Opportunities to improve guideline adherence for the diagnosis and treatment of onychomycosis: Analysis of commercial insurance claims data, United States
Jeremy A.W. Gold, Karen Wu, Brendan R. Jackson, Kaitlin Benedict
Abstract
To the Editor: Onychomycosis, a fungal nail infection most frequently caused by dermatophytes, is an underrecognized public health problem, particularly given the global emergence of terbinafine resistance.1Gu D. Hatch M. Ghannoum M. Elewski B.E. Treatment-resistant dermatophytosis: a representative case highlighting an emerging public health threat.JAAD Case Rep. 2020; 6: 1153-1155Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar,2Singh S. Chandra U. Anchan V.N. Verma P. Tilak R. Limited effectiveness of four oral antifungal drugs (fluconazole, griseofulvin, itraconazole and terbinafine) in the current epidemic of altered dermatophytosis in India: results of a randomized pragmatic trial.Br J Dermatol. 2020; 183: 840-846Crossref PubMed Scopus (30) Google Scholar We estimated onychomycosis prevalence, described risk factors, and assessed adherence to American Academy of Dermatology (AAD) guidelines that recommend confirmatory testing (eg, direct microscopy, histopathology, fungal culture) before prescribing oral antifungal therapy.3Guidelines of care for superficial mycotic infections of the skin: onychomycosis. Guidelines/Outcomes Committee. American Academy of Dermatology.J Am Acad Dermatol. 1996; 34: 116-121Abstract Full Text PDF PubMed Scopus (53) Google Scholar We analyzed IBM MarketScan Commercial and Medicare Supplemental databases. We used International Classification of Diseases, 10th Revision, codes to identify onychomycosis patients and underlying conditions; we used Current Procedural Terminology codes to identify onychomycosis-related tests and procedures (Supplementary Table I, available via Mendeley at https://doi.org/10.17632/33j7s646j7.1). We calculated disease prevalence among all outpatients seen during 2018. We assessed underlying conditions and diagnostic and treatment practices in an analytic cohort; this cohort included outpatients diagnosed with onychomycosis during 2018 who had continuous insurance enrollment 365 days before and after incident diagnosis date and no onychomycosis diagnosis during the 365 days before their 2018 incident diagnosis. During 2018, among 21,298,716 outpatients, onychomycosis prevalence was 1.6% overall and 12.7% among patients aged ≥65 years. In the analytic cohort (n = 121,386), male gender (age-adjusted odds ratio [aOR] = 1.03, 95% confidence interval [CI]: 1.02-1.04) and nonrural residence (aOR = 1.34, 95% CI: 1.31-1.37) were associated with onychomycosis (Table I). Common underlying conditions included diabetes (23.0%), immunosuppressive conditions (21.8%), and non-unguium tinea (12.6%).Table IDemographic features and underlying medical conditions associated with onychomycosis patients (n = 121,386) in a large, commercially insured population—United States, 2018Characteristicn%aOR (95% CI)∗Age-adjusted odds ratios were calculated by comparing patients diagnosed with onychomycosis versus patients not diagnosed with onychomycosis during 2018. Among the 121,386 patients diagnosed with onychomycosis, 2.6% were aged <18 years, 8.8% were aged 18-34 years, 12.6% were aged 35-44 years, 22.2% were aged 45-54 years, 31.2% were aged 55-64 years, and 22.6% were aged ≥65 years. Among the 10,732,239 patients without onychomycosis, 23.6% were aged <18 years, 19.8% were aged 18-34 years, 15.1% were aged 35-44 years, 19.0% were aged 45-54 years, 18.4% were aged 55-64 years, and 4.1% were aged ≥65 years.Age (years), median (IQR)56 (45-63)NAGender Male56,23146.31.03 (1.02-1.04) Female65,15553.7ReferentUrban-rural classification Nonrural110,37190.91.34 (1.31-1.37) Rural10,8478.9Referent Unknown/missing†These missing data were excluded from aOR calculations.1680.1NAUS census region of primary beneficiary's residence Northeast37,09130.61.47 (1.45-1.50) South47,59339.21.04 (1.02-1.05) West13,93311.51.01 (0.99-1.03) Midwest22,52918.6Referent Unknown/missing†These missing data were excluded from