Allergic or Not: Final Interpretation of Doubtful Patch Test Reactions From the North American Contact Dermatitis Group, 2019-2020.

Background: Doubtful patch test reactions generally do not meet criterion for positivity in patch testing. However, the North American Contact Dermatitis Group (NACDG) allows for doubtful reactions to be coded with a final determination of "allergic/positive" based on the temporal pattern, appearance, known characteristics of the allergen, and/or other supportive patch test reactions. Objectives: To analyze NACDG data from the 2019-2020 patch test cycle to identify patterns in the interpretation and relevance of doubtful reactions. Methods: The frequency and proportions of doubtful reactions were tabulated and analyzed for patterns. Statistical analyses were limited to allergens with ≥30 doubtful reactions to ensure adequate sample size. Results: Of patch-tested patients, 31.9% (1315/4121) had ≥1 doubtful reaction. Of 2538 total doubtful reactions, 46% (n = 1167) had a final interpretation of "allergic/positive." The allergens with the highest proportion of doubtful reactions at the final visit were hydroperoxides of linalool 1% (4.5%), fragrance mix I 8.0% (3.9%), and cetrimonium chloride 0.5% (3.4%). Methylchloroisothiazolinone/methylisothiazolinone (MCI/MI) 0.02% (P < 0.001), MI 0.2% (P < 0.001), nickel sulfate hexahydrate 2.5% (P = 0.001), and neomycin sulfate 20.0% (P = 0.003) doubtful reactions were more likely to be interpreted as allergic than nonallergic. Methyldibromoglutaronitrile/phenoxyethanol 0.2% (P < 0.001), oleamidopropyl dimethylamine 0.1% (P < 0.001), formaldehyde 2.0% (P < 0.001), cetrimonium chloride 0.5% (P < 0.001), benzophenone-4 (sulisobenzone) 10% (P < 0.001), iodopropynyl butylcarbamate 0.5% (P < 0.001), cocamidopropyl betaine 1.0% (P = 0.002), and benzisothiazolinone 0.1% (P = 0.012) doubtful reactions were less likely to be interpreted as allergic. Of the 1167 doubtful reactions interpreted as allergic, 84.9% had current relevance. Conclusions: Doubtful reactions were common and approximately one half were coded with a final interpretation of "allergic/positive." Of those, most were clinically relevant. MCI/MI, MI, nickel, and neomycin were more likely to be interpreted as allergic.

Patch Testing With Nickel, Cobalt, and Chromium in Patients With Suspected Allergic Contact Dermatitis.

Background: Allergic contact dermatitis is frequently caused by metals, including multiple metals simultaneously. Objectives: To assess characteristics and associations of positive and clinically relevant patch test (PT) reactions with solitary and concurrent metal sensitization. Methods: A retrospective analysis of PT results for nickel, cobalt, and/or chromium from the North American Contact Dermatitis Group between 2001 and 2018 (n = 43,522). Results: 18.0% had a positive/allergic reaction to nickel sulfate hexahydrate, 7.3% to cobalt chloride hexahydrate, and 3.0% to potassium dichromate. 87.9% patients had a currently relevant reaction to 0, 9.4% to 1, and 2.7% to multiple metals tested. Patients with 1 versus no currently relevant reactions to metal were more likely to have a primary dermatitis site of trunk, feet, and ears; patients with currently relevant reactions to multiple metals had more dermatitis affecting the trunk and ears. Metal sources varied by co-reacting metal, especially for patients with cobalt and chromium allergy. Jewelry was the most commonly identified source of nickel and cobalt for both solitary and concurrent metal allergy. Conclusions: Sensitization to multiple metals occurred in 6% of patients. Allergen sources varied between patients with sensitivity to 1 metal versus those who had concurrent sensitivity to cobalt and/or chromium.

More than just methylisothiazolinone: Retrospective analysis of patients with isothiazolinone allergy in North America, 2017-2020.

