Sex Differences in the Incidence of Skin and Skin-Related Diseases in Olmsted County, Minnesota, United States, and a Comparison With Other Rates Published Worldwide

Abstract

Many skin and skin-related diseases affect the sexes unequally, with attendant implications for public health and resource allocation. To better evaluate the incidence of skin and skin-related diseases affecting males versus females, we reviewed published population-based epidemiology studies of skin disorders performed utilizing Rochester Epidemiology Project (REP) data. Females had a higher incidence of the following diseases: connective tissue diseases (scleroderma, morphea, dermatomyositis, primary Sjögren syndrome, systemic lupus erythematosus [not in all studies]), pityriasis rosea, herpes progenitalis, condyloma acuminatum, hidradenitis suppurativa, herpes zoster (except in children), erythromelalgia, venous stasis syndrome, and venous ulcers. Males had a higher incidence of psoriasis and psoriatic arthritis, basal cell carcinoma (exception, females aged ≤40 years), squamous cell carcinoma, and lentigo maligna. Incidence rates were equal in males and females for cutaneous malignant melanoma (exception, higher in females aged 18–39 years), lower-extremity cellulitis, cutaneous nontuberculous mycobacterial infection, Behçet disease, delusional infestation, alopecia areata, and bullous pemphigoid. Many of the population-based sex-specific incidence rates of skin and skin-related diseases derived from the Rochester Epidemiology Project are strikingly different from those estimated elsewhere. In general, females are more commonly affected by skin and skin-related diseases. The reasons for this imbalance remain to be determined and are likely multifactorial.

Introduction

Over the decades, interest is growing regarding sex differences in disease with respect to epidemiology, pathogenesis, clinical presentation, and response to treatment. Certain skin and skin-related diseases show striking differences between sexes in incidence patterns (a measure of new cases in a population over a given period). For example, females have a higher incidence of connective tissue disease such as primary Sjögren syndrome (pSS) and scleroderma, whereas males more commonly are affected by nonmelanoma skin cancer. The reasons underlying sex-based disparity in the development of skin and skin-related diseases remain largely unknown but are likely multifactorial. Factors that may be considered include differences in skin layers and physiology, sex hormones, genetics, age, ethnicity/race, lifestyle, occupation, and others.^1–3

In Olmsted County, Minnesota, United States, population-based incidence studies can be performed using the Rochester Epidemiology Project (REP). The REP is a unique records linkage research infrastructure system that has captured health care data on almost all residents of Olmsted County since 1966 (described in detail elsewhere).^4,5 In Olmsted County, the majority of the population is white (ie, less racially diverse than the United States in general), and the number of residents is relatively small (146,000 persons).

In a previous manuscript, we summarized all incidence studies of skin and skin-related diseases derived from REP.⁶ The aim of this study is to outline sex disparities in the incidence of the select skin and skin-related diseases in Olmsted County, Minnesota, and compare them with other reported overall sex-specific incidence rates (IRs) worldwide.

Methods

We abstracted a complete list of published REP studies from 1966 (when the REP was established) through April 2014. We reviewed each article and identified those that described IRs of skin and skin-related diseases in Olmsted County. We included incidence studies in which overall sex-specific IRs either were reported or could be calculated (using data from the original publications). Studies reporting 10 or fewer cases (females and males combined) were excluded.

We abstracted the following data from the REP studies: study period, population defined by age, overall IR, overall sex-specific IRs, and P values with corresponding 95% CIs, when available.

We also reviewed the worldwide English-language literature to identify additional reports of overall sex-specific IRs for the entire study period using PubMed. We searched PubMed for incidence studies within the same period as the REP studies (1966–2014). Studies reporting 10 or fewer cases (men and women combined) were again excluded. In a few reports, the overall sex-specific IR was calculated only for one sex; we included these studies in our analysis. Search terms were as follows: basal cell carcinoma OR squamous cell carcinoma OR lentigo maligna OR cutaneous malignant melanoma OR cutaneous LE OR systemic LE OR scleroderma (or systemic sclerosis) OR dermatomyositis OR primary Sjögren syndrome OR psoriasis OR psoriatic arthritis OR pityriasis rosea OR lower-extremity cellulitis OR herpes zoster OR herpes progenitalis OR condyloma acuminatum OR cutaneous nontuberculous mycobacterial infection OR hidradenitis suppurativa OR Behçet disease OR delusional infestation OR erythromelalgia OR alopecia areata OR venous stasis syndrome OR venous ulcer OR bullous pemphigoid AND incidence AND sex-specific incidence. For each of these articles, we abstracted the following data: study period, geographic areas, method, population defined by age, total number of incidents, crude or standardized rates, overall IR, overall sex-specific IRs, and P values with corresponding 95% CIs, when available. If multiple non-REP studies on overall sex-specific IRs were available, a maximum of 5 references for each disease were included; however, for some diseases, only 1 additional study was identified. We did not identify any non-REP incidence studies that met our inclusion criteria for lentigo maligna, pityriasis rosea, lower-extremity cellulitis, herpes progenitalis, cutaneous nontuberculous mycobacterial infection, hidradenitis suppurativa, delusional infestation, erythromelalgia, and alopecia areata; specifically, overall sex-specific IRs for the entire study period were either not reported in the studies that we did find, or they could not be calculated because of insufficient data in the original articles.

