Introduction

Hyperhidrosis (HH), a disorder of excessive sweating, is a very distressing condition for which treatment is currently inadequate. HH is often described as either primary or secondary. Primary HH is localised/focal but may be multifocal. Secondary HH can be focal (due to diabetic neuropathy or an eccrine tumour) but is often generalised when induced by an internal disease or a psychophysiological condition.1 2 Both can lead to significant social and occupational limitations and thus to a marked reduction in the quality of life.3 4 To date, there is no literature on the prevalence and incidence of this disease in Germany, with the exception of a 2013 study of 14 336 workers who underwent skin screening.5 The prevalence rate was 16.3%. International studies show a prevalence of HH of between 1.0% and 4.8% in the USA,6–8 in Brazil 2.07%,9 in Denmark 9.07%,10 in Sweden 5.5%11 and in the UK 1.6%.6 In addition, some studies show that women are more affected than men, and HH increases with age.6 11 Currently, it is unclear how often HH is diagnosed in Germany and, therefore, how often it is treated in the general population and covered by the statutory health insurance (SHI).

The current study thus aimed to address the following questions:

What is the epidemiology of HH in Germany?

Are there regional differences in prevalence’s?

Which levels of severity are identified?

Which medical specialties are involved in health care?

How many patients received hospital care?

MethodsPatient and public involvement

None.

Study design, data source and study population

This is a retrospective analysis of secondary healthcare data based on an anonymised 40% random sample (n=2 885 984; average age 49.1 years, 56.8% women) of all insured persons who were at least 18 years old and insured for at least 1 day with a major national health insurance company (DAK-Gesundheit (DAK-G)) between 01 January 2016 and 31 December 2020. The DAK-G is a large and nationwide operating health insurance company with 5.6 million members, respectively, in 2019. SHI is an essential part of the German healthcare system: about 89% of the German population (about 72 million) is insured by 95 SHI funds (in 2024). The remaining 11% are privately insured.12 The SHI system in Germany is based on the principle of solidarity. Contributions are based on the financial capacity of members, and healthcare services are the same for everyone.13

Frequency of the disease

Persons with HH were identified by diagnoses based on the International Statistical Classification of Diseases (ICD-10-GM [German Modification] codes: R61.0 (localised), R61.1 (generalised) and R61.9 (not further specified)). The prevalence of HH was determined annually from 2016 to 2020. The detailed case definition is: (a) ≥1 R61 principal or secondary inpatient diagnosis in the respective year OR (b) ≥1 confirmed R61 outpatient diagnosis in the respective year.

To calculate the incidence of HH, the previously described definition of prevalence was applied but a diagnosis-free period of 2 years prior to the incidence quarter was defined. As a sensitivity test, longer diagnosis-free periods of (a) 3 years and (b) 4 years prior to the year of incidence were considered.

Possible regional differences were identified by stratifying prevalence and incidence by federal state of the person’s living place.14

Severity of HH

In SHI, clinical information is lacking, impeding the assessment of disease severity. The severity was indirectly determined in the SHI data by the presence of inpatient hospitalisation (main diagnosis of HH), prescription of disease-related systemic medication and days of incapacity to work due to HH. Prevalent enrollees with HH who showed one of the three criteria were considered to have severe disease, and those who did not have any of the criteria were assumed to have a mild or moderate form. The included drugs and treatments were selected based on the German guideline15 on HH and consented with dermatologists according to their suitability and availability in the routine data of the SHI.

For inpatient care, relevant surgery and procedure codes (Operationen- und Prozedurenschlüssel [Operation and Procedure Classification System]: 5–895.6—radical excision of sweat glands; 5–894.2—local excision of sweat glands; 5–911.3—suction curettage of sweat glands and 5–043 .x—sympathectomy) were included if documented in association with the diagnosis of HH. The HH-related drugs were identified using the codes of the German Anatomical Therapeutic Chemical (ATC) Classification System (D11AA04—methanthelinium bromide and N04AA11—bornaprin). A relevant therapy for HH is the use of botulinum toxin type A (ATC Code M03AX21). Despite the reimbursement of the drug by the SHI, the predominant practice is an application in the self-payer area. As a sensitivity test, these prescriptions of M03AX21 in exceptional cases were analysed.

Outpatient care

To determine the proportions of medical specialties treating persons with HH, the number of persons with at least one contact with a specialist group was summarised. The results are described by absolute and relative frequencies per year from 2016 to 2020.

