Epidemiology of alopecia areata in the Hispanic/Latinx community: A cross-sectional analysis of the All of Us database

We read with interest the paper by Phong et al1Phong C.H. McMichael A.J. Mesinkovska N.A. Epidemiology of alopecia areata in Hispanic/Latinx patients.J Am Acad Dermatol. 2023; 88: 916-917https://doi.org/10.1016/j.jaad.2022.10.049Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar and commend them for evaluating the epidemiology of alopecia areata (AA) in a cohort of exclusively Hispanic/Latinx (H/L) patients. Here, we add to their work by presenting a cross-sectional analysis of H/L patients diagnosed with AA in the All of Us database, which is a National Institutes of Health initiative to collate data on >1 million Americans, with an emphasis on recruiting communities historically underrepresented in research.2Denny J.C. Rutter J.L. et al.All of Us research program investigatorsThe “All of Us” research program.N Engl J Med. 2019; 381: 668-676Crossref PubMed Scopus (520) Google Scholar We identified AA cases using International Classification of Diseases, Ninth Revision, Clinical Modification code 704.01 and International Classification of Diseases, Tenth Revision, Clinical Modification code L63; patients who self-identified as H/L were selected for analysis. Each case was matched to the following 4 controls: age, ethnicity, race, and sex matched. Data on demographic characteristics and comorbidities were collected through review of electronic medical records. Epidemiological characteristics of H/L patients with AA and controls were compared using Pearson χ2 test or Fisher exact test for categorical variables and unpaired t test for continuous variables. We extracted data of 280 H/L patients with AA from the All of Us database. Similar to Phong et al,1Phong C.H. McMichael A.J. Mesinkovska N.A. Epidemiology of alopecia areata in Hispanic/Latinx patients.J Am Acad Dermatol. 2023; 88: 916-917https://doi.org/10.1016/j.jaad.2022.10.049Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar we observed a mean age at diagnosis of 35.9 ± 13.8 years with a woman:man predominance of 1.4:1. Altogether, 155 (55.3%) patients in our cohort were diagnosed with AA before the age of 40 years. We found H/L patients with AA to be more likely to have attained <12th grade education compared with controls (20.7% vs 9.3%, P < .001); this emphasizes the need to explain diagnoses and treatments in straightforward terms. However, there was no difference in annual household income between H/L patients with AA and controls. We also found H/L patients with AA to be less likely to have smoked compared with controls (25% vs 38.4%, P < .001) (Table I).Table IDemographics of Hispanic/Latinx patients with alopecia areata in the All of Us database cohort, compared with controlsCharacteristicHispanic/Latinx patients with AA (N = 280)Controls (N = 1120)P valueAge at diagnosis (y)35.9 ± 13.835.9 ± 13.81.0 <30 (%)94 (33.6)376 (33.6) 31-4061 (21.8)244 (21.8) 41-5071 (25.3)284 (25.3) 50-6545 (15.7)180 (15.7) >65<20†Per All of Us data dissemination guidelines, data with fewer than 20 participants should be suppressed from display.36 (3.2)Sex1.0 Female162 (57.9)648 (57.9)Education attainment<.001 <12th grade58 (20.7)104 (9.3) 12th grade or GED52 (18.6)215 (19.2) College79 (28.2)482 (43.0) Unknown91 (32.5)319 (28.5)Ever smoker∗Ever smoker defined as having smoked >100 cigarettes.<.001Yes70 (25)430 (38.4)No203 (72.5)634 (56.6)Unknown<20<20Household income.34 <10K49 (17.5)164 (14.6) 10-25K41 (14.6)134 (12.0) 25-50K41 (14.6)167 (14.9) >50K79 (28.2)318 (28.4) Unknown70 (25.0)337 (30.1)Insurance.02 Yes266 (95)1003 (89.6) No<2076 (6.8) Unknown<2041 (3.7)GED, general equivalency diploma.∗ Ever smoker defined as having smoked >100 cigarettes.