Bias in self-reported height and weight in a cohort of young adults – perspectives for cardio-metabolic risk prevention programmes

Background

Precise assessment of obesity is essential in evaluating and applying preventive strategies to avoid adverse cardio-metabolic risk profiles later in life. Self-reported measures of height and weight is often used in cohort and cross-sectoral studies due to practical and financial reasons. However, bias due to self-reporting is well known with risk of overestimation of height and underestimation of weight. The latter is particular reported among women and persons with obesity. Little is known about the influence of educational factors in the self-reporting of height and weight. To gain insight into the extent to which different factors bias self-reporting, is important to assess the validity of self-reported body mass index (BMI) within groups of people with certain characteristics.

Purpose

The aim was to investigate the validity of self-reported height and weight versus objective measured height and weight. Additionally, the study aimed to examine if sex, measured BMI, and educational factors biased the self-reporting differenially.

Methods

The study included 264 participants, all aged 28 years who were a sub-sample from an ongoing cohort in Central Denmark Region. The participants attended a physical health examination, including measures of height and weight after having self-reported these not knowing they were to be objectively measured later. Data of their own and parental educational levels were drawn from registries. Paired t-tests were used to assess the descriptive differences between self-reported and measured values and linear regression was used to assess the independent effects of sex, education, and BMI on the differences. In model 1 adjustment for all other variables except for BMI was conducted and full adjustments were made in model 2.

Results

Overall, participants overestimated their height by 0.89 (0.67; 1.10) cm and underestimated their weight by 2.47 (1.78; 3.17) kg (Table 1). Female sex was found to independently bias underreporting of weight (β −1.51 (−2.86; −0.15)) and male sex overestimated height (β 0.83 (0.38; 1.27) (Table 2). Additionally, being overweight or obese also independently biased the underestimation of self-reported weight ((β 2.23 (0.64; 3.83) and β 5.22 (3.51; 6.94)) (Table 2). Additionally, participants with overweight were less likely to overestimate their height compared to those with normal weight. Without adjustments for BMI, low level of education also biased underestimation of weight significantly (Table 2).

Conclusion

Women and persons with obesity in particular underestimated weight. Participants with lower level of education also underestimated weight but this was explained by obesity. Men overestimated their height but participants with obesity were less likely to overestimate height compared to normal weighty. When screening for overweight and obesity to initiate interventions to prevent cardio-metabolic diseases, the lack of accuracy in self-reported values must be taken into account.

Funding Acknowledgement

Type of funding sources: None.

Physical activity during adolescence and early adulthood and adult cardiometabolic risk profile: a longitudinal study

Introduction

Physical activity (PA) is important in the prevention of cardiometabolic diseases. However, little is known of the effect of maintaining PA during adolescence and cardiometabolic risk in early adulthood.

Purpose

This 14-year longitudinal prospective cohort study investigated the association of maintaining PA since adolescence and subsequent cardiometabolic risk factors in early adulthood and to what extend adiposity affects this.

Methods

Data were drawn from an ongoing cohort and consisted of 262 participants with longitudinal information on PA and cardiometabolic risk factors in adulthood. The cardiometabolic risk factors were collected by blood samples, anthropometric measures, and blood pressure in a health examination at the age of 28 years (table 1). Subsequently, the risk factors were clustered in four domains (“Inflammation”, “Lipids”, “Glucose metabolism” and “Blood pressure”) and calculated into z-scores. Further, a composite z-score of all domains was calculated. PA was obtained by questionnaires at age 15, 18, 21 and 28. The participants were categorized into “inactive maintainers”, “fluctuaters” and “active maintainers” and linear regression was used to examine the association between PA groups and the cardiometabolic domains in a crude and two adjusted models 1 and 2.

Results

Increasing z-scores from “active maintainers” to “fluctuaters” to “inactive maintainers” were seen in all domains except in “blood pressure” when adjusted for sex, smoking, parental and own education, and parental cardiovascular medical history, and significantly different when comparing “active maintainers” to “inactive maintainers” (table 2, model 1). In the composite z-score outcome including all domains, a significantly adverse z-score was found for “inactive maintainers” z=0.63, p=0.001 when adjusted for the variables in model 1 (table 2). When also adjusted for body fat percentage (model 2), the z-score became insignificant z=0.19, p=0.198 (table 2).

Conclusion

Maintaining PA from adolescence to early adulthood is associated with a significant lower cardiometabolic risk profile compared to inactive maintainers. However, some of the adverse risk profile is explained by being overweight or obese. Given the increasing prevalence of CVD and diabetes these results highlight the importance of maintaining PA at all stages of life.

Funding Acknowledgement

Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): Karen Elise Jensen Foundation and The Steno Partner Collaboration between Goedstrup Hospital and Steno Diabetes Center Aarhus.

Impact of socioeconomic position on coronary artery disease burden in men and women with de-novo symptoms suggestive of chronic coronary syndrome

Background

Low socioeconomic position (SEP) is associated with shorter life expectancy and one of the main drivers is an increase in cardiovascular deaths. A higher prevalence of risk factors only partly explains the complex multifactorial pathogenesis. The aim of this study was to investigate the association between SEP and the development of coronary artery disease (CAD) assessed as calcium score (CACS) at coronary computed tomography angiography (CCTA) as well as stenosis at downstream invasive coronary angiography (ICA) in a population presenting with symptoms suggestive of chronic coronary syndrome (CCS). A secondary aim was to establish whether SEP affects men and women differently.

