The questionnaires share the following characteristics. They are:
• Designed to address covariates (confounding) exposures, not main variables or the principal subject of a study, which might require more detailed questions
• Developed primarily for use in cohort studies, rather than case-control studies, which often require more questions about lifetime and historic, rather than current, exposures
• Predicated on English-speaking literate adult respondents, not special populations
• Designed to be self-explanatory, assuming that the questionnaire will be self-administered, not by interview.
In questionnaire development, there is often a choice between open-ended and categorical questions. We included open-ended questions only when responses were likely to be both easily remembered by the participant and straightforward to analyze.