Inpatient Endometriosis Care in Germany: Hospital Caseloads and their Spatial Distribution

Abstract

Background

Endometriosis is a chronic gynaecological disease with an estimated prevalence of 10–15%. The German guideline provides evidence-based recommendations for diagnosis and treatment, but care provided is inadequate care due to long diagnostic pathways. Recent German research focused on regional variations in outpatient care, however research on inpatient endometriosis care is still lacking.

Aim of the Study

The aim of the study was to examine inpatient endometriosis care – hospital locations and their caseloads. Spatial coverage, caseload distribution patterns and possible clusters, including certified endometriosis centres (CEC) and non-certified hospitals nationwide were analysed.

Method

German hospital quality report data from 2021 was used as data source. The location, certification status and caseload, meaning coded ICD-10 N80 Endometriosis cases, were collected for all hospitals. Then, 20-, 40- and 60-minutes’ drive radius of CEC and non-certified hospitals were determined. Global and Local Moran’s I was calculated to assess spatial clusters in caseload.

Results

A CEC 60-minutes’ drive radius covers 78.15% of the area in Germany. Including all hospital locations that coded endometriosis, a maximum driving time of 40-minutes provides almost nationwide coverage. High caseload clusters appeared in urban areas and low caseload clusters especially in eastern Germany.

Conclusion

The results indicate spatial clusters in providers caseload and difficulties in access to CEC for patients depending on location. Further research with patient-level data is needed to investigate the spatial distribution of patients and precise travel time for inpatient care.

Keywords

endometriosis - health services research - spatial distribution - German hospital quality reports

Introduction

Endometriosis is a benign, chronic gynaecological disease associated with painful menstruation, heavy bleeding or infertility [1] [2]. The disease has a significant impact on women’s lives and they often experience a lack of understanding from the healthcare system, stigmatisation of their pain and delayed diagnoses [2] [3].

However, the precise healthcare situation of endometriosis patients or even reliable rates of endometriosis prevalence are not available in Germany due to a lack of data [4]. It is estimated that 10–15% of all women of reproductive age are affected and 40,000 women develop endometriosis each year [4]. Recent research showed that annual prevalence continued to increase from 2012 to 2022, but remained below epidemiological prevalence estimates [5].

The guideline Diagnosis and Therapy of Endometriosis [6] provides evidence-based recommendations for diagnosis and treatment. Still, on average, it takes 2.3 years from symptom onset to the first contact with providers, and 7.7 years elapse from first contact to diagnosis [7]. This suggests that dysfunctional or inadequate care still seems to exist, particularly due to long diagnostic pathways.

A certification programme and treatment standards have been initiated in several European countries to improve the quality of care for endometriosis patients [6]. The German guideline states that patients should be treated by an interdisciplinary team in a certified structure (e. g. centre) that integrates all necessary specialities across sectors. Therefore, the consensus-based care algorithm includes certified endometriosis centres (CEC) for diagnosis, operative/interdisciplinary laparoscopy, follow-up and conservative therapy [6]. However, the QS ENDO quality assurance programme for German-speaking countries (DACH region) estimates that around 60% of endometriosis patients are not treated in hospitals with a CEC [8].

There is little research on the differences on the quality of patient care in a CEC and a non-certified hospital. In general, in certified centres, e. g. for cancer care, guideline-conforming care is higher in percentage than in non-certified facilities [9]. Specifically for CEC, improved quality of life and lower complication rates were associated with certification status [10].

Although the ESHRE Endometriosis Guideline [11] no longer recommends laparoscopy as the diagnostic gold standard, laparoscopy is still the standard procedure to surgical therapy [6]. Often, laparoscopy is used for diagnostic examination and, if necessary, treatment of endometriosis during the same procedure [6], meaning inpatient care has its irreplaceable importance. However, little is known about the structure of inpatient endometriosis care, in particular about the distribution of hospitals providing such care in Germany. The need for a regional monitoring to evaluate healthcare provision as a first step to healthcare improvement and equity is well known [12] [13], meaning that investigations on the distribution of inpatient care and the certified care structure for endometriosis are needed.

Therefore, the aim of this study was to examine the spatial distribution and accessibility of CEC and non-certified hospitals and to evaluate their caseload in order to show regions with low access to specialised care.

