Project details

or


A validation study on self-reported oncology data in LifeLines

Keywords:
LifeLines Prevalence of cancer Validation study

Researchers:
Prof. dr. G.H. de Bock
Dr. B. van der Vegt

Nature of the research:
A comparison of self-reported prevalence of cancer in the LifeLines baseline data(1), with the cancer prevalence data according to the National Cancer Registry (NKR)(2) with regard to type of cancer, gender and age.

Fields of study:
epidemiology oncology pathology

Background / introduction
It is well understood that there is a worldwide rise in the incidence of cancer, in all areas where life expectancy is rising, as attested by countless articles describing this phenomenon. Large population-based cohort studies and biobanks such as LifeLines can become a useful tool in understanding more about the incidence and prevalence of cancer. LifeLines currently uses self-reported data on general health and medical history, as well as a range of objectively measured biometrics and certain hematological and biochemical measurements.(1) The use of self-reported data could lead to a reporting bias as certain information might be missing or inaccurately reported. It is therefore important to validate the self-reported data by comparing them to objectively measured data.
Research question / problem definition
Does self-reported prevalence of cancer in the baseline data of LifeLines match with objectively observed cancer prevalence within the NKR in the same period, overall, as well as stratified by age, sex and tumor type?
Workplan
Initially, the algorithm that was devised to categorize the type of tumor reported in LifeLines will be manually checked and supplemented where necessary. After the data on type of tumor have been fully categorized, 95% confidence intervals can be established for the prevalence of cancer, using SPSS descriptive statistical analysis, both overall, as well as stratified by tumor type, age and sex. To validate the data, these confidence intervals can then be compared to the ranges from the data published by IKNL on cijfersoverkanker.nl(2). This will be done by performing either a logistic regression or Chi-Square test in SPSS.

Any observed, significant differences could indicate that the self-reported prevalence is unrepresentative of the actual prevalence, or that the LifeLines cohort is unrepresentative of the total Dutch population. To further investigate beyond that, unfortunately falls outside the scope of this TTT study.
References
(1) Scholtens S, Smidt N, Swertz MA, Bakker SJ, Dotinga A, Vonk JM, et al. Cohort Profile: LifeLines, a three-generation cohort study and biobank. Int J Epidemiol 2015 /08/01;44(4):1172-1180.

(2) IKNL. Over de registratie. Available at: http://www.cijfersoverkanker.nl/over-de-registratie-12.html. Accessed January 9th, 2017.
back to toptop