Assessment of a questionnaire for breast cancer case-control studies

Abstract

The aim of the present study was to assess criterion validity and external reliability of a questionnaire on risk factors for breast cancer. Materials and Methods. Women with breast cancer diagnosis (the cases) (N=40) and matched individuals without cancer (the controls) (N=40) were asked to fill in a questionnaire twice: on a day of admission to hospital (Q1) and on a day before discharge (Q2), with a time interval of 4-6 days. The questionnaire included questions (N=150) on demographic and socioeconomic factors, diseases in the past, family history of cancer, woman's health, smoking, alcohol use, diet, physical activity, and work environment. Criterion validity of the questionnaire Q2 relative to reference questionnaire Q1 was assessed with the Spearman correlation coefficient (SCC); external reliability of the questionnaire was measured in terms of the intraclass correlation coefficient (ICC). Statistical analysis was performed with SPSS 16. Results. The responses to most of the questions on socioeconomic factors, family history on cancer, female health, lifestyle risk factors (smoking, alcohol use, physical activity) correlated substantially in both the cases and the controls with SCC and ICC>0.7 (p<0.01). Statistically non significant relationships defined only between the responses on amount of beer the cases drank at the ages up to 25 years and 26-35 years as well as time of use of estrogen and estrogens-progestin during menopause by the cases. Moderate and substantial SCC and ICC were determined for different food items. Only the response of the cases on veal consumption did not correlate significantly. Conclusions. The questionnaire on breast cancer risk factors is valid and reliable for most of the questions included.