Methods

The study uses German hospital quality reports (HQR) 2021 as data source, Isochrones process in OpenRouteService as well as Spatial Autocorrelation (Global Moran’s I) and Cluster and Outlier Analysis (Anselin Local Moran’s I) processes in ArcGIS Pro 2024 for the spatial analysis of coded inpatient caseloads

Data sources

German Hospital Quality Reports

The HQR are made publicly available by the Federal Joint Committee (G-BA), which specified that hospitals must report on inpatient care for each location every year [14].

The coded endometriosis caseload, defined as a discharge diagnosis of ICD-10 N80 Endometriosis and all subgroups, of each hospital location in the 2021 HQR was used as the basic population. Also, endometriosis-related inpatient procedures were identified using the operation and procedure code (OPS). According to the endometriosis guideline, specific OPS exist only in part and mainly for the destruction of endometriosis lesions [6]. Therefore, OPS 5-651.b, 5-702.2 and 5-702.4 were included. Other OPS were excluded as it is not possible to filter out whether they were specifically coded for endometriosis. Each included ICD-10 and OPS Codes are presented in the online-Appendix (Table S1).

The exact case number is not reported in the HQR for privacy reasons, if the caseloads were fewer than four cases [15].Then, the odd value of 1.5 cases was assumed, as already applied for HQR data analysis before [16]. The possible error in the calculation of the caseloads was squared using DESTATIS data. According to DESTATIS, 32,304 discharged patients were coded as endometriosis cases in 2021 [17].

Furthermore, name, hospital location ID and address of the hospital was extracted.

List of certified endometriosis centres

The Endometriosis-Association Germany (Endometriose-Vereinigung Deutschland e.V.) has listed all hospitals certified by the Endometriosis Research Foundation (SEF) on its website [18]. We therefore defined all hospitals listed on this website as CEC. At the time of data collection (15 July 2024), 72 providers were listed as certified, of which 68 were identified in the 2021 HQR. The remaining 4 were either outpatient providers and probably accidentally listed with the inpatient CEC, or they did not provide their HQR for 2021. Therefore, n=68 hospitals were included in our CEC sample.

Data Analysis

Descriptive analysis

Inpatient endometriosis caseloads were analysed descriptively. Case numbers were calculated for each ICD-10 code and, in the case of data protection, estimated using the procedure described. Then, total cases were calculated by summing all N80.x codes to an overall endometriosis caseload for each hospital location. Frequencies, means and standard deviations of total cases were calculated for all hospital locations, separately for CEC and non-certified hospital locations with at least one coded endometriosis case. These steps were repeated for OPS codes 5-651b and 5-702.2 / 5-702.4 summed up.

Data was processed with Excel 2019 and SPSS 29.

Spatial analysis

We analysed spatial coverage to provide an initial general estimate for the care situation in Germany (a.). Then, we carried out spatial statistics within a defined radius to identify regions with particularly high or low clusters, and therefore gaps in care (b.). Based on the addresses in the HQR, the geographical positions of the hospital location ID were determined using OpenStreetMap.

a. The spatial coverage (drive radius) was calculated separately for CEC and for all hospitals with at least one case. A recent study on pain medicine discussed that a 30 to 45 minutes (min) drive is reasonable for specialised pain care in Germany, with up to 60 min being acceptable in individual cases [19]. Applying this to endometriosis care, a maximum driving time (DT) of 60 min was set. Thus, regions with DT less than or equal to 20-, 40- and 60-min to CEC and other hospitals was determined and presented with Isochrones function/module of OpenRouteService.

b. To identify regional spatial clusters in coded cases, Moran’s I analyses were performed [20]. All hospital locations coding endometriosis in 2021 were included. The number of cases at each hospital location was compared with cases at hospitals within a 45 km radius. A distance of 45 km was set for patients based on an assumed driving time of 45 min at an average car speed of 60 km/h. Global Moran’s I analysis [20] was performed to investigate whether caseloads were clustered. If Global Moran's I shows a significant spatial correlation, Local Moran’s I [20] can be used to test for local clusters.

Global and Local Moran’s I analysis was performed using ArcGIS Pro 2024. QGIS Version 3.36.1 was used for visualisation.