BACKGROUND: Isothiazolinones are a common cause of allergic contact dermatitis. OBJECTIVE: To examine the prevalence of positive patch test reactions to isothiazolinones from 2017-2020 and characterize isothiazolinone-allergic (Is+) patients compared with isothiazolinone nonallergic (Is-) patients. METHODS: Retrospective cross-sectional analysis of 9028 patients patch tested to methylchloroisothiazolinone (MCI)/methylisothiazolinone (MI) 0.02% aqueous, MI 0.2% aqueous, benzisothiazolinone (BIT) 0.1% petrolatum, and/or octylisothiazolinone (OIT) 0.025% petrolatum. Prevalence, reaction strength, concurrent reactions, clinical relevance, and source of allergens were tabulated. RESULTS: In total, 21.9% (1976/9028) of patients had a positive reaction to 1 or more isothiazolinones. Positivity to MI was 14.4% (1296/9012), MCI/MI was 10.0% (903/9017), BIT was 8.6% (777/9018), and OIT was 05% (49/9028). Compared with Is-, Is+ patients were more likely to have occupational skin disease (16.5% vs 10.3%, P <.001), primary hand dermatitis (30.2% vs 19.7%, P <.001), and be >40 years (73.1% vs 61.9%, P <.001). Positive patch test reactions to >1 isothiazolinone occurred in 44.1% (871/1976) of Is+ patients. Testing solely to MCI/MI would miss 47.3% (611/1292) of MI and 60.1% (466/776) of BIT allergic reactions. LIMITATIONS: Retrospective cross-sectional study design and lack of follow-up data. CONCLUSION: Sensitization to isothiazolinones is high and concurrent sensitization to multiple isothiazolinone allergens is common.

Patch Test Results Among Older Adults: A Retrospective Analysis of the North American Contact Dermatitis Group Data (2009-2020).

Background: Allergic contact dermatitis (ACD) in older adults (OA) represents a significant health burden, but few studies examine the prevalence and characteristics of contact allergy and ACD in this population. Objective: To compare positive and clinically relevant patch test results in OA versus younger adults (YA) and children. Methods: Retrospective analysis of patch test results obtained in OA (≥65 years), YA (19-64 years), and children (≤18 years) by the North American Contact Dermatitis Group, 2009 to 2020. Results: Of 28,177 patients patch tested, 5366 (19.0%) were OA. OA were more likely to have a final primary diagnosis of ACD as compared with YA (50.8% vs 49.2%, P = 0.035) and children (44.6%, P < 0.0001). The primary site of dermatitis also differed by age group, with OA having a higher proportion of dermatitis affecting the trunk, scalp, anogenital region, and "under clothing," and a lower proportion of dermatitis affecting the face, lips, and feet. Limitations: Retrospective design, lack of follow-up, and referral population. Conclusion: OA were as likely and were statistically even more likely to have a final primary diagnosis of ACD compared with YA and children. Anatomic site of dermatitis also differed by age group. This underscores the need for patch testing in OA when ACD is suspected.

Patch Testing to Paraphenylenediamine: The North American Contact Dermatitis Group Experience (1994-2018).

Background/Objectives: Paraphenylenediamine (PPD) is an aromatic amine dye that may cause allergic contact dermatitis. This study examines the epidemiology of allergic patch test reactions to PPD. Methods: This retrospective analysis characterizes individuals tested to PPD (1% petrolatum) by the North American Contact Dermatitis Group (1994-2018). Demographics and dermatitis site(s) were compared between PPD-allergic and PPD-negative patients. PPD reactions were analyzed by reaction strength, clinical relevance, occupational relatedness, and source as well as coreactivity with structurally related compounds. Results: Of 54,917 patients tested to PPD, 3095 (5.6%) had an allergic patch test reaction. Compared with PPD-negative patients, PPD-allergic patients had significantly greater odds of age >40 years (odds ratio [OR] 1.55 [95% confidence interval; CI 1.43-1.69]) and female gender (OR 1.52 [95% CI 1.41-1.66]), but lower odds of being White (OR 0.66 [95% CI 0.60-0.71]). The most common primary anatomic sites of dermatitis were face (25.5%), hands (21.9%), and scattered/generalized pattern (15.5%). Over half (55.3%) of PPD reactions were ++ or +++ at the final reading and 60.9% were currently relevant. Common exposure sources included hair dye (73.5%) and clothing/shoes/apparel (3.9%). Occupationally related reactions occurred in 8.3%, most commonly in hairdressers/cosmetologists (72.8%). The most common coreactions were benzocaine (11.3%), N-isopropyl-N'-phenyl-p-phenylenediamine (6.7%), disperse dye mix (6.5%), and black rubber mix (5.1%). Conclusions: The 24-year percentage of allergic reactions to PPD was 5.6%. PPD allergy was associated with female gender and age >40 years. PPD allergic patients were less likely to be White. Allergic reactions were usually clinically relevant and hair dye was the most frequently identified source.

Patch Testing to Chlorhexidine Digluconate, 1% Aqueous: North American Contact Dermatitis Group Experience, 2015-2020.

