Nonparametric estimation of risk tracking indices for longitudinal studies

Tracking a subject's risk factors or health status over time is an important objective in long-term epidemiological studies with repeated measurements. An important issue of time-trend tracking is to define appropriate statistical indices to quantitatively measure the tracking abilities of the targeted risk factors or health status over time. We present a number of local and global statistical tracking indices based on the rank-tracking probabilities, which are derived from the conditional distribution functions, and propose a class of kernel-based nonparametric estimation methods. Confidence intervals for the estimators of the tracking indices are constructed through a resampling subject bootstrap procedure. We demonstrate the application of the tracking indices using the body mass index and systolic blood pressure data from the Coronary Artery Risk Development in Young Adults (CARDIA) study. Statistical properties of the estimation methods and bootstrap inference are investigated through a simulation study and an asymptotic development.

[1]  J. Hart,et al.  Consistency of cross-validation when the data are curves , 1993 .

[2]  Colin O. Wu,et al.  Nonparametric Mixed Effects Models for Unequally Sampled Noisy Curves , 2001, Biometrics.

[3]  Damla Şentürk,et al.  Functional Varying Coefficient Models for Longitudinal Data , 2010 .

[4]  D. Jacobs,et al.  Cumulative Blood Pressure in Early Adulthood and Cardiac Dysfunction in Middle Age: The CARDIA Study. , 2015, Journal of the American College of Cardiology.

[5]  J. Hart,et al.  Kernel Regression Estimation Using Repeated Measurements Data , 1986 .

[6]  Robert E. Weiss,et al.  An Analysis of Paediatric Cd4 Counts for Acquired Immune Deficiency Syndrome Using Flexible Random Curves , 1996 .

[7]  J. Ware,et al.  Tracking: Prediction of Future Values from Serial Measurements , 1981 .

[8]  R. Lauer,et al.  Factors affecting tracking of coronary heart disease risk factors in children. The Muscatine Study. , 1991, Annals of the New York Academy of Sciences.

[9]  M. A. Foulkes,et al.  An index of tracking for longitudinal data , 1981 .

[10]  G. Molenberghs,et al.  Models for Discrete Longitudinal Data , 2005 .

[11]  Jianhua Z. Huang,et al.  Joint modelling of paired sparse functional data using principal components. , 2008, Biometrika.

[12]  Li Ping Yang,et al.  Nonparametric smoothing estimates of time-varying coefficient models with longitudinal data , 1998 .

[13]  R. Lauer,et al.  Tracking of Blood Lipids and Blood Pressures in School Age Children: The Muscatine Study , 1978, Circulation.

[14]  Donald Hedeker,et al.  Longitudinal Data Analysis , 2006 .

[15]  Tapabrata Maiti,et al.  Analysis of Longitudinal Data (2nd ed.) (Book) , 2004 .

[16]  C. A. McMahan.  An Index of Tracking , 1981 .

[17]  R. Carroll,et al.  Semiparametric Regression for Clustered Data Using Generalized Estimating Equations , 2001 .

[18]  Xin Tian,et al.  Statistical Indices for Risk Tracking in Longitudinal Studies , 2017 .

[19]  A. Dyer,et al.  Cardiovascular risk factors in young adults. The CARDIA baseline monograph. , 1991, Controlled clinical trials.

[20]  Colin O. Wu,et al.  Nonparametric Estimation of Conditional Distributions and Rank-Tracking Probabilities With Time-Varying Transformation Models in Longitudinal Studies , 2013 .

[21]  Colin O. Wu,et al.  Nonparametric Estimation of Conditional Distribution Functions and Rank-Tracking Probabilities With Longitudinal Data , 2013 .

[22]  Colin O. Wu,et al.  Estimation of rank-tracking probabilities using nonparametric mixed-effects models for longitudinal data , 2014 .

[23]  C. Lewis,et al.  Body mass index trajectories in young adulthood predict non‐alcoholic fatty liver disease in middle age: The CARDIA cohort study , 2018, Liver international : official journal of the International Association for the Study of the Liver.

[24]  W. Härdle.  Applied Nonparametric Regression , 1992 .

[25]  Catherine A. Sugar,et al.  Principal component models for sparse functional data , 1999 .

[26]  Juned Siddique,et al.  Blood pressure trajectories in early adulthood and subclinical atherosclerosis in middle age. , 2014, JAMA.

[27]  E. Arnesen,et al.  Tracking of cardiovascular risk factors: the Tromsø study, 1979-1995. , 2001, American journal of epidemiology.

[28]  S B Hulley,et al.  CARDIA: study design, recruitment, and some characteristics of the examined subjects. , 1988, Journal of clinical epidemiology.

[29]  Jianqing Fan,et al.  Two‐step estimation of functional linear models with applications to longitudinal data , 1999 .

[30]  R. Lauer,et al.  Factors Affecting Tracking of Coronary Heart Disease Risk Factors in Children a , 1991 .