Results

Caseload

Overall, 32,440.5 endometriosis cases were coded in 2021. In CEC (n=68), a total of 13,709.5 cases (42.26%) were observed. In hospital locations without CEC and at least one endometriosis case (n=807), 18,731 cases (57.74%) were coded. [Table 1] shows the caseload (including privacy calculations) for CEC and non-CEC, [Table 2] the caseload for each included ICD-10 and OPS code.

Table 1 Total caseload.

	n (%)	total cases (%)	M	SD	Md	[Min–Max]
All hospitals with at least 1 coded endometriosis case	875 (100%)	32,440.5 (100%)	37.07	79.93	17.5	[1.5–1135]
Non-certified hospital locations & at least 1 coded endometriosis case	807 (92.23%)	18,731 (57.74%)	23.21	30.67	15	[1.5–388.5]
Hospitals with certified centre	68 (7.77%)	13,709.5 (42.26%)	201.61	205.53	149.52	[15.5–1135]

total cases, number of coded ICD-10-GM N80 endometriosis cases including all subcategories and data protection calculation; M, Mean; SD, standard deprivation; Md, Median; Min, minimum; Max, maximum.

Table 2 Caseload for each ICD-10 and OPS code (see appendix for code explanations).

Caseload calculations (including data protection)	in hospitals with certified centre (n=68)	in hospitals without certified centre (n=807)	total (%)
ICD-10-GM N80.x total	13,709.5 (42.26%)	18,731 (57.74%)	32,440.5 (100%)
→ N80.-	0	7	7 (0.02%)
→ N80.0	2,724.5	4,954	7,678.5 (23.67%)
→ N80.1	2,638.5	6,322.5	8,961 (27.62%)
→ N80.2	54.5	187	241.5 (0.74%)
→ N80.3	6,545.5	4,477	11,022.5 (33.98%)
→ N80.4	355.5	255	610.5 (1.88%)
→ N80.5	383	285.5	668.5 (2.06%)
→ N80.6	199.5	604	723.5 (2.23%)
→ N80.8	729	1,268.5	1,997.5 (6.16%)
→ N80.9	159.5	370.5	530 (1.63%)
OPS 5-651.b	1,342.5 (30.19%)	3,104 (69.81%)	4,446.5 (100%)
OPS 5-702.2+5-702.4	13,686.5 (51.86%)	12,704 (48.14%)	26,390.5 (100%)

The included OPS codes are not solely specified for endometriosis cases. This suggests that other OPS codes were probably used for endometriosis cases, or that the included OPS were also used for other diseases. Consequently, the determined caseload of endometriosis and selected OPS codes do not correspond.

Spatial distribution

Driving time

Caseloads and DT were analysed and cartographically presented.

[Fig. 1] shows cases and DT for CEC, with most cases coded in Cologne and Berlin. In Munich, medium caseload was coded in two locations. Less than 500 cases were coded in each remaining CEC (n=64). The distribution of CEC varies from state to state, ranging from zero CEC in Bremen, Mecklenburg-Western Pomerania (M-WP) and Saxony-Anhalt (S-A) to 21 in North Rhine-Westphalia (NRW).

Fig. 1 Driving time and caseload of certified centres.

With a maximum DT of 60 min, the CEC alone do not cover the entire country, but 78.15% of the total area and 91.32% of the female population. Only the west of M-WP is covered within 60 min, as are the south and eastern parts of S-A. In Brandenburg, the 60 min radius around Berlin is covered, the outskirts mainly not. North and south Thuringia are also not covered, with one CEC in the east. Gaps also exist in the north, east and south of Saxony, as well as in southern Baden-Württemberg (B-W), eastern Bavaria, northern Schleswig-Holstein and western Rhineland-Palatinate (R-P). Major cities in NRW, e. g. Cologne, Duisburg and Münster, can be reached within 20 min DT, with most of the state being accessible within 40 min.

[Fig. 2] shows all non-certified hospitals with at least one endometriosis case. Most cases were coded in the cities of Neuss, Frankfurt (n=2), Munich and Wuppertal. Caseload>37.5−158 were coded in 138 hospital locations, while caseloads≤37.5 were coded in the remaining 664 hospital locations.

Fig. 2 Driving time to all hospitals and caseload of non-certified hospitals.

Considering all hospitals that coded endometriosis, almost the entire country is covered within a 40 min DT, expect for a few areas where it takes 60 min to reach a hospital (e. g. Ortenau district, the Brandenburg border or in frontier regions). NRW, Berlin, Hamburg and Bremen are mainly covered with a DT of 20 min and medium or large case numbers.