Background: Chlorhexidine is an antiseptic that may cause allergic contact dermatitis. Objectives: To describe the epidemiology of chlorhexidine allergy and characterize positive patch test reactions. Methods: This retrospective study analyzed patients patch tested to chlorhexidine digluconate 1% aqueous by the North American Contact Dermatitis Group, 2015-2020. Results: Of 14,731 patients tested to chlorhexidine digluconate, 107 (0.7%) had an allergic reaction; of these, 56 (52.3%) reactions were currently clinically relevant. Most (59%) reactions were mild (+), followed by strong (++, 18.7%) and very strong (+++, 6.5%). Common primary dermatitis anatomic sites in chlorhexidine-positive patients were hands (26.4%), face (24.5%), and scattered/generalized distribution (17.9%). Compared with negative patients, chlorhexidine-positive patients were significantly more likely to have dermatitis involving the trunk (11.3% vs 5.1%; P = 0.0036). The most commonly identified source category was skin/health care products (n = 41, 38.3%). Only 11 (10.3%) chlorhexidine reactions were occupationally related; of these, 81.8% were in health care workers. Conclusions: Chlorhexidine digluconate allergy is uncommon, but often clinically relevant. Involvement of the hands, face, and scattered generalized patterns was frequent. Occupationally related reactions were found predominantly in health care workers.

Patch Testing With Cobalt in Adults: A North American Contact Dermatitis Group Experience, 2001-2018.

Background: Cobalt is a recognized cause of allergic contact dermatitis (ACD); however, detailed information on patient characteristics, sites, and sources is lacking. Objective: The aim of the study is to assess trends in patch test reactions to cobalt and associated patient characteristics, common sources, and body sites affected. Methods: The study used a retrospective analysis of adult patients who were patch tested to cobalt by the North American Contact Dermatitis Group between 2001 and 2018 (n = 41,730). Results: Overall, 2986 (7.2%) and 1362 (3.3%) had allergic or currently relevant patch test reaction to cobalt, respectively. Patients with versus without an allergic patch test reaction to cobalt were more likely to be female, employed, have a history of eczema or asthma, be Black, Hispanic, or Asian, and have occupational-related dermatitis. The most commonly identified sources of cobalt in allergic patients included jewelry, belts, and cement, concrete, and mortar. Affected body site(s) varied by cobalt source among patients with currently relevant reactions. Occupational relevance was found in 16.9% of patients with positive reactions. Conclusions: Positive patch test reactions to cobalt were common. The most common body sites were the hands, and affected site varied by the source of cobalt.

North American Contact Dermatitis Group Patch Test Results: 2019-2020.

Background: Patch testing is an important diagnostic tool for assessment of allergic contact dermatitis (ACD). Objective: This study documents the North American Contact Dermatitis Group (NACDG) patch testing results from January 1, 2019, to December 31, 2020. Methods: At 13 centers in North America, patients were tested in a standardized manner with a screening series of 80 allergens, and, as indicated, supplemental allergens. Results: Overall, 4121 patients were tested; 2871 (69.7%) had at least 1 positive/allergic patch test reaction and 2095 patients (51.2%) had a primary diagnosis of ACD. The most commonly positive allergens were nickel (18.2%), methylisothiazolinone (MI) (13.8%), fragrance mix (FM) I (12.8%), hydroperoxides of linalool (HPL) (11.1%), and benzisothiazolinone (BIT) (10.4%). Compared with that of 2017-2018, prevalence of top 20 allergens statistically increased for FM I, HPL, BIT, propolis, and hydroperoxides of limonene (3.5%). For the first time, MI positivity did not increase between reporting periods. Approximately one-fifth of patients (20.3%) had ≥1 clinically relevant reaction(s) to allergens/substances not on the NACDG series. Conclusions: The epidemic of MI contact allergy in North America may have reached a plateau. Patch testing using a robust screening series, and supplemental allergens as indicated, is necessary for comprehensive evaluation of ACD.

Patch Testing With Benzophenone-3 and -4: The North American Contact Dermatitis Group Experience, 2013-2020.

