This population-based analysis establishes standardized overall and sex- and age-specific prevalence estimates for hidradenitis suppurativa in the United States.
Key Points
Questions
What is the prevalence of hidradenitis suppurativa in the United States, and does prevalence vary by demographic group?
Findings
In this population-based analysis, the overall point prevalence of hidradenitis suppurativa was 0.10%, or 98 per 100 000 persons. Prevalence was highest among women (137 per 100 000), those aged 30 to 39 years (172 per 100 000), and African American (296 per 100 000) and biracial (218 per 100 000) patient groups.
Meaning
Hidradenitis suppurativa is an uncommon, but not rare, disease in the United States that disproportionately affects female patients, young adults, and African American and biracial patients.
Abstract
Importance
The true prevalence of hidradenitis suppurativa (HS) is unknown.
Objective
To establish standardized overall and group-specific prevalence estimates for HS in the United States.
Design, Setting, and Participants
This retrospective analysis included a demographically heterogeneous population-based sample of more than 48 million unique patients across all US census regions. As of October 27, 2016, a total of 47 690 patients with HS were identified using electronic health record data.
Main Outcomes and Measures
Standardized overall point prevalence for HS and sex-, age-, and race-specific prevalence estimates of HS in the general US population.
Results
Of the 47 690 patients with HS (26.2% men and 73.8% women), the overall HS prevalence in the US population sample was 0.10%, or 98 per 100 000 persons (95% CI, 97-99 per 100 000 persons). The adjusted prevalence in women was 137 per 100 000 (95% CI, 136-139 per 100 000), more than twice that of men (58 per 100 000; 95% CI, 57-59 per 100 000; P < .001). The prevalence of HS was highest among patients aged 30 to 39 years (172 per 100 000; 95% CI, 169-275 per 100 000) compared with all other age groups (range, 15-150 per 100 000; P < .001). Adjusted HS prevalences among African American (296 per 100 000; 95% CI, 291-300 per 100 000) and biracial (218 per 100 000; 95% CI, 202-235 per 100 000) patients were more than 3-fold and 2-fold greater, respectively, than that among white patients (95 per 100 000; 95% CI, 94-96 per 100 000; P < .001).
Conclusions and Relevance
Hidradenitis suppurativa is an uncommon, but not rare, disease in the United States that disproportionately affects female patients, young adults, and African American and biracial patients.
Introduction
Hidradenitis suppurativa (HS) is a debilitating, painful chronic inflammatory disease for which the inherent unpredictability with respect to the course of disease and response to treatment poses significant challenges for patients. With sparse epidemiologic data for HS, disease burden has not yet been established. The prevalence of HS has been estimated using claims analyses, examinations, medical record reviews, or patient-administered questionnaires. Most prevalence studies have involved selected samples and/or small cohorts, which limit generalizability. Because of the different data acquisition methods, study settings, and geographic samples in these analyses, prevalence estimates have, not surprisingly, varied significantly across the world. Overall prevalence estimates for HS in Europe and in the United States have ranged from 0.00033% to as high as 4.1%. Few adjusted population analyses on which HS prevalence may be accurately estimated have been performed. The true overall prevalence of HS is therefore unknown. The purpose of our study was to establish standardized prevalence estimates for HS in the overall US population and in specific demographic subgroups.
Methods
Case Identification
We performed a retrospective case cohort analysis by querying a multi-institutional data analytics and research platform (Explorys) developed by IBM Watson Health. Clinical information from electronic medical records, laboratories, practice management systems, and claims systems is matched using the single set of Unified Medical Language System ontologies to create longitudinal records for each unique patient. Human Subjects Committee review was waived by the Feinstein Institute for Medical Research at Northwell Health because no idendifiers are associated with the data.