Spatial cluster

Global Moran’s I test revealed a slightly positive, significant spatial correlation (I=0.028, z=1.722; p=0.085) of coded cases, meaning the distribution of caseloads is clustered.

Local Moran’s I was performed to assess clusters of high and low case numbers in hospitals. The cartographic visualisation is shown in [Fig. 3].

Fig. 3 Local Moran’s I (radius=45 km) for all hospital locations.

Low-Low clusters can be identified in Saxony, M-WP, S-A, Lower Saxony, Thuringia, Brandenburg, Bavaria, Flensburg city, the Hochsauerland district and R-P. Those clusters are mainly located in eastern Germany, with the most clusters in Saxony.

High-High clusters can be found in the city of Munich and its environs, Berlin, Brandenburg, the districts of Cologne and Düsseldorf, in B-W in the city of Freiburg and the district of Stuttgart, and in Frankfurt.

Discussion

Inpatient endometriosis care in Germany shows regional variation. A tendency towards clusters with more coded cases in urban areas and fewer coded cases in eastern Germany was observed. Particularly areas in the eastern German states, which largely correspond to the territory of the former GDR, were not covered with a driving time (DT) of 60 min to certified endometriosis centres (CEC). The evaluation of all hospital locations that coded at least one endometriosis case in the HQR 2021 showed an almost nationwide DT of 40 min.

To our knowledge, there is little research on spatial clusters of inpatient endometriosis care in Germany. So far, prevalence or incidence studies [21] [22] or regional trends in outpatient care [5] [23] have mainly been analysed.

Data on outpatient care provide evidence of regional variations in prevalence [5] and healthcare structures [23], suggesting that endometriosis is more frequently diagnosed in regions with a CEC. Each additional centre is associated with 0.28 more cases per 1,000 women with statutory health insurance [23]. Greater diagnostic expertise at CEC may explain some differences, but certified hospitals alone do not fully account for regional variation. Clusters with high incidence were identified in NRW and southern Germany, where most of the CEC are located [23]. Regional clusters with high prevalence of diagnosed endometriosis were also found in southern Germany and Lower Saxony [5]. There was no clustering in regions with many CEC, but a cluster in Saxony with low diagnosis prevalence and districts close to CEC had a higher prevalence [5]. This is consistent with our findings, as we identified high clusters of inpatient cases in southern Germany, NRW and Berlin and low clusters of caseloads in eastern Germany.

With regard to CEC, the QS ENDO quality assurance programme was developed for the DACH region to provide further insights, which showed that about 60% of endometriosis patients are not treated in CEC [8]. Based on our results, it is possible that accessibility of centres may play a role. We were able to show that CEC alone are not accessible to all patients within 60 min DT, so we assume that patients either accept a longer DT or seek care in non-certified hospitals and may receive less specialised care.

However, specialised care is guideline-conform as the care algorithm includes a CEC for diagnosis, laparoscopy, and follow-up or conservative therapy [6]. CEC were also associated with improved quality of life and low major complication rates [10]. Therefore, access to CEC should be available to all patients. A possible solution would be a better spatial distribution or more CEC. However, we assume that staff shortages, financial barriers (e. g. the cost of the certification process) and/or structural factors (e. g. a lack of regional cooperation) may influence hospitals in regions without a CEC to decide not to pursue or achieve certification. Telemedicine networks may hereby be a possible tool for providing medical infrastructure and improving quality, presumably also for endometriosis care, particularly in regions without a CEC [24].

Research on certified centres for other gynaecological conditions underlines their importance. In oncology care, centres can improve cost-effectiveness while optimising quality [25], which advocates treatment in certified centres to meet quality indicators [9] and in terms of health economics [25]. These factors may also apply to CEC.

International research also found heterogeneous incidence pattern in hospitalisation, suggesting changing practices, awareness, inequalities and environmental factors [26]. Similarly in Italy, endometriosis incidence showed a spatial gradient, with a cluster of high-risk municipalities [27].

In our study, environmental factors and socio-economic structures are likely to explain some of the spatial patterns. However, HQR data cannot be used to identify individual patients, so we do not have information on, e. g., socioeconomic status (SES). Yet SES may have an impact on patients’ choice of hospital, as patients with higher SES may have less difficulty in choosing or travelling to specialised care, while those with lower SES may lack access or information [28].