Background: Benzophenone (BZP)-3 and BZP-4 are ultraviolet (UV) absorbers used in sunscreens and personal care products (PCPs) and may cause allergic contact dermatitis. Objective: To characterize positive patch test reactions to BZP-3 (10% in petrolatum [pet]) and BZP-4 (2% pet) in a screening allergen series. Methods: Retrospective analysis of patients tested to BZP-3 and BZP-4 was conducted by the North American Contact Dermatitis Group from 2013 to 2020. Results: Of 19,618 patients patch tested to BZP-3 and BZP-4, 103 (0.5%) and 323 (1.6%) had positive reactions, respectively: 413 (2.1%) reacted to at least 1 BZP (BZP-positive patient). As compared with BZP-negative patients, BZP-positive patients were significantly more likely to have a history of hay fever (39.3% vs 33.4%, P = 0.0134), history of atopic dermatitis (39.8% vs 30.7%, P = 0.0001), and facial involvement (37.4% vs 32.2%, P = 0.0272). Most reactions were currently clinically relevant (BZP-3: 90.4%; BZP-4: 65.8%). Common identified sources included PCPs and sunscreens. Coreactivity between BZP-3 and BZP-4 was low: 13.5% (14/104) of BZP-3-positive patients were allergic to BZP-4 and 4.3% (14/322) of BZP-4-positive patients were allergic to BZP-3. Conclusions: Eight-year prevalence of BZP positivity was 2.1%. Reactions were frequently clinically relevant and linked to PCPs and sunscreens.

Trends in the Prevalence of Methylchloroisothiazolinone/Methylisothiazolinone Contact Allergy in North America and Europe.

Importance: The common use of isothiazolinones as preservatives is a global cause of allergic contact dermatitis. Differences in allowable concentrations of methylisothiazolinone (MI) exist in Europe, Canada, and the US. Objective: To compare the prevalence of positive patch test reactions to the methylchloroisothiazolinone/methylisothiazolinone (MCI/MI) combination and MI alone in North America and Europe from 2009 to 2018. Design, Setting, and Participants: This retrospective analysis of North American Contact Dermatitis Group, European Surveillance System on Contact Allergies (ESSCA), and the Information Network of Departments of Dermatology (IVDK) databases included data from patients presenting for patch testing at referral patch test clinics in North America and Europe. Exposures: Patch tests to MCI/MI and MI. Main Outcomes and Measures: Prevalence of allergic contact dermatitis to MCI/MI and MI. Results: From 2009 to 2018, participating sites in North America and Europe patch tested a total of 226 161 individuals to MCI/MI and 118 779 to MI. In Europe, positivity to MCI/MI peaked during 2013 and 2014 at 7.6% (ESSCA) and 5.4% (IVDK) before decreasing to 4.4% (ESSCA) and 3.2% (IVDK) during 2017 and 2018. Positive reactions to MI were 5.5% (ESSCA) and 3.4% (IVDK) during 2017 and 2018. In North America, the frequency of positivity to MCI/MI increased steadily through the study period, reaching 10.8% for MCI/MI during 2017 and 2018. Positive reactions to MI were 15.0% during 2017 and 2018. Conclusions and Relevance: The study results suggest that in contrast to the continued increase in North America, isothiazolinone allergy is decreasing in Europe. This trend may coincide with earlier and more stringent government regulation of MI in Europe.

Prevalence and trend of allergen sensitization in patients with cheilitis referred for patch testing, North American Contact Dermatitis Group data, 2001-2018.

BACKGROUND: An updated understanding of allergic contact cheilitis is needed. OBJECTIVES: To characterize clinical characteristics and allergen relevance in patients with cheilitis referred for patch testing. METHODS: Retrospective analysis of 43 772 patients patch tested with the North American Contact Dermatitis Group (NACDG) screening series from 2001 to 2018. RESULTS: Overall, 2094 patients (4.8%) had lips as one of three sites of dermatitis, 1583 (3.6%) had lips as the primary site and 1167 (2.7%) had lips as the sole site of dermatitis. Prevalences of cheilitis at any, primary, and sole sites significantly increased throughout the study cycle from 2001-2002 (2.7%, 2.2% and 1.7%) to 2017-2018 (7.8%, 5.2% and 3.7%). Approximately 60% of patients with any, a primary, or a sole site of cheilitis had one or more positive allergic patch-test reactions compared to 65% of those without cheilitis. CONCLUSION: Patients with cheilitis who were referred for patch testing had high rates of positive and relevant allergens. More than one in four patients with any, primary, or sole cheilitis had a positive reaction to non-NACDG screening allergens (28.0%, 26.8%, 31.1% vs. 21.6%) compared to patients without cheilitis, emphasizing the need for expanded patch test series in cheilitis.

Hand and foot dermatitis in patients referred for patch testing: Analysis of North American Contact Dermatitis Group Data, 2001-2018.

