Agreement between interview data and a self-administered questionnaire on dietary supplement use

Area of Science:

• Epidemiology
• Nutritional Science
• Public Health

Background:

• Dietary supplement consumption is widespread.
• Accurate assessment of supplement use is crucial for epidemiological studies.
• Self-administered questionnaires are common tools for data collection in large cohorts.

Purpose of the Study:

• To assess the relative validity of an open-ended question on dietary supplement consumption.
• To evaluate the accuracy of a self-administered questionnaire in the Netherlands Cohort Study (NLCS).
• To determine the reliability of recall for supplement use over a five-year period.

Main Methods:

• Compared questionnaire data with reference information from three personal interviews.
• Utilized a randomly selected subgroup of 109 participants (59 men, 50 women).
• Participants were from 12 municipalities in The Netherlands.

Main Results:

• Overall sensitivity for any dietary supplement use was 65.9% with 98.5% specificity.
• Kappa measure of agreement was estimated at 0.69, indicating substantial agreement.
• High recall percentages were observed for women, long-term users, and older individuals.
• Recall for specific supplements varied, with garlic preparations at 77.8% and 'other' supplements at 11.8%.

Conclusions:

• The questionnaire demonstrates good overall validity for assessing dietary supplement use.
• Estimates for specific supplements like garlic and vitamin preparations may be sufficiently precise for classifying users.
• The findings support the use of this questionnaire in large-scale dietary studies.
• Further research may be needed to improve recall for less common supplement types.