and population registry data (SMR). Exposure data are needed to characterize exposure effects on cancer risk and/or mortality. However, analyzing exposure data is not straightforward and often relies on surrogate measures [1], [2], [3]. Losses to follow-up (LTFU) may limit the interpretation of results. If a high proportion of LTFUs occur due to exposure and subsequent cancer and/or death, the results may be severely underestimated when censoring LTFUs at the last available date. This work presents a detailed, dynamic analysis of cumulative occupational exposure and quantifies the effect of LTFUs on SIRs and SMRs from a study of workers at the Ihlenberg toxic waste landfill in northern Germany.

Methods: This retrospective cohort study compared cancer risk and mortality of 590 landfill workers with those of the general population [4]. Cumulative employment duration at the landfill was used as a proxy for exposure to toxic waste. Because SIRs and SMRs were calculated as an aggregated measure for the time period 2009-2021, and cohort entry was possible 1983-2018, simple categorization of years of total employment (e.g., into quartiles) was inadequate. Therefore, a dynamic approach was utilized, where workers moved between different employment duration categories by accumulating employment duration. SIRs and SMRs were estimated yearly, separately for each employment duration category (0 - <5, 5 - <15, 15 - <25, 25-36 years) and aggregated for the period 2009-2021. To estimate a range of possible SIRs and SMRs, we performed sensitivity analyses by assigning hypothetical events to the LTFUs (n=48): 1) all LTFUs were cancer-free and alive at the end of follow-up, 2) 30% of LTFUs were assumed to have had cancer and died, 3) all LTFUs were assumed to have had cancer and died. The 30% figure reflects a probability well above the actual probability of cancer among the workers (approximately 5%).

Results: With an SIR of 0.69 (95% confidence interval (CI): 0.47-0.98) and SMR of 0.51 (CI: 0.35-0.73), cancer incidence and all-cause mortality were not increased. The sensitivity analysis of LTFU yielded SIRs ranging from 0.62 (CI: 0.42-0.88) to 1.13 (CI: 0.85-1.48) and SMRs from 0.45 (CI: 0.31-0.64) to 0.80 (CI: 0.60-1.06). No pattern of increasing cancer risk with increasing employment duration was found. For SMR, a healthy worker effect was observed.

Discussion: Time-dependent analysis of occupational exposure is required to adequately evaluate effects of exposure on cancer risk and/or mortality. Increasing employment duration was not associated with increasing cancer risk or mortality in this study. Distribution of hypothetical events to LTFU in sensitivity analyses allowed estimation of a range of possible SIRs and SMRs, however, without changing the overall conclusion of no statistically significant increase in cancer incidence or mortality. Conclusion

Dynamic analysis of exposure and estimation of SIR and SMR ranges in the presence of LTFU are helpful tools for analyzing occupational cohort studies.

The authors declare that they have no competing interests.

The authors declare that a positive ethics committee vote has been obtained.

References

[1] Hammer GP, Zeeb H, Tveten U, Blettner M. Comparing different methods of estimating cosmic radiation exposure of airline personnel. Radiat Environ Biophys. 2000;39(4):227-31. DOI: 10.1007/s004110000071
[2] Blettner M, Zeeb H, Langner I, Hammer GP, Schafft T. Mortality from cancer and other causes among airline cabin attendants in Germany, 1960-1997. Am J Epidemiol. 2002;156(6):556-65. DOI: 10.1093/aje/kwf083
[3] Zeeb H, Blettner M, Langner I, Hammer GP, Ballard TJ, Santaquilani M, et al. Mortality from cancer and other causes among airline cabin attendants in Europe: a collaborative cohort study in eight countries. Am J Epidemiol. 2003;158(1):35-46. DOI: 10.1093/aje/kwg107
[4] Manz K, Weitmann K, Theen S, Robers G, Pritzkuleit R, Katalinic A, et al. Cancer incidence and mortality in the occupational cohort of a German toxic waste landfill: a retrospective cohort study. BMC Public Health. 2024;24(1):3560. DOI: 10.1186/s12889-024-21134-z