The Incidence of Endometriosis, 2014–2022

An Analysis of Nationwide Claims Data From Physicians in Private Practice

Background: The epidemiological characterization of endometriosis, particularly with regard to its incidence, has been inadequate to date both in Germany and other countries. The goal of this study was to determine trends in the incidence of diagnosed endometriosis and changes in age structure at the time of first diagnosis over the period 2014–2022.

Methods: Nationwide claims data from physicians in private practice, obtained according to relevant German law (§ 295 SGB V), were used to identify the population at risk for a first assured diagnosis of endometriosis (ICD-10-GM: N80) during each year of the study period, consisting of women and girls aged 10–52 who were insured by the statutory health insurance system and for whom at least two years of prior observation were possible. Patients were defined as incident if they were documented as having received a first confirmed diagnosis of endometriosis, according to the case definition, during the study year. The case definition comprised multiple options for validating the diagnosis.

Results: The incidence of diagnosed endometriosis rose over the period of the study from 2.8 per 1000 persons at risk in 2014 to 4.1 per 1000 in 2022, corresponding to a 44% relative increase. There was also a marked shift in age-specific incidence toward higher values at younger ages: the median age at diagnosis fell from 37 years (2014) to 34 (2022).

Conclusion: This is the first study providing nationwide population-based data on the incidence of endometriosis in Germany. The observed rise in newly diagnosed cases is presumably mainly due to an increased awareness of endometriosis and to the growing recognition of the disease.

Even though endometriosis is considered one of the most common gynecological conditions, its epidemiological characterization is still inadequate. There is considerable variation in prevalence rates with different study settings; furthermore, population-based incidence rates are almost completely unavailable (1). This situation is a direct consequence of the stigmatization of menstruation-related symptoms (2) and the complexity of confirming the diagnosis (3) (Box).

Box

Confirming the diagnosis of endometriosis

Only three more recent studies, determining population-based incidence rates over prolonged periods of time, are to be highlighted:

An Israeli study based on health insurance data found only a minor variation in annual incidence rates, with a slight upward trend since 2000 to most recently 0.8 cases per 1000 women aged between 15 and 55 in 2015 (4).

An Australian secondary data study, which was supplemented by administrative health data of participants of a longitudinal survey over the period 2000–2018, showed an incidence of 4.5 cases of clinically confirmed or suspected endometriosis per 1000 person-years in women aged 20 to 44 years; no temporal trend information is reported (5).

In an Icelandic study with comprehensive individual linking of a wide variety of data sources, an age standardized incidence rate of 1.3 women aged 15 to 49 years with surgically confirmed pelvic endometriosis per 1000 person-years was determined for the period 2001–2015 (6). After an initial increase, the incidence in this study showed a falling trend after 2005 (6).

The only data available for Germany are older and based on a sample of insured persons of the Statutory Health Insurance Sample AOK Hesse/KV Hesse. An analysis of these data for 2007 found an age-standardized cumulative incidence between 1.9 and 3.5 diagnosed cases per 1000 women aged 15 to 54 years (7). Other than these studies, only the German Hospital Statistics provide robust data for Germany on an annual basis. In 2022, approximately 32 000 hospital stays were recorded with a primary diagnosis of endometriosis, which corresponds to 0.8 cases per 1000 persons (8).

With regard to age-specific incidence, these studies arrived at comparable results with a median age of 34–35 years; the highest incidence rates were found in the age groups 25–44, with an age peak in the age group 30–34 (4, 5, 6, 7).

In a German-Austrian cross-sectional study, patients with histologically confirmed endometriosis were aged 20 on average at the time of onset of symptoms. The mean interval from the onset of symptoms to final diagnosis was 10 years, indicating a considerable diagnostic delay (9). Older data from the international Endometriosis Association already showed similar intervals of diagnostic delay earlier; they also indicate that a large proportion of patients already experienced first, especially pain-related symptoms before the age of 20 (10).

The aim of our study was to determine the nationwide incidence of diagnosed endometriosis as well as temporal and age-specific trends observed in the setting of physicians in private practice in Germany over the period 2014–2022 (Box).

