Optimal selection of individuals for repeated covariate measurements in follow-up studies

Abstract

Repeated covariate measurements bring important information on the time-varying risk factors in long epidemiological follow-up studies. However, due to budget limitations, it may be possible to carry out the repeated measurements only for a subset of the cohort. We study cost-efficient alternatives for the simple random sampling in the selection of the individuals to be remeasured. The proposed selection criteria are based on forms of the D-optimality. The selection methods are compared with the simulation studies and illustrated with the data from the East–West study carried out in Finland from 1959 to 1999. The results indicate that cost savings can be achieved if the selection is focused on the individuals with high expected risk of the event and, on the other hand, on those with extreme covariate values in the previous measurements.

Keywords

optimal design data collection repeated measurements follow-up study missing covariate data

Get full access to this article

View all access options for this article.

References

Clarke

Shipley

Lewington

et al.

Underestimation of risk associations due to regression dilution in long-term follow-up of prospective studies. Am J Epidemiol 1999; 150: 341–353.

Frost

Thompson

. Correcting for regression dilution bias: comparison of methods for a single predictor variable. J R Stat Soc Ser A Stat Soc 2000; 163: 173–189.

Reilly

. Optimal sampling strategies for two-stage studies. Am J Epidemiol 1996; 143: 92–100.

Liu

. Statistical power and optimum sample allocation ratio for treatment and control having unequal costs per unit of randomization. J Educ Behav Stat 2003; 28: 231–248.

Atkinson

Donev

Tobias

. Optimum experimental designs, with SAS, Oxford: Oxford University Press, 2007.

Pukelsheim

. Optimal design of experiments, New York, NY: Wiley, 1993.

Karvanen

Kulathinal

Gasbarra

. Optimal designs to select individuals for genotyping conditional on observed binary or survival outcomes and non-genetic covariates. Comput Stat Data Anal 2009; 53: 1782–1793.

Mehtälä

Auranen

Kulathinal

. Optimal designs for epidemiologic longitudinal studies with binary outcomes. Stat Meth Med Res 2011. DOI: 10.1177/0962280211430663.

Wulfsohn

Tsiatis

. A joint model for survival and longitudinal data measured with error. Biometrics 1997; 53: 330–339.

10.

Diggle

Sousa

Chetwynd

. Joint modelling of repeated measurements and time-to-event outcomes: The fourth Armitage lecture. Stat Med 2008; 27: 2981–2998.

11.

Rathbun

Song

Neustifter

et al.

Survival analysis with time varying covariates measured at random times by design. J R Stat Soc Ser C Appl Stat 2013; 62: 419–434.

12.

Njagi

Molenberghs

Rizopoulos

et al.

A flexible joint modeling framework for longitudinal and time-to-event data with overdispersion. Stat Meth Med Res 2013. DOI: 10.1177/0962280213495994.

13.

Keys

. Coronary heart disease in seven countries. Circulation 1970; 41: 186–195.

14.

Little

Rubin

. Statistical analysis with missing data, 2nd ed. New York, NY: Wiley, 2002.

15.

Robert

Casella

. Monte Carlo statistical methods, New York, NY: Springer, 1999.

16.

Elfving

. Optimum allocation in linear regression theory. Ann Math Stat 1952; 23: 255–262.

17.

Wright

Bailer

. Optimal experimental design for a nonlinear response in enviromental toxicology. Biometrics 2006; 62: 886–892.

18.

Dykstra

. The augmentation of experimental data to maximize

| X' X |

. Technometrics 1971; 13: 682–688.

19.

Retout

Comets

Samson

et al.

Design in nonlinear mixed effects models: optimization using the Fedorov–Wynn algorithm and power of the Wald test for binary covariates. Stat Med 2007; 26: 5162–5179.

20.

Rubin

. Inference and missing data. Biometrika 1976; 63: 581–592.

21.

Dempster

Laird

Rubin

. Maximum likelihood estimation from incomplete data via the EM algorithm (with discussion). J R Stat Soc Ser B 1977; 39: 1–38.

22.

Tanner

Wong

. The calculation of posterior distributions by data augmentation. J Am Stat Assoc 1987; 82: 528–540.

23.

Rubin

. Multiple imputation for nonresponse in surveys. Wiley Ser Probab Math Stat, New York, NY: John Wiley & Sons, Inc., 1987.

24.

Moons

Donders

Stijnen

et al.

Using the outcome for imputation of missing predictor values was preferred. J Clin Epidemiol 2006; 59: 1092–1101.

25.

White

Royston

. Imputing missing covariate values for the Cox model. Stat Med 2009; 28: 1982–1998.

26.

R Core Team. R: A language and environment for statistical computing. Vienna, Austria, http://www.R-project.org/ (2012, accessed 26 August 2013)..

27.

Broström G. eha: Event History Analysis. R package version 2.2-0, http://CRAN.R-project.org/package=eha (2012, accessed 26 August 2013).

28.

Gilbert P and Varadhan R. numDeriv: Accurate Numerical Derivatives. R package version 2012.9-1, http://CRAN.R-project.org/package=numDeriv (2012, accessed 26 August 2013).

29.

Ferreira da Silva

Ferrari

Cribari-Neto

. Improved likelihood inference for the shape parameter in Weibull regression. J Stat Comput Simul 2008; 78: 789–811.

30.

Karvonen

Blomqvist

Kallio

et al.

Men in rural East and West Finland. Acta Med Scand 1966; 180: 169–190.

31.

Pekkanen J. Coronary heart disease during a 25-year follow-up: risk factors and their secular trends in the Finnish cohorts of the Seven Countries Study. PhD Thesis, University of Helsinki, Government Printing Centre, 1987.

32.

Saarela

Kulathinal

Karvanen

. Secondary analysis under cohort sampling designs using conditional likelihood. J Probab Stat 2012; 2012: Article ID 931416. DOI: 10.1155/2012/931416–Article ID 931416. DOI: 10.1155/2012/931416.