Variations in sociodemographic features, exposure characteristics and genetic polymorphisms between community and hospital controls: a validation study of control selection in case-control designs

[Truncated abstract] Background The choice of an appropriate control group in case-control studies is widely acknowledged as a crucial methodological issue, since inappropriate selection can pose a serious challenge to the internal validity of case-control study results. In particular, little information exists on differences between community controls and hospital outpatient controls when used in case-control studies of cancers in the Chinese hospital setting. With respect to characteristics associated with willingness to participate in genetic cancer research in China, there has been no previous examination at all. Methods Three parallel case-control studies on adult leukemia, breast and colorectal cancers were conducted in Shenyang, Northeast China, between 2009 and 2010, using community and hospital outpatient control series, each separately matched by age, gender and residency in a 1:1 ratio to 560 incident cases. A 7-point 'willingness to participate' scale was developed for use by the interviewer to record the levels of ease or difficulty experienced in recruiting each hospital participant. A reliable, structured questionnaire was used to collect demographic and lifestyle characteristics, diet, medication, and hormone related factors by face-to-face interview. A quantitative Food Frequency Questionnaire (FFQ) and a short food habit questionnaire (SFHQ) were used to measure foods intake. The Goldberg formula (ratio of energy intake to basal metabolic rate, EI/BMR) was used to assess the validity of the FFQ. Blood samples were collected from participants after obtaining their additional consent. Odds ratios (ORs) and 95% confidence intervals (CIs) were obtained using conditional logistic regression analyses adjusted for potential confounders. The willingness to participate was compared between different subgroups of participants.