• Asian Pacific Journal of Cancer Prevention
• /
• 제15권16호
• /
• pp.6811-6817
• /
• 2014

Background: Breast cancer risk prediction models are widely used in clinical practice. They should be useful in identifying high risk women for screening in limited-resource countries. However, previous models showed poor performance in derived and validated settings. Therefore, we aimed to develop and validate a breast cancer risk prediction model for Thai women. Materials and Methods: This cross-sectional study consisted of derived and validation phases. Data collected at Ramathibodi and other two hospitals were used for deriving and externally validating models, respectively. Multiple logistic regression was applied to construct the model. Calibration and discrimination performances were assessed using the observed/expected ratio and concordance statistic (C-statistic), respectively. A bootstrap with 200 repetitions was applied for internal validation. Results: Age, menopausal status, body mass index, and use of oral contraceptives were significantly associated with breast cancer and were included in the model. Observed/expected ratio and C-statistic were 1.00 (95% CI: 0.82, 1.21) and 0.651 (95% CI: 0.595, 0.707), respectively. Internal validation showed good performance with a bias of 0.010 (95% CI: 0.002, 0.018) and C-statistic of 0.646(95% CI: 0.642, 0.650). The observed/expected ratio and C-statistic from external validation were 0.97 (95% CI: 0.68, 1.35) and 0.609 (95% CI: 0.511, 0.706), respectively. Risk scores were created and was stratified as low (0-0.86), low-intermediate (0.87-1.14), intermediate-high (1.15-1.52), and high-risk (1.53-3.40) groups. Conclusions: A Thai breast cancer risk prediction model was created with good calibration and fair discrimination performance. Risk stratification should aid to prioritize high risk women to receive an organized breast cancer screening program in Thailand and other limited-resource countries.

• 한국의료질향상학회지
• /
• 제10권2호
• /
• pp.216-231
• /
• 2003

Objectives: To develop a model that predicts a death probability of acute myocardial infarction(AMI) patient, and to evaluate a performance of hospital services using the developed model. Methods: Medical records of 861 AMI patients in 7 general hospitals during 1996 and 1997 were reviewed by two trained nurses. Variables studied were risk factors which were measured in terms of severity measures. A risk model was developed by using the logistic regression, and its performance was evaluated using cross-validation and bootstrap techniques. The statistical prediction capability of the model was assessed by using c-statistic, $R^2$ as well as Hosmer-Lemeshow statistic. The model performance was also evaluated using severity-adjusted mortalities of hospitals. Results: Variables included in the model building are age, sex, ejection fraction, systolic BP, congestive heart failure at admission, cardiac arrest, EKG ischemia, arrhythmia, left anterior descending artery occlusion, verbal response within 48 hours after admission, acute neurological change within 48 hours after admission, and 3 interaction terms. The c statistics and $R^2$ were 0.887 and 0.2676. The Hosmer-Lemeshow statistic was 6.3355 (p-value=0.6067). Among 7 hospitals evaluated by the model, two hospitals showed significantly higher mortality rates, while other two hospitals had significantly lower mortality rates, than the average mortality rate of all hospitals. The remaining hospitals did not show any significant difference. Conclusion: The comparison of the qualities of hospital service using risk-adjusted mortality rates indicated significant difference among them. We therefore conclude that risk-adjusted mortality rate of AMI patients can be used as an indicator for evaluating hospital performance in Korea.

• Journal of Preventive Medicine and Public Health
• /
• 제38권2호
• /
• pp.132-140
• /
• 2005

Objectives: To propose a risk-adjustment model from insurance claims data, and analyze the changes in cesarean section rates of healthcare organizations after adjusting for risk distribution. Methods: The study sample included delivery claims data from January to September, 2003. A risk-adjustment model was built using the 1st quarter data, and the 2nd and 3rd quarter data were used for a validation test. Patients' risk factors were adjusted using a logistic regression analysis. The c-statistic and Hosmer-Lemeshow test were used to evaluate the performance of the risk-adjustment model. Crude, predicted and risk-adjusted rates were calculated, and compared to analyze the effects of the adjustment. Results: Nine risk factors (malpresentation, eclampsia, malignancy, multiple pregnancies, problems in the placenta, previous Cesarean section, older mothers, bleeding and diabetes) were included in the final risk-adjustment model, and were found to have statistically significant effects on the mode of delivery. The c-statistic (0.78) and Hosmer-Lemeshow test ($x^2$=0.60, p=0.439) indicated a good model performance. After applying the 2nd and 3rd quarter data to the model, there were no differences in the c-statistic and Hosmer-Lemeshow $x^2$. Also, risk factor adjustment led to changes in the ranking of hospital Cesarean section rates, especially in tertiary and general hospitals. Conclusion: This study showed a model performance, using medical record abstracted data, was comparable to the results of previous studies. Insurance claims data can be used for identifying areas where risk factors should be adjusted. The changes in the ranking of hospital Cesarean section rates implied that crude rates can mislead people and therefore, the risk should be adjusted before the rates are released to the public. The proposed risk-adjustment model can be applied for the fair comparisons of the rates between hospitals.