For the incident population in 2019, the proportions of attending specialists were analysed quarterly over time. The population was followed for a period of 4 quarters after the initial diagnosis. The quarter of the first diagnosis was taken as the first quarter. For each quarter, all contacts with a diagnosis of HH were selected.

Inpatient care

From the prevalent cohort, the proportion of persons with at least one hospital stay due to a principal R61 diagnosis was determined by a specialist group. In addition, the length of hospital stay (in days) was analysed.

Statistics

Annual prevalence and incidence rates were reported in absolute and relative (per cent) terms with 95% CI. Baseline characteristics (age and sex) are presented as percentages for categorical variables and as mean, SD, median and range (minimum (min.) and maximum (max.)) for continuous variables. Age was calculated by dividing the year of birth by the year of observation, and age-related data are presented in 10-year intervals. The results were directly standardised to the adult population of Germany according to Destatis (2016–2020) by age, sex and federal state by means of direct standardisation. All analyses were carried out using the software package SAS for Windows V.9.4 (SAS Institute, Cary, North Carolina, USA).

ResultsDatabase

The analysed cohort included a total of 2 513 860 insured adult persons insured with DAK-G for a minimum of 1 day between 01 January 2016 and 31 December 2020. The proportion of women was 58.0%, with a mean age of 55.1 years (SD 20.9, median 56, min. 18 and max. 114).

Prevalence and incidence

In 2020, 0.70% (CI 0.70 to 0.70; table 1) of insured adults were diagnosed with HH. Extrapolated to the German population, these rates correspond to 500 000 persons with HH in medical care. In 2016, the prevalence rate was slightly higher at 0.80% (CI 0.80 to 0.80). The mean age of the prevalent persons was 59.5 years (SD 18.9, median 62, min. 18 and max. 101), 61.6% were female. Women were slightly more likely to suffer from HH than men (2020: 0.82% vs 0.52%, ratio 70.43% vs 29.57%). There was no change in prevalence between 2016 and 2020 in terms of gender and age.

Table 1

Standardised prevalence and incidence of hyperhidrosis over the years

The largest proportion of prevalent persons was identified by outpatient diagnoses (98.43%). Inpatient diagnoses accounted for a small proportion of prevalent persons (1.16%), and 0.41% of individuals had both an inpatient and outpatient diagnosis of HH in 2016. This distribution was observed in almost equal proportions both within subtypes and in subsequent years. Of these, 9.24% had localised HH, 8.65% had generalised HH and 84.80% had an unspecified form of HH.

In 2020, the highest standardised prevalence rate of localised HH was found in the younger age group (18 to <30 years with 0.10%, CI 0.09 to 0.11) (figure 1). Generalised HH was significantly more common in the older age groups (70 to <80 years with 0.09%, CI 0.08 to 0.11). In men, the age group 80 to <90 years (0.08%, CI 0.06 to 0.10) and women between 70 years and <80 years had the highest prevalence rate of generalised HH (0.11%, CI 0.10 to 0.13). Only in the 18 to <30 age group were men slightly more likely to be affected than women. The age distribution of unspecified HH is the same as that of generalised HH and increases with age.

Figure 1

Standardised prevalence of localised hyperhidrosis (R61.0) (top) and of generalised hyperhidrosis (R61.1) (down) by age and gender in 2020 (per 100 000). ICD, International Statistical Classification.

In 2020, 0.35% (CI 0.35 to 0.35) of insured adults developed new cases of HH. In 2019, the incidence rate was slightly higher at 0.41% (CI 0.41 to 0.41) (table 1). The sensitivity test showed that a longer washout time only slightly reduced the incidence in 2020 to (a) 0.33% (CI 0.33 to 0.33) and (b) 0.32% (CI 0.32 to 0.32). The distributions of incidence by age group and sex were similar to those of prevalence for localised and unspecified HH. Only for generalised HH was the number of cases too small.

Regional variations

On a federal state level, the highest prevalence of HH (standardised by age, gender and federal state) was observed in Saxony-Anhalt (0.94%), Mecklenburg-Western Pomerania (0.87%) and Brandenburg (0.82%). The lowest prevalence rates were found in Baden-Württemberg (0.53%), Schleswig-Holstein (0.59%) and Hamburg (0.64%) (figure 2).

Figure 2

Annual prevalence rate of hyperhidrosis by federal state in 2020 in percentage.