Statistical Analysis

IRs with corresponding 95% CIs and P values are presented as they were reported in the original publications, when available. Otherwise, for REP studies, we performed additional calculations using age- and sex-specific counts and IRs from the publications, assuming that the number of cases followed a Poisson distribution. For consistency, all IRs from REP studies were standardized to the 2000 US white population. For non-REP studies, the quantities were calculated as follows using the Person-Time module of Open Epi.⁷ If the 95% CI for an IR was not reported but the observed number of cases and the IR were reported, then the 95% CI was calculated using the normal approximation to the Poisson distribution if (N≥10) or using the Mid-P exact test if (N<10) for either males or females. If the P value for the comparison of the IRs between males and females was not reported but both N and IR were reported, then the P value was calculated using the z-score test if both (N≥10) or using the Mid-P exact test if (N<10) for either males or females.

Results

The skin and skin-related diseases were grouped as follows for both REP studies and non-REP studies: 1) skin cancer; 2) connective tissue diseases; 3) papulosquamous diseases (including psoriatic arthritis [PsA]); 4) skin infections; and 5) other skin diseases.

Table 1 summarizes data abstracted from each REP study. Females in Olmsted County were more affected by connective tissue diseases such as scleroderma, morphea, dermatomyositis (DM), pSS, systemic lupus erythematosus (SLE; not in all studies), pityriasis rosea, herpes progenitalis, condyloma acuminatum, hidradenitis suppurativa, herpes zoster (except in children), erythromelalgia, venous stasis syndrome, and venous ulcers. Males were slightly more affected by psoriasis and PsA. Males were more commonly affected by basal cell carcinoma (except for females aged ≤40 years), squamous cell carcinoma, and lentigo maligna. Males and females were equally affected by cutaneous malignant melanoma (CMM), except for a significantly higher incidence among females aged 18 to 39 years. Differences by sexes were not statistically significant for lower-extremity cellulitis, cutaneous nontuberculous mycobacterial infection, Behçet disease, delusional infestation, alopecia areata, or bullous pemphigoid (BP).

Table 1.

Studies of of Sex-Specific Incidence of Skin and Skin-Related Disease in Olmsted County, Minnesota