Methods

Our analysis is based on nationwide outpatient claims data from physicians in private practice, obtained according to relevant German law (section 295 of Book V of the German Social Code, SGB V) of all persons insured by the statutory health insurance system (SHI, gesetzliche Krankenversicherung) from first quarter 2012 to first quarter 2023.

Study population

The study population comprises the SHI-insured female population in the reproductive period between menarche and menopause in each study year (13). The information on the sex of the insured person is based on the data stored on the electronic health card (eGK). Transgender persons were included in this study if the entry “female” was still stored on their cards. The inclusion criteria were:

• at least one utilization of services that were billed to the respective Association of Statutory Health Insurance Physicians
• SHI-insured
• female sex, and
• age 10–52 years (period between menarche and menopause).

Population at risk (denominator determination)

For the purpose of identifying the population at risk of being diagnosed with endometriosis for the first time during a study year, it had to be possible to observe the insured persons of the study population described above for at least two years prior to the respective study year (Figure 1). Thus, the study period starts with the year 2014 and the corresponding pre-observation period 2012–2013. Insured persons with at least one billable outpatient contact in the respective study year were deemed to be “pre-observable”. It was a requirement that no diagnosis of endometriosis (ICD-10-GM: N80, including adenomyosis) was documented in combination with the mandatory additional identifier “confirmed” during the defined two-year pre-observation period. The annual population at risk was made up of all insured persons who fulfilled these inclusion criteria and thus formed the denominator of the incidence calculation.

Figure 1

Determination of the annual populations at risk after 2-year / 5-year diagnosis-free pre-observation period

Case definition and validation of the diagnosis (numerator determination)

Those insured persons of the population at risk were defined as incident cases for whom a diagnosis of endometriosis (N80) with the additional identifier “confirmed” was documented for the first time during the period 2014–2022. For these insured persons, the index quarter was determined; they were assigned to the respective study year as incident cases and formed the numerator for the incidence calculation for this year. One of the following criteria had to be fulfilled for the validation of the diagnosis:

• Repeated coding of the diagnosis in one of the patient-individual three subsequent quarters (M2QR)
• Documentation of the diagnosis by a gynecologist (M1Q + FG)
• In the same treatment case: Billing of a specific diagnostic service (fee schedule item, GOP) of the German Uniform Value Scale (EBM, Einheitlicher Bewertungsmaßstab) (M1Q + GOP), or
• Documentation of a corresponding OPS code (OPS, operations and procedures key, a German adaptation of the ICPM) for surgery performed on an outpatient basis or by an external doctor using hospital facilities (M1Q + OPS) in the quarter of diagnosis.

For information on specialty group assignments, fee schedule items and OPS codes please refer to Kohring et al. (14). The additional identifier “confirmed“ provides no information on how the diagnosis was confirmed, making it impossible to determine the proportion of patients with histologically confirmed diagnosis. As part of the validation of diagnosis, which is essential for routine data, the validation criteria “M1Q + GOP“ and “M1Q + OPS“ provide information on whether appropriate diagnostic and, if necessary, surgical methods have been used. One feature inherent to outpatient claims data from physicians in private practice is that only those patients can be recorded for whom a diagnosis documented by a physician on an outpatient basis is available.

Incidence calculation

For each study year, the (administrative) cumulative and age-specific incidence per 1000 female SHI-insured persons aged 10–52 was calculated for the period 2014–2022. The age structure of the patients with incident endometriosis in the respective study years was described using median and interquartile range (IQR). In a supplementary analysis, the study population was limited to the age group 15–39 years to obtain results that are comparable to those of other studies in which a narrower definition of the reproductive period was used. Directly age-standardized incidence values were also calculated (reference: female German population aged 10 to 52 years, as of 31 December 2014 [15]).

Sensitivity analyses

We repeated the analyses based on a diagnosis-free pre-observation period of five years to determine any effect of the selected pre-observation period on the calculated incidence. Due to the limited availability of data, this analysis covered the period 2017–2022 (Figure 1). The same criteria for inclusion in the risk population, definition of cases and validation of the diagnosis applied.