This shows case numbers from the HQR have limitations. First, they can only be used to determine general coding frequencies, not person-specific codes, and coding practices may influence caseload. It is not possible to estimate diagnoses prevalence using the HQR. The 2021 data may also be biased by reduced use of health services during the pandemic [29]. We did not consider population or female density for spatial case clusters. Besides, this investigation does not allow conclusions about waiting times for treatment, which seems likely in regions with high clusters.

Our case calculations resulted in a slightly higher total than reported by DESTATIS [17] for 2021 (32,440.5 estimated cases vs. 32,304 reported cases), which may have led to an overestimate in some hospitals. However, it also shows that our calculations are within a good range, meaning our study provides a robust overview of caseloads and spatial patterns. Another strength is the analysis of DT to CEC, which has not been investigated before.

Conclusion

The spatial distribution of coded endometriosis cases from the HQR varied, with higher clusters in urban areas and lower clusters in eastern Germany. Assuming a driving time of 60 min to a certified endometriosis centre, there is no nationwide coverage, mainly for the eastern German states. Including the non-certified hospitals, a driving time of 40 min was found almost nationwide. The results show the importance of either more certified centres or a better distribution to cover all areas. Further research with person-related data linked to population is needed for precise statements about the spatial distribution of inpatient cases, diagnoses and endometriosis-related surgery. These investigations could ensure barrier-free access to inpatient endometriosis care, regardless of the person or location, in line with the principle of equity in health care.

Data Availability

The dataset generated in this study is available from the corresponding author on reasonable request. German Hospital Quality Report data is publicly available.

Contributions

LB, LJ and MG developed the study concept and design. Analyses were carried out by LB and LJ. The results were interpreted by LB. LB wrote the first draft of the manuscript; LJ and MG revised and approved it.

Conflict of Interest

The authors declare that they have no conflict of interest.

