A statistical method for studying correlated rare events and their risk factors

Abstract

Longitudinal studies of rare events such as cervical high-grade lesions or colorectal polyps that can recur often involve correlated binary data. Risk factor for these events cannot be reliably examined using conventional statistical methods. For example, logistic regression models that incorporate generalized estimating equations often fail to converge or provide inaccurate results when analyzing data of this type. Although exact methods have been reported, they are complex and computationally difficult. The current paper proposes a mathematically straightforward and easy-to-use two-step approach involving (i) an additive model to measure associations between a rare or uncommon correlated binary event and potential risk factors and (ii) a permutation test to estimate the statistical significance of these associations. Simulation studies showed that the proposed method reliably tests and accurately estimates the associations of exposure with correlated binary rare events. This method was then applied to a longitudinal study of human leukocyte antigen (HLA) genotype and risk of cervical high grade squamous intraepithelial lesions (HSIL) among HIV-infected and HIV-uninfected women. Results showed statistically significant associations of two HLA alleles among HIV-negative but not HIV-positive women, suggesting that immune status may modify the HLA and cervical HSIL association. Overall, the proposed method avoids model nonconvergence problems and provides a computationally simple, accurate, and powerful approach for the analysis of risk factor associations with rare/uncommon correlated binary events.

Keywords

Get full access to this article

View all access options for this article.

References

Liang

Zeger

. Longitudinal data analysis using generalized linear models. Biometrika 1986; 73: 13–22.

Albert

Anderson

. On the existence of maximum likelihood estimates in logistic regression models. Biometrika 1984; 71: 1–10.

Schaefer

. Bias correction in maximum likelihood logistic regression. Stat Med 1983; 2: 71–78.

Cordeiro

McCullagh

. Bias correction in generalized linear models. J Roy Stat Soc Ser B 1991; 53: 629–643.

Bull

Greenwood

CMT

Hauck

. Jackknife bias reduction for polychotomous logistic regression. Stat Med 1997; 16: 545–560.

Cordeiro

Cribari-Neto

. On bias reduction in exponential and non-exponential family regression models. Commun Stat 1998; 27: 485–500.

Leung

DH-Y

Wang

. Bias reduction using stochastic approximation. Aust NZ J Stat 1998; 40: 43–52.

Anderson

Blair

. Penalized maximum likelihood estimation in logistic regression and discrimination. Biometrika 1982; 69: 123–136.

Self

Liang

K-Y

. Asymptotic properties of maximum likelihood estimators and likelihood ratio tests under nonstandard conditions. J Am Stat Assoc 1987; 82: 605–610.

10.

Agresti

. Exact inference for categorical data: recent advances and continuing controversies. Stat Med 2001; 20: 2709–2722.

11.

Potter

. A permutation test for inference in logistic regression with small- and moderate sized data sets. Stat Med 2005; 24: 693–708.

12.

Heinze

Schemper

. A solution to the problem of separation in logistic regression. Stat Med 2002; 21: 2409–2419.

13.

Firth

Generalized linear models and Jeffreys priors: An iterative weighted least-squares approach. In: Dodge

Whittaker

(eds). Computational statistics, Heidelberg: Physica-Verlag, 1992, pp. 553–557.

14.

Firth

. Bias reduction of maximum likelihood estimates. Biometrika 1993; 80: 27–38.

15.

Tang

M-L

Tang

N-S

Rosner

. Statistical inference for correlated data in ophthalmologic studies. Stat Med 2006; 25: 2772–2783.

16.

Hungsberger

Graubard

Korn

. Testing logistic regression coefficients with clustered data and few positive outcomes. Stat Med 2008; 27: 1305–1324.

17.

Corcoran

Ryan

Senchaudhuri

. An exact trend test for correlated binary data. Biometrics 2001; 57: 941–948.

18.

Molenberghs

Ryan

. An exponential family model for clustered multivariate binary data. Environmetrics 1999; 10: 279–300.

19.

Ryan

Molenberghs

Statistical methods for developmental toxicity: Analysis of clustered multivariate binary data. In: Bailer

Maltoni

Bailar

(eds). Uncertainty in the risk assessment of environmental and occupational hazards: An international workshop (Annals of the New York Academy of Sciences), New York: New York Academy of Sciences, 1999, pp. 196–211.

20.

Mehta

Patel

Senchaudhuri

. Exact stratified linear rank tests for ordered categorical and binary data. J Comput Graph Stat 1992; 1: 21–40.

21.

Gelman

Jakulin

Pittau

. A weakly informative default prior distribution for logistic and other regression models. Ann Appl Stat 2008; 2: 1360–1383.

22.

Abrahantes

Aerts

. A solution to separation for clustered binary data. Stat Model 2012; 12: 3–27.

23.

Lancaster

. Significance tests in discrete distributions. J Am Stat Assoc 1961; 56: 223–234.

24.

Bernard

. Must clinical trials be large? The interpretation of p-values and the combination of test results. Stat Med 1990; 9: 601–614.

25.

Hirji

Tan

S-J

Elashoff

. A quasi-exact test for comparing two binomial proportions. Stat Med 1990; 9: 601–614.

26.

Still

White

. The approximate randomization test as an alternative to the F-test in the analysis of variance. Br J Math Stat Psychol 1981; 34: 243–252.

27.

Ter Braak CJF. Permutation versus bootstrap significance tests in multiple regression and ANOVA. In: Jockel

(ed). Bootstrapping and related techniques, Berlin: Springer, 1992, pp. 79–86.

28.

Jung

Jhun

Song

. A new random permutation test in ANOVA models. Stat Papers 2007; 48: 47–62.

29.

Good

. Extension of the concept of exchangeability and their applications. J Modern Appl Stat Meth 2002; 1: 243–247.

30.

Barkan

Melnick

Preston-Martin

. The Women's Interagency HIV Study. WIHS Collaborative Study Group. Epidemiology 1998; 9: 117–125.

31.

Strickler

Palefsky

Shah

. Human papillomavirus 16 and immune status in human immunodeficiency virus-seropositive women. J Natl Cancer Inst 2003; 95: 1062–1071.

32.

Strickler

Burk

Fazzari

. HPV natural history and possible HPV reactivation in HIV-positive women. J Natl Cancer Inst 2005; 97: 577–586.

33.

Xue

Gange

Zhong

. Marginal and mixed effects models in the analysis of HPV natural history data. CEBP 2010; 19: 159–169.

34.

Kuniholm

Gao

Xue

. The relation of HLA genotype to hepatitis C viral load and markers of liver fibrosis in HIV-infected and HIV-uninfected women. J Infect Dis 2013; 12: 1807–1814.