Pyoderma gangrenosum (PG) is a rare ulcerative disorder affecting nearly 3–10 patients per million yearly and is associated with high morbidity. 1 Epidemiology studies investigating PG in patients historically underrepresented in medicine may give further information regarding disease pathophysiology and may aid in the identification of ‘at‐risk’ populations. The aim of this study is to elucidate patterns of PG diagnoses in populations that have been historically underrepresented in biomedical research.
Participants of the ‘All of Us’ (AoU) registered tier version 7 database were stratified based on PG diagnosis (SNOMED 133283) and the characteristics of the patients populations were compared with non‐PG patients using Pearson's chi‐squared tests (or Fisher's exact test if low n). Adjusted odds ratios for PG diagnoses were calculated using multivariate logistic regression. Various demographic factors including race/ethnicity, age, gender identity, etc. and selected comorbidities of PG were also evaluated (Tables 1 and 2).
TABLE 1.
Characteristics and adjusted odds ratios of pyoderma gangrenosum (PG) versus non‐PG cohorts in the All of Us (AoU) electronic health record (EHR).
| Parameter | PG, n = 165 (%) | Non‐PG, n = 266 447 (%) | p value | Rate of PG diagnosis % (95% CI) | Age‐adjusted OR (95% CI) | Multivariable adjusted OR (95% CI)* |
|---|---|---|---|---|---|---|
| Race | 0.340 | |||||
| White | 101 (61.21) | 145 528 (54.61) | 0.07 (0.06–0.08) | Reference | Reference | |
| Asian | 2 (1.21) | 7382 (2.77) | 0.03 (0.00–0.11) | 0.40 (0.10–1.62) | 0.43 (0.10–1.74) | |
| Black/African American | 31 (18.78) | 52 846 (19.83) | 0.06 (0.04–0.08) | 0.86 (0.57–1.29) | 0.71 (0.46–1.11) | |
| Other a | ≤20 h | 9296 (3.48) | N/A | 0.47 (0.15–1.49) | 0.38 (0.09–1.56) | |
| Unknown b | ≤20 h | 7374 (2.76) | N/A | ‐ | ‐ | |
| Hispanic only c | 26 (15.75) | 44 021 (16.52) | 0.06 (0.04–0.09) | 0.87 (0.56–1.35) | 0.38 (0.17–0.85) | |
| Ethnicity | 0.562 | |||||
| Not Hispanic | 128 (77.57) | 206 438 (77.47) | 0.06 (0.05–0.07) | Reference | Reference | |
| Hispanic/Latino | 34 (20.6) | 49 875 (18.71) | 0.07 (0.05–0.10) | 1.13 (0.77–1.66) | 2.25 (1.09–4.65) | |
| None of these | ≤20 h | 2759 (1.03) | N/A | 0.59 (0.08–4.23) | 1.27 (0.11–14.07) | |
| Unknown b | ≤20 h | 7375 (2.76) | N/A | ‐ | ‐ | |
| Age | 0.474 | |||||
| ≤65 | 109 (66.06) | 183 692 (68.94) | 0.06 (0.05–0.07) | Reference | Reference | |
| >65 | 56 (33.93) | 82 755 (31.05) | 0.07 (0.05–0.09) | 1.14 (0.83–1.58 | 0.91 (0.63–1.31) | |
| Gender identity | 0.056 | |||||
| Male | 46 (27.87) | 98 287 (36.88) | 0.05 (0.03–0.06) | Reference | Reference | |
| Female | 114 (69.09) | 160 757 (60.33) | 0.07 (0.06–0.09) | 1.51 (1.07–2.12) | 1.51 (1.07–2.13) | |
| Other d | ≤20 h | 7403 (2.77) | N/A | 2.39 (0.95–6.03) | 2.14 (0.83–5.52) | |
| LGBTQ identification | 0.858 | |||||
