Effect of Screening Out Implausible Energy Intake Reports on Relationships between Diet and BMI
2005
Huang, Terry T.-K. | Roberts, Susan B. | Howarth, Nancy C. | McCrory, Megan A.
OBJECTIVE: We present an updated method for identifyingphysiologically implausible dietary reports by comparing reportedenergy intake (rEI) with predicted energy requirements (pER), and weexamine the impact of excluding these reports. RESEARCH METHODS AND PROCEDURES: Adult data from theContinuing Survey of Food Intakes by Individuals 1994 to 1996were used. pER was calculated from the dietary reference intakeequations. Within-subject variations and errors in rEI [coefficient ofvariation (CV) [approximately] 23%] over 2 days (d), pER (CV [approximately] 11%),and measured total energy expenditure (mTEE; doubly labeled water,CV [approximately] 8.2%) were propagated, where ±1 SD =[Formula: see text] = ±22%. Thus, a report was identified as implausible if rEI wasnot within 78% to 122% of pER. Multiple cut-offs between ±1 and ±2SD were tested. RESULTS: %rEI/pER = 81% in the total sample(n = 6499) and progressively increased to 95% inthe ±1 SD sample (n = 2685). The ±1 to 1.4 SDsamples yielded rEI-weight associations closest to the theoreticalrelationship (mTEE to weight). Weak or spurious diet-BMI associationswere present in the total sample; ±1 to 1.4 SD samples showed thestrongest set of associations and provided the maximum nwhile maintaining biological plausibility. DISCUSSION: Our methodology can be applied to different data setsto evaluate the impact of implausible rEIs on health outcomes.Implausible rEIs reduce the overall validity of a sample, and notexcluding them may lead to inappropriate conclusions about potentialdietary causes of health outcomes such as obesity.
显示更多 [+] 显示较少 [-]