Source	Study Period	Skin Disease	Age, y	No. of Cases	Age- and Sex-Adjusted Incidence Rate Per 100,000 Person-Years (95% CI)a			P Valueb
					Overall	Female	Male
Skin Cancer
Chuang et al8	1976–1984	Basal cell carcinoma	All	657	171.2 (157.8–184.5)	146.1 (130.6–161.6)	203.6 (179.7–227.4)	<.001
Christenson et al9	1976–2003	Basal cell carcinoma	≤40	417	23.4 (21.1–25.6)	25.9 (22.6–29.2)	20.9 (17.8–23.9)	.03
Chuang et al8	1976–1984	Squamous cell carcinoma	All	169	47.9 (40.6–55.3)	27.7 (20.9–34.4)	77.0 (61.5–92.5)	<.001
Christenson et al9	1976–2003	Squamous cell carcinoma	≤40	68	3.9 (3.0–4.8)	3.5 (2.3–4.7)	4.3 (2.9–5.7)	.40
Gray et al10	1984–1992	Squamous cell carcinoma	All	511	114.6 (104.5–124.7)	77.2 (67.0–87.4)	166.2 (146.2–186.1)	<.001
Mirzoyev et al11	1970–2007	Lentigo maligna	≥18	145	6.3 (5.3–7.4)	3.8 (2.9–5.1)	9.9 (7.9–11.9)	<.001
Resseguie et al12	1950–1974	Cutaneous malignant melanoma	All	42	4.9 (3.4–6.5)	4.5 (2.6–6.4)	5.7 (3.1–8.3)	.47
Popescu et al13	1950–1985	Cutaneous malignant melanoma	All	107	8.2 (6.6–9.8)	7.6 (5.7–9.6)	9.2 (6.4–12.0)	.34
Reed et al14	1970–2009	Cutaneous malignant melanoma	18–39	256	16.9 (14.8–19.0)	23.2 (19.7–26.6)	10.8 (8.4–13.3)	<.001
Lowe et al15	1970–2009	Cutaneous malignant melanoma	40–60	383	37.1 (33.4–40.8)	34,5 (29.5–39.5)	39.8 (34.3–45.3)	.16
Connective Tissue Diseases
Durosaro et al16	1965–2005	Cutaneous LE	All	156	4.3 (3.6–5.0)	5.2 (4.2–6.3)	3.3 (2.4–4.2)	.006
Peterson et al17	1960–1993	Morphea	All	82	2.7 (2.1–3.3)	3.8 (2.8–4.8)	1.7 (0.9–2.5)	<.001
Bendewald et al18	1976–2007	Dermatomyositis and its subtypes	All	29	1.0 (0.6–1.3)	1.4 (0.8–2.0)	0.5 (0.1–0.8)	.007
Pillemer et al19	1976–1992	Primary Sjögren syndrome	All	53	4.1 (2.9–5.2)	7.2 (5.2–9.3)	0.6 (0.0–1.4)	<.001
Nannini et al20	1976–2005	Primary Sjögren syndrome	≥18	105	5.1 (4.1–6.1)	8.7 (6.9–10.4)	1.1 (0.4–1.9)	<.001
Uramoto et al21	1980–1992	SLE	All	48	6.4 (4.5–8.2)	10.1 (7.0–13.3)	2.2 (0.4–4.0)	<.001
Kurland et al22	1951–1967	SLE	All	29	5.6 (3.5–7.6)	7.3 (4.1–10.4)	3.5 (1.1–6.0)	.07
Nobrega et al23	1950–1965	Combined SLE	All	25	3.3 (2.0–4.6)	4.0 (2.1–5.9)	2.5 (0.8–4.2)	.27
Michet et al24	1950–1979	SLE	All	25	2.3 (1.4–3.2)	3.0 (1.7–4.4)	1.5 (0.3–2.7)	.12
Michet et al24	1950–1979	Suspected SLE	All	21	1.9 (1.1–2.7)	3.5 (2.0–5.0)	0	Not calculatedc
Michet et al24	1950–1979	Discoid LE	All	24	2.3 (1.4–3.3)	2.5 (1.2–3.8)	2.2 (0.8–3.6)	.77
Michet et al24	1950–1979	Scleroderma	All	13	1.2 (0.5–1.8)	2.0 (0.9–3.2)	0.3 (0.0–0.9)	.03
Papulosquamous Diseases (Including Psoriatic Arthritis)
Tollefson et al25	1970–1999	Psoriasis	≤18	357	40.8 (36.6–45.0)	43.9 (37.6–50.2)	37.9 (32.2–43.6)	.17
Icen et al26	1970–2000	Psoriasis	≥18	1,633	78.9 (75.0–82.9)	73.2 (68.0–78.4)	85.5 (79.5–91.6)	.002
Bell et al27	1980–1983	Psoriasis	All	132	61.2 (50.2–72.1)	63.8 (48.8–78.9)	60.0 (43.2–76.8)	.72
Wilson et al28	1970–1999	Psoriatic arthritis	≥18	147	7.2 (6.0–8.4)	5.4 (4.0–6.9)	9.1 (7.1–11.0)	.002
Shbeeb et al29	1982–1991	Psoriatic arthritis	≥18	66	9.3 (7.0–11.6)	9.3 (6.0–12.5)	9.3 (6.0–12.7)	.97
Chuang et al30	1969–1978	Pityriasis rosea	All	939	140.7 (131.1–150.3)	165.3 (151.1–179.4)	117.0 (103.9–130.0)	<.001
Infections
McNamara et al31	1999	Lower-extremity cellulitis	≥18	176	213.0 (181.2–244.9)	193.6 (153.1–234.1)	226.1 (181.5–245.2)	.30
Guess et al32	1960–1981	Herpes zoster	≤20	173	48.6 (41.3–55.9)	42.6 (33.5–51.6)	48.2 (38.1–58.4)	.41
Yawn et al33	1996–2001	Herpes zoster	≥22	1,669	374.0 (355.8–392.2)	407.4 (382.0–432.9)	336.6 (310.2–363.1)	<.001
Chuang et al34	1965–1979	Herpes progenitalis	All	392	49.8 (44.7–54.9)	48.9 (42.5–55.3)	33.2 (27.2–39.2)	<.001
Chuang et al35	1950–1978	Condyloma acuminatum	All	746	48.7 (44.9–52.5)	56.5 (51.2–61.8)	41.2 (35.9–46.6)	<.001
Wentworth et al36	1980–2009	Cutaneous nontuberculous mycobacterial infection	All	40	1.3 (0.9–1.7)	1.4 (0.8–2.0)	1.0 (0.5–1.6)	.50
Vazquez et al37	1968–2008	Hidradenitis suppurativa	All	268	6.1 (5.4–6.9)	7.9 (6.8–9.1)	3.8 (3.0–4.7)	<.001
Other Skin Diseases
Calamia et al38	1960–2005	Behçet disease	≥18	13	0.4 (0.2–0.5)	0.5 (0.2–0.8)	0.2 (0.0–0.5)	.26
Bailey et al39	1976–2010	Delusional Infestation	All	64	1.9 (1.5–2.4)	2.3 (1.6–3.0)	1.7 (1.0–2.4)	.16
Reed and Davis40	1976–2005	Erythromelalgia	All	33	1.3 (0.8–1.7)	2.0 (1.2–2.7)	0.6 (0.1–1.1)	.004
Safavi et al41	1975–1989	Alopecia areata	All	292	19.9 (17.6–22.3)	20.0 (16.7–23.4)	19.7 (16.3–23.1)	.88
Mirzoyev et al42	1990–2009	Alopecia areata	All	530	20.9 (19.1–22.6)	21.3 (18.8–23.9)	20.2 (17.7–22.6)	.77
Heit et al43	1966–1990	Venous stasis syndrome	≥15	1,131	87.0 (81.8–92.1)	96.5 (89.3–103.6)	76.2 (68.4–83.9)	<.001
Heit et al43	1966–1990	Venous ulcer	≥15	263	23.0 (18.4–23.5)	24.2 (20.6–27.7)	16.7 (13.0–20.4)	.004
Brick et al44	1960–2009	Bullous pemphigoid	All	87	2.4 (1.9–2.9)	2.2(1.6–2.8)	2.8(1.8–3.7)	.25

Adapted from Andersen and Davis (6). Used with permission.

Table 2 summarizes data abstracted from each of the included non-REP studies. In general, females worldwide also were more affected by connective tissue diseases such as SLE, cutaneous LE, scleroderma (or systemic sclerosis), pSS and DM (DM was not in all studies), venous leg ulcers (only 1 study found), and herpes zoster (except for a US study in which males were slightly more affected). Males worldwide seemed to be more commonly affected by basal cell carcinoma and squamous cell carcinoma. For cutaneous malignant melanoma, PsA, and condyloma acuminatum, the overall sex-specific IR varied by geographic area (one sex was not predominant in all areas). Differences by sex were not statistically significant for Behçet disease and bullous pemphigoid worldwide (except for one German study, which reported male predominance with bullous pemphigoid).