We conducted a second sensitivity analysis to determine the extent to which the standardized procedure with a fixed pre-observation period had an impact on the incidence calculation of the primary analysis. In contrast to the primary analysis, all patients with a documented confirmed diagnosis of endometriosis were successively excluded from the annual risk populations, as of the year 2012. This means that, for example, for determining the 2014 population at risk, all patients diagnosed with endometriosis in the two preceding years (2012–2013) were excluded. Finally, for determining the 2022 population at risk, all patients diagnosed with endometriosis from all available ten preceding years (2012–2021) were excluded.

Results

In 2022, the size of the study population was 18.7 million girls and women (2014: 19.7 million). The annual population at risk differed depending on the diagnosis-free pre-observation period selected and decreased in size over time (2022: 15.6 million with 2 years of diagnosis-free pre-observation period; 13.8 million with 5 years of diagnosis-free pre-observation period) (Table 1).

Table 1

Size of the study populations and populations at risk, the number of patients as well as the cumulative incidence of diagnosed endometriosis per 1000 girls and women aged 10–52 for each study year, by the respective diagnosis-free pre-observation period selected

For 2022, the crude cumulative incidence of diagnosed endometriosis for a two-year diagnose-free pre-observation period was 4.1 per 1000 female SHI-insured persons aged 10–52 (Figure 2). Compared to the incidence in 2014 at the start of the study (2.8/1000), this corresponds to a relative increase of 44% (mean annual increase: 4.7%). Particularly from 2020 to 2021, a more pronounced relative increase in incidence was observed (13%) which leveled off again the following year. A parallel trend, albeit at a slightly lower level, was observed for the longer disease-free pre-observation period of five years (Figure 2). Even after standardizing the age structure, the increase in incidence could still be observed (Table 1). In the supplementary analysis with restriction of the age group to 15–39 years, the incidence was slightly higher (2014: 3.0/1000; 2022: 5.0/1000) and showed an identical development over time. In the sensitivity analysis with successive exclusion from the population at risk as of 2012, the incidence rates were only marginally lower with an identical course of the trend (Table 2). Likewise, the trends for the various validation criteria ran in parallel, except of the “M1Q + OPS“ criterion which remained constant during the study period (eFigure).

Figure 2

Development of the annual cumulative incidence of diagnosed endometriosis during the period 2014–2022 with diagnosis-free pre-observation periods of different lengths

Table 2

Cumulative incidence of diagnosed endometriosis per 1000 girls and women aged 10–52 years for each study year with two-year diagnosis-free pre-observation period and successive exclusion of all patients with confirmed diagnosis of endometriosis (ICD-10-GM: N80) from the population at risk, as of 2012 (sensitivity analysis)

eFigure

Development of the annual cumulative incidence of diagnosed endometriosis during the period 2014–2022 with diagnosis-free pre-observation periods of two years for each of the individual validation criteria used for the applied case definition as well as two additional validation criteria that are commonly used in health services research

The comparison of the years 2014 and 2022 revealed changes in the age group-specific incidence of diagnosed endometriosis (Figure 3). In the 2014 diagnosis year, the incidence rose from the youngest age group to a peak in the age group 35–39 (4.4/1000), followed by a successive decline. The median age was 37 years (IQR: 30–44 years). In 2022, the age peak was in the age group 30–34 years (6.1/1000) and the median age dropped to 34 years (IQR: 27–42). In addition, the incidence in the younger age groups, particularly among the 10– to 39-year-olds, was many times higher than it had been in 2014, while in the age groups 40 years and over only minor increases were observed.

