| Self-reported food intake may thwart
research
By Amy Norton
NEW YORK (Reuters Health) - A new study of overweight, diabetic
women casts doubt on the reliability of the self-reported dietary
habits often used in medical research.
Researchers found that most of the 200 women they studied, as many
as 81 percent, reported eating fewer calories than they actually
had, based on objective measures.
Many also claimed to have eating habits that closely matched the
recommended diet for diabetics, which suggests the women were really
reporting what they thought they should be eating, according to
the researchers.
The problem with fibbing about or underestimating calories is that
it makes it hard to measure the true effectiveness of dietary interventions--a
key component of diabetes treatment. The new findings point to the
importance of using some objective measure to back up research participants'
dietary claims, the study authors report in the March issue of the
journal Diabetes Care.
Their study included middle-aged and older African-American women
with type 2 diabetes, most of whom were overweight or obese. Past
research has shown calorie underreporting to be common among women,
people who are overweight, and those who want to lose weight, lead
study author Dr. Carmen D. Samuel-Hodge told Reuters Health.
She said people may, for instance, have a hard time remembering
what or how much they ate, or may feel pressured to report eating
habits that are "socially acceptable."
For the current study, Samuel-Hodge and her colleagues at the University
of North Carolina at Chapel Hill compared diabetic women's reported
food intake with objective estimates of their calorie expenditure.
To get these estimates, they had the women wear small electronic
devices called accelerometers, which gauge the number of calories
burned during physical activity, for one week. The researchers also
measured the women's base metabolic rates. These measures were compared
with participants' self-reported dietary intake on three days.
The idea is that in the absence of weight loss or gain, a person's
calorie intake should roughly equal calorie expenditure. If someone
takes in fewer calories than she burns, she should lose weight.
Samuel-Hodge and her colleagues found that most of the women in
their study reported calorie intakes that were lower than their
estimated calorie expenditure.
Based on the accelerometer data, the researchers estimate that
81 percent underreported their calorie intake. This estimate dipped,
but remained high at 58 percent, when the researchers compared calorie
intake with base metabolic rates.
The researchers also found that the heavier a woman was, the more
likely she was to underreport calories.
It's possible, the investigators acknowledge, that many of these
women, who were part of a larger study on managing diabetes with
diet and exercise, truly were cutting calories. But, they note,
six months after the current results were compiled, the women were
showing no significant weight loss.
The "major implication," the researchers conclude, is
that such self-reports need to be independently validated.
Samuel-Hodge said the findings are particularly relevant to studies
of people with type 2 diabetes because of their high prevalence
of obesity.
SOURCE: Diabetes Care, March 2004.
Previous Diabetes
News 
|