† Per All of Us data dissemination guidelines, data with fewer than 20 participants should be suppressed from display. Open table in a new tab GED, general equivalency diploma. We also analyzed the prevalence of many of the comorbidities explored by Phong et al.1Phong C.H. McMichael A.J. Mesinkovska N.A. Epidemiology of alopecia areata in Hispanic/Latinx patients.J Am Acad Dermatol. 2023; 88: 916-917https://doi.org/10.1016/j.jaad.2022.10.049Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar H/L patients with AA had increased odds of the following autoimmune diseases: hypothyroidism (odds ratio [OR], 2.21; 95% CI, 1.16-4.19), rheumatoid arthritis (OR, 4.68; 95% CI, 2.55-8.59), systemic lupus erythematosus (OR, 8.13; 95% CI, 3.09-22.35), and type I diabetes (OR, 4.68; 95% CI, 1.68-13.01) (Table II). Notably, we report rheumatoid arthritis to be the most common autoimmune comorbidity in H/L patients with AA, a finding that concurs with Phong et al’s study.1Phong C.H. McMichael A.J. Mesinkovska N.A. Epidemiology of alopecia areata in Hispanic/Latinx patients.J Am Acad Dermatol. 2023; 88: 916-917https://doi.org/10.1016/j.jaad.2022.10.049Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar We also report H/L patients with AA to have increased odds of anxiety (OR, 3.05; 95% CI, 2.19-4.25) and depression (OR, 2.72; 95% CI, 1.91-3.88). Increased odds of hypertension (OR, 1.86; 95% CI, 1.30-2.65) and hyperlipidemia (OR, 3.67; 95% CI, 2.52-5.34) were also observed. We have previously shown H/L patients with AA to carry a significant burden of atopic disease, more so than White patients with AA.3Joshi TP, Fernandez B, Friske S, et al. Burden of atopic disease in Black and Hispanic patients with alopecia areata: a case-control study in the All of Us research program. Int J Dermatol. Published online November 28, 2022. https://doi.org/10.1111/ijd.16528Google ScholarTable IIComorbidities present in Hispanic/Latinx patients with alopecia areata in the All of Us database cohort, compared with controlsComorbidityHispanic/Latinx patients with AA (N = 280)Controls (N = 1120)OR (95% CI)P valueAnxiety (%)72 (25.7)114 (10.2)3.05 (2.19-4.25)<.001Depression59 (21.1)100 (8.9)2.72 (1.91-3.88)<.001GERD122 (43.6)203 (18.1)3.49 (2.63-4.62)<.001Hyperlipidemia57 (20.4)73 (6.5)3.67 (2.52-5.34)<.001Hypertension51 (18.2)120 (10.7)1.86 (1.30-2.65)<.001Hypothyroidism<20∗Per All of Us data dissemination guidelines, data with fewer than 20 participants should be suppressed from display.28 (2.50)2.21 (1.16-4.19).016IBD<20<200.14 (0.01-2.46).18Rheumatoid arthritis23 (8.2)21 (1.9)4.68 (2.55-8.59)<.001SLE<20<208.13 (3.09-22.35)<.001Type I diabetes<20<204.68 (1.68-13.01).0031Type II diabetes32 (11.4)137 (12.2)0.93 (0.61-1.39).71GERD, gastroesophageal reflux disease; IBD, inflammatory bowel disease; OR, odds ratio; SLE, systemic lupus erythematosus.∗ Per All of Us data dissemination guidelines, data with fewer than 20 participants should be suppressed from display. Open table in a new tab GERD, gastroesophageal reflux disease; IBD, inflammatory bowel disease; OR, odds ratio; SLE, systemic lupus erythematosus. Limitations of our analysis include inability to comment on the clinical characteristics/severity of AA or establish temporality between AA and the comorbidities studied. Altogether, we show H/L patients with AA, compared with controls, to have lower educational attainment, lower prevalence of smoking, and increased rates of autoimmune disease, hyperlipidemia, hypertension, and psychiatric illness. We hope that our findings, together with those of Phong et al,1Phong C.H. McMichael A.J. Mesinkovska N.A. Epidemiology of alopecia areata in Hispanic/Latinx patients.J Am Acad Dermatol. 2023; 88: 916-917https://doi.org/10.1016/j.jaad.2022.10.049Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar galvanize further research into the epidemiology of AA in the H/L community. None disclosed.