Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

RE: “USE OF TWO-SEGMENTED LOGISTIC REGRESSION TO ESTIMATE CHANGE-POINTS IN EPIDEMIOLOGIC STUDIES”

RE: “USE OF TWO-SEGMENTED LOGISTIC REGRESSION TO ESTIMATE CHANGE-POINTS IN EPIDEMIOLOGIC STUDIES” We thank Ulm and Küchenhoff (1) for their valuable comments on our paper (2) about the estimation of change-points in epidemiologic studies. We agree with them that several methods for change-point estimation in generalized linear models have already been described in the statistical literature (3–6). However, it is equally important to recognize that their epidemiologic application to dose-response assessment has been limited by the lack of an algorithm to simultaneously estimate the change-point and the other parameters of effect. In the above approaches, the maximum likelihood estimates are obtained from the profile log-likelihood of sequential models at different fixed change-points, and hence, this estimation procedure does not provide appropriate error estimates for model parameters, except for the change-point (see the Discussion section of our paper (2) for a detailed description). To overcome this problem, the algorithm proposed in our article (2) provides simultaneous estimation of all model parameters and realistic measurements of statistical error. An alternative method recently proposed by Daniels et al. (7) is now also available. As stated in our paper (2), all inferences presented, such as confidence intervals or tests of hypothesis, were based on the assumption of existence of the change-point. Although we did not explicitly address the problem of testing for this assumption, we recognized, as stated in the Discussion section of our paper, that further research is needed in this area. Ulm (3) suggested a test procedure based on a quasi one-sided χ21 distribution. The justification of this method was supported by a simulation study, but it would be desirable to have a more rigorous theoretical justification. In our opinion, alternative methods to test formally for the existence of change-points are still needed. With respect to the example of alcohol intake and risk of myocardial infarction, several points need to be clarified. This example illustrates the applicability of two-segmented logistic models to estimate and provide inferences about the location of the change-point and the magnitude of other parameters of effect, when the change-point actually exists. If there is no change-point (i.e., β2 = 0), the model reduces to a standard logistic regression, and hence, the change-point is not well defined. In such circumstances, the standard asymptotic properties of Wald and likelihood ratio statistics do not hold (8). Finally, we would like to stress that, although several models are used to display the dose-response relation of alcohol with the risk of myocardial infarction, only the quadratic-linear model provides a statistical estimation of the change-point. We concur with Ulm and Küchenhoff (1) on the importance of careful modeling and interpretation in assessing change-points and, as already discussed in our article (2), we suggest using nonparametric regression to check the appropriate parameterization of segmented models. Alternative parametric change-point models, which can accommodate many epidemiologic dose-response relations, are also described in the Discussion section of our paper. In conclusion, we believe that the two-segmented logistic regression model provides valuable inference procedures when threshold effects are anticipated and that it deserves wider use in epidemiologic dose-response analyses. REFERENCES 1. Ulm K, Küchenhoff H. Re: “Use of two-segmented logistic regression to estimate change-points in epidemiologic studies.” (Letter). Am J Epidemiol  2000; 152: 289. Google Scholar 2. Pastor R, Guallar E. Use of two-segmented logistic regression to estimate change-points in epidemiologic studies. Am J Epidemiol  1998; 148: 631–42. Google Scholar 3. Ulm K. A statistical method for assessing a threshold in epidemiological studies. Stat Med  1991; 10: 341–9. Google Scholar 4. Stasinopoulos DM, Rigby RA. Detecting break points in generalized linear models. Comput Stat Data Anal  1992; 13: 461–71. Google Scholar 5. Goetghebeur E, Pocock SJ. Detection and estimation of J-shaped risk-response relationships. J R Stat Soc (A)  1995; 158: 107–21. Google Scholar 6. Küchenhoff H. An exact algorithm for estimating breakpoints in segmented generalized linear models. Comput Stat  1997; 12: 235–47. Google Scholar 7. Daniels MJ, Dominici F, Samet JM, et al. Estimating particulate matter-mortality dose-response curves and threshold levels: an analysis of daily time-series for the 20 largest US cities. Am J Epidemiol  2000; 152:397–406. Google Scholar 8. Feder PI. The log likelihood ratio in segmented regression. Ann Stat  1975; 3: 84–97. Google Scholar http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png American Journal of Epidemiology Oxford University Press

RE: “USE OF TWO-SEGMENTED LOGISTIC REGRESSION TO ESTIMATE CHANGE-POINTS IN EPIDEMIOLOGIC STUDIES”

Download PDF
3 pages

Loading next page...
 