• References

• 1 Wren G, Mercer J. Dismissal, distrust, and dismay: A phenomenological exploration of young women's diagnostic experiences with endometriosis and subsequent support. J Health Psychol 2022; 27: 2549-2565
• 2 Pettersson A, Berterö CM. How Women with Endometriosis Experience Health Care Encounters. Womens Health Rep (New Rochelle) 2020; 1: 529-542
• 3 Nielsen LJ, Poulsen K, Funch AL. et al. The lived experiences of endometriosis in adolescence – A critical hermeneutic perspective. Scand J Caring Sci 2023; 37: 1038-1047
• 4 Robert Koch-Institute. Hrsg. Gesundheitliche Lage der Frauen in Deutschland. Gesundheitsberichterstattung des Bundes. Berlin: RKI;; 2020
• 5 Kohring C, Holstiege J, Heuer J. et al. Endometriosis in outpatient care – regional and temporal trends in the period 2012 to 2022. Berlin: Zi 2024;
• 6 DGGG, OEGGG, SGGG. Guideline: Diagnosis and Therapy of Endometriosis [S2k, Version 1.0]; 2020
• 7 Hudelist G, Fritzer N, Thomas A. et al. Diagnostic delay for endometriosis in Austria and Germany: causes and possible consequences. Hum Reprod 2012; 27: 3412-3416
• 8 Zeppernick F, Zeppernick M, Janschek E. et al. QS ENDO Real – A Study by the German Endometriosis Research Foundation (SEF) on the Reality of Care for Patients with Endometriosis in Germany, Austria and Switzerland. Geburtshilfe Frauenheilkd 2020; 80: 179-189
• 9 Schulz S, Lange C, Emrich K. et al. Quality Indicators Show Higher Fulfilment in Centers Certified by the German Cancer Society. Gesundheitswesen 2024; 86: 783-787
• 10 Byrne D, Curnow T, Smith P. et al. Laparoscopic excision of deep rectovaginal endometriosis in BSGE endometriosis centres: a multicentre prospective cohort study. BMJ Open 2018; 8: e018924
• 11 Becker CM, Bokor A, Heikinheimo O. et al. ESHRE guideline: endometriosis. Hum Reprod Open 2022; 2022: hoac009
• 12 Wennberg J. Gittelsohn Small area variations in health care delivery. Science 1973; 182: 1102-1108
• 13 Nuti S, Bini B, Ruggieri TG. et al. Bridging the Gap between Theory and Practice in Integrated Care: The Case of the Diabetic Foot Pathway in Tuscany. Int J Integr Care 2016; 16: 9
• 14 Federal Joint Committee. Regelungen zum Qualitätsbericht der Krankenhäuser. Online https://www.g-ba.de/downloads/62-492-3634/Qb-R_2024-09-19_iK-2024-09-19.pdf accessed: Dezember 5, 2024
• 15 Federal Joint Committee. Qualitätsberichte der Krankenhäuser. Was sie bieten und wie sie sich nutzen lassen. 2016 Online https://www.g-ba.de/downloads/17-98-4123/2016-06-16_Erlaeuterung-Qb-R_Langfassung.pdf accessed: Juli 16, 2024
• 16 Ji L, Geraedts M, de Cruppé W. Internal validation of self-reported case numbers in hospital quality reports: preparing secondary data for health services research. BMC Med Res Methodol 2024; 24: 325
• 17 DESTATIS. GENESIS. Krankenhauspatienten: Deutschland, Jahre, Hauptdiagnose ICD-10 (1-3-Steller Hierarchie). Online: https://www-genesis.destatis.de/genesis/online accessed: Juli 15, 2024
• 18 Endometriose-Vereinigung Deutschland e.V. Medizinisch zertifizierte Einrichtungen. Online: https://www.endometriose-vereinigung.de/zertifizierte-einrichtungen/?certificate=medizinisch&type=klinik-endometriosezentrum accessed: Juli 15, 2024
• 19 Erlenwein J, Buchholz J, Weißmann C. et al. Regional comparison of specialized outpatient and (partial) inpatient pain medicine care in Germany. Schmerz 2024;
• 20 Anselin L. Local Indicators of Spatial Association—LISA. Geographical Analysis 1995; 27: 93-115
• 21 Göhring J, Drewes M, Kalder M. et al. Germany Endometriosis Pattern Changes; Prevalence and Therapy over 2010 and 2019 Years: A Retrospective Cross-Sectional Study. Int J Fertil Steril 2022; 16: 85-89
• 22 Kohring C, Akmatov MK, Holstiege J. et al. The Incidence of Endometriosis, 2014–2022. An Analysis of Nationwide Claims Data From Physicians in Private Practice. Dtsch Arztebl Int 2024;
• 23 Kohring C, Akmatov MK, Holstiege J. et al. Wird Endometriose in Regionen mit zertifizierten Endometriosezentren häufiger diagnostiziert? [GMS Publishing House]. Gesundheit – gemeinsam. Kooperationstagung der GMDS, DGSMP, DGEpi, DGMS und der DGPH 2024;
• 24 Schmidhuber C, Strotbaum V, Beckers R. et al. Potential of a Telemedical, Inpatient-Outpatient Care Concept to Improve the Quality of Healthcare from the User's Perspective – An Acceptance Analysis of the TELnet@NRW Study. Gesundheitswesen 2024; 86: 723-729
• 25 Lux MP. 10. Zertifizierung. In: Kolberg-Liedtke C, Hrsg. Konservative Tumortherapie Beim Mammakarzinom. Berlin/Boston: Walter de Gruyter GmbH; 2020: 279-288
• 26 Le Moal J, Goria S, Chesneau J. et al. Increasing incidence and spatial hotspots of hospitalized endometriosis in France from 2011 to 2017. Sci Rep 2022; 12: 6966
• 27 Catelan D, Giangreco M, Biggeri A. et al. Spatial Patterns of Endometriosis Incidence. A Study in Friuli Venezia Giulia (Italy) in the Period 2004-2017. Int J Environ Res Public Health 2021; 18
• 28 Rochon J, Du Bois A, Lange T. Mediation analysis of the relationship between institutional research activity and patient survival. BMC Med Res Methodol 2014; 14: 9
• 29 Poppe A, Ansmann L, Meyer I. et al. Spatial and Socioeconomic Patterns of Mental Health and Healthcare Utilization in Cologne, Germany. Gesundheitswesen 2024; 86: S267-S274

Publication History

Received: 06 March 2025

Accepted: 12 June 2025

Article published online:
03 November 2025