At 1A, 1B, 1C, 2A, and 3A type of cancer (tumor) was reported by participants in an open text field. From 2A onwards a limited number of cancer types (i.e., colon cancer, lung cancer, skin cancer, brain tumor, bone cancer, breast cancer, and prostate cancer) are also directly assessed via yes/no questions. At 3B only these direct cancer type questions are used, there is no open text field. Researchers from the University Medical Center Groningen coded these self-reported cancer variables into 35 cancer type categories. The secondary cancer type variables will be added to the Lifelines catalogue very soon.

Tumors were coded using the directly assessed cancer variables or terms mentioned in the open text field and categorized according to type, location, and stage. If participants mentioned a term related to metastasis in the open text field an additional cancer metastasis variable was coded as 1.

For 1A the coding of all open text fields was checked by an MD/PhD student of the Oncology Epidemiology department. When in doubt a pathologist was asked to make a final decision. If the student or pathologist did not agree with the coding of an open text field, the text field was manually recoded based on their input. From 1B onwards the algorithm and manual correction have been applied by the Lifelines data management team. Only the open text fields that were also recoded at 1A have manually been corrected. Additionally, open text fields that were not coded by the algorithm have manually been coded.

The data processing described above was validated by Cortés-Ibáñez et al. 2022 on the 1A data.

Results
6611 (4.50%) participants had at least one self-reported diagnosis of cancer, where 9960 (6.97%) participants had at least one cancer diagnosis in PALGA. The sensitivity of self-reported cancer was 64.68% [95%CI:63.71–65.66], and 70.18% [95%CI:68.83–71.56] after excluding skin and cervical cancers. Skin and cervical cancers represented 61.24% of non-self-reported cancers. The overall positive predictive value was 97.45% [95%CI:97.45–97.81], and 97.33% [95%CI:96.72–97.82] after the exclusion of skin and cervical cancers. Participants who did not self-report their cancer were more likely to be male, had longer time since diagnosis and lower educational level. Besides the different sum scores, the number of valid items is provided. Enabling everyone to decide how many valid answers is enough to include a sum score.

Conclusion
Overall, the reports of cancer in Lifelines have a high positive predictive value and moderate sensitivity. One third of the cancers were not reported, mainly skin and cervical cancers. Male participants, those with a lower educational level and those with longer time since diagnosis were less likely to self-report a diagnosed cancer.

When the secondary cancer variables have been used in your research, you will have to include a reference to the validation paper.

• Cortés-Ibáñez, F.O., et al. The validity of self-reported cancer in a population-based cohort compared to that in formally registered sources. Cancer Epidemiology (2022): 1022-1068.