Discriminatory analysis in determination of high risk of gastric cancer
1540 Background: Determination of the high risk of gastric cancer and target examination of patients from high-risk groups makes early detection of gastric cancer possible and contributes to the decrease of mortality rates as the outcome of gastric cancer treatment is primarily dependant on the stage of disease. Methods: 700 patients with gastric cancer and 1069 persons without any oncopathology referred to Altay oncological center for consultation in 2002–2004 were interviewed and examined to get the data on different factors known as risk factors for gastric cancer (endogenic, environmental, lifestyle, nutritive factors, results of blood-tests and non-oncological stomach diseases, etc.). Results: By means of discriminatory analysis from 131 investigated factors there were 26 factors distinguished and corresponding coefficients of canonical function of discrimination calculated, on the basis of which the prognosis of gastric cancer development may be ascertained (p<0.001): age, body mass, education, duration of stress, insomnia, extrasleep during daylight hours, family history of gastric cancer, family history of cancer, presence of clinical symptoms, lifetime history of smoking, usage of strong spirits, amount of spirits used per month, regularity of nutrition (eating patterns), time intervals between food intakes, variety of food allowance, usage of food and drinks of high temperature, animal fats, spicy food, canned and preserved food, bakery, strong black tea and coffee, fresh vegetables and fruit, sour milk products, green tea, level of haemoglobin, ESR. Multiplying the interval values of those significant factors and corresponding coefficients of canonical function of discrimination and subsequent summarizing of the results and the constant gives an <<integral rate>>, the positive value of which is the evidence of high risk of gastric cancer either to be developed or already exists. Negative integral rate is the evidence of low risk. Cross check up showed the 95.1% test-sensitivity and 95.0% test-specificity. Conclusions: Thus, the value of the integral rate can be used as a criterion for selection of concrete patients with high risk of gastric cancer to be deeply examined and to be under monitoring for detection of early gastric cancer. No significant financial relationships to disclose.