Global assessment of migraine severity measure: preliminary evidence of construct validity

DOI: https://doi.org/10.1186/s12883- 019- 1284- 8

Abstract-Summary This study investigated how migraine patients rate the severity of their disease and how these ratings correlate with their socio-demographic, clinical, and psycho- social characteristics.

We obtained a broad range of clinical and patient-reported measures (e.g., patients’ ratings of migraine severity using the Global Assessment of Migraine Severity (GAMS), and migraine-related disability, as measured by the Migraine Disability Scale (MIDAS)).

Median regression analysis was used to examine the predictors of patient ratings

of migraine severity.

Patients’ report of severity on the GAMS was strongly correlated with patients’ ratings of MIDAS global severity question, overall MIDAS score, migraine type, PHQ-9 score, and frequency of migraine attacks.

Mediation analyses revealed that MIDAS mediated the effect of depression on patient ratings of migraine severity, accounting for about 32% of the total effect of depression.

Migraine subtype, frequency of migraine, employment status, depression, and migraine-related disability were statistically significant predictors of patient-ratings of migraine severity.

This study highlights the impact of clinical and psychosocial determinants of

patient-ratings of migraine severity.

1.2

Personal and Societal Burden

45

Extended: Median regression analysis was used to model the association between influence of migraine disability assessment, socio-demographic characteristics (sex, age, employment status, education and marital status), self-reported side effects from medication (yes or no), migraine subtype, migraine frequency classification, and self-reported depression and anxiety symptoms (PHQ-9).

Future research will also need to investigate the validity of GAMS in community-

dwelling persons with migraine.

A single-item global rating scale of migraine severity may be a rapid and effi- cient way for clinicians to obtain information about disease severity, but requires validation in future prospective studies.

Background Migraine represents one-third of all neurological disease burden [140] and is one of the top 15 conditions with the most substantially increased disease burden ranking in the past decade; it is among the top 25 causes of years lived with disability (YLDs)2.

Several standardized and validated scales have been developed to assess patient- reported or physician-reported migraine severity, a concept closely linked to dis- ability [141–144].

The limitations of longer severity measures are addressed in a brief single item patient-reported measure, the Global Assessment of Migraine Severity (GAMS), developed to assess patients’ perception of their disease severity.

The objectives of this current study were to (1) explore how patients with migraine perceive the severity of their disease and its inter-relationship with patients’ demographic, clinical, and psychosocial characteristics, and (2) assess how GAMS compares to other measures of migraine severity and disability.

We hypothesize that patients’ ratings of migraine severity would be strongly cor- related with other validated measures of migraine severity and patients’ psychoso- cial characteristics.

Methods Fisher’s exact test was used to assess the association between patients’ ratings on GAMS and sex, marital status, employment status, education, medication side effects, migraine subtype, MIDAS (No/little/mild vs moderate/severe).

The association between patients’ ratings on the GAMS and MIDAS, PHQ9, HADS-D, HADS-A, and migraine frequency was assessed using polyserial correlation.

Associations between patients’ ratings on the GAMS and binary or ordinal clini- cal and self-reported characteristics (e.g., frequency of migraine attacks, intensity of migraine pain, migraine subtype, use of psychotropic medications, side-effects from medication, employment status) were assessed using biserial correlations.

Median regression analysis was used to model the association between influence of migraine disability assessment, socio-demographic characteristics (sex, age, employment status, education and marital status), self-reported side effects from medication (yes or no), migraine subtype, migraine frequency classification, and self-reported depression and anxiety symptoms (PHQ-9).

46

1 Public Health

Results Pearson’s chi square test revealed statistically significant univariate associations between patients’ ratings of migraine severity, as measured by GAMS, and patients’ rating on migraine-related disability, migraine subtype (chronic versus episodic), and employment status (employed versus not employed).

Patients who report more severe migraines are likely to be unemployed individu- als who report more disabling, chronic migraine with aura, and endorse clinically elevated levels of depression symptoms.

Mediation analysis revealed that migraine-related disability (measured by MIDAS) partially mediated the effects of depression and employment status on patient-reported GAMS ratings.

Patients with clinically elevated levels of depression symptoms were likely to self-report a 0.63 unit increase on the GAM, after adjusting for the indirect effects through migraine-related disability as measured by MIDAS, as compared to an unadjusted effect of 0.47.

The indirect effects of depression mediated through the MIDAS accounted for 9.5% of the employment status patients’ ratings of migraine severity (measured by GAMS).

Discussion In our headache outpatient clinic, we investigated the validity of GAMS, a brief patient-reported measure of migraine severity in migraine patients, and explored the determinants of patient-ratings of migraine severity in this outpatient cohort.

Our study demonstrates a moderate to strong correlation between patient-ratings of migraine severity and clinical characteristics and validated measures such as MIDAS score, MIDAS migraine severity question, PHQ9, HADS-A, migraine sub- type, frequency of migraine attacks.

The novelty of our study finding lies in the use of a single item global rating of migraine severity to identify clinical and psychosocial determinants of patient- reported severity of migraine.

A future prospective longitudinal study to assess the reliability and other psycho- metric properties of the GAMS in relation to other validated scales such as the Migraine Severity Scale and the Headache Impact Test is warranted.

Patients with chronic migraine were more likely to report a higher severity of

headache compared to patients with episodic migraine.

Conclusion We demonstrate the validity of GAMS as a brief measure of patient-reported migraine severity and identified clinical and psychosocial correlates and/or media- tors of patient-reported severity of migraine in a cohort of migraine patients seen in an outpatient Headache clinic.

Acknowledgement A machine generated summary based on the work of Sajobi, Tolulope T.; Amoozegar, Farnaz; Wang, Meng; Wiebe, Natalie; Fiest, Kirsten M.; Patten, Scott B.; Jette, Nathalie. 2019 in BMC Neurology.

1.2

Personal and Societal Burden

47

Perceived stress in patients with migraine: a case-control study