Table 2.

Studies of Sex-specific Incidence Rates of Skin and Skin-related Disease Worldwide

Skin Disease	Source	Study Period	Method	Geographic Area	Age, y	Total Incidents, No.a	Crude or Standardized Rates	Incidence Rate per 100,000 Person Years (95% CI)			P Valueb
								Overall	Female	Male
Skin Cancer
Basal cell carcinoma	Bath-Hextall et al45	1996–2003	Cases of persons registered with a general practitioner, with data maintained in The Health Improvement Network database	United Kingdom	≥18	11,113	World population	60.0 (58.9–61.1)c	53.0 (51.6–54.4)c	69.0 (67.2–70.8)c	<.001d
	Osterlind et al46	1973–1982	Cases reported to the Danish Cancer Registry	Denmark	All	10,846	World population	Not reported	23.7 (23.1–24.3)c	30.4 (29.6–31.2)c	<.001d
	Green et al47	1986–1992	Clinical examination and postal survey of a random sample of residents	Nambour, Queensland	All	250	World population	Not reported	1,579.0e	2,074.0e	Not possible to calculate
	English et al48	1987–1992	Clinical examination and postal survey of residents; medical record review of persons reporting any treatment	Geraldton, Australia	45–64	717	Crude	Not reported	3,379 (2,813–4,059)	7,067 (6,015–8,304)	<.001d
	Staples et al49	2002	Face-to-face interviews to identify persons previously treated for skin cancer using a stratified sampling of households	Australia	All	612	World population	884 (816–957)	745 (662–839)	1,041 (936–1,158)	Not possible to calculate
Squamous cell carcinoma	Osterlind et al46	1978–1982	Cases reported to the Danish Cancer Registry	Denmark	All	2,005	World population	Not reported	2.5 (2.3–2.7)c	6.7 (6.3–7.1)c	<.001d
	Serrano et al50	June 1979–May 1980	Survey sent to health professionals to identify cases among state residents; details abstracted by study staff	New Hampshire and Vermont, United States	All	277	US population	Not reported	8.0 (6.2–9.8)c	32.0 (27.6–36.4)c	<.001d
	Green et al47	1986–1992	Clinical examination and postal survey of a random sample of residents	Nambour, Queensland	≥18	94	World population	Not reported	472.0	1035.0	Not possible to calculate
	English et al48	1987–1992	Clinical examination and postal survey of residents; medical record review of persons reporting any treatment	Geraldton, Australia	45–64	88	Crude	Not reported	501 (334–668)	775 (567–984)	.04d
	Staples et al49	2002	Face-to-face interviews to identify persons previously treated for skin cancer using a stratified sampling of households	Australia	All	286	World population	387.0 (344.0–434.0)	291.0 (242.0–349.0)	499.0 (430.0–580.0)	Not possible to calculate
Cutaneous malignant melanoma	Osterlind et al46	1978–1982	Cases reported to the Danish Cancer Registry	Denmark	All	2,376	World population	Not reported	8.4 (8.0–8.8)c	6.1 (5.7–6,5)c	<.001d
	MacKie et al51	1979–1994	Cases reported by pathologists to the Scottish Melanoma Group	Scotland	All	6,288	World population	Not reported	6.7 (6.5–6.9)c	4.5 (4.3–4.7)c	<.001d
	Noorbala et al52	1988–2008	Cancer registry of cases among persons referred to public health centers in the province	Province of Yazd, Iran	All	71	Crude	Not reported	0.27 (0.17–0.37)c	0.40 (0.28–0.52)c	.11d
	Jelfs et al53	1989	Cases captured by all state or territory cancer registries	Australia	All	5,494	World population	Not reported	23.9 (23.0–24.8)c	30.2 (29.1–31.3)c	<.001d
	Jones et al54	1995	Cases from the Auckland region reported to the New Zealand Cancer Registry or collected by the Auckland melanoma unit	Auckland region, New Zealand	All	525	World population	56.2	55.2 (NA)	58.0 (NA)	Not possible to calculate
Connective Tissue Diseases
Systemic lupus erythematosus	Johnson et al55	1991	Cases ascertained from a survey sent to attending or primary care physicians, a specialist clinic, and members of a patient support group and from hospital inpatient and laboratory data; medical record review and in-person standardized examination	Metropolitan districts of Birmingham and Solihull, England	≥18	33	West Midlands population	3.8 (2.5–5.1)	6.8 (4.4–9.2)	0.5 (0.1–1.7)	<.001d
	Somers et al56	1990–1999	Cases captured in the General Practice Research Database, covering 5% of persons served by the National Health Service	United Kingdom	All	1,638	European population	4.71 (4.48–4.94)	7.89 (7.46–8.31)	1.53 (1.34–1.71)	<.001d
	Nossent57	1978–1996	Cases captured by inpatient and outpatient registries; medical record review	Counties of Finnmark and Troms, Norway	≥16	83	Crude	2.6 (2.1–3.2)	4.6 (3.6–5.8)	0.6 (0.3–1.3)	<.001d
	Hochberg58	1970–1977	Cases with a hospital discharge diagnosis	City of Baltimore, Maryland	All	W: 73 B: 233	Baltimore population	Not reported	W: 3.93 (3.0–4.8)c B: 11.4 (9.8–13.0)c	W: 0.42 (0.11–0.73)c B: 2.50 (1.7–3.4)c	<.001 for W and Bd
