OBJECTIVES: Toenail onychomycosis (fungal infection) affects approximately 7% of Canadians. Comorbid diabetes may increase risk of infection and associated complications. Treatments include topical therapies and oral agents with varying modes of administration, efficacy, and safety profiles. The objective was to understand patient preferences regarding onychomycosis treatment features and how preferences vary among those with and without diabetes.

METHODS: A discrete choice experiment was conducted. Attributes were informed by qualitative interviews with 14 individuals with onychomycosis; seven had diabetes. Derived attributes included: efficacy, regimen details, risk of mild and severe adverse events (AEs), and interactions with medications and alcohol. Participants were shown infographics describing attribute features and completed 16 choice scenarios. Scenarios were randomized; data were analyzed using conditional logit regression (presented as odds ratios [ORs] with 95% confidence intervals [CIs]) and stratified by diabetes status.

RESULTS: The sample comprised 309 Canadians with toenail onychomycosis (n=160 with diabetes). Mean (standard deviation) age was 57.7 (14.6) years and 56.3% were male. Among those with diabetes, a statistically significant preference was observed for treatments that were: more efficacious (OR=1.06; CI=1.04-1.08 [per 1% increase]), not associated with severe AEs (0.82; 0.76-0.89), and did not interact with other medications (0.76; 0.70-0.82) or alcohol (0.83; 0.77-0.90). Those without diabetes preferred treatments that: used a squishable applicator (48 weeks) (1.28; 1.09-1.50) or one oral pill (12 weeks) (1.16; 1.01-1.34) over the reference regimen (nail polish applicator, 48 weeks, weekly removal of polish/monthly removal of separated nail), were not associated with severe AEs (0.88; 0.81-0.95), and did not interact with other medications (0.86; 0.79-0.93).

CONCLUSIONS: Among Canadians with toenail onychomycosis, presence of diabetes was associated with differences in treatment-related preferences. Those with diabetes were more likely to select a treatment with higher efficacy and had no risk of medication or alcohol interactions; those without diabetes were more concerned with mode of administration.