BACKGROUND: Dermatitis localized to hands (HD), feet (FD), or both hands and feet (HFD) has multiple etiologies, including atopic dermatitis, irritant contact dermatitis, and allergic contact dermatitis. Unfortunately, little is known about clinical differences between patients with HD, FD, and HFD. OBJECTIVE: To characterize differences in demographics, etiology, and patch testing results among patients presenting with HD, FD, or HFD referred for patch testing. METHODS: A retrospective analysis of patients patch tested by the North American Contact Dermatitis Group between 2001 and 2018. RESULTS: Of 43,677 patients who were patch tested, 22.8% had HD, 2.9% had FD, and 3.7% had HFD. Allergic and currently relevant patch test reactions to ≥1 North American Contact Dermatitis Group screening allergen occurred in similar proportions in all 3 study groups. However, HD (18.0%) had higher proportions of occupationally relevant reactions than HFD (8.9%) or FD (4.0%). Nickel and fragrance mix I were in the top 5 currently relevant allergens for HD, FD, and HFD. Other top allergens, as well as allergen sources, differed between HD, FD, and HFD. LIMITATIONS: No data on HD or FD morphology or distribution. CONCLUSION: HD, FD, and HFD have several differences with respect to patient characteristics, etiologies, and clinically relevant allergens.

Patch testing with cobalt in children and adolescents: North American contact dermatitis group experience, 2001-2018.

BACKGROUND: Allergic contact dermatitis (ACD) to cobalt is more common in children and adolescents than adults. However, detailed information on sites and sources of cobalt ACD is limited. OBJECTIVES: To assess trends in positive and clinically relevant patch test reactions to cobalt in children and associated patient characteristics, common sources and body sites affected. METHODS: A retrospective analysis of children (<18 years) patch tested to cobalt by the North American Contact Dermatitis Group between 2001 and 2018. RESULTS: Of 1919 children patch tested, 228 (11.9%) and 127 (6.6%) had a positive/allergic or currently relevant patch test reaction to cobalt, respectively. The most common primary body sites affected were scattered generalized (30.0%), face, not otherwise specified (10.6%) and trunk (10.1%). Patients with allergic and currently relevant allergic patch test reactions were more likely to have a primary site of trunk (p = 0.0160 and p = 0.0008) and ears (p = 0.0005 and p < 0.0001). Affected body site(s) varied by cobalt source among patients with currently relevant reactions, especially for less common sources. The most commonly identified sources of cobalt included jewellery, belts and clothing. CONCLUSIONS: Positive patch test reactions to cobalt were common in children. The most common body site was scattered generalized and the sources of cobalt varied by body site.

Patch testing with glucosides: The North American Contact Dermatitis Group experience, 2009-2018.

BACKGROUND: Alkyl glucosides are nonionic surfactants that are increasingly used in personal care products. OBJECTIVE: To characterize positive patch test reactions to decyl glucoside (5% petrolatum, tested 2009-2018) and lauryl glucoside (3% petrolatum, tested 2017-2018). METHODS: Retrospective analysis of patients tested by the North American Contact Dermatitis Group. RESULTS: Of 24,097 patients patch tested to decyl and/or lauryl glucoside, 470 (2.0%) had positive reactions. Compared with glucoside-negative patients, glucoside-positive patients had higher odds of occupational skin disease (13.4% vs 10.1%; P = .0207), history of hay fever (38.5% vs 31.6%; P = .0014), atopic dermatitis (39.0% vs 28.6%; P < .0001), and/or asthma (21.8% vs 16.5%; P = .0023). Most glucoside reactions (83.9%) were currently relevant. The most common source was personal care products (63.0%), especially hair products (16.5%) and skin cleansers (15.2%). Of 4933 patients tested to decyl and lauryl glucoside, 134 (2.7%) were positive to 1 or both; 43.4% (43 of 99) of decyl-positive patients were also positive to lauryl glucoside and 55.1% (43/78) of lauryl glucoside patients were also positive to decyl glucoside. LIMITATIONS: The cohort predominantly reflects a referral population, and follow-up after testing was not captured. CONCLUSION: Glucoside positivity occurred in 2.0% of the tested patients. Reactions were often clinically relevant and linked to personal care products. Cross-reactivity was >40%.

Lanolin Allergic Reactions: North American Contact Dermatitis Group Experience, 2001 to 2018.