Cutaneous lupus erythematous	Deligny et al59	1995–1999	Cases captured by dermatologists in public and private practice, all rheumatology and internal medicine in public and private practice	French Guiana	All	20	Crude	2.0 (1.1–3.2)	3.4 (1.8–5.8)	0.5 (0.1–1.9)	.06
Scleroderma	Silman et al60	1980–1985	Cases captured by 7 separate sources. 1) Consultants (vascular surgeons, general physicians, dermatologists, rheumatologists, renal physicians, chest physicians); 2) relevant regional physicians; 3) 2 rheumatology units in the area; 4) the RegionalHealth Authority; 5) the Royal College of General Practitioners Research Unit; 6) the Raynaud’s Association; and 7) consultants of the United Kingdom Scleroderma Study Group	The West Midlands Region, United Kingdom	All	156	Crude	0.37 (NA)	0.62 (NA)	0.11 (NA)	Not possible to calculate
	Herrick et al61	2005–2007	Cases captured by members of 5 specialist medical associations (including pediatricians, dermatologists, and rheumatologists)	United Kingdom and Ireland	<16	94	Crude	0.34 (0.27–0.41)	0.44 (0.33–0.58)	0.24 (0.17–0.35)	.005c
	Arias-Nunez et al62	1988–2006	Cases identified by the Department of Medicine of the Hospital Xeral-Calde of Lugo	Spain	≥15	78	European population	2.3 (1.6–2.5)	3.5 (2.3–3.9)	1.0 (0.5–1.4)	.001
	Yu et al63	2000–2008	Cases registered with the Taiwan National Health Insurance Database containing records of 1,000,000 beneficiaries	Taiwan	All	118	Crude	1.5 (1.2–1.7)	2.5 (2.1–3.0)	0.4 (0.2–0.6)	<.001d
	See et al64	2005–2009	Cases registered with the Taiwan National Health Insurance Database containing records of 1,000,000 beneficiaries	Taiwan	All	54	Crude	1.1 (0.8–1.4)	1.6 (1.1–2.1)	0.6 (0.3–0.9)	<.001d
Dermatomyositis	Oddis et al65	1963–1982	Cases with a hospital discharge disease (35 hospitals)	Allegheny County, Pennsylvania, United States	All	W: 22 B: 5	US population	Not reported	W: 0.8 (0.01–1.6)e B: 2.8 (0.1–5.5)e	W: 0.7 (0–1.4)e B: 0	W: .758d B: Not calculatedf
	See et al64	2005–2009	Cases registered with the Taiwan National Health Insurance Database containing records of 1,000,000 beneficiaries	Taiwan	All	36	Crude	0.7 (0.5–1.0)	0.8 (0.5–1.2)	0.6 (0.3–0.9)	.40d
	Kuo et al66	2003- 2007	Cases registered with the Taiwan National Health Insurance Research Dataset and the National Death Registry of Taiwan. Coverage rate, >99%, of Taiwanese citizens	Taiwan	All	803	Crude	7.1 (6.6–7.6)	10.0 (9.2–10.9)	4.3 (3.8–4.9)	<.001
Primary Sjögren syndrome	See et al64	2005–2009	Cases registered with the Taiwan National Health Insurance Database containing records of 1,000,000 beneficiaries	Taiwan	All	583	Crude	11.8 (10.8–12.7)	20.1 (18.3–21.8)	3.3 (2.6–4.1)	<.001d
	Yu et al63	2000–2008	Cases registered with the Taiwan National Health Insurance Database containing records of 1,000,000 beneficiaries	Taiwan	All	855	Crude	10.6 (9.9–11.4)	18.5 (17.2–19.9)	2.9 (2.4–3.5)	<.001d
	Weng et al67	2005–2007	Cases registered with the Bureau of National Health Insurance of Taiwan, covering 99.5% of the Taiwan population (23 million) by 2008	Taiwan	≥15	3,352	Taiwan population	6.0 (5.8–6.2)	11.0 (10.6–11.4)	1.1 (1.0–1.2)	<.001d
	Alamanos et al68	1982–2003	Cases captured by inpatients and outpatients referred to rheumatology clinics of the Ioannina University Hospital and the Ioannina General Hospital; patients referred to private rheumatologists practicing in the study area	Ioannina, Greece	All	422	Greek population	5.3 (4.8–5.8)c	10.1 (9.2–11.1)c	0.5 (0.3–0.7)c	<.001d
Papulosquamous Diseases, Including Psoriatic Arthritis
Psoriasis	Setty et al69	1991–2005	Cases captured by the Nurses’ Health Study II	United States (15 States)	25–42	892	Crude	Not studied	82 (77–89)	Not studied	Not calculatedf
Psoriatic arthritis	Soriano et al70	2000–2006	Cases captured by the membership of the Hospital Italiano Medical Care Program (2 main hospitals and 24 medical office buildings to around 140,000 members)	Buenos Aires, Argentina	>18	35	Crude	6.26 (4.2–8.3)	3.64 (1.6–5.7)	10.02 (5.9–14.1)	.005d
	Soderlin et al71	1999–2000	Cases captured by rheumatology department or to the 1 private rheumatologist	Växjö, Sweden	≥16	11	Crude	8 (4–15)	12 (5–24)	5 (1–13)	.20d
	Savolainen et al72	2000	Cases captured by all physicians practicing in the Kuopio Health Centre, private clinics, and occupational health services or in various departments of the Kuopio University Hospital	Kuopio, Finland	All	16	Crude	23.1 (13.2–37.5)	27.2 (13.1–50.0)	18.4 (6.8–40.1)	.45d
	Nossent and Gran73	1978–1996	Cases captured by departments of the University Hospital	Norway	≥16	232	Crude	6.9 (4.3–9.8)	5.2 (3.8–6.7	7.9 (5.2–10.9)	.002d