The data were standardized and curated according to common controlled vocabularies and classifications systems, including International Classification of Diseases, Ninth Revision (ICD-9); Systemized Nomenclature of Medicine–Clinical Terms (SNOMED-CT); RxNorm; Logical Observation Identifiers Names and Codes (LOINC); and Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. At present, the database encompasses 27 participating integrated health care organizations. More than 48 million unique lives, representing approximately 15% of the population across all 4 census regions of the United States, are captured. Patients with all types of insurance and those who self-pay are represented. The earliest clinical observations in the data system are from 1999. The database is dynamic and performs periodic automatic updates. Population counts are reported by the database to the nearest 10 or represented as less than 10 if they range from 0 to 9 to prevent patient identification. The platform has been applied to validation and epidemiologic studies across other disciplines.
In this analysis, all active nondeceased patients in the database at the time of our query (October 1, 2016) were included in this study. The SNOMED-CT term hidradenitis has one-to-one mapping to the ICD-9 code for HS (705.83) and was used to identify patients with the ICD-9 code. To validate use of this administrative coding to identify the HS case cohort in the database, we reviewed a random sample of 150 patient records linked to the ICD-9 diagnosis code for HS at Northwell Health, which is a member of the data consortium. Two trained medical students extracted clinical information documented by the health care professionals in each medical record. Two dermatologists experienced in evaluating HS (including A.G.) independently reconciled the clinical information for each patient against a criterion checklist for the case definition adopted by the 2nd International Conference on Hidradenitis Suppurativa. Cases with disagreement on HS diagnosis were reconciled by discussion between the dermatologists. In this cohort, a single ICD-9 code for HS yielded a positive predictive value of 79.3% for diagnosis. This case identification method was also validated previously and independently validated, and it was shown to have a positive predictive value of 77%.
Statistical Analysis
We calculated frequencies describing the demographic characteristics of the HS cohort identified from January 1, 1999, through October 1, 2016. Sex- and age-specific HS prevalences were calculated for the following 4 racial groups: white, African American, biracial (ie, white and African American), and other (including Asian, Pacific Islander, Native American, Alaskan Native, Latin American, Native Hawaiian, and unknown). The entire nondeceased population for each racial group was considered at risk. To calculate sex- and age-specific HS prevalences for the overall population, the number of HS cases and total population size were collapsed across the 4 racial groups. We applied direct standardization to allow for subgroup comparisons using the projected US population in 2000 as the standard population and the following 6 age groups: 0 to 17, 18 to 29, 30 to 39, 40 to 49, 50 to 59, and 60 years or older. Estimates were age adjusted for sex comparisons, sex adjusted for age group comparisons, and sex and age adjusted for race comparisons. Crude and standardized prevalence estimates were calculated based on patients with available race, sex, and age information. Confidence intervals for crude and standardized prevalences were computed using the Wilson score method and gamma distribution, respectively. Standardized prevalences were compared assuming the prevalence ratio follows a log-normal distribution. Age-specific prevalences were compared using the χ2 test. A 2-sided α level of .05 was applied to determine statistical significance. All analyses were performed using SAS software (version 9.4; SAS Institute, Inc).
Results
Table 1 lists the demographic characteristics of patients with HS in the United States. We identified 47 690 patients with HS as of October 27, 2016. Women with HS outnumbered men with HS by nearly 3 to 1 (73.8% female and 26.2% male). Patients with HS were most frequently in the fourth decade of life. Patients aged 0 to 17 years represented 2.2% of the HS cohort. Of note, 30.1% of the HS cohort was 50 years or older. Patients with HS were most frequently white (54.1%), and of importance, nearly one-third were African American.
Table 1. Demographic Characteristics of Patients With Hidradenitis Suppurativa in the United States.
| Characteristic | No. (%) of Patients (N = 47 690)a |
|---|---|
| Sex | |
| Male | 12 480 (26.2) |
| Female | 35 210 (73.8) |
| Age, y | |
| 0-17 | 1060 (2.2) |
| 18-29 | 10 460 (21.9) |
| 30-39 | 12 160 (25.5) |
| 40-49 | 9650 (20.2) |
| 50-59 | 8030 (16.8) |
| ≥60 | 6330 (13.3) |
| Raceb | |
| White | 25 780 (54.1) |
| African American | 15 540 (32.6) |
| Biracialc | 740 (1.6) |
| Otherd | 5640 (11.8) |
Percentages have been rounded and may not total 100.
Sum of all race counts totals 47 000 because the database rounds counts to the nearest 10.
Includes patients who are white and African American.
Includes Asian, Pacific Islander, Native American, Alaskan Native, Latin American, Native Hawaiian, and unknown.
Overall and adjusted group-specific HS prevalences in the United States are listed in Table 2. We observed an overall standardized point prevalence of 0.10%, or 98 patients with HS per 100 000 persons in the United States (95% CI, 97-99 per 100 000 persons). Female patients had a greater than 2-fold prevalence of HS (137 per 100 000; 95% CI, 136-139 per 100 000) than male patients (58 per 100 000; 95% CI, 57-59 per 100 000; P < .001). The prevalence of HS among patients aged 30 to 39 years (172 per 100 000; 95% CI, 169-175 per 100 000) was significantly higher than that among all other age groups (range, 15-150 per 100 000; P < .001 for all pairwise comparisons). The standardized prevalence of HS in the pediatric age group was 15 per 100 000 (95% CI, 14-16 per 100 000). The standardized prevalences of HS among African American (296 per 100 000; 95% CI, 291-300 per 100 000) and biracial (218 per 100 000; 95% CI, 202-235 per 100 000) patients were 3-fold and 2-fold greater, respectively, than that among white patients (95 per 100 000; 95% CI, 94-96 per 100 000; P < .001 for both comparisons).
Table 2. Overall and Adjusted Group-Specific Prevalence Estimates for HS in the United States.
| Characteristic | No. of HS Cases | Population Size | Prevalence per 100 000 Persons (95% CI) | Standardized Prevalence, % | |
|---|---|---|---|---|---|
| Crude | Standardizeda,b | ||||
| All | 47 620 | 45 855 960 | 104 (103-105) | 98 (97-99) | 0.10 |
| Sexc | |||||
| Male | 12 430 | 20 832 510 | 60 (59-61) | 58 (57-59) | 0.06 |
| Female | 35 190 | 25 023 450 | 141 (139-142) | 137 (136-139) | 0.14 |
| Age, yc | |||||
| 0-17 | 1020 | 7 058 400 | 14 (14-15) | 15 (14-16) | 0.02 |
| 18-29 | 10 480 | 6 750 400 | 155 (152-158) | 150 (147-152) | 0.15 |
| 30-39 | 12 160 | 6 691 550 | 182 (178-185) | 172 (169-175) | 0.17 |
| 40-49 | 9650 | 6 202 300 | 156 (152-159) | 150 (147-153) | 0.15 |
| 50-59 | 8030 | 6 596 900 | 122 (119-124) | 119 (117-122) | 0.12 |
| ≥60 | 6280 | 12 556 410 | 50 (49-51) | 49 (48-51) | 0.05 |
| Racec | |||||
| White | 25 790 | 26 318 800 | 98 (97-99) | 95 (94-96) | 0.10 |
| African American | 15 520 | 4 695 420 | 331 (325-336) | 296 (291-300) | 0.30 |
| Biraciald | 720 | 317 300 | 227 (210-243) | 218 (202-235) | 0.22 |
| Othere | 5590 | 14 524 440 | 38 (37-39) | 36 (35-37) | 0.04 |
Abbreviation: HS, hidradenitis suppurativa.
Sex comparisons are adjusted for age. Age group comparisons are adjusted for sex. Race comparisons are adjusted for sex and age. The projected 2000 US population was used as the standard population, with 6 groups aged 0 to 17, 18 to 29, 30 to 39, 40 to 49, 50 to 59, and 60 years or older.
All comparisons between standardized subgroups were significant with a P < .001. Reference groups for sex, age, and race comparisons were female, aged 30 to 39 years, and white, respectively.
Group-specific results are based on records for which age and sex information was available. As a result, group-specific counts may not add to totals for all patients in the database.
Includes patients who are white and African American.
Includes Asian, Pacific Islander, Native American, Alaskan Native, Latin American, Native Hawaiian, and unknown.
