Development and validation of the Migraine Screen Questionnaire (MS-Q)

Headache. Nov-Dec 2005;45(10):1328-38. doi: 10.1111/j.1526-4610.2005.00265.x.

Abstract

Aim: To develop and evaluate the clinimetric properties of a new migraine screening questionnaire: the Migraine Screen Questionnaire (MS-Q).

Background: Migraine is a public health problem requiring screening programs and tools to ensure early detection.

Methods: A questionnaire was developed based on the criteria of the International Headache Society (IHS) and a review of the literature by a committee of experts. Stage I: The original version of the MS-Q was distributed by mail and completed by Pfizer employees and self-administered to neurological patients; all subjects were afterward evaluated by a neurologist who was blinded to the MS-Q results, to establish an independent IHS diagnosis. Stage II: A final version of the MS-Q was administered to neurological patients to confirm clinimetric properties. Logistic regression and receiver-operator characteristic curve statistical methods were used and the 95% confidence interval, sensitivity, specificity, and positive (PPV) and negative (NPV) predictive values, were estimated.

Results: Of the 605 subjects recruited, 465 were evaluable (325 in stage I and 140 in stage II). Of the original 15 items, 5 conformed the final version of the MS-Q: frequency and intensity of headache; a duration of between 4 hours and 3 days; nausea; sensitivity to light/noise; and disability. A cutoff point of > or = 4 points showed a sensitivity of 0.93 (95% CI = 0.87 to 0.99), specificity of 0.81 (95% CI = 0.72 to 0.91), PPV of 0.83 (95% CI = 0.75 to 0.91), and NPV of 0.92 (95% CI = 0.85 to 0.99). Cronbach's alpha coefficient = 0.82.

Conclusions: The MS-Q showed adequate validity and reliability, and it could be a good screening tool for application to clinical practice and research.

Publication types

  • Clinical Trial

MeSH terms

  • Adult
  • Female
  • Humans
  • Logistic Models
  • Male
  • Migraine Disorders / diagnosis*
  • Reproducibility of Results
  • Surveys and Questionnaires