DOI: https://doi.org/10.4414/smw.2020.20180
We welcome the comment by Feller and colleagues [1] about our study on ovarian cancer epidemiology and treatment in Switzerland [2]. Their comment contributes to a much-needed discussion on the strengths and limitations of the various data sources available for epidemiological research in Switzerland.
Our study estimated the incidence of ovarian cancer based on a hospital registry covering all inpatient care episodes. This mandatory registry was first introduced in 1998 by the Federal Office of Statistics with the primary goal of epidemiological surveillance [3] and allows patients to be tracked over time and across treatment sites. We identified incident ovarian cancer patients based on a C56 (malignant neoplasm of the ovary) main diagnosis and by applying a preceding ovarian cancer event-free period of 6 years. In comparison with data of the National Institute for Cancer Epidemiology and Registration (NICER), we found a similar slight decline in age-adjusted incidence of ovarian cancer between 2004 and 2012, but at a substantially higher level (+35%).
Feller and colleagues have probed our results with two approaches. Firstly, they replicated our incidence estimation based on hospital registry data for ovarian cancer and five other selected cancers sites for the period 2011–2014. Their estimates for ovarian cancer are in line with our results. However, the differences between NICER rates and hospital registry-based rates differ substantially in magnitude and sign of the difference. Secondly, they verified the accuracy of diagnostic coding of ovarian cancer cases in the hospital registry with related pathology reports and found numerous coding errors. After eliminating these coding errors their estimate of hospital registry-based incidence was much more in line with NICER rates.
These results are quite puzzling, both because of the heterogeneous differences between NICER and hospital registry-based cancer incidence estimates and because of the substantial coding errors detected. The diagnostic coding in the hospital registry is, for example, repeatedly evaluated by auditors, as diagnostic coding is also relevant for the SwissDRG hospital reimbursement scheme. We see an urgent need to better understand the causes of these discrepancies. The hospital registry has several strengths, which are complementary to the strengths of the NICER cancer registry, as it not only covers all inpatient-treated diseases but also allows monitoring of patients over subsequent inpatient stays. Combining these important data sources might substantially contribute to a better understanding of cancer epidemiology, treatment and health outcomes in Switzerland.
The original study of Wieser et al. [2] was financially supported by AstraZeneca AG, Switzerland. The funders of the study had no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication. The authors declare that they have no competing interests.
1 Feller A , Bopp M , Lorez M , Zellweger U , Adam M , Curjuri I , et al. Comment on: Ovarian cancer in Switzerland: incidence and treatment according to hospital registry data. Swiss Med Wkly. 2020;150:w20179. doi:https://doi.org/10.4414/smw.2020.20179
2 Wieser S , Schmidt M , Kind AB , Heinzelmann-Schwarz VA . Ovarian cancer in Switzerland: incidence and treatment according to hospital registry data. Swiss Med Wkly. 2018;148:w14647. doi:.https://doi.org/10.4414/smw.2018.14647
3Swiss Federal Office of Statistics. Medizinische Statistik der Krankenhäuser - Detailkonzept 1997 (Version: 12. Dezember 2005). https://www.bfs.admin.ch/bfs/de/home/statistiken/gesundheit/erhebungen/ms.assetdetail.230430.html.
The original study of Wieser et al. [2] was financially supported by AstraZeneca AG, Switzerland. The funders of the study had no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication. The authors declare that they have no competing interests.