[Validation of a 4-Item-Questionnaire on Familial and Hereditary Colorectal Cancer Risk in General Practice]

A 4-item-questionnaire has been developed by the "Netzwerk gegen Darmkrebs e. V." (Network against colorectal cancer) to identify familial or hereditary susceptibility to colorectal cancer. The aim of this inquiry was to validate the questionnaire in 40 to 54 year old persons in a general practice setting.

Four general practices from the "Forschungsnetzwerk Allgemeinmedizin Frankfurt" database generated a sample of 100 index patients whose first-degree relatives were known to the general practitioner. The general practitioners prepared pedigrees in accordance with the four items of the questionnaire. Subsequently, the index patients were surveyed by a healthcare assistant on the basis of the questionnaire. The level of agreement between the patients' and general practitioners' responses was measured using Cohen's kappa coefficient.

An almost perfect level of agreement was found for question 1 on whether a first-degree relative had been diagnosed with colorectal cancer (kappa coefficient 0,82; CI: 0,58-1,10). Question 2 on the existence of a first-degree relative under the age of 50 diagnosed with colorectal cancer was not answered positively by general practitioners for any patient, whereas one patient answered the question positively. Interpreting the level of agreement using the kappa coefficient was meaningless in this case. General practitioners did not have information to answer question 3 on the occurrence of a colorectal polyp in a first-degree relative under the age of 50. Likewise, it was impossible to find the necessary information on second-degree relatives in practice files (question 4). Overall, 18,8% of all patients (15/80) answered "yes" to one or more questions on the existence of a hereditary colorectal cancer risk.

Question 1 allows the identification of a familial colorectal cancer risk among 40 to 54 year old persons in a general practice setting. Questions 2 to 4 of the questionnaire could not be validated using the chosen methodology. However, the valid answer of 40 to 54 year old persons to question 1 allows familial risk to be identified. More detailed family history-taking regarding hereditary risk should be conducted in this group.