Severity of HH

In 2020 4.88% were identified as having a severe form of the condition. In 2016, the proportion of people with a severe form was 3.55%. There were no gender differences (women: 4.73% and men: 4.49%). The most frequently prescribed medication was a systemic anticholinergic (93.28%). A total of 6.58% were off work due to HH (with 13.18 mean days, SD 18.4 days, 5.5 median days per certificate, min. 1 day and max. 99 days) and 0.96% had a principal inpatient diagnosis or specific inpatient procedure. In 2016, fewer people were identified by medication (86.02%). However, more had work incapacity (11.06%) and more had an inpatient stay or procedure (5.07%).

A sensitivity test of botulinum toxin type A prescriptions in 2019 revealed that 0.81% of persons with HH had at least one prescription. Among insured persons with severe HH, the medication was prescribed in 1.60% of those with severe HH, indicating that the proportion of severely ill persons remained unchanged.

Course of treatment (1-year follow-up)

In 2019, 0.40% (CI 0.40 to 0.42) of outpatients were diagnosed with HH for the first time. Almost half of them were treated by a general practitioner (48.47%), a quarter by an internist (27.81%), 13.46% by a dermatologist and 12.45% by a gynaecologist. 8.81% of incident cases were seen by another specialist in the first quarter. The proportions do not change much between 2018 and 2020. This proportion increased slightly during the observation period and was 49.79% in the fourth quarter. Slightly less than a third of the persons visited an internist (27.81%). This proportion remained unchanged in the year after diagnosis. The proportion of people seeing a dermatologist increased to 14.33% in the second quarter after diagnosis and decreased to 9.71% two quarters later. 12.45% of incident cases were diagnosed by gynaecologists and 8.81% by other specialists (figure 3).

Figure 3

Proportion of attended physicians four quarters after the initial quarter with the incident hyperhidrosis (1-year follow-up, including initial quarter) diagnosis in 2019 (all forms, (R61)). ICD, International Statistical Classification.

Similar to the prevalence, the non-specific diagnosis code (R61.9) was predominantly coded by a general practitioner (51.47%) and an internist (30.42%). Only dermatologists were more likely to code R61.0 (53.02%) or R61.1 (25.71%).

Healthcare providers in outpatient care

Related to HH, most people consulted a general practitioner (81.40%) and an internist (68.08%). About half consulted a gynaecologist (43.98%) and 35.43% a dermatologist. These proportions have not changed over the years.

Inpatient care

0.11% of prevalent persons had at least one hospital admission with a principal diagnosis of HH in 2016. The rate of hospital stays increased to 0.14% in 2019 and decreased to 0.04% in 2020. The average length of stay per person increased slightly between 2016 and 2020 (4.50 days vs 4.67 days). In 2016, 47.83% of insured persons with HH and at least one inpatient stay were treated in a surgical department, 34.78% in an internal medicine department and 17.39% in a dermatology department (figure 4). In 2020, the proportion increased to 66.67% in the surgical department and decreased to 16.67% in both the internal medicine and dermatology departments. The longest average length of stay per person was in the internal medicine department (2016: mean 5.63 days, SD 4.17 days, median 4 days; min. 2 days and max. 14 days).

Figure 4

Distribution of hospitalisations of persons with hyperhidrosis by hospital specialty department in 2016 and 2020.

Discussion

Compared with the international literature, the prevalence of HH based on SHI data seems to be underestimated. The current prevalence of 0.7% is lower than the international published literature with 1%–9.7%.6–11 Possible reasons for the large differences in the internationally published literature may lie in the methods used and the study population (age, gender, occupation and setting). The only available German study,5 found a significantly higher prevalence of 16.3% through direct questioning of employees of 52 companies. Another study showed that only half of the patients who reported their symptoms to a doctor were ultimately diagnosed with HH.16 This suggests that (a) HH may be overdiagnosed if based on self-assessments but (b) in spite of this HH probably is a more common condition than suggested by claims data. The most important explanation for this discrepancy is the fact that a significant proportion of people with HH do not seek medical care regularly because, for most of them, there seems to be no cure or not even coverage by payers. The mean age of 59.5 years indicates that younger individuals, in particular, are less likely to seek medical advice. This may be due to a lack of perception of their complaints as diseases or a lack of knowledge about potential treatment options, which may result in a lack of expectations of relief. Regional differences within Germany may possibly reflect differences in the utilisation behaviour of the insured, differences in the coding habits of physicians or the general morbidity burden.

