Abstract
Knowledge of the regression reiaticm between dietary intake reported on a food frequency questionnaire and true average intake is useful in interpreting resutts from nutritional epidemiologic studies and in planning such studies. Studies which validate a questionmire against a food record may be used to estimate this regression relation provided the food record is compkted by each subject on at least two occasions. Using data collected from women aged 45-69 years during 1985-1 986 in the pilot study of the Women's Health Trial, the authors show how variation in diet over time and intralndividual correlation between a questionnaire and food r m d obtained dose ogether in time affects the estimatim of the regression. The authors' method provides estimates of the regression slope and the questionnaire 'bias' that are corrected for hese effects, together with standard m. A computer program in the SAS language, for carrying out the analysis, is provided. Am J Epidemiol 1991 ;134;310-20.
| Original language | English |
|---|---|
| Pages (from-to) | 310-320 |
| Number of pages | 11 |
| Journal | American Journal of Epidemiology |
| Volume | 134 |
| Issue number | 3 |
| DOIs | |
| State | Published - 1 Aug 1991 |
| Externally published | Yes |
Bibliographical note
Funding Information:Dr. Carroll's research was supported by NHH grant no. GM-39015 and partaly completed during a visit to Purdue University's Center for Decision Sciences, which is supported by NSF grant nos. DMS-8702620 and DMS-8717799.
Funding
Dr. Carroll's research was supported by NHH grant no. GM-39015 and partaly completed during a visit to Purdue University's Center for Decision Sciences, which is supported by NSF grant nos. DMS-8702620 and DMS-8717799.
| Funders | Funder number |
|---|---|
| NHH | |
| National Science Foundation | DMS-8717799, DMS-8702620 |
| National Institute of General Medical Sciences | R01GM039015 |
Keywords
- Measurement error
- Questionnaores
- Regression analysis