PO-39 | Leeds Dependence Questionnaire (LDQ) in migraine: diagnostic utility without predictive value for anti-CGRP (R) response

Antonio Munafò,¹ Andrea Burgalassi,¹ Giulia Vigani,² Francesco De Cesaris,¹ Alberto Chiarugi^1,2 | ¹Headache Center and Clinical Pharmacology Unit, Careggi University Hospital, Florence, Italy; ²Section of Clinical Pharmacology and Oncology, Department of Health Sciences, University of Florence, Florence, Italy

Background: Medication overuse (MO) represents a clinically significant complication in migraine, often undermining preventive treatment success. Previous studies have shown that medication dependence behaviors contribute to treatment resistance. The Leeds Dependence Questionnaire (LDQ), originally designed for substance dependence assessment, has demonstrated diagnostic utility in identifying medication overuse, though its predictive value for treatment outcomes remains controversial. While anti-CGRP (R) therapies have shown efficacy in migraine prevention, the role of baseline medication dependence behaviors in predicting anti-CGRP (R) treatment outcomes has not been specifically investigated.

Methods: We conducted a prospective observational study involving 71 patients (22 episodic migraine, 49 chronic migraine) undergoing anti-CGRP(R) medications. MO was defined according to ICHD-3 criteria. Baseline assessments included LDQ, HIT-6, MIDAS, and headache frequency measures. Treatment response was defined as a ≥50% reduction in monthly headache days (MHD) at 3 and 6 months. Statistical analyses included Fisher's exact test for categorical associations, Welch's t-test or Mann-Whitney U test for group comparisons based on normality assumptions, and Spearman or Pearson correlations for continuous variables, with correction for multiple comparisons.

Results: Patients with MO exhibited significantly higher LDQ scores at baseline compared to those without MO (12.6 ± 6.3 vs 9.2 ± 4.1; p = 0.009), supporting the LDQ’s diagnostic validity. However, baseline LDQ scores were not predictive of treatment response at either 3 months (responders: 12.0 vs non-responders: 10.1; p = 0.269) or 6 months (responders: 11.5 vs non-responders: 11.6; p = 0.954). Exploratory correlation analyses revealed that higher LDQ scores were associated with smaller reductions in symptomatic medication use at 6 months (rho = –0.302), and in monthly migraine days at both 3 months (rho = –0.305) and 6 months (rho = –0.308), all statistically significant after correction. No significant correlations were found with changes in medication use frequency, disability (MIDAS), or headache impact (HIT-6).

Conclusion: In the context of anti-CGRP(R) therapy, the LDQ confirms its diagnostic utility in identifying medication overuse behaviors. However, it lacks predictive value for therapeutic response. Exploratory findings suggest that higher baseline LDQ scores may be associated with a reduced ability to decrease the use of symptomatic medications, indicating that medication dependence behaviors may interfere with effective therapeutic management. These preliminary observations should be confirmed in larger, prospective studies.