RESUMO
Objective To establish models to predict individual risk of essential hypertension and to evaluate and explore new forecasting methods. Methods To select data of 3054 community residents from a epidemiological survey and divided them into 4 : 1 (2438 cases and 616 cases) ratio in accordance with the balance of age and sex to filter variables, and to establish, test and evaluate the prediction models. Using artificial neural network (ANN) and logistic regression analysis to establish models while applying ROC to evaluate the prediction models. Results Forecast results of the models applying to the test set proved that ANN had lower specificity but better veracity and sensitivity than logistic regression.In particular, the Youden's index of the ANN2 came up to 0. 8399 which was distinctly higher than the other two models.When the area was under the ROC curve of logistic regression, the ANN1 and ANN2 models equaled to 0. 732±0. 026,0. 900±0. 014 and 0. 918±0. 013 respectively, which proved that the ANN model was better in the prediction about individual health risk of essential hypertension. Conclusion Our results showed that ANN method seemed better than logistic regression in terms of predicting the individual risk from hypertension thus supplied a new method to solve the forecast of individual risk.
RESUMO
OBJECTIVES: The objectives of this study were to develop a health risk appraisal (HRA) model of breast cancer and to evaluate the model's ability to accurately predict ndividual breast cancer risk. METHOD: The sample data were from a prospective cohort study in Korea with a follow-up period of 8 years in Korea (1995-2002). The cohort was composed of 108,708 Korean women, aged 30 to 64 years, who received health insurance from the Korea Medical Insurance Corporation and who underwent a medical evaluation in 1994. Between 1995 and 2002 (1,138,349 person-year), there were 772 incident cases of breast cancer (67.8/100,000 person-years). First, we developed the HRA based on the Cox proportional hazard models using a prospective cohort data with half-split data (50% random sample). Second, using Cox proportional hazard models, we compared the probability of breast cancer predicted by the model to the actual number of cases observed in the other 50% of the random sample of the study population. RESULTS: The HRA Cox proportional hazard model of breast cancer developed in the study included age, age2, age at menarche, and lactation. Multivariate Cox proportional hazards models were used to test the HRA model validation. Compared with women in the lowest 10 percentile of subjects, women in the highest 10 percentile had an increased risk of breast cancer (relative risk, 3.701; 95% confidence interval, 2.554 to 5.364). CONCLUSION: The model's performance was satisfactory for estimating breast cancer risk for individual women aged 30-64 years.