Calcium intake may explain colorectal cancer risk reduction by dietary selenium – a case-control study from Poland | BMC Nutrition

Study population

The current study is based on a hospital-based case-control study that was carried out between 2000 and 2012. The study design, data collection methods and main objectives have been described elsewhere. [17]. In short, only incident cases of CRC were identified and recruited in cooperation with the [blinded for the review]. They must be histologically confirmed adenocarcinomas, either of the colon (ICD-X: C18) or of the rectum (ICD-X: C20). Only sporadic cancers met the eligibility criteria, which means that all cases of CRC suspected of being hereditary cancers were excluded (ICD-X: C18.9, D12.6, Q85.8, Z80.0 , Z80.3). Additionally, to limit the possibility of genetically determined CRC for the current study, cases under 40 years of age were also excluded. The controls were patients admitted to the emergency room and subsequently hospitalized at Krakow University Hospital and Narutowicz Municipal Hospital in Krakow, Poland. Primarily, the controls were individually matched to the cases on age (range +/- 5 years) and gender with the frequency ratio 1:2, meaning we tried to identify two controls for one case being of the same sex and of similar age. The course of the study is shown in Fig. 1. Exclusion criteria for cases and controls included: age over 75, cognitive limitations and verbal communication problems that prevented examiners from recalling, diagnosis of secondary cancer or CRC being a metastasis to the colon or rectum, and any other than CRC cancer (current or past), any type of surgery to the gastrointestinal tract in the past, a current or past diagnosis of gastrointestinal disease chronic bowel disease, and any other disease requiring dietary restrictions (such as diabetes, kidney failure, liver failure), as well as the presence of prolonged gastrointestinal symptoms (lasting more than one month) were checked.

Fig. 1

The study was conducted in accordance with the guidelines set forth in the Declaration of Helsinki and all procedures involving research study participants were approved by the Jagiellonian University Ethics Board (IRB no 1072.6120.347.2020) . Written informed consent was obtained from each study participant.

Data collection and evaluation

The data used for the study presented were collected by standardized questionnaire. Trained interviewers collected information on sociodemographic characteristics, lifetime smoking habits, lifetime physical activity, and diet during face-to-face recall. In detail, a validated semi-quantitative food frequency questionnaire prepared in cooperation with the German Cancer Research Center, Heidelberg, Germany, during the preparation phase of the European Prospective Cancer and Nutrition Study [18, 19] was set up to assess eating habits. A total of 148 dietary items (food and beverages) were used. There were questions about the consumption of cereals, dairy products, bread, type and cuts of meat and fish (including preparation technique), fresh fruit (in summer and winter separately), salads, fresh and cooked vegetables, rice and pasta, soups, sweets, bakery products and others. Study participants estimated commonly consumed portion sizes, for each food and drink, using standardized photographs, and then a frequency of consumption was reported. The dietary data covered the period of one calendar year. Case participants were asked about the year that passed 5 years before the onset of gastrointestinal symptoms (if symptoms were present) or before the start of a diagnostic process. Control patients reported their usual diet which took place 5 years before the interview. Then, to obtain data on macro and micronutrient intake, Polish food composition tables were used. Since the selenium content was not available in the main tables, for the current study, information on the intake of each food element was also used to calculate the dietary selenium content. Selenium concentrations, including cooking losses, for each food item assessed were presented in Supplementary Materials (see Supplementary Materials).


Several covariates likely to contribute to the odds of colorectal cancer were collected. These included adults’ lifetime leisure-time physical activity, which was assessed by recall as the average weekly time spent in different types of activity (including walking or hiking, cycling, gardening, sports and household activities) during the summer and winter seasons separately. Additionally, respondents were asked about time spent on recreational activities requiring at least moderate effort. Then, to obtain metabolic equivalents, the time claimed for each activity was multiplied by its typical energy expenditure requirement as published in the 2011 Compendium of Physical Activities. [20]. Since participants were not asked about barriers in their activities, it was assumed that 70% of reported time was spent efficiently and these were presented and used as a covariate in the analyses. Other covariates include age, sex, respondent’s body mass index calculated from weight and height measured on admission to a clinic or hospital, exposure to cigarette smoke ( categorized as non-smoker, former smoker, current smoker) was taken into account. In addition, the full set of dietary covariates were used, and these included total dietary fiber, dietary vitamin c and vitamin e intake, fish consumption (categorized as yes/no; “yes” for participants who reported consuming any fish, having a serving size of at least 20 g canned fish or 45 g cooked fish, per month), and taking mineral supplements (yes/no category).

statistical analyzes

There were two groups included in the study analyses, colorectal cancer cases and controls, presented and compared to assess the role of dietary selenium. Groups were characterized by providing means and standard deviations, and in addition, as the majority of variables had skewed distributions (tested by the Shapiro-Wilk normality test), the median and interquartile range were provided in the descriptive part. Study groups were compared using the chi-square test for categorized variables (all expected cell values ​​in the analyzes met the assumption of being greater than 5) and the U-Mann-Whitney test (because the distribution of variables in the compared groups was skewed). To assess the role of dietary selenium intake on the odds of CRC, logistic regression was used. We decided to run 3 consecutive models: a univariate, then a multivariate adjusted for major confounders such as age, gender, body mass index (BMI), leisure-time physical activity during average adult lifespan, alcohol consumption and smoking. The third model additionally included as covariates major dietary components such as total dietary fiber, dietary vitamin c and vitamin e content, mineral supplement intake, dietary calcium intake, and fish consumption. There were models to assess odds associated with selenium considered as a continuous variable (results shown for a 10 μg/day increase in dietary selenium) and also across selenium intake categories (

Comments are closed.