The incidence of pemphigus considerably differs between different populations. It ranges from 0.6 to 0.76 per million per year in Switzerland and Finland to 8.0 and 10.0 in Greece and Iran [9, 25–27]. The highest incidences, 16.1 and 32/million/year, were found in the Jewish population [28, 29] (Table 21.1) [8, 9, 11, 13–19, 23, 25–39]. The variability is most likely explained by different genetic backgrounds and trigger factors (see below). In addition, data quality of most epidemiological studies is hampered by their retrospective design and the inability to define the study populations, i.e., to ensure that all patients within a certain area are included. The latter shortcoming does not apply for three recent studies from Switzerland, the UK, and Taiwan that reported incidences of 0.6, 6.8, and 4.7/million/year, respectively [26, 37, 38]. Limitations of the latter studies, however, include the relatively low number of pemphigus patients in the Swiss study and the use of a general practitioner-based registry that may have included inadequately diagnosed patients in the UK population [26, 37, 40]. The Taiwanese National Health Insurance database has previously been the source of ample epidemiological studies. Pemphigus patients, however, were only studied after they had applied for a catastrophic illness certificate [38].

Mortality of PV and PF has been dramatically reduced by the use of corticosteroids in the early 1950s from 75 to 30 % [41]. The adjuvant use of immunosuppressants in the 1980s probably contributed to the further decrease in mortality to below 5 % of the study populations and 0.021 per 100,000 inhabitants of the USA [17, 41, 42]. In two recent studies from the UK and Taiwan, the risk of death was calculated to be two- to threefold higher compared to the control population [37, 38]. Infections, in particular pneumonia and septicemia, were the most frequent causes of death in different study populations [9, 38, 43]. Cardiovascular diseases and peptic ulcer disease were also significant reasons for death compared to the control population [38].

In two controlled studies an increased association of pemphigus was found with autoimmune thyroid disease and rheumatoid arthritis [44]. A case-control study revealed a higher incidence of autoimmune diseases in first-degree family members of PV patients [45]. A dramatically higher percentage of osteoporosis was diagnosed in pemphigus patients (40 %) compared to controls (6.5 %) [46] highlighting the importance of monitoring affected patients. This notion was reconfirmed in a controlled study in patients with immunobullous disorders, including pemphigus [47].

Paraneoplastic pemphigus is very rare with about 300 reported cases in the literature [55, 56]. The disease is invariably associated with a neoplasm [57]. While in Chinese patients, Castleman’s disease is the predominate underlying neoplasm, in Europe, Japan, and the USA, non-Hodgkin’s lymphoma and chronic lymphocytic leukemia account for about two third of the neoplasms followed by Castleman’s disease, thymoma, and other diseases [58–60]. The prognosis of paraneoplastic pemphigus is usually considered poor with a mortality rate ranging from 75 to 90 %. Its prognosis has been recently reevaluated in large series of 53 patients whose 1-, 2-, and 5-year overall survival rates were 49, 41, and 38 %, respectively [56]. Infections and evolution of the neoplasm were identified as the main cause of death with erythema multiforme-like skin lesions being associated with a twofold increased risk of death following multivariate analysis [56].

The incidence has been estimated between 0.9 and 1.3/million/a in Germany and France. In the early 1990s, 10 years later, the incidence appeared to have increased to 2.0/million/a [34, 61, 63]. The disease onset appeared to be earlier as in BP with a mean age of 60–65 years [94–98]. Several investigators provided evidence indicating that about 30 % of patients with anti-laminin 332 (formerly known as laminin 5 and epiligrin) mucous membrane pemphigoid present with a solid cancer [99–101]. No association with a malignancy was found in all mucous membrane pemphigoid irrespective of the target antigen as well as in patients with anti-α6β4 reactivity [102–104]. Thus, the identification of the target antigen may be useful.

The incidence of pemphigoid gestationis has been estimated in the range of 0.8 and 2.0/million/year in France, Germany, and Kuwait [12, 34, 61, 63]. A recent study in two English tertiary referral centers, however, found pemphigoid gestationis to arise in 4.2 % of pregnancies [105]. Previous data revealed frequencies of 1:60,000 (USA) [106], 1:50,000 (USA) [107], 1:44,000 (England) [108], 1:7,000 (Switzerland) [109], 1:3,000–4,000 (England) [110], and 1:3,000 (England, 1921–1955) pregnancies [111]. Individual cases have been described in association with hydatidiform moles and choriocarcinomas, without pregnancy [112].

Patients with pemphigoid gestationis show apparently an increased risk for other autoimmune disorders, such as Grave’s disease (in 10 %), hypothyroidism, alopecia areata, vitiligo, and autoimmune thrombocytopenia [112–115].

An incidence of 0.25–1.0 patients with linear IgA disease/million/year was reported in Central Europe, Singapore, and Kuwait [12, 34, 63, 65]. Linear IgA disease is the most frequent pemphigoid disorder in children with two peaks of onset, below the age of 5 and after the age of 60 years [116]. The incidence appeared to be higher in developing countries such as Malaysia, India, Thailand, Tunisia, Mali, South Africa, and Uganda [7, 31, 117–121]. This may be however related to the different age distribution of the population with up to half of the population being minors in developing countries. Since linear IgA disease sera, in addition to IgA, often also contain IgG antibodies to BP180, whereas most BP sera also have IgA anti-BP180 antibodies, these two diseases may represent different ends of the same spectrum [122]. In a study from Uganda, younger patients preferentially developed IgA anti-basal membrane zone autoantibodies, whereas the IgG isotype predominated in older patients [117]. A high frequency of lymphoproliferative disorders and ulcerative colitis compared to the general population has been found in linear IgA disease [123, 124].