TY - JOUR
T1 - Longitudinal analysis of sibling correlation on blood pressure using mixed modeling
AU - Tan, Qihua
AU - Duan, Hongmei
AU - Wang, Ancong
AU - Zhu, Dongyi
AU - Li, Shuxia
PY - 2019
Y1 - 2019
N2 - Purpose: Although moderate to high genetic contribution to blood pressure variation have been estimated in numerous studies, the genetic control over the longitudinal change in blood pressure has been less frequently investigated because of the requirement of longitudinal design. Methods: Based on blood pressure data from a large-scale family-based longitudinal survey, we introduced hierarchical modeling of longitudinal family data in combination with fractional polynomials for fitting nonlinear age patterns of blood pressure and the mixed-effect models for estimating sibling correlation on blood pressure to assess the genetic and shared environmental effects on blood pressure level as well as on the rate of change in blood pressure over ages. Results: Significant sibling correlations were estimated on the levels of systolic blood pressure (0.2, 95% CI: 0.10–0.30) and diastolic blood pressure (0.28, 95% CI: 0.18–0.38), whereas for the longitudinal change or the rate of change, significant correlation was estimated only for diastolic blood pressure (0.13, 95% CI: 0.04–0.23). In the sex-specific analysis, similar pattern is observed, but statistical significance was only reached in female siblings with correlation estimates higher than the overall sample. Conclusion: The rate of change in blood pressure is mainly influenced by individual's unique environment; and the genetic and common family environment may play a role in regulating the longitudinal change of diastolic but not systolic blood pressure.
AB - Purpose: Although moderate to high genetic contribution to blood pressure variation have been estimated in numerous studies, the genetic control over the longitudinal change in blood pressure has been less frequently investigated because of the requirement of longitudinal design. Methods: Based on blood pressure data from a large-scale family-based longitudinal survey, we introduced hierarchical modeling of longitudinal family data in combination with fractional polynomials for fitting nonlinear age patterns of blood pressure and the mixed-effect models for estimating sibling correlation on blood pressure to assess the genetic and shared environmental effects on blood pressure level as well as on the rate of change in blood pressure over ages. Results: Significant sibling correlations were estimated on the levels of systolic blood pressure (0.2, 95% CI: 0.10–0.30) and diastolic blood pressure (0.28, 95% CI: 0.18–0.38), whereas for the longitudinal change or the rate of change, significant correlation was estimated only for diastolic blood pressure (0.13, 95% CI: 0.04–0.23). In the sex-specific analysis, similar pattern is observed, but statistical significance was only reached in female siblings with correlation estimates higher than the overall sample. Conclusion: The rate of change in blood pressure is mainly influenced by individual's unique environment; and the genetic and common family environment may play a role in regulating the longitudinal change of diastolic but not systolic blood pressure.
KW - Blood pressure
KW - Fractional polynomials
KW - Longitudinal
KW - Mixed-effects model
KW - Sibling correlation
U2 - 10.1016/j.annepidem.2019.02.006
DO - 10.1016/j.annepidem.2019.02.006
M3 - Journal article
C2 - 30904389
AN - SCOPUS:85062940376
SN - 1047-2797
VL - 33
SP - 49
EP - 53
JO - Annals of Epidemiology
JF - Annals of Epidemiology
ER -