Use of Two-segmented Logistic Regression to Estimate Change-points in Epidemiologic Studies

Pastor, Roberto Viciano; Güallar, Eliseo

doi:10.1093/aje/148.7.631

Cited by 77 publications

(61 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1 Cox, 1987;Schwartz et al, 1995). In epidemiology, logistic regressions with change-points are used to model the relationship between the continuous exposure variable and disease risk (see Pastor and Guallar, 1998;Pastor-Barriuso et al, 2003). In economics, …nance, and management, random utility models that are nonlinear in income and/or price are commonly employed (see, e.g.…”

Section: Introductionmentioning

confidence: 99%

Semiparametric estimation of a binary response model with a change-point due to a covariate threshold

Lee

Seo

2008

Journal of Econometrics

View full text Add to dashboard Cite

This paper is concerned with semiparametric estimation of a threshold binary response model. The estimation method considered in the paper is semiparametric since the parameters for a regression function are finite-dimensional, while allowing for heteroskedasticity of unknown form. In particular, the paper considers Manski (1975Manski ( , 1985's maximum score estimator. The model in this paper is irregular because of a change-point due to an unknown threshold in a covariate. This irregularity coupled with the discontinuity of the objective function of the maximum score estimator complicates the analysis of the asymptotic behavior of the estimator. Sufficient conditions for the identification of parameters are given and the consistency of the estimator is obtained. It is shown that the estimator of the threshold parameter is n-consistent and the estimator of the remaining regression parameters is cube root n-consistent. Furthermore, we obtain the asymptotic distribution of the estimators. It turns out that a suitably normalized estimator of the regression parameters converges weakly to the distribution to which it would converge weakly if the true threshold value were known and likewise for the threshold estimator.

show abstract

Section: Introductionmentioning

confidence: 99%

Semiparametric estimation of a binary response model with a change-point due to a covariate threshold

Lee

Seo

2008

Journal of Econometrics

View full text Add to dashboard Cite

show abstract

“…ORs were elevated among those whose first exposure was before age 20 (OR = 2.0; 95% CI, 1.4-3.0) and those who began later in life but to a lesser extent (OR for first use at 20-35 years = 1.4; 95% CI, 1.0-2.0; and OR for first use at $36 years = 1.6; 95% CI, 1.0-2.6) ( Table 2). On a continuous scale our data fit a 2-segment model 31 with a change point at 23 years of age at first use of tanning lamps (95% CI, 22-25 years). In this model, the ORs remained fairly constant by age for first use after 23 years (OR for each year of age younger =1.0; 95% CI, 0.9-1.0) and increased with each age younger at first exposure at or before 23 years (OR for each year of age younger = 1.1; 95% CI,…”

Section: Resultsmentioning

confidence: 98%

Early-Onset Basal Cell Carcinoma and Indoor Tanning: A Population-Based Study

2014

PEDIATRICS

View full text Add to dashboard Cite

WHAT'S KNOWN ON THIS SUBJECT: Indoor tanning has gained widespread popularity among adolescents and young adults. Incidence rates of early-onset basal cell carcinoma also appear to be rising. Scant evidence exists on the impacts of early exposure and whether it leads to early occurrence of this malignancy. WHAT THIS STUDY ADDS:In a US population-based study, indoor tanning was associated with an elevated risk of basal cell carcinomas occurring at or before the age of 50 years, with an increasing trend in risk with younger age at exposure among adolescents and young adults. abstract OBJECTIVE: Indoor tanning with UV radiation-emitting lamps is common among adolescents and young adults. Rising incidence rates of basal cell carcinoma (BCC) have been reported for the United States and elsewhere, particularly among those diagnosed at younger ages. Recent epidemiologic studies have raised concerns that indoor tanning may be contributing to early occurrence of BCC, and younger people may be especially vulnerable to cancer risk associated with this exposure. Therefore, we sought to address these issues in a population-based case-control study from New Hampshire. METHODS:Data on indoor tanning were obtained on 657 cases of BCC and 452 controls #50 years of age.RESULTS: Early-onset BCC was related to indoor tanning, with an adjusted odds ratio (OR) of 1.6 (95% confidence interval, 1.3-2.1). The strongest association was observed for first exposure as an adolescent or young adult, with a 10% increase in the OR with each age younger at first exposure (OR per year of age #23 = 1.1; 95% confidence interval, 1.0-1.2). Associations were present for each type of device examined (ie, sunlamps, tanning beds, and tanning booths). CONCLUSIONS:Our findings suggest early exposure to indoor tanning increases the risk of early development of BCC. They also underscore the importance of counseling adolescents and young adults about the risks of indoor tanning and for discouraging parents from consenting minors to this practice. Pediatrics 2014;134:e4-e12 AUTHORS:

show abstract

“…In the second, broader definition, the threshold is seen as an inflection point, i.e. the slope of the dose-response curve is assumed to change after the threshold, implying a bi-linear model (Chu et al, 1999;Pastor and Guallar, 1998). In the context of multivariable linear regression, these two models can be written, respectively as:…”

Section: Threshold Definitionmentioning

confidence: 99%

Using Generalized Additive Models to Detect and Estimate Threshold Associations

Benedetti¹,

Abrahamowicz²,

Leffondré³

et al. 2009

The International Journal of Biostatistics

View full text Add to dashboard Cite

In a variety of research settings, investigators may wish to detect and estimate a threshold in the association between continuous variables. A threshold model implies a non-linear relationship, with the slope changing at an unknown location. Generalized additive models (GAMs) (Hastie and Tibshirani, 1990) estimate the shape of the non-linear relationship directly from the data and, thus, may be useful in this endeavour.We propose a method based on GAMs to detect and estimate thresholds in the association between a continuous covariate and a continuous dependent variable. Using simulations, we compare it with the maximum likelihood estimation procedure proposed by Hudson (1966).We search for potential thresholds in a neighbourhood of points whose mean numerical second derivative (a measure of local curvature) of the estimated GAM curve was more than one standard deviation away from 0 across the entire range of the predictor values. A threshold association is declared if an F-test indicates that the threshold model fit significantly better than the linear model.For each method, type I error for testing the existence of a threshold against the null hypothesis of a linear association was estimated. We also investigated the impact of the position of the true threshold on power, and precision and bias of the estimated threshold.Finally, we illustrate the methods by considering whether a threshold exists in the association between systolic blood pressure (SBP) and body mass index (BMI) in two data sets.

show abstract

Use of Two-segmented Logistic Regression to Estimate Change-points in Epidemiologic Studies

Cited by 77 publications

References 21 publications

Semiparametric estimation of a binary response model with a change-point due to a covariate threshold

Semiparametric estimation of a binary response model with a change-point due to a covariate threshold

Early-Onset Basal Cell Carcinoma and Indoor Tanning: A Population-Based Study

Using Generalized Additive Models to Detect and Estimate Threshold Associations

Contact Info

Product

Resources

About