BACKGROUND: Lanolin is an important cause of allergic contact dermatitis. OBJECTIVES: The aims of this study were to describe the epidemiology of lanolin allergy and to assess trends in patch test reactions to lanolin over time. METHODS: This study used a retrospective analysis of patients patch tested with lanolin alcohol 30% or Amerchol L-101 50% in petrolatum by the North American Contact Dermatitis Group between 2001 and 2018 (n = 43,691). RESULTS: Overall, 1431 (3.3%) had a positive reaction, and 1238 (2.8%) were currently relevant. Prevalence of lanolin allergy was 4.63% between 2011 and 2018 (P < 0.0001). Most lanolin-allergic patients had + (52%) reactions; 18%, and 6% had ++ and +++ reactions, respectively. Common primary anatomic sites of dermatitis were the hands (20.7%), scattered/generalized distribution (19.6%), and face (17.0%). Allergic reactions to lanolin were more common in children (4.5%) than in adults (3.2%, P = 0.0018). Compared with nonallergic patients, lanolin-allergic patients were more likely to have history of eczema or hay fever, male sex, older than 40 years, or Black race (P < 0.05). Common lanolin sources were personal care products and drugs/medications. Only 2.24% of the positive reactions were linked to occupation. CONCLUSIONS: Lanolin sensitivity was common. Reactions were often clinically relevant and linked to personal care products and medications.

Natural Is Not Always Better: The Prevalence of Allergenic Ingredients in "Clean" Beauty Products.

BACKGROUND: Consumers seek "clean" and "natural" products based on their perception of safety. However, there is no standard, scientific basis, or regulatory oversight in the marketing or ingredient use for "clean" products. OBJECTIVE: The aim of this study was to determine the prevalence of allergenic ingredients in "clean" products. METHODS: Target and Walgreens Web sites were queried for "clean" products with inclusion of 1470 products. Ingredient lists were analyzed for potential allergens. Analysis included descriptive statistics and χ2 test. RESULTS: The most common allergens were fragrances/botanicals (1218/1470, 82.9%), phenoxyethanol (591/1470, 40.2%), tocopherol (545/1470, 40.2%), benzoic acid and benzoates (434/1470, 29.5%), propylene glycol (369/1470, 25.1%), alkyl glucosides (305/1470, 20.7%), ethylhexylglycerin (304/1470, 20.7%), cetyl alcohol (282/1470, 19.2%), cocamidopropyl betaine (258/1470, 17.6%), and benzyl alcohol (232/1470, 15.8%). Among fragrances/botanicals, the most common ingredients found were fragrance/perfume/aroma (911/1470, 68.2%), citrus derivatives (375/1470, 25.5%), linalool (305/1470, 20.7%), limonene (279/1470, 19.0%), and benzyl alcohol (231/1470, 15.7%). A total of 93.8% of the products (1379/1470) contained at least 1 potential allergen. CONCLUSIONS: Most "clean" products contain a potential allergen, predominately fragrances and botanicals.

Shoe Allergens: A Retrospective Analysis of Cross-sectional Data From the North American Contact Dermatitis Group, 2005-2018.

BACKGROUND: Shoe contact allergy can be difficult to diagnose and manage. OBJECTIVE: The aim of the study was to characterize demographics, clinical characteristics, patch test results, and occupational data for the North American Contact Dermatitis Group patients with shoe contact allergy. METHODS: This is a retrospective study of 33,661 patients, patch tested from 2005 to 2018, with a shoe source, foot as 1 of 3 sites of dermatitis, and final primary diagnosis of allergic contact dermatitis. RESULTS: Three hundred fifty-two patients met the inclusion criteria. They were more likely to be male (odds ratio = 3.36, confidence interval = 2.71-4.17) and less likely to be older than 40 years (odds ratio = 0.49, confidence interval = 0.40-0.61) compared with others with positive patch test reactions. The most common relevant North American Contact Dermatitis Group screening allergens were potassium dichromate (29.8%), p-tert-butylphenol formaldehyde resin (20.1%), thiuram mix (13.3%), mixed dialkyl thioureas (12.6%), and carba mix (12%). A total of 29.8% (105/352) had positive patch test reactions to supplemental allergens, and 12.2% (43/352) only had reactions to supplemental allergens. CONCLUSIONS: Shoe contact allergy was more common in younger and male patients. Potassium dichromate and p-tert-butylphenol formaldehyde resin were the top shoe allergens. Testing supplemental allergens, personal care products, and shoe components should be part of a comprehensive evaluation of suspected shoe contact allergy.