Skip to main content
ARS Home » Research » Publications at this Location » Publication #172214
Title: DIETARY RESTRAINT AND DISINHIBITION DO NOT AFFECT THE ACCURACY OF 24-HOUR RECALL IN A MULTIETHNIC POPULATION

Author
item ARD, JAMY - U OF ALABAMA, BIRMINGHAM
item DESMOND, RENEE - U OF ALABAMA, BIRMINGHAM
item ALLISON, DAVID - U OF ALABAMA, BIRMINGHAM
item CONWAY, JOAN

Submitted to: Journal Of The American Dietetic Association
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/29/2005
Publication Date: 2/24/2006
Citation: Ard, J.D., Desmond, R.A., Allison, D.B., Conway, J.M. 2006. Dietary restraint and disinhibition do not affect the accuracy of 24-hour recall in a multiethnic population. Journal Of The American Dietetic Association. 106:434-437.

Interpretive Summary: Assessment of dietary intake is an important task for nutritionists, health care professionals and government agencies; however problems exist with under- and over-reporting of food intake. Specifically some psychological predictors of eating behaviors have been shown to affect usefulness of methods for dietary assessment. This study was conducted to determine the association of dietary restraint (the tendency to limit food intake to control body weight) and disinhibition (the tendency to overindulge when food is readily available, e.g., at parties or buffet lines) with recall accuracy for food intake. Data were obtained from 79 male and 71 female volunteers, who selected and consumed all foods for a 1-day period under observation. Each participant completed a telephone 24-hour recall using the USDA 5-Step Multiple-Pass method to elicit responses. Actual and recalled energy (E), protein (PRO), carbohydrate (CHO), and fat (FAT) intakes were determined by direct observation and a 24-h dietary recall respectively. The Three-Factor Eating Inventory, which measures dietary restraint and disinhibition, was administered prior to eating any food in the study. Mean age and BMI was 43 +/- 12 yrs and 29 +/- 5.5 kg/m2, respectively. African Americans represented 48% of the sample. Men over-reported E intake by 265 kcal and women by 250 kcal; both groups over-reported intake of PRO, CHO, and FAT. When controlling for BMI, sex and race, restraint was a significant independent predictor of E and FAT intake, i.e., people with high restraint scores had lower E and FAT intakes. Dietary restraint did not affect the accuracy of E, FAT, CHO or PRO dietary recall. Disinhibition was not related to intake or accuracy in similar models. Dietary restraint was significantly associated with intake of E and FAT; however, dietary restraint does not appear to be associated with dietary recall accuracy when using the USDA 5-Step Multiple Pass 24-hour recall method. Nutritionists and health care professionals should consider dietary restraint a possible reason for a lower than expected estimate of energy intake when using 24-hour recalls.

Technical Abstract: Some psychological predictors of eating behaviors have been shown to affect usefulness of methods for dietary assessment. Therefore this study was conducted to determine the association of dietary restraint and disinhibition with recall accuracy for total energy (E), fat (FAT), carbohydrate (CHO), and protein (PRO). Data were obtained from 79 male and 71 female European and African American volunteers. Participants selected and consumed all foods for a 1-day period under observation. The following day, each participant completed a telephone 24-hour recall using the USDA 5-Step Multiple-Pass method to elicit responses. Actual and recalled energy (E), protein (PRO), carbohydrate (CHO), and fat (FAT) intakes were determined by direct observation and a 24-h dietary recall respectively. The Three-Factor Eating Inventory, which measures dietary restraint and disinhibition, was administered prior to eating any food in the study. Repeated measures analyses of variance were used to determine if dietary restraint or disinhibition were independent predictors of recall accuracy. The mean age and BMI of the 150 participants was 43 +/- 12 yrs and 29 +/- 5.5 kg/m2, respectively. African Americans represented 48% of the sample, and approximately half of the sample received a college education. On average, men overreported E by 265 kcal and women by 250 kcal; both groups also overreported intake of PRO, CHO, and FAT. When controlling for BMI, sex and race, restraint was a significant independent predictor of E intake (p= 0.004) and negatively correlated with E intake (r= -0.23, p less than 0.001). Unlike intake of CHO or PRO, FAT intake was significantly and negatively associated with dietary restraint (r = -0.3; p less than 0.001). A non-significant interaction (restraint*(actual-recalled)) revealed that dietary restraint did not affect accuracy of E, FAT, CHO or PRO dietary recall. Disinhibition was not related to intake or accuracy in similar models. Dietary restraint was significantly associated with intake of E and FAT; however, dietary restraint does not appear to be associated with dietary recall accuracy when using the USDA 5-Step Multiple Pass 24 h recall method. Dietitians should consider dietary restraint a possible reason for a lower than expected estimate of energy intake when using 24h recalls.