Inaccurate self-report of height and its impact on misclassification of body mass index in postmenopausal women.

OBJECTIVE: Self-reported height is commonly used in population obesity research. Evidence has also shown a positive association between depression and obesity. We examined the extent of height misreporting and its impact on body mass index (BMI) calculations and classification, and explored whether depression is associated with height misreporting. METHODS: The Buffalo Osteoporosis and Periodontal Disease Follow-up Study enrolled 1,015 postmenopausal women between 2002 and 2006. Participants self-reported their height on a questionnaire before stadiometer measurement at the clinical visit. Depressive symptoms were assessed using the Center for Epidemiologic Studies Depression Scale. Odds ratios and 95% CI for association between depression and height misreporting were estimated using logistic regression. RESULTS: Overall, 446 women (43.9%) misreported height by greater than 1/2 inch, of which 296 (29.2%) underestimated and 150 (14.8%) overestimated their height. Height misreporting influenced BMI calculations by ≥1 unit in 12% of women, and influenced classification into WHO BMI categories in 8% of women. After adjusting for age, race, education, and measured BMI, women with significant depressive symptoms were more likely to misreport their height (odds ratio = 1.65, 95% CI, 1.04-2.61). CONCLUSIONS: Height misreporting was common in older women and significantly influenced BMI calculations and classification. Obtaining objective data is thus important for studies investigating obesity-disease associations in this population, especially in those with significant depressive symptoms.

Reproducibility of probing depth measurement using a constant-force electronic probe: analysis of inter- and intraexaminer variability.

BACKGROUND: Probing depth (PD) is a commonly used method to determine periodontal disease severity in both treating and evaluating disease progression. Agreement among examiners collecting data in scientific investigations is necessary to establish reliable criteria for determining levels of periodontal attachment loss. The objective of our study was to evaluate inter- and intraexaminer variability of PD measurements among study examiners using a constant force periodontal probe, and to compare the variability of tooth-mean and quadrant-mean. METHODS: Three examiners, who had been previously trained and calibrated, performed measurements on 20 volunteers. Intra- and interexaminer variability of sites was determined by means of standard error of measurement (SE). Data analysis included determination of error for both quadrant mean and tooth mean. RESULTS: PD measurements for the quadrant-mean were used to calculate the intraexaminer variability, resulting in a mean (SD) value for an SE of 0.40 mm (+/- 0.02). Interexaminer variability for quadrant mean was 0.16 mm (+/- 0.02). For tooth-mean SE, the intraexaminer variability values were equal to 0.38 mm (+/- 0.07), and interexaminer variability equal to 0.24 mm (+/- 0.05). CONCLUSIONS: All three examiners participating in our study were able to obtain reliable measurements for PD, using the constant force electronic probe. Reproducibility did not vary appreciably when using the whole quadrant mean compared to the tooth mean. These trained examiners were able to provide reproducible measures under 0.5 mm.