Associations of social and economic and pregnancy exposures with blood pressure in UK White British and Pakistani children age 4/5

Abstract

South Asians have higher rates of coronary heart disease (CHD) than White European individuals. Blood pressure (BP) is one of the most important risk factors for CHD and ethnic differences in BP have been identified in childhood. Early life exposures could explain some of these differences. We examined associations of family social and economic and maternal pregnancy exposures and BP at age 4/5 in 1644 White British and 1824 Pakistani mother-offspring pairs from the Born in Bradford study. We found that systolic BP was similar but diastolic BP was higher, in Pakistani compared to White British children (adjusted mean differences were -0.170 mmHg 95% CI -0.884, 0.543 for systolic BP; 1.328 mmHg 95% CI 0.592, 2.064 for diastolic BP). Social and economic exposures were not associated with BP in either ethnic group. Maternal BMI was positively associated with BP in both groups but this association was mediated by child BMI. Only gestational hypertension was associated with child systolic and diastolic BP and this was only identified in Pakistani mother-offspring pairs. These findings suggest that Pakistani populations may have a different BP trajectory compared to White British groups and that this is already evident at age 4/5 years.

Figure 1

Flow chart of study sample.

Figure 2

Conceptual model for selecting confounders. Notes: (a) All arrows indicate that we believe it is plausible that the characteristic(s) at the base of the arrow influence the characteristic(s) at the head of the arrow. The associations that we focus on here are maternal characteristics that could be modified during or before pregnancy and their potential impact on offspring blood pressure in childhood (highlighted in bold and heavily weighted boxes and arrows). (b) This is not a complete graph of all plausible relations between all of the characteristics shown; it is an illustration of our conceptual model used to decide what to consider for each of the maternal early life risk factors (the first 4 left side boxes) for child BP that we considered to be confounders (i.e. influencing the risk factor and child BP). Thus, for e.g. for maternal smoking in the best confounder adjusted model we adjusted for maternal parity, age at delivery, education and family housing tenure and receipt of benefits, but not other characteristics.

Figure 3

Associations between family social and economic exposures and offspring BP at age 4/5 (Model 1). ●White British ▲Pakistani. *Model 1: Adjusted for sex; age at measurement. Values are differences in means (95% CI) of outcome per maternal exposure unit or category.

Figure 4

Associations between maternal pregnancy characteristics (BMI and smoking) and offspring BP at age 4/5 (Models 1 and 2). ●White British ▲Pakistani Filled shapes on graph (○): model 1; Clear dots on graph (●): model 2. *Model 1: Adjusted for sex; age at measurement; *Model 2: Adjusted for sex; age at measurement; maternal age; parity; maternal education; family housing tenure; family receipt of benefits. Maternal BMI model 2 additionally adjusted for smoking in pregnancy. Values are differences in means (95% CI) of outcome per maternal exposure unit or category, maternal BMI is difference in means per 5 kg/m².

Figure 5

Associations between maternal pregnancy characteristics (glycaemia and HDP) and offspring BMI at age 4/5 (Models 1 and 2). ●White British ▲Pakistani. Filled shapes on graph (○): model 1; Clear dots on graph (●): model 2. *Model 1: Adjusted for sex; age at measurement; *Model 2: Adjusted for sex; age at measurement; maternal age; parity; smoking in pregnancy; maternal education; family housing tenure; family receipt of benefits. Values are differences in means (95% CI) of outcome per maternal exposure unit or category.