| Straight | 144 (87.27) | 234 550 (88.02) | 0.06 (0.05–0.07) | Reference | Reference | |
| Not straight | 21 (12.72) | 31 897 (11.97) | 0.07 (0.04–0.10) | 1.23 (0.78–1.95) | 1.11 (0.69–1.78) | |
| Education | 0.874 | |||||
| College graduate | 66 (40.0) | 112 256 (42.13) | 0.06 (0.05–0.07) | Reference | Reference | |
| Grade 12/GED + College 1–3 year | 78 (47.27) | 120 411 (45.19) | 0.06 (0.05–0.08) | 1.13 (0.81–1.57) | 0.87 (0.60–1.26) | |
| Less than high School | 17 (10.3) | 25 436 (9.54) | 0.07 (0.04–0.11) | 1.19 (0.69–2.03) | 0.84 (0.45–1.55) | |
| Other b | ≤20 h | 8344 (3.13) | N/A | 1.30 (0.47–3.57) | 1.24 (0.43–3.54) | |
| Occupation | <0.001 | |||||
| Employed | 44 (26.66) | 112 209 (42.11) | 0.04 (0.03–0.05) | Reference | Reference | |
| Not employed | 113 (68.48) | 133 192 (49.98) | 0.08 (0.07–0.10) | 2.26 (1.57–3.25) | 2.00 (1.34–2.98) | |
| Other b | ≤20 h | 21 046 (7.89) | N/A | 1.00 (0.45–2.23) | 0.94 (0.42–2.12) | |
| Health insurance | 0.060 | |||||
| Yes | 159 (96.36) | 242 777 (91.11) | 0.07 (0.06–0.08) | Reference | Reference | |
| No | ≤20 h | 14 766 (5.54) | N/A | 0.43 (0.16–1.16) | 0.42 (0.15–1.14) | |
| Other b | ≤20 h | 8904 (3.34) | N/A | 0.52 (0.13–2.09) | 0.52 (0.13–2.14) | |
| Disability status | 0.039 | |||||
| No disability e | 32 (19.39) | 70 099 (26.3) | 0.05 (0.03–0.06) | Reference | Reference | |
| Disability f | 23 (13.93) | 25 438 (9.54) | 0.09 (0.06–0.14) | 2.03 (1.18–3.49) | 1.61 (0.92–2.79) | |
| Unknown g | 110 (66.66) | 170 910 (64.14) | 0.06 (0.05–0.08) | 1.45 (0.97–2.16) | 1.33 (0.89–2.00) | |
| Income | 0.001 | |||||
| >25k | 74 (44.84) | 141 522 (53.11) | 0.05 (0.04–0.07) | Reference | Reference | |
| <25k | 63 (38.18) | 68 406 (25.67) | 0.09 (0.07–0.12) | 1.83 (1.30–2.57) | 1.63 (1.08–2.47) | |
| Unknown d | 28 (16.96) | 56 519 (21.21) | 0.05 (0.03–0.07) | 1.03 (0.67–1.60) | 1.01 (0.63–1.63) |
Note: ‐, dropped variable in age and multivariate regression calculations due to high multicollinearity.
Adjusted for all demographic factors and comorbidities assessed.
Another single population, more than one population, none of these.
Prefer not to answer, skip and/or N/A.
Participants who selected Hispanic only for Race question.
Not man only, not woman only, prefer not to answer or skipped.
Answered ‘no’ to each and every disability question including deaf, blind, difficulty concentrating, difficulty walking/climbing, difficulty dressing/bathing, and difficulty doing errands alone.
Answered ‘yes’ to at least one of the disability questions.
Did not answer at least one of the disabilities questions.
As per AoU database reporting guidelines, all values under 20 must be represented as ≤20.
TABLE 2.
Prevalence of selected comorbidities of pyoderma gangrenosum (PG) vs. non‐PG cohorts in the All of Us (AoU) electronic health record (EHR).