	Hoff et al74	2000–2008	Cases registered in the Nord–Trøndelag Health Study 3	Norwayd	≥20	188	Crude	41.3 (35.8–47.6)	43.4 (36.0–52.5)	38.7 (31.10–48.20)	.44d
Infections
Herpes zoster	Opstelten et al75	2001	Cases captured from primary care physicians	The Netherlands	All	1,080	Crude	320.0 (300.9–339.1)c	389.0 (359.2–418.8)c	254.0 (229.8–278.2)c	<.001d
	Toyama and Shiraki76	1997–2006	Cases identified by 39 dermatology clinics and the dermatology departments of 7 general hospitals associated with the Miyazaki Dermatologic Society	Miyazaki, Japan	All	43,388	Crude	415.0 (411.3–418.7)c	458.0 (452.7–463.3)c	367.0 (361.9–372.1)c	<.001
	Donahue et al77	1990–1992	Cases captured from the Harvard Community Health Plan	Boston, Mass. United States	All	1,075	Crude	215.0 (202.2–227.9)	211.0 (193.7–228.4)	219.0 (199.9–238.1)	.54d
	Insinga et al78	2000–2001	Cases captured from the Health care claims database (Medstat MarketScan) of individuals from all states	United States	All	9,152	US population	320.0 (313.4–326.6)c	380.0 (370.0–390.0)c	260.0 (251.5–268.5)c	<.001
	Jih et al79	2000–2006	Cases registered in the National Health Insurance program; covering 96% of the population (2000)	Taiwan	All	34,280	Crude	489.0 (483.8–494.2)c	509.0 (501.6–516.5)c	471.0 (463.8–478.2)c	<.001d
Condyloma acuminatum	Hoy et al80	2004	Cases registered in the HealthCore Integrated Research Database	United States	All	4,264	US population	120.0e	120.0e	110.0e	Not possible to calculate
	Lin et al81	January 5–18, 2009	Cases recorded in 170 private doctors and all doctors working in the 5 local Social Hygiene Clinics	Hong Kong, China	≥18	73	Crude	203.7 (157.0–250.4)c	124.9 (61.7–188.1)c	292.2 (217.0–367.4)c	.003d
	Castellsague et al82	2005	Cases identified by gynecologists, dermatologists, and urologists in 6 geographic regions; covering 68% of the total population	The regions of Spain (Andalucia, Catalonia, Valencia, Galicia, Madrid, Pais Vasco)	14–64	36,565	Spain population	117.9 (116.7–119.1)c	99.6 (98.0–101.2)c	136.6 (134.8–138.4)c	<.001d
	Desai et al83	2006–2008	Cases recorded in the UK General Practice Research Database (Genitourinary Medicine Clinic activity dataset); covering 6% of persons registered with general practitioners	United Kingdom	All	80,531	England population	156.5 (155.4–157.6)c	142.0 (140.6–143.4)c	167.7 (166.1–169.3)c	<.001d
	Hillemanns et al84	2005 2 month	Cases identified by 130 gynecologists, 50 dermatologists, and 20 urologists from 8 geographic regions	Germany	All	303	Crude	Not studied	76.0 (72.6–79.4)	Not studied	Not calculatedf
Other Skin Diseases
Behçet disease	Mohammad et al85	1997–2010	Cases captured from 3 health care districts (covering 1.2% of the total area of Sweden)	Skåne, southern Sweden	≥15	20	Crude	0.2 (0.1–0.3)	0.1 (0.0–0.2)	0.3 (0.1–0.4)	.14
	Salvarani et al86	1998–2004	Cases identified from physicians at Reggio Emilia Hospital, medical practitioners, and community–based specialists	Reggio Emilia area, northern Italy	All	18	Crude	0.24 (0.14–0.38)	0.24 (0.11–0.45)	0.25 (0.11–0.47)	.93d
	Yu et al63	2000–2008	Cases registered with the Taiwan National Health Insurance Database containing records of 1,000,000 beneficiaries	Taiwan	All	64	Crude	0.8 (0.6–1.0)	1.0 (0.7–1.3)	0.6 (0.4–0.8)	.048d
	See et al64	2005–2009	Cases registered with the Taiwan National Health Insurance Database containing records of 1,000,000 beneficiaries	Taiwan	All	42	Crude	0.9 (0.6–1.1)	1.1 (0.7–1.5)	0.6 (0.3–0.9)	.06d
Venous leg ulcer	Margolis et al87	1988–1996	Cases recorded in the General Practice Research Database, covering 6% of persons registered with general practitioners	United Kingdom	65–95	>2,000	Crude	Not reported	1,420.0 (1,350.0–1,480.0)	760.0 (711.0–830.0)	Not possible to calculate
Bullous pemphigoid	Jung et al88	1989–1997	Cases identified in the Departments of Dermatology	Mannheim and Würzburg, Germany	All	94	German population	0.61 (0.49–0.73)c	0.49 (0.35–0.63)c	0.87 (0.62–1.12)c	.005
	Bertram et al89	2001–2002	Cases identified in the Department of Dermatology at the University of Wu_rzburg	Würzburg, Germany	≥21	27	Crude	1.34 (0.83–1.85)c	1.46 (0.72–2.20)c	1.22 (0.53–1.91)c	.64
	Gudi et al90	1991–2001	Cases registered by the Pathology Department, Aberdeen Royal Infirmary	Grampian Region of northeast Scotland	All	83	Crude	1.4 (1.1–1.7)c	1.7 (1.2–2.2)c	1.1 (0.7–1.5)c	.05d
	Marazza et al91	2001–2002	Study protocol was distributed to all dermatology clinics, immunopathology laboratories, and practicing dermatologists	Switzerland	All	140	Crude	1.2 (1.0–1.4)c	1.33(1.0–1.6)c	1.1 (0.8–1.3)c	.19d