Methods

We included 50,561 patients (Mean age 57.35±11.50, 53.7% women) from the Western Denmark Heart Registry (WDHR) with no previous CAD undergoing CCTA from 2008–2019 for suspected CCS. ICA was conducted in patients where obstructive CAD was not excluded at CCTA. Outcome measures was level of CACS and haemodynamically significant stenosis at ICA defined as either fractional flow reserve <0.80 or visually assed diameter stenosis of ≥50% stenosis. Odds Ratio of haemodynamically significant stenosis at ICA was calculated using multiple logistic regression and models adjusted for risk factors (smoking, medical treatment for hypertension, medical treatment for high cholesterol, diabetes and family history of CAD). Information on SEP was obtained from national registries. We included mean individual income at age 30–60 or until CCTA (quintiles); and length of education (<10 years, 10–13 years or >13 years). Information on risk factors was obtained from the WDHR registry.

Results

Mean number of risk factors are presented in each educational group in Figure 1. Median CACS for women with <10 years of education is 2 [0–82] vs. 0 [0–15] for women with >13 years of education (p<0.001). For men <10 years of education median CACS is 10 [0–143] vs. 8 [0–118] for men with >13 years of education (p=0.05) (Figure 1). Mean number of risk factors are presented at each level of income in Figure 2. For women with low income median CACS was 6 [0–103] vs. 0 [0–3] for women with high income (p<0.001). For men with low income median CACS is 8 [0–144] vs. 5 [0–105] for men with high income (p=0.002) (Figure 2). The odds ratio (OR) of a stenosis at downstream ICA was 1.47 (p=0.004) for women with <10 years of education vs. >13 years of education and 1.17 for men (p=0.122). OR of stenosis at ICA was 2.40 (p<0.001) for women with low income (1 quintile) using high income (5 quintile) as reference and 1.12 for men (p=0.321).

Conclusion

In de-Novo patients referred for CAD rule-out both coronary calcium score and the prevalence of stenosis at ICA is strongly correlated to low income and short education. The correlation seems to be stronger in women compared to men and calls for further research into the mechanism behind low SEP and atherosclerosis.

Funding Acknowledgement

Type of funding sources: Public hospital(s). Main funding source(s): Hospital Unit West Jutland Research foundation Figure 2

The experience of demanding work environments in younger workers

BACKGROUND

Investigating whether certain individual or background characteristics are associated with an increased risk of experiencing an excessively demanding work environment in younger workers may help to reduce future inequality in health and maximize their labour market participation.

AIMS

To describe the work environment of Danish 20- to 21-year olds and to investigate the influence of family socioeconomic background and individual characteristics at age 14-15 on later experience of physical and psychosocial work environments.

METHODS

We obtained information on subjects' school performance, vulnerability, health and parental socioeconomic status from registers and a questionnaire completed in 2004. A questionnaire concerning eight measures of subjects' psychosocial and physical work environment in 2010 was used to determine the outcomes of interest.

RESULTS

The study population consisted of 679 younger workers aged 20-21. The psychosocial work environment was in general good but younger workers experienced more demanding physical work than the general working population. Overall, individual as well as family factors had a limited impact on their assessment of the work environment. Low self-esteem at age 14-15 was associated with experiencing high demands and lack of trust and fairness at work, whereas low parental socioeconomic status was associated with a demanding physical work environment.

CONCLUSIONS

This study showed a social gradient in experiencing a demanding physical work environment at age 20-21. The psychosocial work environment experienced by younger workers was generally good, but vulnerable young people may need special attention to protect them from or prepare them for psychosocially demanding jobs later in life.

Low PAPP-A in the first trimester is associated with reduced fetal growth rate prior to gestational week 20

OBJECTIVE

To evaluate the association between maternal pregnancy-associated plasma protein-A (PAPP-A) and fetal growth from the first to the second trimester.

METHODS

A prospective cohort study including 8347 pregnant women attending prenatal care at Aarhus University Hospital were conducted. PAPP-A was measured during 8 to 14 gestational weeks. Fetal growth between the two scans in the first and second trimesters was estimated by (GA(20)- GA(12))/Days(calendar), where GA(12) reflects gestational age in days calculated from crown-rump length at a 12-week scan, GA(20) reflects gestational age in days calculated from biparietal diameter at a 20-week scan, and Days(calendar) reflects the number of calendar days between the two scans.

RESULTS

Fetal growth rate from the first to the second trimester was correlated with PAPP-A, with a regression coefficient of 0.009 (95% CI, 0.007-0.012, P < 0.001). PAPP-A below 0.30 MoM was associated with a fetal growth rate below the tenth centile, with an adjusted OR of 2.05 (95% CI, 1.24-3.38).

CONCLUSION

Low levels of PAPP-A are associated not only with low birth weight at term but also with slower fetal growth prior to 20 weeks of gestation.