• Journal of Preventive Medicine and Public Health
• /
• 제34권1호
• /
• pp.21-27
• /
• 2001

Objectives : To develop a model that will predict the mortality of patients undergoing Coronary Artery Bypass Graft (CABG) and evaluate the perfermance of hospitals. Methods : Data from 564 CABGs peformed in six general hospitals were collected through medical record abstraction by registered nurses. Variables studied involved risk factors determined by severity measures. Risk modeling was performed through logistic repression and validated with cross-validation. The statistical performance of the developed model was evaluated using c-statistic, $R^2$, and Hosmer-Lemeshow statistic. Hospital performance was assessed by severity-adjusted mortalities. Results : The developed model included age, sex, BUN, EKG rhythm, Congestive Heart Failure at admission. acute mental change within 24 hours, and previous angina pectoris history. The c-statistic and $R^2$ were 0.791 and 0.001, respectively. Hosmer-Lemeshow statistic was 10.3(p value=0.2415). One hospital had a significantly higher mortality rate than the average mortality rate, while others were net significantly different. Conclusion : Comparing the quality of service by severity adjusted mortality rates, there were significant differences in hospital performance. The severity adjusted mortality rate of CABG surgery may He an indicator for evaluating hospital performance in Korea.

• Communications for Statistical Applications and Methods
• /
• 제9권1호
• /
• pp.229-240
• /
• 2002

Whether to use linear or quadratic model in the analysis of regression data is one of the important problems in classical regression model and measurement error model (MEM). In MEM, four goodness of fit test statistics are available In solving that problem. Two are from the derivation of estimators of quadratic MEM, and one is from that of the general $k^{th}$-order polynomial MEM. The fourth one is derived as a variation of goodness of fit test statistic used in linear MEM. The purpose of this paper is to find the most powerful test statistic among them through the small-scale simulation.

• 전기학회논문지
• /
• 제61권7호
• /
• pp.1007-1013
• /
• 2012

Modeling of photovoltaic power systems is significant to analytically predict its dynamics in practical applications. This paper presents a novel modeling algorithm of such system by using probability and statistic theories. We first establish a linear model basically composed of Fourier parameter sets for mapping the input/output variable of photovoltaic systems. The proposed model includes solar irradiation and ambient temperature of photovoltaic modules as an input vector and the inverter power output is estimated sequentially. We deal with these measurements as random variables and derive a parameter learning algorithm of the model in terms of statistics. Our learning algorithm requires computation of an expectation and joint expectation against solar irradiation and ambient temperature, which are analytically solved from the integral calculus. For testing the proposed modeling algorithm, we utilize realistic measurement data sets obtained from the Seokwang Solar power plant in Youngcheon, Korea. We demonstrate reliability and superiority of the proposed photovoltaic system model by observing error signals between a practical system output and its estimation.

• 산업경영시스템학회지
• /
• 제18권33호
• /
• pp.93-102
• /
• 1995

In this study we are concerned with the diagnostics method of cross-classified categorical data using logistic regression model of binary response models for cell proportions. under this model, we could examine the goodness-of-fit of the models using Pearson's $x^2$test statistic and likelihood ratio statistic. Under this model, these statistics are assumed that sample survey schemes are with replacement sampling model. But these statistics are often inappropriate for analysing contingency tables consists of complex sampling schemes obtained sample survey data. In this study we are examined diagnostics procedures detecting any outlying cell proportions and influential observations on design space in logistic regression modeltake account of the survey design effects.

• Journal of the Korean Data and Information Science Society
• /
• 제5권2호
• /
• pp.95-106
• /
• 1994

A proposed test statistic is obtained by multiplying constant weights by the Neumann smooth type statistic discussed by Eubank and Hart(1993) in order to observe the effect of weight. It has very good results of power studies. Another advantage of this test is that it simultaneously provides an important diagnostic tools that can be used in many cases to determine how the model should be adjusted.

• 품질경영학회지
• /
• 제21권2호
• /
• pp.156-161
• /
• 1993

Limiting distributions of Studentized test statistics have been shown for testing the slope parameter in a simple linear structural model. Since the limiting distribution of Studentized one appears to yield inaccurate inference, this paper suggests adjustment of critical value and normalization of the Studentized one. As results, we can have procedures for refined inference based on our approximate distrbution instead of the limiting distribution.

• Communications for Statistical Applications and Methods
• /
• 제8권2호
• /
• pp.483-497
• /
• 2001

When one fits a parametric density function to a data set, it is usually advisable to test the goodness of the postulated model. In this paper we study the nonparametric tests for testing the null hypothesis against general alternatives, when the null hypothesis specifies the density function up to unknown parameters. We modify the test statistic which was proposed by the first author and his colleagues. Asymptotic distribution of the modified statistic is derived and its performance is compared with some other tests through simulation.