Abbreviations: B, black; NA, not available; W, white.

^aBoth sexes.

^bWomen vs men.

^cConfidence interval calculated.

^dP value calculated.

^eIncidence rate is reported per million.

^fP value was not calculated because the incidence was 0 in the male group.

Discussion

We describe the results of multiple population-based epidemiologic studies performed in Olmsted County, Minnesota, examining sex-based differences in disease IRs. Intriguingly, most of the skin and skin-related diseases studied in Olmsted County were more common in females than males (Figure). Others have also noted sex-based disparities in the overall IRs of skin and skin-related diseases worldwide (Table 2). The reasons underlying sex-based disparities in the incidence of skin and skin-related diseases remain largely unknown but are likely multifactorial. Factors that ought to be considered include 1) sex differences in the structure and function of skin; 2) genetic predisposition; 3) effects of sex hormones; 4) race/ethnicity; 5) sociocultural behavior; and 6) environmental or geographic factors.

Figure.

Female:Male Incidence Ratio and 95% CIs of Studies From the Rochester Epidemiology Project, Presented on a Log Scale. A, Skin cancer. B, Connective tissue diseases. C, Papulosquamous diseases (including psoriatic arthritis). D, Skin infections. E, Other skin diseases. AA indicates alopecia areata; BCC, basal cell carcinoma; BD, Behçet disease; BP, bullous pemphigoid; CA, condyloma acuminatum; CLE, cutaneous lupus erythematosus; DI, delusional infestation; DM, dermatomyositis (and its subtypes); HP, herpes progenitalis; HS, hidradenitis suppurativa; HZ, herpes zoster; LE-cellulitis, lower-extremity cellulitis; LM, lentigo maligna; MM, malignant melanoma; NMI, nontuberculous mycobacterial infection; PR, pityriasis rosea; PsA, psoriatic arthritis; pSS, primary Sjögren syndrome; SCC, squamous cell carcinoma; SLE, systemic lupus erythematosus; VSS, venous stasis syndrome; VU, venous ulcer.

A direct comparison of the overall sex-specific IR data derived from the REP with data from other sources could, however, be complicated by several factors such as type of study (eg, population-based, survey), differences in the methods of disease ascertainment (eg, diagnostic classification criteria vs self-reported), whether the sex-specific IR was crude or standardized, or the age groups studied (eg, all ages, adults only, children only).

Skin Cancer

The incidence of basal cell carcinoma was higher in males than females in Olmsted County, except for females 40 years or younger^8,9; the same was true for squamous cell carcinoma in Olmsted County.^9,10 Other studies have shown that, compared with females, males had significantly greater risk of acquiring basal cell carcinoma or squamous cell carcinoma,^46–50 with the highest IRs reported among Australian males.⁴⁷

The incidence of lentigo maligna, a rare form of malignant melanoma in situ, is higher in males than females in Olmsted County. The incidence increased significantly with age and was higher for males (particularly age ≥50 years and highest between age 70–79 years) compared with females throughout the entire study period.¹¹

In Olmsted County, females aged 18 to 39 years had a significantly higher incidence of cutaneous malignant melanoma than males,¹⁴ but when considering all ages combined, as well as patients aged 40 to 60 years, males and females were equally affected by cutaneous malignant melanoma.^12,13,15 In Europe, females had a significantly greater risk of acquiring cutaneous malignant melanoma compared with males.^46,51 In Australia and New Zealand, the incidence of cutaneous malignant melanoma was much higher in both males and females.^53,54 In Iran, the reported IRs of cutaneous malignant melanoma were lowest for both sexes.⁵²

Connective Tissue Diseases

The incidence of lupus was higher in females than males in Olmsted County,^16,21,22 although this difference was not statistically significant for combined SLE and discoid SLE. Other studies have also shown that females had a significantly greater risk than males of acquiring SLE^55–58 and cutaneous LE.⁵⁹ In Olmsted County, for example, females had a 5- to 6-fold higher risk than males.²¹ In a US study (Baltimore, Maryland), the IR of SLE was highest among black females and lowest among white males.⁵⁸ In the United Kingdom, Afro-Caribbeans had the highest female IR, followed by Asian and white females, suggesting a racial disparity.⁵⁶

The incidence of scleroderma in Olmsted County is higher in females than in males for both localized scleroderma (morphea) and diffuse scleroderma (systemic sclerosis).^17,24 Other studies have also found that females had a significantly greater risk of acquiring scleroderma compared with males, showing that estradiol concentrations may have a role in the pathogenesis.^60–64

