Zografos LJ, Andrews E, Whalley D, Petraro P, Vassilev Z. Use of cognitive testing to optimize questionnaire wording and mode of administration in the evaluation of risk minimization activities. Poster presented at the 32nd International Conference on Pharmacoepidemiology & Therapeutic Risk Management (ICPE); August 27, 2016. Dublin, Ireland. [abstract] Pharmacoepidemiol Drug Saf. 2016 Aug; 25(Suppl 3):319-20.


BACKGROUND: As part of the evaluation of risk minimization measures for aflibercept, for intravitreal injection, a questionnaire was developed to assess patient knowledge and understanding of key safety information contained in aflibercept's EU educational materials. Interviews were conducted to test the questionnaire with patients prior to the start of data collection.

OBJECTIVES: To ensure that patients understood and consistently interpreted the questions and response options and to determine the most appropriate mode of data collection given the potential for visual impairment in the target population.

METHODS: Two rounds of interviews were first conducted in English (in the UK) with 11 patients to identify issues and optimize wording. Interviewers trained in cognitive debriefing methods asked patients to complete the questionnaire while describing their thought processes aloud. Additional probe questions elicited more information on how patients interpreted and chose their answers and the format and usability of the questionnaire. The questionnaire was revised after each round of interviews and translated into 4 additional languages (for France, Germany, Italy, and Spain). Cognitive interviews were then conducted with 4 patients per country to confirm wording and cultural acceptability.

Results: Across all countries, 56% of patients were aged 50 to 75 years, and 37% were 75 years or older. Early results indicated that some patients likely could have difficulty completing the questionnaire without support due to visual impairment and cognitive difficulties. Based on these findings, the questionnaire was shortened, the language simplified, and the format changed to be interviewer administered. Subsequent interviews in the UK and other countries supported the length and wording of the revised questionnaire, as well as the mode of administration.

CONCLUSIONS: Appropriate questionnaire design is essential to optimize data quality. Careful pretesting is critical to ensure appropriate wording and administration format, particularly when there is potential for visual and/or cognitive impairment within the target population.

Share on: