Prevalence Estimates for Pemphigus in the United States

Abstract

This cross-sectional analysis using Explorys data examines the prevalence of pemphigus in US adults across age, sex, and race groups.

Pemphigus represents a subset of bullous disease in which autoantibodies disrupt intercellular keratinocyte adhesion.¹ Loss of epidermal integrity clinically translates to flaccid blisters and erosions of skin and mucous membranes. Although treatment with systemic corticosteroids has mitigated its life-threatening course, pemphigus remains associated with significant morbidity and mortality.² Although the national prevalence of pemphigoid has recently been described,³ the burden of pemphigus is not well characterized, particularly in the United States.

Methods

We conducted a cross-sectional analysis using Explorys, a multi-institutional data analytics and research platform developed by IBM Watson Health, to establish standardized overall and subgroup prevalence estimates for pemphigus in the United States. At present, Explorys captures more than 55 million unique lives across all 4 census regions of the United States. To identify the case cohort, we used 2 counts of the Systematized Nomenclature of Medicine–Clinical Terms word pemphigus, which is associated with the International Classification of Diseases, Ninth Revision (ICD-9) code for pemphigus (694.4) that includes diagnoses codes for pemphigus vulgaris, pemphigus foliaceus, pemphigus vegetans, pemphigus erythematosus, paraneoplastic pemphigus, and drug-induced pemphigus. Cases were excluded if they included the Systematized Nomenclature of Medicine–Clinical Terms word pemphigoid, which is associated with the ICD-9 code for bullous pemphigoid (694.5). This method of case cohort identification for pemphigus has been previously validated.⁴

The study population included adults 18 years and older with an active status in the database within the past 5 years and nonmissing information on age, sex, and race. Prevalence among subgroups was standardized using the 2010 US Census population. This investigation was determined by Northwell Health to not to require institutional review board approval nor written informed consent from participants because data were deidentified.

Results

Using Explorys, we identified 1795 patients with pemphigus. A majority of patients were women (n = 1095 [61%]), 50 years and older (n = 1410 [78.5%]), and white (n = 1330 [74.1%]) (Table 1). Crude and standardized prevalences are summarized in Table 2. We observed an overall standardized point prevalence of 5.2 (95% CI, 4.9-5.4; 0.005%) cases per 100 000 adults. Prevalence was increased in successively older age groups. Overall standardized prevalence was similar among women (5.7; 95% CI, 5.3-6.0; 0.005%) and men (4.6; 95% CI, 4.3-5.0; 0.004%), and across race groups.

Table 1. Patient Demographics in Study Cohort.

Characteristic	No. (%)
Total patients, No.	1795
Sex
Male	700 (39.0)
Female	1095 (61.0)
Age, y
18-29	70 (3.9)
30-39	110 (6.1)
40-49	205 (11.4)
50-59	300 (16.7)
60-69	400 (22.3)
70-79	325 (18.1)
80-89	270 (15.0)
≥90	115 (6.4)
Race
White	1330 (74.1)
African American	215 (12.0)
Othera	125 (7.0)
Unknown	125 (7.0)

^aOther includes identification as Asian, Hispanic or Latin American, multiracial, and other.

Table 2. Crude and Standardized Prevalence of Pemphigus by Age, Sex, and Race^a.

Characteristic	No. of Pemphigus Casesb	Population Sizeb	Crude Prevalence per 100 000 (95% CI)	Crude Prevalence, %	Standardized Prevalence per 100 000 (95% CI)c	Standardized Prevalence, %
Overall population	1795	29 396 600	6.1 (5.8-6.4)	0.006	5.2 (4.9-5.4)	0.005
Age, y
18-29	70	5 179 440	1.4 (1.1-1.7)	0.001	1.4 (1.1-1.7)	0.001
30-39	110	4 702 070	2.3 (1.9-2.8)	0.002	2.3 (1.9-2.7)	0.002
40-49	205	4 355 030	4.7 (4.1-5.4)	0.005	4.6 (4.0-5.3)	0.005
50-59	300	4 889 630	6.1 (5.5-6.9)	0.006	6.1 (5.4-6.8)	0.006
60-69	400	4 627 480	8.6 (7.8-9.5)	0.009	8.6 (7.8-9.5)	0.009
70-79	325	3 150 390	10.3 (9.2-11.5)	0.010	10.3 (9.2-11.5)	0.010
80-89	270	1 777 240	15.2 (13.4-17.1)	0.015	14.9 (13.2-16.8)	0.015
≥90	115	715 320	16.1 (13.3-19.3)	0.016	16.1 (13.2-19.5)	0.016
Sex
Male	700	12 788 300	5.5 (5.1-5.9)	0.006	4.6 (4.3-5.0)	0.004
Female	1095	16 608 300	6.6 (6.2-7.0)	0.007	5.7 (5.3-6.0)	0.005
Race
White	1330	18 346 330	7.2 (6.9-7.6)	0.007	5.8 (5.5-6.1)	0.006
African American	215	2 907 470	7.4 (6.4-8.5)	0.007	7.2 (6.3-8.3)	0.007
Otherd	125	2 030 330	6.2 (5.1-7.3)	0.006	5.8 (4.8-7.0)	0.006
Unknown	125	6 112 470	2.0 (1.7-2.4)	0.002	1.9 (1.6-2.3)	0.002

^aPrevalence estimates are based on records for which age and sex information were available.

^bSum of group-specific counts may not equal total because Explorys data is rounded to the nearest 10.

^cSex comparisons are adjusted for age. Age group comparisons are adjusted for sex. Race comparisons are adjusted for sex and age. The sex and age distribution of the 2010 US Census population was used as the standard population, including 8 age groups: 18-29 years, 30-39 years, 40-49 years, 50-59 years, 60-69 years, 70-79 years, 80-89 years, and ≥90 years.

^dOther includes identification as Asian, Hispanic or Latin American, multiracial, and other.

Discussion

In this study we observed that pemphigus is rare among adults in the United States. Although peak onset of pemphigus is observed in the fourth or fifth decade of life and is more prevalent in white persons of Mediterranean or Jewish background,¹ we observed an increasing prevalence with age and a similar racial prevalence across all groups. Given the size, heterogeneity, and overall representative nature of the Explorys platform, we believe the racial and age-specific prevalence observations may be generalizable to the US population. However, we recognize that exact prevalence data may be difficult to calculate owing to the various forms of pemphigus encompassed within the 694.4 ICD-9 code.

There are few adjusted population-based analyses evaluating pemphigus disease burden. A German claims-based analysis identified 7699 patients with pemphigus vulgaris and observed point prevalence of 0.009%.⁵ This study only included patients with insurance and did not include a description of race; therefore, some demographic groups may have been underrepresented. A Danish analysis estimated the pemphigus prevalence at 0.006%, but no information on age, sex, and race was included.⁶

Despite using a validated method to identify the case cohort, use of administrative data in this population-based analysis on pemphigus burden in the United States has the potential to result in misclassification bias. An additional limitation is the inability to stratify prevalence by pemphigus subtype because the 694.4 ICD-9 code included various forms of pemphigus. We could not capture patients who did not seek care at health systems included in Explorys; therefore, the extent to which the analysis underestimates the prevalence is unknown. These results may prompt areas of future study, including evaluation of potential disease associations and treatment outcome.