Linear spline multilevel models for summarising childhood growth trajectories: A guide to their application using examples from five birth cohorts

Statistical Methods in Medical Research - Tập 25 Số 5 - Trang 1854-1874 - 2016
Laura D Howe1, Kate Tilling2, Alícia Matijasevich3, Emily Petherick4, Ana Cristina Santos5, Lesley Smith4, John Wright4, Iná S. Santos3, Aluísio J. D. Barros3, Richard M. Martin6,2, Michael S. Kramer7, Natalia Bogdanovich8, Lidia Matush8, Henrique Barros5, Debbie A. Lawlor1
1MRC Integrative Epidemiology Unit at the University of Bristol, School of Social and Community Medicine, University of Bristol, UK
2School of Social and Community Medicine, University of Bristol, UK
3Postgraduate Programme in Epidemiology, Federal University of Pelotas, Pelotas, Brazil .
4Bradford Teaching Hospitals NHS Foundation Trust, Bradford, UK
5Department of Clinical Epidemiology, Predictive Medicine and Public Health, University of Porto Medical School, University of Porto Institute of Public Health, Porto, Portugal
6National Institute for Health Research Bristol Nutrition Biomedical Research Unit, University of Bristol / University Hospitals Bristol NHS Foundation Trust, Bristol, UK
7Department of Pediatrics, Faculty of Medicine, McGill University, Montreal, Canada
8Belarusian Ministry of Health and Belarussian Maternal and Child Health Research Institute, Minsk, Belarus

Tóm tắt

Childhood growth is of interest in medical research concerned with determinants and consequences of variation from healthy growth and development. Linear spline multilevel modelling is a useful approach for deriving individual summary measures of growth, which overcomes several data issues (co-linearity of repeat measures, the requirement for all individuals to be measured at the same ages and bias due to missing data). Here, we outline the application of this methodology to model individual trajectories of length/height and weight, drawing on examples from five cohorts from different generations and different geographical regions with varying levels of economic development. We describe the unique features of the data within each cohort that have implications for the application of linear spline multilevel models, for example, differences in the density and inter-individual variation in measurement occasions, and multiple sources of measurement with varying measurement error. After providing example Stata syntax and a suggested workflow for the implementation of linear spline multilevel models, we conclude with a discussion of the advantages and disadvantages of the linear spline approach compared with other growth modelling methods such as fractional polynomials, more complex spline functions and other non-linear models.

Từ khóa


Tài liệu tham khảo

Cameron N, 2006, Human growth and development, 1

10.1056/NEJMoa072515

10.1056/NEJMoa044160

10.1136/jech.54.2.97

10.1016/j.ehb.2009.06.004

WHO Multicentre Growth Reference Study Group, 2006, Acta Paediatr, 56, 10.1111/j.1651-2227.2006.tb02376.x

Steckel RH, 1995, J Econom Lit, 33, 1903

10.1136/jech.2010.113068

10.1111/j.1365-3016.2011.01251.x

10.1093/jn/134.1.201

10.1093/aje/kwp047

10.1093/aje/kwi002

Anderson EL, Tilling K, Fraser A, et al. Estimating trajectories of energy intake through childhood and adolescence using linear-spline multilevel models. Epidemiology. Epub ahead of print 21 May 2013. DOI: 10.1097/EDE.0b013e318295af33.

10.1093/ije/dyr119

10.1161/HYPERTENSIONAHA.108.114256

Howe LD, Tilling K, Benfield L, et al. Changes in ponderal index and body mass index across childhood and their associations with fat mass and cardiovascular risk factors at age 15. PLoS One 2010;5:e15186.

Tilling K, Davies NM, Nicoli E, et al. Associations of growth trajectories in infancy and early childhood with later childhood outcomes. Am J Clin Nutr 2011; 94: 1808S–1813S.

10.1002/sim.5385

Steele F, 2008, J R Stat Soc (A), 171, 5, 10.1111/j.1467-985X.2007.00509.x

10.1002/9781119013563

10.1214/ss/1177011926

Goldstein H, 1995, Multilevel statistical models, 2

10.1002/sim.4780141303

10.1093/ije/28.5.964

10.1177/1536867X0400400301

Wen X, 2011, Am J Epidemiol, 173, S157, 10.1093/aje/kwq391

Marsh LC, 2001, Spline regression models

10.1186/1471-2458-8-327

10.1001/jama.285.4.413

Boyd A, Golding J, Macleod J, et al. Cohort profile: The ‘Children of the 90s'–the index offspring of the Avon Longitudinal Study of Parents and Children. Int J Epidemiol 2013; 42: 111–127.

Fraser A, Macdonald-Wallis C, Tilling K, et al. Cohort profile: The Avon Longitudinal Study of Parents and Children: ALSPAC mothers cohort. Int J Epidemiol 2013; 42: 97–110.

10.1093/ije/dyq130

10.1080/01621459.1994.10476806

Centre for Multilevel Modelling. runmlwin: Stata module for fitting multilevel models in the MLwiN software. Centre for Multilevel Modelling, University of Bristol (computer program), 2011.

StataCorp. Stata 12.0 [computer program]. Texas: StataCorp, 2011.

Centre for Multilevel Modelling. MLwiN Version 2.24. Centre for Multilevel Modelling, University of Bristol (computer program), 2011.

10.1016/j.csda.2006.05.021

10.1002/(SICI)1097-0258(19981030)17:20<2381::AID-SIM926>3.0.CO;2-S

10.1002/sim.4780131605

10.1038/ijo.2009.108

10.1093/ije/dyq115

10.1080/03014467800002601

10.1371/journal.pone.0029118

10.1136/archdischild-2012-302778