Age-specific HS prevalences by race are listed in Table 3. The prevalence of HS was greater among African American (range, 40-541 per 100 000) and biracial (range, 20-428 per 100 000) patients than among white patients (range, 14-176 per 100 000) across all age-groups (P < .001), except for biracial patients aged 0 to 17 years (20 per 100 000; 95% CI, 11-37 per 100 000; P = .20). The highest HS prevalence among African American (541 per 100 000; 95% CI, 525-559 per 100 000) and biracial (428 per 100 000; 95% CI, 370-495 per 100 000) patients occurred in the group aged 30 to 39 years.
Table 3. Age-Specific Prevalence of HS by Racea.
| Age Group, y | White | African American | Biracialb | Otherc | ||||
|---|---|---|---|---|---|---|---|---|
| No. of HS Cases | Prevalence per 100 000 Persons (95% CI) | No. of HS Cases | Prevalence per 100 000 Persons (95% CI) | No. of HS Cases | Prevalence per 100 000 Persons (95% CI) | No. of HS Cases | Prevalence per 100 000 Persons (95% CI) | |
| 0-17 | 500 | 14 (12-15) | 340 | 40 (36-44) | 10 | 20 (11-37) | 170 | 7 (6-8) |
| 18-29 | 5290 | 145 (141-149) | 3570 | 436 (422-450) | 160 | 326 (279-380) | 1460 | 66 (62-69) |
| 30-39 | 6240 | 176 (172-180) | 4050 | 541 (525-558) | 180 | 428 (370-495) | 1690 | 72 (68-75) |
| 40-49 | 5290 | 155 (151-160) | 3110 | 455 (439-471) | 160 | 390 (334-455) | 1090 | 53 (50-56) |
| 50-59 | 4550 | 116 (112-119) | 2640 | 391 (376-406) | 120 | 261 (218-311) | 720 | 37 (35-40) |
| ≥60 | 3920 | 49 (47-50) | 1810 | 197 (188-206) | 90 | 100 (82-123) | 460 | 13 (12-14) |
Abbreviation: HS, hidradenitis suppurativa.
P < .001 for all age-specific comparisons with white race as the reference, excluding only biracial vs white persons aged 0 to 17 years (P = .20).
Includes patients who are white and African American.
Includes Asian, Pacific Islander, Native American, Alaskan Native, Latin American, Native Hawaiian, and unknown.
Discussion
In this study, we have determined overall and group-specific adjusted prevalence estimates for HS in the United States. On the basis of sample size, demographic diversity of the prevalence cohort, and standardization of prevalence estimates using 12 age- and sex-specific strata, we believe that these results may be generalized to the US population.
The overall point prevalence of HS is 0.10%, or 98 HS patients per 100 000 persons in the United States. The prevalence of HS among female patients was more than 2-fold greater than in male patients. The prevalence of HS was greatest among patients aged 30 to 39 years and similar to the prevalence in patients aged 18 to 29 years and 40 to 49 years. The standardized prevalence of HS in the pediatric population was significantly lower than in the overall population or in all other age groups. An additional finding in this study is that HS disproportionately affects African American and biracial patients, who have 3-fold and 2-fold greater prevalences, respectively, than white patients. Although this disparity was observed across all age groups except biracial patients aged 0 to 17 years, the highest prevalence for African American and biracial patients was in the group aged 30 to 39 years. To date, we are not aware of other population-based prevalence estimates of HS among nonwhite patients.
Few adjusted population-based studies evaluating HS prevalence in the United States have been performed. One claims-based study identified 7927 patients with HS among a population of 15 million persons in 2007. The overall prevalence of HS was estimated to be 53 per 100 000, or 0.053%, which was similar to prevalence estimates adjusted for sex (0.052%) and age (0.051%). The prevalence of HS among women (75 per 100 000; 0.075%) was 3 times that among men (29 per 100 000; 0.029%). Prevalence estimates for patients with HS aged 18 to 29 years was 85 per 100 000 (0.085%); for patients aged 30 to 44 years, 87 per 100 000 (0.087%). Prevalence in the pediatric population was 17 per 100 000 (0.017%). However, that analysis estimated prevalence based on 4 case definition criteria, 3 of which included Current Procedural Terminology codes for procedures that were not all specific to a diagnosis of HS. The case cohort captured only commercially insured patients, and thus many demographic groups may have been underrepresented in their sample. Moreover, their analysis did not include a description of race.
