Factors associated with errors in self-reported height and weight in older adults.

OBJECTIVES: Describe the distribution and direction of self-reported versus measured height and weight using variables associated with aging such as cognition, health status, age, and bone mineral density (BMD), and examine the effect of these measurement differences on body mass index (BMI) classification.
DESIGN: Data was derived from the third National Health and Nutrition Examination Survey (NHANESIII) conducted from 1988-1994, a nationwide probability sample. PARTICIPANTS: 4,590 non-institutionalized older adults aged 60 and older. MEASUREMENTS: Self-reported and measured height and weight, demographic and lifestyle characteristics, BMD, and subscales from the Mini Mental State Exam were used. Values were considered correct if self-reported height was within one inch of measured height, self-reported weight was within 5 lbs of measured weight, and self-reported BMI was within the same classification as measured BMI.
RESULTS: Over-reported height increased with age in both men and women, occurring in 70% of those aged 80 and older. Compared to people with normal BMD, a significantly higher proportion of osteoporotic men (76% versus 47%, P<0.001) and women (52% versus 35%, P<0.001) over-reported their height. Additionally, significant misclassifications of self-reported height and weight occurred among people in poor health and those with poor performances on memory and calculation tests. Nevertheless, there was agreement in BMI classification among almost 80% of the population and among 90% of individuals in the healthy BMI category.
CONCLUSION: This study suggests that among an older population, self-reported height and weight may be strongly related to age-associated changes in health status, cognition and BMD.