aOR calculations.2400.2NAUnderlying conditions Tinea (non-unguium)15,23712.613.46 (13.21-13.72)Tinea pedis13,29211.026.75 (26.18-27.33)Tinea manuum2310.215.43 (13.31-17.88) Psoriasis25142.11.58 (1.52-1.65) Hallux valgus53734.45.26 (5.11-6.42) Overweight or obesity21,88318.01.64 (1.61-1.66) Diabetes27,91023.02.03 (2.00-2.05) Immunosuppressive conditions26,46121.81.56 (1.54-1.58)Cancer21,14117.41.54 (1.52-1.57)Immune-mediated inflammatory disease61535.11.48 (1.44-1.52)HIV4990.41.57 (1.44-1.72)Solid organ or stem cell transplantation5130.41.69 (1.55-1.85) Chronic venous insufficiency35052.92.58 (2.49-2.67) Peripheral arterial disease52484.33.64 (3.53-3.75) Tobacco use/nicotine dependence61205.01.13 (1.10-1.16)aOR, Age-adjusted odds ratio; CI, confidence interval; IQR, interquartile range; NA, not applicable.∗ Age-adjusted odds ratios were calculated by comparing patients diagnosed with onychomycosis versus patients not diagnosed with onychomycosis during 2018. Among the 121,386 patients diagnosed with onychomycosis, 2.6% were aged <18 years, 8.8% were aged 18-34 years, 12.6% were aged 35-44 years, 22.2% were aged 45-54 years, 31.2% were aged 55-64 years, and 22.6% were aged ≥65 years. Among the 10,732,239 patients without onychomycosis, 23.6% were aged <18 years, 19.8% were aged 18-34 years, 15.1% were aged 35-44 years, 19.0% were aged 45-54 years, 18.4% were aged 55-64 years, and 4.1% were aged ≥65 years.† These missing data were excluded from aOR calculations. Open table in a new tab aOR, Age-adjusted odds ratio; CI, confidence interval; IQR, interquartile range; NA, not applicable. Most patients were initially diagnosed by podiatrists (n = 62,177, 51.2%), followed by general practitioners (n = 28,223, 23.3%) and dermatologists (n = 15,910, 13.1%) (Table II). Across specialties, confirmatory laboratory testing was infrequent (15.3%); 12.0% of patients received a histopathology test, 2.8% a fungal culture, 2.1% direct microscopy, and 2.1% fungal polymerase chain reaction; 0.5% received antifungal susceptibility testing. Patients seen by dermatologists more frequently received confirmatory testing (31.0%) than those seen by podiatrists (16.9%) or general practitioners (5.2%). Overall, of the 18,128 patients prescribed an oral antifungal drug (most frequently terbinafine [88.2%]), 1756 (9.7%) received confirmatory diagnostic testing.Table IIDiagnostic and treatment practices by provider type for patients with onychomycosis in a large, commercially insured population—United States, 2018Characteristic∗Patients could receive more than 1 type of diagnostic test and more than 1 type of treatment. Diagnostic tests were considered onychomycosis related if they were documented within 7 days before, on, or after the incident onychomycosis visit date. Antifungal drug prescriptions and non-pharmaceutical therapies were considered onychomycosis related if they were documented within 0-7 days after the incident onychomycosis visit date. In total, 57.3% of onychomycosis patients had 1 visit for onychomycosis, 20.4% had 2 visits, 9.4% had 3 visits, and 12.9% of patients had ≥4 visits.OverallDermatologistPodiatristGeneral practitionerOther or unknown†Provider type was missing for 1229 patients; otherwise, the most common provider type visited on the incident diagnosis date among these patients included physician assistants (specialty unknown) (n = 1742) and nurse practitioners (specialty unknown) (n = 1733).P-value‡P-values were calculated using χ2 tests to compare practices among provider types.