| Comorbidity | PG, n = 165 (%) | Non‐PG, n = 266 447 (%) | p value | Rate of PG diagnosis % (95% CI) |
|---|---|---|---|---|
| PG specific | ||||
| Acute renal failure | 69 (41.82) | 23 420 (8.79) | <0.001 | 0.29 (0.23–0.37) |
| Anaemia | 116 (70.30) | 71 282 (26.75) | <0.001 | 0.16 (0.14–0.19) |
| Crohn's disease | 54 (32.73) | 2943 (1.10) | <0.001 | 1.80 (1.38–2.35) |
| Gastroesophageal reflux disease | 99 (60.00) | 77 531 (29.10) | <0.001 | 0.13 (0.10–0.16) |
| Immunodeficiency disorder | 42 (25.45) | 9015 (3.38) | <0.001 | 0.46 (0.34–0.63) |
| Neoplasm | 124 (75.15) | 120 467 (45.21) | <0.001 | 0.10 (0.09–0.12) |
| Neuropathy | 94 (56.97) | 88 525 (33.22) | <0.001 | 0.11 (0.09–0.13) |
| Rheumatoid arthritis | 35 (21.21) | 8029 (3.01) | <0.001 | 0.43 (0.31–0.60) |
| Sepsis | 54 (32.73) | 15 201 (5.71) | <0.001 | 0.35 (0.27–0.46) |
| Ulcerative colitis | 39 (23.64) | 2829 (1.06) | <0.001 | 1.36 (0.99–1.86) |
| General | ||||
| Anxiety | 77 (46.67) | 83 478 (31.33) | <0.001 | 0.09 (0.07–0.12) |
| Arthritis | 76 (46.06) | 43 888 (16.47) | <0.001 | 0.17 (0.14–0.22) |
| Asthma | 50 (30.30) | 46 890 (17.60) | <0.001 | 0.11 (0.08–0.14) |
| Atrial fibrillation | 23 (13.94) | 17 626 (6.62) | <0.001 | 0.13 (0.09–0.20) |
| CKD | 51 (30.91) | 25 436 (9.55) | <0.001 | 0.20 (0.15–0.26) |
| Heart disease | 105 (63.64) | 83 581 (31.37) | <0.001 | 0.13 (0.10–0.15) |
| Hyperlipidaemia | 91 (55.15) | 107 629 (40.39) | <0.001 | 0.08 (0.07–0.10) |
| Hypertension | 116 (70.30) | 117 617 (44.14) | <0.001 | 0.10 (0.08–0.12) |
| Hyperthyroidism | ≤20 a | 7394 (2.78) | <0.001 | |
| Hypothyroidism | 35 (21.21) | 37 039 (13.90) | 0.007 | 0.09 (0.07–0.13) |
| Insomnia | 46 (27.88) | 37 719 (14.16) | <0.001 | 0.12 (0.09–0.16) |
| Major depressive disorder | 78 (47.27) | 62 710 (23.54) | <0.001 | 0.12 (0.10–0.16) |
| Obesity | 95 (57.58) | 76 314 (28.64) | <0.001 | 0.12 (0.10–0.15) |
| Type 2 diabetes | 72 (43.64) | 52 494 (19.70) | <0.001 | 0.14 (0.11–0.17) |
Abbreviation: CKD, chronic kidney disease.
As per AoU database reporting guidelines, all values under 20 must be represented as ≤20.
There were 165 PG patients out of 266 612 total patients with electronic health record (EHR) data (overall prevalence 0.05%, 95% confidence interval [CI] 0.005–0.007). PG was more prevalent in lower income brackets (odds ratio [OR] 1.63, 95% CI 1.08–2.47), unemployed (OR 2.00, 95% CI 1.34–2.98), females (1.51, 95% CI 1.07–2.13), and Hispanic/Latinx (OR 2.25, 95% CI 1.09–4.65) populations (Table 1). There was no relationship between PG and age, LGBTQ identity, education, and health insurance status. Therewas a higher rate of all comorbidities assessed in PG including gastrointestinal diseases, arthritis, cardiovascular diseases, and psychiatric illness (p < 0.001) (Table 2). There was a much higher prevalence of arthritis, hypertension, obesity, and type 2 diabetes mellitus in PG patients (p < 0.001).
