Does using a short dietary questionnaire instead of a food frequency improve response rates to a health assessment survey?