Linking Differences in Eating Behavior with Appetitive Hormones in Women

Abstract

Abstract Objectives Examine and compare the hormones related to satiety in women identified as 1) restrained + disinhibited eaters, 2) restrained, non-disinhibited eaters, or 3) unrestrained, non-disinhibited eaters. We hypothesized restrained eaters with disinhibition would have a different pattern of appetitive hormones involved in food intake regulation. Methods 178 women from a cross-sectional study completed the three factor eating questionnaire (TFEQ), and were classified as restrained + disinhibited eaters (RD, n = 38), restrained, non-disinhibited eaters (R, n = 52), or unrestrained, non-disinhibited eaters (LL, n = 88). Women fasted for 12 h prior to baseline blood collection, and postprandial blood was collected after a meal challenge at 30 min, 3 h, and 6 h. Data were evaluated for outliers and missingness, and transformed and imputed using multivariate normal imputation. Differences in glucose, insulin and ghrelin were analyzed using a linear mixed model comparing groups. Time to peak and nadir for glucose, insulin and ghrelin were calculated from response curves. Chi square tests of proportions were used to determine whether the distribution of participants in the three groups were different in their times to peak and nadir. Results There were no significant differences between groups in ghrelin or insulin concentrations over time. However, overall glucose levels in RD were significantly lower than R and LL groups (P = 0.05 and P = 0.03, respectively). Chi-squared tests indicated that a greater proportion of women in the RD group reached peak glucose at 3 h, compared to women in R and LL with peak glucose occurring at 30 min (P = 0.04). Glucose nadir displayed a bimodal distribution in R, with more women at baseline and 3 h compared to RD and LL (P = 0.05). Ghrelin nadir was reached at 3 h for more participants in RD and LL, but occurred later at 6 h in more participants in R (P = 0.04). Conclusions Dietary restraint and disinhibition are associated with circulating glucose and ghrelin concentrations; however, prospective research is needed to determine if there is a causal relationship. Funding Sources USDA CRIS 2032-51530-022, 2032-51530-025, and 2032-51530-026-00-D, and a grant from Arla Foods Inc.