Figure 3

Development of the annual age-specific cumulative incidence

Discussion

For the first time, differentiated analyses of the incidence of endometriosis in Germany were performed, based nationwide outpatient claims data from all SHI-insured persons. Two key developments were observed over the period from 2014 up to and including 2022:

The first development was the steady moderate increase in incident endometriosis diagnoses. This trend remained robust even when the period of prior observation was changed. Abbas et al. (7) found an incidence of 1.9 new cases per 1000 women in Germany in 2007, using a comparable case definition. This finding indicates that the increase in the incidence of diagnosed endometriosis may already have been occurring for some time and has intensified in the recent past. This increase is likely to be primarily attributable to a greater awareness of the condition and not so much to changes in the risk of developing the disease in the study population. It can be assumed that a change in the method of diagnosis in favor of ultrasound already began during the study period, particularly in the specialist care setting, but only became established in primary care after the end of the study period with the inclusion and establishment of corresponding guideline recommendations, as of 2022 (11).

During the study period, office-based physicians had no known financial incentives to document the a diagnosis of endometriosis (16). At the same time, awareness of the condition has increased significantly both in the general public and in the specialist community, for example after national endometriosis strategies were adopted in Australia (2018) and in France (2022) (17, 18). In Germany, the Endometriosis Association played a particularly important role in these efforts by organizing highly effective public information and education campaigns and being active in the discourse on health policy (19). This study’s result of 4.1 new cases per 1000 female SHI-insured persons (2022) are at a similar level as the results of international studies, reporting incidence rates of 0.7 to 4.5 endometriosis cases per 1000 women (4, 5, 6).

The second key development is the marked shift in the age peak of age-specific incidence to the younger age groups at the time of first diagnosis. Our results for Germany are in line with current international data (4, 5, 6) on the age distribution at the time of first diagnosis. This shift and the marked increase in the number of first diagnoses indicate that the time to final diagnosis has decreased somewhat between 2014 and 2022. However, the time from symptom onset to diagnosis cannot be readily stated based on outpatient claims data from physicians in private practice, which exclusively contain ICD-10-encoded diagnoses on a quarterly basis. Assuming symptom onset on average in the first half of the third decade of life (9) or even before the age of 20 (10) and the median age of 34 at the time of first diagnosis, as determined in this study for the year 2022, a very long delay in diagnosis is still likely. The diagnosis should be established in less than five years after the onset of symptoms in order to minimize the risk of unfavorable disease progression (20).

In absolute figures, more than 53,000 newly diagnosed cases of endometriosis were calculated for 2022, based on a diagnosis-free pre-observation period of five years. The much-cited assumption of 40 000 new cases of endometriosis per year (21) can hence be replaced by a more up-to-date rate. At the same time, it should be taken into account that this updated absolute number of patients is likely to rather correspond to a minimum number for girls and women insured by the statutory health insurance system due to the known multifactorial challenge of diagnostic delay (2, 9), complex confirmation of diagnosis (3) and decrease in the utilization of medical care during the corona pandemic (22, 23). Furthermore, an estimate of the absolute number of affected girls and women for the whole of Germany would have to also include the data of persons who are not insured by the statutory health insurance system.

Limitations

The results of this study should be interpreted in particular in view of the following limitations:

Due to the fact that data availability was overall limited in time, it cannot be ruled out that, especially in older women, a diagnosis of endometriosis may already have been documented prior to the start of the study period in 2012 so that these patients possibly were incorrectly defined as incidental in this study. However, given the special care situation of patients with endometriosis, which is characterized by underreporting of the diagnosis especially in the past, we believe that the risk of misclassification and consequently overestimation of the numerator in this respect is low. The systematic approach with fixed observation periods results in an overestimation of the risk for the denominator population; however, the sensitivity analysis showed that this effect on the cumulative incidence was marginal.

The creation of annual study populations and populations at risk is inherently subject to demographic influences and factors relating to the utilization of medical care provided by physicians in private practice which are not always stable over time, for example, due to reduced utilization during the corona pandemic. For this reason, the systematic approach was given priority for the primary analysis. The observation that in the primary analysis and in the supplementary calculations of directly age-standardized incidence values the trends run parallel with both the two- and five-year pre-observation periods suggests that the direction of the development over time is real and only slightly overestimated by a shorter pre-observation period. Overall, we consider the underestimation of the incidence as the result of underrecording of diagnoses to be more critical.