PG has been associated with lower socio‐economic status (SES), and our findings indicate a higher risk of PG among patients with low income, unemployment, and disability. Similar to other studies, we find higher odds of PG in females; however, we also find higher odds in Hispanic/Latinx populations. This may be another manifestation of lower SES, as Hispanics comprise a significantly lower socio‐economic status compared with non‐Hispanic Whites. 2 Social determinants of health, such as access to care, income, education, and health insurance, amplify both the burden of disease and healthcare costs. 3 As PG is a diagnosis of exclusion, misdiagnosis potentiates unnecessary management, leading to longer time to resolution and increases in medical costs. 4 These findings may aid in future studies elucidating the pathophysiology of disease and identification of vulnerable patients.
Low SES is also linked with a higher rate of cardiovascular disease, and we report higher rates of hypertension, heart disease, and atrial fibrillation in PG patients. Given the disease burden and disability of PG, depression is more commonly reported in PG patients than in the general population. Our study presents the first evidence of a higher prevalence of anxiety and insomnia in PG patients compared with non‐PG controls. 5
Limitations of the present study include being dependent on EHRs, which may introduce selection bias, misclassification bias, or confounding by indication, leading to decreased generalizability of these findings. However, the size and diversity of the AoU cohort is an important strength. Increased rates of PG in patients with medical comorbidities may delay diagnosis, thus clinicians should be mindful of underdiagnosis in these populations. In light of this, dermatologists are encouraged to adopt an interdisciplinary approach. Collaborating with general practitioners and social workers can be particularly beneficial for patients with PG who are of low SES and/or unemployed.
FUNDING INFORMATION
No funding was provided for this study.
CONFLICT OF INTEREST STATEMENT
Dr. Rao is a speaker for Incyte. The other authors have no conflicts of interest to disclose.
ACKNOWLEDGEMENTS
The All of Us Research Program is supported by the National Institutes of Health, Office of the Director: Regional Medical Centers: 1 OT2 OD026549; 1 OT2 OD026554; 1 OT2 OD026557; 1 OT2 OD026556; 1 OT2 OD026550; 1 OT2 OD 026552; 1 OT2 OD026553; 1 OT2 OD026548; 1 OT2 OD026551; 1 OT2 OD026555; National Institutes of Health, IAA #: AOD 16037; Federally Qualified Health Centers: HHSN 263201600085U; Data and Research Center: 5 U2C OD023196; Biobank: 1U24 OD023121; The Participant Center: U24 OD023176; Participant Technology Systems Center: 1U24 OD023163; Communications and Engagement: 3 OT2 OD023205; 3 OT2 OD023206; and Community Partners: 1 OT2 OD025277; 3 OT2 OD025315; 1 OT2 OD025337; 1 OT2 OD025276. In addition, the All of Us Research Program would not be possible without the partnership of its participants.
DATA AVAILABILITY STATEMENT
Datasets related to this article can be found at https://allofus.nih.gov, hosted at the National Institutes of Health All of Research Program All of Us Research Program.” National Institutes of Health, U.S. Department of Health and Human Services, https://allofus.nih.gov/.
REFERENCES
- 1. Monari P, Moro R, Motolese A, et al. Epidemiology of pyoderma gangrenosum: results from an Italian prospective multicentre study. Int Wound J. 2018;15(6):875‐879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Morales LS, Lara M, Kington RS, Valdez RO, Escarce JJ. Socioeconomic, cultural, and behavioral factors affecting Hispanic health outcomes. J Health Care Poor Underserved. 2002;13(4):477‐503. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Braveman P, Gottlieb L. The social determinants of health: it's time to consider the causes of the causes. Public Health Rep. 2014;129(Suppl 2):19‐31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Na HD, Yan H, Park CH. Misdiagnosis of pyoderma gangrenosum increases medical costs and prolongs hospital stay: a case report. Int J Low Extrem Wounds. 2023;22(2):428‐433. [DOI] [PubMed] [Google Scholar]
- 5. McPhie ML, Fletcher J, Machado MO, Carvalho AF, Piguet V, Alavi A. A systematic review of depression and anxiety in adults with pyoderma gangrenosum. Adv Skin Wound Care. 2021;34(8):432‐436. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Datasets related to this article can be found at https://allofus.nih.gov, hosted at the National Institutes of Health All of Research Program All of Us Research Program.” National Institutes of Health, U.S. Department of Health and Human Services, https://allofus.nih.gov/.