The age distribution shows that the focal form of HH is more common in younger people, whereas the generalised form occurs with increasing age, which is similar to other studies.11 17 The age distribution of unspecified HH suggests, on the one hand, that HH is not perceived as a problem until a relatively late age or that a doctor is consulted at a later age. On the other hand, it corresponds to the age distribution of generalised HH, which may indicate that the unspecified form is a generalised form. Other studies have also shown that 48.9% of patients wait 10 years or more before seeking medical advice.16

The slight decrease in inpatient cases in 2020 compared with previous years is most likely due to the start of the corona pandemic in 2020. In Germany, 13.5% fewer inpatient cases were registered in 2020 than in the previous year.18 The number of inpatient stays due to HH from the DAK-G population extrapolated to the German population corresponds quite well with the actually registered cases in the years 2016 to 2019 (2016: 623 vs 679; 2017: 619 vs 662; 2018: 761 vs 741 and 2019: 771 vs 749).19 Only in 2020 were there fewer stays than reported by the Federal Statistical Office (204 vs 388).

According to the S1 guideline, treatment of HH should be gradual, individualised and localised.15 The therapies recommended to date include topical therapy with aluminium chloride hexahydrate, anticholinergics, iontophoresis with tap water, injection therapy with botulinum toxin A, interventions with radiofrequency, microwaves or focused ultrasound, surgical therapies and systemic therapy with anticholinergics. However, when using SHI-based treatments as a surrogate for disease severity, there are limitations in the claims data set, since several treatments which may indicate severe disease do not appear in the SHI data. This accounts for therapies which are not reimbursed by the SHI or not supplied by the physicians on a SHI basis.

The present analyses show that the majority of patients are treated by general practitioners and internists. To date, dermatologists play a minor role in the provision of care, with proportions of prevalent and incidental HH of 35% and 14%, respectively. For other dermatological conditions, evidence shows that guideline-based treatment is more often provided by dermatologists.20 21 Interdisciplinary care, particularly involving dermatologists, would therefore improve healthcare. Additional studies are required to investigate potential differences in treatment behaviour between specialist area and the quality of care provided for HH.

A small but relevant proportion of the persons shows a high disease burden reflected by the severity measured in the claim data. Systemic treatments, missed work days and hospital treatment including surgical measures contribute to this high burden. In order to gain a full spectrum of disease burden, these data need to be completed with patient-reported outcomes from persons affected.

Strengths and limitations

The fundamental strength of this analysis is the large number of SHI data and their scientific added value in terms of population coverage—around 89% of the German population has SHI.12 The cohort, from one of the largest health insurance funds (DAK-G with a total of 5.6 million out of 73 million insured persons in Germany), provides a comprehensive database without significant selection bias. However, there are data and methodological limitations that need to be considered when interpreting the results.

In general, data are collected for billing purposes and not for research purposes. As the database only contains claims data, it can only cover services that are actually reimbursed by the SHI system. Consequently, there is no information on services that are not billable and are paid for by individuals themselves. Data from SHI claims analyses may underestimate the reality of care, as services can only be billed when people use the health system. In addition, the data do not provide information on whether and when people actually take their prescribed medicines after obtaining them from a pharmacy. In addition, the proportion of people with HH may be higher or lower, limiting the reliability of the findings.22 This could be due, for example, to insufficient or inadequate differential diagnosis, misclassification or the coding behaviour of the physician.

In addition, claims data do not include clinical information on disease severity, quality of life, or detailed personal information such as weight, lifestyle, and education.23 Consequently, the current treatment guideline for HH was reviewed with a view to identifying potential surrogate parameters that might be used to identify severe forms and indirectly determined by the presence of these parameters. These surrogate parameters are therefore only an approximation of the actual magnitude, which may result in an underestimation.

Furthermore, as claims for outpatient treatments and diagnoses are only recorded quarterly, it is impossible to precisely correlate a given diagnosis with a specific time point or therapy measure.13

The populations of the different health insurance funds differ.24 To minimise these differences, prevalence and incidence rates were directly standardised by age, sex and federal state. Furthermore, a study on a dermatological disease (psoriasis) demonstrated that the epidemiological findings of the DAK-G data can be extrapolated without limitation to the SHI population, provided that they are standardised.25