Furthermore, it should be noted that these results are administrative incidences and that the primary purpose of diagnosis documentation in the outpatient-SHI physician setting is to support billing for medical services. In contrast to the coding of diagnoses in the hospital sector, it has no immediate effect on remuneration and information on diagnoses is exclusively available on a quarterly basis. With regard to a validation based on a diagnosis identified as “confirmed” in combination with a corresponding EBM fee schedule item (GOP) (“M1Q + GOP“), we cannot assume with absolute certainty that this service was performed explicitly to confirm the diagnosis of endometriosis, despite limiting the assessment to the documentation in the same treatment case.

Conclusion

The results of this study provide for the first time up-to-date data on the development of newly diagnosed cases of endometriosis over time for the majority of the female population in Germany, thus creating a robust foundation for advancing the medical care for endometriosis patients. In addition, future population-based primary surveys with individual linking of relevant SHI routine/claims data may provide the opportunity to gain further insights into various aspects of care and the chance to quantify the extent of underrecording of the diagnosis.

Acknowledgement

The authors thank the 17 Associations of Statutory Health Insurance Physicians for making the data available.

Ethics committee approval

The Ethics Committee of Charité—Universitätsmedizin Berlin approved the study (20 May 2024, reference number EA4/247/23).

Conflict of interest
The authors declare no conflict of interest.

Manuscript received on 24 April 2024, revised version accepted on
29 July 2024

Translated from the original German by Ralf Thoene, M.D.

Corresponding author
Claudia Kohring, M. Sc.
Fachbereich Epidemiologie und Versorgungsatlas
Zentralinstitut für die kassenärztliche Versorgung in Deutschland
Salzufer 8
10587 Berlin, Germany

Cite this as:
Kohring C, Akmatov MK, Holstiege J, Brandes I, Mechsner S: The incidence of endometriosis, 2014–2022. An analysis of nationwide claims data from physicians in private practice. Dtsch Arztebl Int 2024; 121: 619–26. DOI: 10.3238/arztebl.m2024.0160

References

1. Ghiasi M, Kulkarni MT, Missmer SA: Is endometriosis more common and more severe than it was 30 years ago? J Minim Invasive Gynecol 2020; 27: 452–61

2. Seear K: The etiquette of endometriosis: stigmatisation, menstrual concealment and the diagnostic delay. Soc Sci Med 2009; 69: 1220–7

3. Mechsner S: Endometriose: Eine oft verkannte Schmerzerkrankung. Schmerz 2016; 30: 477–90

4. Eisenberg V, Weil C, Chodick G, Shalev V: Epidemiology of endometriosis: a large population—based database study from a healthcare provider with 2 million members. BJOG Int J Obstet Gynaecol 2018; 125: 55–62

5. Rowlands I, Abbott J, Montgomery G, Hockey R, Rogers P, Mishra G: Prevalence and incidence of endometriosis in Australian women: a data linkage cohort study. BJOG Int J Obstet Gynaecol 2021; 128: 657–65

6. Kristjansdottir A, Rafnsson V, Geirsson RT: Comprehensive evaluation of the incidence and prevalence of surgically diagnosed pelvic endometriosis in a complete population. Acta Obstet Gynecol Scand 2023; 102: 1329–37

7. Abbas S, Ihle P, Köster I, Schubert I: Prevalence and incidence of diagnosed endometriosis and risk of endometriosis in patients with endometriosis-related symptoms: findings from a statutory health insurance-based cohort in Germany. Eur J Obstet Gynecol Reprod Biol 2012; 160: 79–83

8. Gesundheitsberichterstattung des Bundes: Diagnosedaten der Krankenhäuser ab 2000 (Eckdaten der vollstationären Patienten und Patientinnen). Gliederungsmerkmale: Jahre, Behandlungs-/Wohnort, ICD10. ICD10: N80 Endometriose. Datenstand: 15.12.2023. www.gbe-bund.de (last accessed on 8 March 2024).