/lp/oxford-university-press/re-use-of-two-segmented-logistic-regression-to-estimate-change-points-m0AyAh5Den

References (23)

Publisher
Oxford University Press
ISSN
0002-9262
eISSN
1476-6256
DOI
10.1093/aje/153.6.615
Publisher site
See Article on Publisher Site

Abstract

We thank Ulm and Küchenhoff (1) for their valuable comments on our paper (2) about the estimation of change-points in epidemiologic studies. We agree with them that several methods for change-point estimation in generalized linear models have already been described in the statistical literature (3–6). However, it is equally important to recognize that their epidemiologic application to dose-response assessment has been limited by the lack of an algorithm to simultaneously estimate the change-point and the other parameters of effect. In the above approaches, the maximum likelihood estimates are obtained from the profile log-likelihood of sequential models at different fixed change-points, and hence, this estimation procedure does not provide appropriate error estimates for model parameters, except for the change-point (see the Discussion section of our paper (2) for a detailed description). To overcome this problem, the algorithm proposed in our article (2) provides simultaneous estimation of all model parameters and realistic measurements of statistical error. An alternative method recently proposed by Daniels et al. (7) is now also available. As stated in our paper (2), all inferences presented, such as confidence intervals or tests of hypothesis, were based on the assumption of existence of the change-point. Although we did not explicitly address the problem of testing for this assumption, we recognized, as stated in the Discussion section of our paper, that further research is needed in this area. Ulm (3) suggested a test procedure based on a quasi one-sided χ21 distribution. The justification of this method was supported by a simulation study, but it would be desirable to have a more rigorous theoretical justification. In our opinion, alternative methods to test formally for the existence of change-points are still needed. With respect to the example of alcohol intake and risk of myocardial infarction, several points need to be clarified. This example illustrates the applicability of two-segmented logistic models to estimate and provide inferences about the location of the change-point and the magnitude of other parameters of effect, when the change-point actually exists. If there is no change-point (i.e., β2 = 0), the model reduces to a standard logistic regression, and hence, the change-point is not well defined. In such circumstances, the standard asymptotic properties of Wald and likelihood ratio statistics do not hold (8). Finally, we would like to stress that, although several models are used to display the dose-response relation of alcohol with the risk of myocardial infarction, only the quadratic-linear model provides a statistical estimation of the change-point. We concur with Ulm and Küchenhoff (1) on the importance of careful modeling and interpretation in assessing change-points and, as already discussed in our article (2), we suggest using nonparametric regression to check the appropriate parameterization of segmented models. Alternative parametric change-point models, which can accommodate many epidemiologic dose-response relations, are also described in the Discussion section of our paper. In conclusion, we believe that the two-segmented logistic regression model provides valuable inference procedures when threshold effects are anticipated and that it deserves wider use in epidemiologic dose-response analyses. REFERENCES 1. Ulm K, Küchenhoff H. Re: “Use of two-segmented logistic regression to estimate change-points in epidemiologic studies.” (Letter). Am J Epidemiol  2000; 152: 289. Google Scholar 2. Pastor R, Guallar E. Use of two-segmented logistic regression to estimate change-points in epidemiologic studies. Am J Epidemiol  1998; 148: 631–42. Google Scholar 3. Ulm K. A statistical method for assessing a threshold in epidemiological studies. Stat Med  1991; 10: 341–9. Google Scholar 4. Stasinopoulos DM, Rigby RA. Detecting break points in generalized linear models. Comput Stat Data Anal  1992; 13: 461–71. Google Scholar 5. Goetghebeur E, Pocock SJ. Detection and estimation of J-shaped risk-response relationships. J R Stat Soc (A)  1995; 158: 107–21. Google Scholar 6. Küchenhoff H. An exact algorithm for estimating breakpoints in segmented generalized linear models. Comput Stat  1997; 12: 235–47. Google Scholar 7. Daniels MJ, Dominici F, Samet JM, et al. Estimating particulate matter-mortality dose-response curves and threshold levels: an analysis of daily time-series for the 20 largest US cities. Am J Epidemiol  2000; 152:397–406. Google Scholar 8. Feder PI. The log likelihood ratio in segmented regression. Ann Stat  1975; 3: 84–97. Google Scholar

Journal

American Journal of Epidemiology – Oxford University Press

Published: Mar 15, 2001

There are no references for this article.