Among 178 HS cases in Olmstead County, the total sex- and age-adjusted point prevalence was 127.8 per 100 000 (0.13%). Disease prevalence among women (189.5 per 100 000; 0.19%) was roughly 3 times that among men (65.5 per 100 000; 0.066%). The age range with the highest overall prevalence was 60 to 69 years (201.8 per 100 000; 0.20%), with only 24 cases represented in this subgroup. The highest subgroup prevalence was among women aged 20 to 29 years (307.7 per 100 000; 0.31%). With only 2 cases in the group aged 0 to 19 years, the observed prevalence was 5.1 per 100 000 (0.005%). These results may have limited generalizability because the analysis was limited to patients in southeastern Minnesota and nearly 88% of the patients with HS were white. Small case cohort size may also have limited subgroup analyses by age and race.
Population-based prevalence estimates from Europe are generally 1 order of magnitude higher than those from the United States. Two study distinctions include smaller sample sizes and use of interviews or questionnaires to identify cases in European analyses. In a French study, 1 survey question that had not been previously validated was used to identify cases in a representative sample of the population. Of the 10 000 persons surveyed, 68.9% completed the survey. Sixty-seven cases of HS were identified, yielding an adjusted prevalence of 1%. In a Danish analysis, prevalence was evaluated by survey of a representative sample of 8000 persons. Among 585 respondents, 6 self-reported a history compatible with HS, yielding a 1-year prevalence of 1%. In another Danish population analysis, a validated self-administered questionnaire was used to identify HS cases among 17 454 representative participants aged 30 to 89 years. The overall prevalence in that analysis was observed to be 2.1%. The prevalence of HS among women (2.56%) was roughly 1.5 times that among men (1.58%). Prevalence was greatest among patients aged 30 to 39 years (3.91%) and declined thereafter by the decade. Data for patients younger than 30 years were not included, and almost all patients with HS were white. In another cohort in the Danish analysis, the prevalence of HS was observed to be 4.1%. Cases were identified by examination of 507 consecutive patients evaluated in a screening clinic for sexually transmitted diseases.
Limitations
Some limitations to our analysis bear consideration. We could not capture patients with undiagnosed HS or who did not seek care in health systems included in the database. Diagnosis of HS could not be verified in medical records for the entire universe of data because they are deidentified. The extent to which our analysis may have overestimated or underestimated true prevalence is not known. However, we are reassured that our overall prevalence approximates that of the Olmstead County analysis in which cases were validated.
Conclusions
We have estimated that HS is an uncommon, but not rare, disease in the United States that disproportionately affects female patients, young adults, and African American and biracial patients. However, costs and utilization of health care services among patients with HS rival or exceed those of patients with psoriasis, a common skin disease.