9. Hudelist G, Fritzer N, Thomas A, et al.: Diagnostic delay for endometriosis in Austria and Germany: causes and possible consequences. Hum Reprod 2012; 27: 3412–6

10. Ballweg ML: Big picture of endometriosis helps provide guidance on approach to teens. J Pediatr Adolesc Gynecol 2003; 16: S21–6

11. ESHRE Endometriosis Guideline Development Group: Endometriosis. Guideline of European Society of Human Reproduction and Embryology. 2022. www.eshre.eu/-/media/sitecore-files/Guidelines/Endometriosis/ESHRE-GUIDELINE-ENDOMETRIOSIS-2022_1.pdf (last accessed on 7 July 2024)

12. DGGG, OEGGG, SGGG: S2k-Leitlinie Diagnostik und Therapie der Endometriose (AWMF-Register-Nr.: 015–045). Gültig bis 31.08.2025 (in Revision). 2020. https://register.awmf.org/assets/guidelines/015-045l_S2k_Diagnostik_Therapie_Endometriose_2020-09.pdf (last accessed on 7 July 2024).

13. Robert Koch-Institut: Gesundheitliche Lage der Frauen in Deutschland. 2020. https://doi.org/10.25646/6585 (last accessed on 26 October 2023).

14. Kohring C, Holstiege J, Heuer J, et al.: Endometriose in der vertragsärztlichen Versorgung – Regionale und zeitliche Trends im Zeitraum 2012 bis 2022. Zentralinstitut für die kassenärztliche Versorgung in Deutschland (Zi). Versorgungsatlas-Bericht Nr. 24/01, Berlin 2024. https://doi.org/10.20364/VA-24.01 (last accessed on 13 February 2024).

15. Statistisches Bundesamt (Destatis): GENESIS-Tabelle: 12411–0006, Bevölkerung: Deutschland, Stichtag, Altersjahre, Nationalität/Geschlecht/Familienstand. Fortschreibung des Bevölkerungsstandes Deutschland. 31.12.2014, 2015. www-genesis.destatis.de/genesis/online (last accessed on 1 July 2024).

16. AGEM, AGE: Endometriose. Ein Positionspapier der Arbeitsgemeinschaft Endometriose e.V. (AGEM) und der Arbeitsgemeinschaft Gynäkologische Endoskopie e.V. (AGE). 2023. www.dggg.de/fileadmin/data/Stellungnahmen/GBCOG/2023/Positionspapier_Endometriose_2023.pdf (last accessed on 14 December 2023).

17. Australian Government, Department of Health: National Action Plan for Endometriosis. Canberra, Australien 2018. www.health.gov.au/sites/default/files/national-action-plan-for-endometriosis.pdf (last accessed on 23 June 2024).

18. Ministère des Solidarités et de la Santé: Stratégie nationale de lutte contre l’endométriose. Paris, Frankreich 2022. https://sante.gouv.fr/IMG/pdf/strategie-endometriose.pdf (last accessed on 23 June 2024).

19. Endometriose-Vereinigung Deutschland e. V.: Chronik. 2024. www.endometriose-vereinigung.de/chronik/ (last accessed on 23 June 2024).

20. Brandes I, Kleine-Budde K, Heinze N, et al.: Cross-sectional study for derivation of a cut-off value for identification of an early versus delayed diagnosis of endometriosis based on analytical and descriptive research methods. BMC Womens Health 2022; 22: 521

21. Schweppe KW: Endometriose – Eine Erkrankung ohne Lobby. Zentralblatt Für Gynäkol 2003; 125: 233

22. Mangiapane S, Kretschmann J, Czihal T, von Stillfried D: Veränderung der vertragsärztlichen Leistungsinanspruchnahme während der COVID-Krise. Tabellarischer Trendreport bis zum 1. Halbjahr 2022. Zentralinstitut für die kassenärztliche Versorgung (Zi). Berlin 2022. www.zi.de/fileadmin/Downloads/Service/Publikationen/Trendreport_7_Leistungsinanspruchnahme_COVID_2022-12-08.pdf (last accessed on 10 March 2024).

23. Keilmann L, Beyer S, Meister S, et al.: Trends among patients with endometriosis over a 7-year period and the impact of the COVID-19 pandemic: experience from an academic high-level endometriosis centre in Germany. Arch Gynecol Obstet 2023; 307: 129–37