Skip to main content
ARS Home » Northeast Area » Boston, Massachusetts » Jean Mayer Human Nutrition Research Center On Aging » Research » Publications at this Location » Publication #211862
Title: Don’t Always Believe What You Are Told: A Comparison of Self-Reported with Measured Weight and Height

Author
item NUITE, MEYLYNN - TUFTS-NEMC
item WANG, JUN - TUFTS-NEMC
item WHEELER, LAURA - TUFTS-NEMC
item FLETCHER, JEREMIAH - TUFTS-NEMC
item BADIANI, PRUTHA - TUFTS-NEMC
item MCADAMS, ERICA - TUFTS-NEMC
item JOAS, JOHANE - TUFTS-NEMC
item LAVALLEY, MICHAEL - BOSTON UNI SCH PUB HEALTH
item Dawson-Hughes, Bess
item MCALINDON, TIMOTHY - TUFTS-NEMC

Submitted to: American College of Rheumatology
Publication Type: Abstract Only
Publication Acceptance Date: 7/1/2007
Publication Date: 11/6/2007
Citation: Nuite, M., Wang, J., Wheeler, L.M., Fletcher, J., Badiani, P., Mcadams, E.L., Joas, J., Lavalley, M.P., Dawson-Hughes, B., Mcalindon, T.E. 2007. Don’t Always Believe What You Are Told: A Comparison of Self-Reported with Measured Weight and Height.In: American College of Rheumatology Scientific Meeting. 11/6/07-11/11/07. Bosotn, MA. 56(9 Supplement): S779.

Interpretive Summary:

Technical Abstract: Self-reported weight and height are often used for eligibility screening in osteoarthritis (OA) clinical trials. However, prior studies suggest that individuals give biased reports of their weight and height. Our purpose was to evaluate the accuracy of self-reported weight and height and calculated body mass index (BMI) for trial eligibility. Applicants (n=100, mean age 63.5 (+/-8.7) years) completed a computer-assisted telephone interview which solicited self-reported weight and height. Weight and height were later measured utilizing a Seca 770 Alpha Scale and Holtain Stadiometer, respectively. Student’s t-test was used to compare self-reported and measured weight and height. BMIs were calculated from self-report (srBMI) and measured (mBMI) weight and height. Multiple regression was used to examine the factors associated with the difference between self-reported and measured data. Age, gender, race, education, mBMI, and days elapsed (from self-report to measured) were included as covariates in all regression models. Mean time between self-report and measurement was 72 (+/-30) days. On average, women (n=65) under-reported their weight by 2 kg (2.5% of weight) (p<0.0001) and men (n=35) by 1 kg (0.9% of weight) (p=0.08). Height was over-reported by both women (1.9 cm, 1.2%) and men (3.3 cm, 1.9%) (p<0.0001). The mean srBMI and mBMI for women was 29.0 (+/-5.5) and 30.5 (+/-5.7), respectively. On average, women 'under-reported' their BMI by 1.5 kg/m2 (4.7%). Similarly, the mean srBMI and mBMI for men was 28.9 (+/-4.1) and 30.2 (+/-4.3), respectively. On average, men 'under-reported' their BMI by 1.4 kg/m2 (4.5%). srBMI was significantly lower than mBMI for both women and men (p<0.0001). In multivariate analyses, under-report of weight was positively associated with education (0.82 kg per increment in education level, p=0.004) and mBMI (0.15 kg per 1 kg/m2 increment in mBMI, p=0.02), and appeared to be negatively associated with age (0.068 kg per yr decrement in age, p=0.09). Men over-reported their height more than women by 1.7 cm (p=0.004). Women and men with less education were more likely to over-report their height (0.5 cm per decrement in education level, p=0.02). After adjusting for age, race, gender, days elapsed (from self-report to measured), and education, under-report of BMI was positively associated with mBMI (0.075 kg/m2 under-report per 1 kg/m2 increase in mBMI, p=0.01). Therefore, there is a bias toward under-reporting of weight and over-reporting of height among individuals screening for OA trials. This bias is related to gender, age, education, and BMI. Although the magnitude of this bias is modest, researchers should be cautious in accepting individuals with reported measures close to the screening cut points.