References
- 1.Lookingbill DP. Yield from a complete skin examination: findings in 1157 new dermatology patients. J Am Acad Dermatol. 1988;18(1, pt 1):31-37. [DOI] [PubMed] [Google Scholar]
- 2.Jemec GB, Heidenheim M, Nielsen NH. The prevalence of hidradenitis suppurativa and its potential precursor lesions. J Am Acad Dermatol. 1996;35(2, pt 1):191-194. [DOI] [PubMed] [Google Scholar]
- 3.Jemec GB. The symptomatology of hidradenitis suppurativa in women. Br J Dermatol. 1988;119(3):345-350. [DOI] [PubMed] [Google Scholar]
- 4.von der Werth JM, Williams HC. The natural history of hidradenitis suppurativa. J Eur Acad Dermatol Venereol. 2000;14(5):389-392. [DOI] [PubMed] [Google Scholar]
- 5.Revuz JE, Canoui-Poitrine F, Wolkenstein P, et al. . Prevalence and factors associated with hidradenitis suppurativa: results from two case-control studies. J Am Acad Dermatol. 2008;59(4):596-601. [DOI] [PubMed] [Google Scholar]
- 6.Jemec GB. Clinical practice: hidradenitis suppurativa. N Engl J Med. 2012;366(2):158-164. [DOI] [PubMed] [Google Scholar]
- 7.Vinding GR, Miller IM, Zarchi K, Ibler KS, Ellervik C, Jemec GB. The prevalence of inverse recurrent suppuration: a population-based study of possible hidradenitis suppurativa. Br J Dermatol. 2014;170(4):884-889. [DOI] [PubMed] [Google Scholar]
- 8.Cosmatos I, Matcho A, Weinstein R, Montgomery MO, Stang P. Analysis of patient claims data to determine the prevalence of hidradenitis suppurativa in the United States. J Am Acad Dermatol. 2013;69(5):819. [DOI] [PubMed] [Google Scholar]
- 9.Shlyankevich J, Chen AJ, Kim GE, Kimball AB. Hidradenitis suppurativa is a systemic disease with substantial comorbidity burden: a chart-verified case-control analysis. J Am Acad Dermatol. 2014;71(6):1144-1150. [DOI] [PubMed] [Google Scholar]
- 10.Shahi V, Alikhan A, Vazquez BG, Weaver AL, Davis MD. Prevalence of hidradenitis suppurativa: a population-based study in Olmsted County, Minnesota. Dermatology. 2014;229(2):154-158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Explorys Inc The IBM Explorys Platform. Somers, NY: IBM Corporation; June 2016. https://public.dhe.ibm.com/common/ssi/ecm/hp/en/hps03052usen/HPS03052USEN.PDF. Accessed January 18, 2016.
- 12.US National Library of Medicine Unified Medical Language System (UMLS) Systematized Nomenclature of Medicine—ClinicalsTerms (SNOMED CT). https://www.nlm.nih.gov/research/umls/Snomed/snomed_main.html. Published March 1, 2016. Accessed October 10, 2016.
- 13.Nelson SJ, Zeng K, Kilbourne J, Powell T, Moore R. Normalized names for clinical drugs: RxNorm at 6 years. J Am Med Inform Assoc. 2011;18(4):441-448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.McDonald CJ, Huff SM, Suico JG, et al. . LOINC, a universal standard for identifying laboratory observations: a 5-year update. Clin Chem. 2003;49(4):624-633. [DOI] [PubMed] [Google Scholar]
- 15.Kaelber DC, Foster W, Gilder J, Love TE, Jain AK. Patient characteristics associated with venous thromboembolic events: a cohort study using pooled electronic health record data. J Am Med Inform Assoc. 2012;19(6):965-972. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kim CH, Al-Kindi SG, Jandali B, Askari AD, Zacharias M, Oliveira GH. Incidence and risk of heart failure in systemic lupus erythematosus. Heart. 2017;103(3):227-233. [DOI] [PubMed] [Google Scholar]
- 17.US Census Bureau Projections of the Total Resident Population by 5-Year Age Groups, and Sex with Special Age Categories: Middle Series, 1999 to 2000 (NP-T3-A): http://www.census.gov/population/projections/files/natproj/summary/np-t3-a.pdf Published January 13, 2000. Accessed October 31, 2016.
- 18.Wilson EB. Probable inference, the law of succession, and statistical inference. J Am Stat Assoc. 1927;22(158):209-212. [Google Scholar]
- 19.Fay MP, Feuer EJ. Confidence intervals for directly standardized rates: a method based on the gamma distribution. Stat Med. 1997;16(7):791-801. [DOI] [PubMed] [Google Scholar]
- 20.Khalsa A, Liu G, Kirby JS. Increased utilization of emergency department and inpatient care by patients with hidradenitis suppurativa. J Am Acad Dermatol. 2015;73(4):609-614. [DOI] [PubMed] [Google Scholar]
- 21.Kirby JS, Miller JJ, Adams DR, Leslie D. Health care utilization patterns and costs for patients with hidradenitis suppurativa. JAMA Dermatol. 2014;150(9):937-944. [DOI] [PubMed] [Google Scholar]