Relationship Between Mean Leucocyte Telomere Length and Measures of Allostatic Load in US Reproductive-Aged Women, NHANES 1999-2002

Abstract

Background: Reproductive health disparities may be partly explained by the cumulative effects of chronic stress experienced by socially disadvantaged groups. Although, telomere length (TL) and allostatic load score have each been used as biological markers of stress, the relationship between these two measures is unknown.

Methods: We investigated the association between leucocyte TL and allostatic load score in 1503 non-pregnant women (20-44 years) participating in the National Health and Nutrition Examination Survey, 1999-2002. We constructed six different allostatic load scores using either quartile- or clinical-based cut-points for 14 biomarkers based on previously published methods. We estimated associations between TL and allostatic load scores and component biomarkers using linear regression, also assessing interactions by race/ethnicity.

Results: After adjustment for age, longer TL was associated with higher HDL cholesterol and lower C-reactive protein and creatinine clearance; TL was not associated with the other component biomarkers. Shorter TL was associated with higher allostatic load scores for the two clinical cut-point-based scores after adjustment for age, but not the four scores based on quartile cut-points. Significant interactions by race/ethnicity were observed for TL and HbA1c and triglycerides, but not for other component biomarkers or allostatic load scores.

Conclusions: Although TL and allostatic load score are both considered measures of cumulative stress, most component biomarkers and scores using quartile-based cut-points were not associated with TL. In reproductive-aged women, allostatic load scores using clinical-based cut-points were more strongly associated with TL compared with quartile-based scores.

Keywords: allostatic load; health disparity; perinatal epidemiology; telomere length.

Figure 1.

Interactions in the linear relationship with mean log telomere length by race for glycohaemoglobin (a) and triglycerides (b). P-value for slopes from Wald test using linear regression model adjusted for age.

Figure 2.

Percent difference in mean telomere length per 1 unit increase in allostatic load score adjusted for age among non-pregnant reproductive-aged women (20–44 years old) in NHANES, 1999–2002. NHANES, National Health and Nutrition Examination Survey. Allostatic load scores were constructed using biomarkers and cut-point methods previously implemented. See text, Table S2 and reference list for details. P-value for slopes from Wald test using linear regression model adjusted for age. Of the 1503 observations with telomere data, the following biomarkers and number of observations were used for allostatic load score construction for each method: scoring method 1 – SBP, DBP, BMI, A1C, ALB, CRU, TRI, CRP, HOM, TC (n = 627); scoring method 2 – SBP, DBP, BMI, A1C, ALB, CRP, TC, HDL, PLS (n = 1417); scoring method 3 – SBP, DBP, BMI, A1C, ALB, CRP, HOM, TC, HDL, PLS (n = 1416); scoring method 4 – SBP, DBP, BMI, A1C, ALB, CRP, TC, HDL, PLS (n = 1417); scoring method 5 – SBP, DBP, A1C, ALB, CRS, CRP, HOM, TC, HDL, PLS (n = 1428); and scoring method 6 – SBP, DBP, BMI, A1C, ALB, CRC, CRP, TC, HDL (n = 1422). SBP, systolic blood pressure; DBP, diastolic blood pressure; BMI, body mass index; A1C, glycosylated haemoglobin; ALB, serum albumin; TRI, triglycerides; CRP, C-reactive protein; HOM, homocysteine; TC, total cholesterol; HDL, high-density lipoprotein cholesterol; PLS, pulse; CRU, urine creatinine; CRS, serum creatinine; CRC, creatinine clearance.