The incidence of DM is higher in females than males in Olmsted County.¹⁸ The tendency toward female predominance in DM was also noted in a Taiwanese study,⁶⁶ although others have reported similar IRs for DM in females and males in Taiwan.⁶⁴ In the United States, white females and males were equally affected by DM, whereas black females had a higher incidence than black males (DM was absent in black males).⁶⁵

The incidence of pSS was higher in females than males. In Olmsted County, white females were 13 times more affected by pSS than males.¹⁹ In Taiwan, females had a 10-fold higher risk of acquiring pSS than males.⁶⁷ In Greece, females were 20 times more affected by pSS compared with males,⁶⁸ again showing that estradiol concentrations may affect the pathogenesis of pSS.^63,64

Papulosquamous Diseases

The incidence of psoriasis was similar between males and females in Olmsted County,^25,27 but from 1970 to 2000, males had a higher incidence of psoriasis, except for during the sixth decade of life.²⁶ In another US study, the incidence of psoriasis among females aged 25 to 42 years was similar to that observed in Olmsted County among females aged 18 years or older.⁶⁹

The incidence of PsA was higher in males than females in Olmsted County,²⁸ except from 1982 to 1991, when females and males were equally affected.²⁹ The male predominance in PsA was also noted in Argentina.⁷⁰ In Norway, one study also reported that males had greater risk of acquiring PsA compared with females,⁷³ but another Norwegian study showed that females were more commonly affected by PsA than males.⁷⁴ In Sweden and Finland, female preponderance was reported among patients with PsA.^71,72 In Olmsted County, the incidence of pityriasis rosea was also studied and showed a female predominance.³⁰

Skin Infections

The incidence of lower-extremity cellulitis in Olmsted County showed that females were slightly more affected compared with males.³¹ The incidence of herpes zoster also showed that females older than 22 years had a higher risk compared with males in Olmsted County.³³ Among children in Olmsted County, boys had a higher risk of acquiring herpes zoster than girls.³² Other studies have also showed that females have greater risk of acquiring herpes zoster compared with males,^75,76,78,79 although others have found that females and males were equally affected.⁷⁷ The incidence of herpes progenitalis was higher in females than males in Olmsted County, with the highest IR for both sexes in those aged 20 to 24 years.³⁴ The incidence of condyloma acuminatum was higher in females than males in Olmsted County.³⁵ Others have reported that females and males are similarly affected in the United States.⁸⁰ In China, Spain, and England, males had a significantly higher incidence of condyloma acuminatum compared to females with highest IR among Chinese males.^81–83 The incidence of condyloma acuminatum among females in Germany were found more similar as the one reported in Olmsted County.⁸⁴ The incidence of cutaneous nontuberculous mycobacterial infection was approximately equal between the sexes in Olmsted County.³⁶ In Olmsted County, the incidence of hidradenitis suppurativa was higher in females than males, with a sharp decline after the fifth decade of life in females.³⁷

Other Skin Diseases

The incidence of Behçet disease was approximately equal between the sexes in Olmsted County.³⁸ The same was true in Italy and Sweden.^85,86 In Taiwan, females had nonsignificantly greater risk than males of acquiring Behçet disease.^63,64 The incidence of venous stasis syndrome and venous ulcerations was higher in females than males in Olmsted County.⁴³ In the United Kingdom, females also had a higher incidence of venous ulcerations, except for those aged 86 to 95 years (when males were more likely to have venous ulceration).⁸⁷

In Olmsted County, the incidence of delusional infestation was higher in females than males, although this difference was not statistically significant.³⁹ Erythromelalgia was also more common among females in Olmsted County.⁴⁰ The incidence of alopecia areata was approximately equal between females and males in Olmsted County for more than 4 decades.^41,42 In Olmsted County, the incidence of BP was slightly higher in males compared with females.⁴⁴ In 1 German study, males were twice as often affected by BP compared with females⁸⁸; however, another German study noted that females were slightly more affected by BP than males.⁸⁹ In Scotland and Switzerland, females had a slightly higher risk of getting BP compared with males.^90,91

Strengths and Limitations

The strengths of studies using REP data are the availability of complete medical records since 1966 (the date the REP started) and the ability to examine trends in the incidence of diseases using consistent eligibility criteria. We acknowledge, however, that each of the included REP studies had potential limitations (described in detail in the original publications). A direct comparison of REP vs non-REP studies was complicated by differences in the types of studies performed (population-based, survey, etc). Other limitations could be due to differences in sample sizes and age distributions. Some studies did not provide standardized rates but considered crude rates only. It is unclear whether the results would have been the same if all studies were assessed using standardized rates or crude rates and same methods.

Conclusion

We describe the results of multiple population-based epidemiologic studies performed in Olmsted County, Minnesota, particularly examining sex-based differences in the incidence of skin and skin-related diseases. We observed notable differences between sexes in incidence patterns. Despite methodologic limitations, we showed that females were more commonly affected by skin and skin-related diseases in Olmsted County and worldwide. To further elucidate the global sex disparities in incidence of skin and skin-related diseases, more studies are needed, particularly from regions not represented well in the dermatology literature such as Latin America and Africa. A better insight into sex-based disparities may also facilitate better development for diagnosis, therapy, and prevention of these diseases in the future.

Abbreviations

BP

bullous pemphigoid

DM

dermatomyositis

IR

incidence rate

LE

lupus erythematosus

PsA

psoriatic arthritis

pSS

primary Sjögren syndrome

REP

Rochester Epidemiology Project

SLE

systemic lupus erythematosus