Frontal fibrosing alopecia: survey of severity assessment methods in routine clinical practice and validation of the International Frontal Fibrosing Alopecia Cooperative Group measurement guidance.

Abstract

Background: The lack of validated and responsive outcome measures in the management of frontal fibrosing alopecia (FFA) significantly limits assessment of disease progression and treatment response over time.

Aim: To understand how FFA extent and progression is currently assessed in UK specialist centres, to validate components of the International FFA Cooperative Group (IFFACG) statement on FFA assessment, and to identify pragmatic advice to improve FFA management in clinic.

Methods: Consultant dermatologists with a specialist interest in hair loss (n = 17) were invited to take part. Preferred FFA assessment methods were explored using questionnaires and clinical scenarios. Participants were asked to identify and mark the current hairline in 10 frontal and 10 temporal hairline images (Questionnaire 1), with assessment repeated 3 months later to assess intraindividual variability (Questionnaire 2) and 12 months later to test whether interindividual accuracy could be improved with simple instruction (Questionnaire 3).

Results: All 17 clinicians (100%) completed the questionnaire at each time interval. We identified a wide variation in assessment techniques used by our experts. Measurements were perceived as the most accurate method of assessing frontal recession whereas photography was preferred for temporal recession. Inter-rater reliability between clinicians measuring the frontal hairline scenarios indicated a moderate strength of agreement [intraclass coefficient (ICC) = 0.61; 95% CI 0.40-0.85], yet intrarater reliability was found to be poor with wide limits of agreement (-8.71 mm to 9.92 mm) on follow-up. Importantly, when clear guidance was provided on how the hairline should be identified (Questionnaire 3), inter-rater reliability improved significantly, with ICC = 0.70, suggesting moderate agreement (95% CI 0.51-0.89; P < 0.001). A similar pattern was seen with temporal hairline measurements, which again improved in accuracy with instruction.

Conclusion: We found that accuracy of measurements in FFA can be improved with simple instruction and we have validated components of the IFFACG measurement recommendations.