(N = 121,386)(n = 15,910)(n = 62,177)(n = 28,223)(n = 15,076)N%n%n%n%N%Diagnostic testing18,57915.3493931.010,53716.914695.2163410.8<.0001 Histopathology14,60212.0404625.4886614.36092.210817.2 Fungal culture33612.87794.915902.66082.23842.5 Direct microscopy25132.17004.411711.93751.32671.8 Polymerase chain reaction24962.1470.322433.6890.31170.8 Antifungal susceptibility testing5640.5720.52740.41310.5870.6Prescription antifungal drugs29,83324.6515332.4757112.212,22643.3488332.4<.0001 Topical12,39210.2311519.639046.3357612.7179711.9<.0001Ciclopirox97258.0204312.829634.8321611.4150310.0Efinaconazole22821.99305.87501.23411.22611.7Tavaborole4440.41711.12090.3260.1380.3 Oral18,12814.9220413.938976.3884531.3318221.1<.0001Terbinafine15,98513.2169210.635225.7800428.4276718.4Fluconazole16911.44692.93160.55782.03282.2Itraconazole2720.2300.2400.11440.5580.4Griseofulvin1840.2120.1260.01190.4270.2Ketoconazole1020.1110.150.0590.2270.2Posaconazole20.000.010.000.010.0Both oral and topical antifungal therapy6870.61661.02300.41950.7960.6<.0001Prescribing and testing practices Prescribed oral antifungal therapy without a confirmatory test16,37213.515619.833465.4851630.2294919.6<.0001 Prescribed topical antifungal therapy without a confirmatory test§Overall, 8369 of 9725 (86.1%) patients were prescribed ciclopirox, 1805 of 2282 (79.1%) patients were prescribed efinaconazole, and 322 of 444 (72.5%) were prescribed tavaborole without receiving a confirmatory diagnostic test.10,4548.6221513.931915.1343512.2161310.7<.0001Nonpharmaceutical therapies Nail debridement24,83920.51050.722,93736.94661.713318.8<.0001 Nail avulsion or excision49984.1540.341726.73501.24222.8<.0001 Photodynamic therapy180.0120.150.010.000.0∗ Patients could receive more than 1 type of diagnostic test and more than 1 type of treatment. Diagnostic tests were considered onychomycosis related if they were documented within 7 days before, on, or after the incident onychomycosis visit date. Antifungal drug prescriptions and non-pharmaceutical therapies were considered onychomycosis related if they were documented within 0-7 days after the incident onychomycosis visit date. In total, 57.3% of onychomycosis patients had 1 visit for onychomycosis, 20.4% had 2 visits, 9.4% had 3 visits, and 12.9% of patients had ≥4 visits.† Provider type was missing for 1229 patients; otherwise, the most common provider type visited on the incident diagnosis date among these patients included physician assistants (specialty unknown) (n = 1742) and nurse practitioners (specialty unknown) (n = 1733).‡ P-values were calculated using χ2 tests to compare practices among provider types.§ Overall, 8369 of 9725 (86.1%) patients were prescribed ciclopirox, 1805 of 2282 (79.1%) patients were prescribed efinaconazole, and 322 of 444 (72.5%) were prescribed tavaborole without receiving a confirmatory diagnostic test. Open table in a new tab We found a lower onychomycosis prevalence (1.6%) than previous European and North American studies (2% to 14%),4Scher R.K. Tavakkol A. Sigurgeirsson B. et al.Onychomycosis: diagnosis and definition of cure.J Am Acad Dermatol. 2007; 56: 939-944Abstract Full Text Full Text PDF PubMed Scopus (179) Google Scholar potentially reflecting underreporting, lack of clinical nail examination and testing, or differences in study design. Compared with other oral antifungals, terbinafine was more commonly prescribed, likely because it is of low cost and generally covered by health insurance without a requirement for laboratory testing. Compared with an urban academic medical center,5Geizhals S. Cooley V. Lipner S.R. Diagnostic testing for onychomycosis: a retrospective study over 17 years.J Am Acad Dermatol. 2020; 83: 239-241Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar patients in our study less frequently received confirmatory diagnostic testing (15.3% vs 39.3%), possibly because fewer providers in our study were dermatologists (13.1% vs 62.1%) and practices at an academic institution likely differ from other settings. Despite the limitations inherent to administrative data, including potential disease misclassification and undercoding, our study provides an update regarding US onychomycosis epidemiology and a concerning assessment of adherence to AAD guidelines for onychomycosis diagnosis and treatment. Confirming the diagnosis of onychomycosis with laboratory testing is important for ensuring appropriate therapy and avoiding unnecessary antifungal exposure. In the era of antifungal-resistant dermatophytosis, a renewed, cross-specialty emphasis on guideline-based onychomycosis treatment is needed, emphasizing antifungal stewardship to preserve available treatment options. None disclosed.