Cognitive Testing and Readability of an Item Bank for Measuring the Impact of Headache on Health-Related Quality of Life

• Published in: The patient : patient-centered outcomes research 2012, Vol.5 (2), p.89-99
• Main Authors: Turner-Bowker, Diane M., Saris-Baglama, Renee N., DeRosa, Michael A., Paulsen, Christine A.
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; Comprehension; Female; Headache - psychology; Health Administration; Health Economics; Health Status Indicators; Humans; Interviews as Topic; Male; Medicine; Medicine & Public Health; Middle Aged; Original Research Article; Patient Preference; Pharmacoeconomics and Health Outcomes; Psychological Tests; Public Health; Quality of Life; Quality of Life Research; Reproducibility of Results; Self Report; Socioeconomic Factors; Treatment Outcome; Young Adult
• ISSN: 1178-1653; 1178-1661
• DOI: 10.2165/11592170-000000000-00000

Background: Patient-reported outcomes (PRO) measures should be valid and accessible to a wide audience. Objective: Cognitive item testing and readability studies were conducted to evaluate how adult headache sufferers (N = 9) understood and responded to the Headache Impact Test (HIT™) item bank, a PRO measure for headache that serves as the source of item content for the HIT-6™ (a widely used six-item short-form measure of headache impact with more than 30 language translations), and the Dynamic Health Assessment Headache Impact Test (DYNHA® HIT™) [a computerized adaptive test (CAT) of headache impact]. Methods: During cognitive interviews, participants were asked to ‘think aloud’ as they read survey instructions, completed items, and formulated responses. Data analyses evaluated item comprehension, memory recall of relevant information, and decision and response processes; compared various item attributes; and tested shortened item versions. Results: Survey readability was at the seventh-grade level. Respondents understood most revised items as intended, and found shorter items comparable to longer items with some exceptions. When recall period was included in instructions but not within the items themselves, respondents often expanded the recall period to answer the item. Some response scales (e.g. “Never” to “Always”) were more readily understood than others (e.g. “Definitely true” to “Definitely false”). Conclusion: Qualitative research can improve the validity and accessibility of PRO measures that are used to monitor health conditions and aid patient-provider communication.