A persistent diagnostic gap: awareness and workup of bleeding disorders in heavy menstrual bleeding among Dutch gynecologists and hematologists.

An online questionnaire was developed using the web-based platform Qualtrics XM. The Medical Ethics Committee of Máxima Medical Centre (MMC) reviewed the study and confirmed that the Medical Research Involving Human Subjects Act (WMO) did not apply. The Local Feasibility Research Committee at MMC granted feasibility approval (2023-MMC-056). The questionnaire was developed based on the expertise of a working group consisting of fifteen researchers and gynecologists specialized in benign gynecology or hematology, as well as three hematologists. A draft version of the questionnaire was presented to the working group and was subsequently revised based on their recommendations. Two distinct versions of the questionnaire were developed and distributed; one tailored to gynecologists and one to hematologists. The questionnaire was distributed to all gynecologists, hematologists and residents registered with either the NVOG ( n = 1508) or the Dutch Association for Hematologists (NVVH) ( n = 646). This email invitation included a cover letter explaining the study and a link to the survey. To broaden participation, a QR code linking to the survey was also distributed during relevant working group meetings. Respondents were included between June and October 2023. Due to the anonymity of participation, the affiliated centers of respondents could not be identified, and multiple respondents from the same center may have participated. The questionnaire for gynecologists consisted of 22 questions, whereas the version for hematologists included 19 questions. Both primarily consisted of multiple-choice questions, some with an “Other, please specify” option for free-text responses, and concluded with an open-ended question inviting additional comments or suggestions. Different response formats were used depending on the nature of the question (e.g., Yes/No for factual questions, Likert‑type scales for frequency items, and percentage ranges for estimated proportions). All analyses were conducted using the original response formats, and the response scale for each item is reported in the Results. Open-ended responses were independently reviewed by two researchers using a simple content‑review approach. Responses were screened for new themes, unexpected findings, or information not captured by the closed-ended items. As no novel or analytically relevant themes emerged, these responses were not included in the quantitative analysis. Item-level non‑response resulted in varying denominators across analyses. Analyses were performed using all available data per item (pairwise deletion), and the number of respondents included in each analysis is indicated in the Results. The survey was structured into three sections: [ 1 ] general information) [ 2 ], organization of care, and [ 3 ] analysis and diagnostics related to HMB, with a specific focus on underlying bleeding disorders (see Supplementary data for the complete questionnaires ). We included respondents who confirmed that thwy were a gynecologist, hematologist or a resident in one of these specialties, and who completed at least 90% of the close-ended questions. Questionnaires were considered invalid if they were left blank or if the respondent worked in another medical specialty. Questionnaires were classified as incomplete and excluded from analysis if less than 90% of the close-ended questions were completed. An HMB clinic was defined as a department where patients with HMB can typically undergo a see-and-treat process. A screening tool was defined as the use of a pre-visit questionnaire, medical history including bleeding history and family history with attention to bleeding disorders, or the use of the International Society on Thrombosis and Hemostasis Bleeding Assessment Tool (ISTH BAT). Gynecologists were asked about the laboratory tests they request with the question: “Do you use laboratory tests in the context of bleeding disorders in cases of heavy menstrual bleeding?”, with the following response options: never, seldom, regularly, often or always. This was followed by the question: “Do you use a fixed set of laboratory tests for analysis?” Respondents who answered affirmatively were then asked: “Which laboratory tests do you use in the first step?” (multiple answers allowed). Data was extracted from Qualtrics and imported into Statistical Packages for Social Sciences (SPSS). The statistical analyses were performed with International Business Machines Corporation (IBM) SPSS statistics version 29. Descriptive statistics were employed on all variables. In addition to descriptive statistics, between‑group comparisons were evaluated using chi‑square test as appropriate based on data type and distribution. A significance level of α = 0.05 was used.

A total of 261 surveys were returned. Of these, 181 (69.3%) respondents were working in the field of gynecology and 80 respondents (30.7%) were working in the field of hematology. The analyses were limited to 211 surveys. Nineteen surveys were consideredinvalid, defined as quastionnaires that were left blank or completed by a respondent working in another specialty. Eighteen surveys were not filled in, one survey was completed by a respondent working from another specialty. In addition, 31 surveys were incomplete-defined as completion of less than90% of the close-ended questions-, and were excluded from analysis (Fig.  1 ). The response rate was 10%. We defined response rate was defined as the number of completed questionnaires divided by the total number of eligible participants invited to participate. Of the 211 included questionnaires, 149 surveys were completed by gynecologists (71%) and 62 by hematologists (29%). The characteristics of the respondents are presented in Table  1 . Gynecologists reported involvement in the following subspecialties (multiple options possible): benign gynecology ( n  = 117, 79%), obstetrics ( n  = 66, 44%), fertility ( n  = 41, 28%), uro-gynecology ( n  = 48, 32%), and oncologic gynecology ( n  = 40, 27%). Among the participating hematologists, 20 (32%) were employed at a hemophilia treatment center. Overall, 97% of gynecologists reported managing HMB on a regular basis, compared with 60% of hematologists. 13% of all respondents (28/211) participated in a multidisciplinary team meeting on bleeding disorders in which HMB patients were discussed. This was more common among hematologists (34%) than gynecologists (5%). The number of respondents varies per item due to missing responses; exact denominators are shown for each analysis. Fig. 1 Study flowchart Study flowchart Table 1 Characteristics of respondents Characteristics Gynecologists n  = 149 Hematologists n  = 62 Gynecologist 117 (78.5%) Resident gynecology 32 (21.5%) General internal medicine physician 3 (4.8%) Internist-hematologist 52 (83.9%) Internist vascular medicine 2 (3.2%) Resident internal medicine 5 (8.1%) Hospital type University hospital 19 (12.8%) 18 (29.0%) General teaching hospital 88 (59.0%) 36 (58.1%) General non-teaching hospital 35 (23.5%) 8 (12.9%) Independent treatment centre 7 (4.7%) 0 (0%) Focussed HMB clinic present Yes 30 (20.1%) 9 (14.5%) No 119 (79.9%) 53 (85.5%) Structured multidisciplinary team discussion regarding bleeding disorders and HMB Yes 7 (4.7%) 21 (33.9%) No 142 (95.3%) 41 (66.1%) Values are n (%) Characteristics of respondents Values are n (%)

Of 149 respondents who work as a gynecologist, 26 reported that they ‘never’ refer patients with HMB to a hematologist, 114 gynecologists refer ‘seldom’, 7 refer ‘regularly’ and 2 refer ‘often’. None of the responders reported to ‘always’ refers patients. Figure  5 provides an overview of all referral rates. Fig. 5 Referral rate from gynecologist to hematologist of patients with HMB Referral rate from gynecologist to hematologist of patients with HMB Finally, we examined whether gynecologists believe that the Dutch guideline for HMB (NVOG) providessufficient information on the background and diagnostics of bleeding disorders, as this may influence awareness and clinical practice. Among respondents, 49% reported the information provided in the guideline is insufficient.

The estimates of the prevalence of bleeding disorders as an underlying cause of HMB varied considerably among respondents (Fig. 2 ). Most respondents selected categories below the literature-based prevalence of 30% [ 8 ]. The majority of gynecologists (89%) estimated the prevalence to be below 15%, with 40% placing it under 5%. In comparison, 63% of hematologists estimated the prevalence to be below 15%, and most hematologists (77%) estimated it to fall between 5% and 30%. Among respondents working in centers with a dedicated HMB clinic ( n = 39), estimated prevalence rates of underlying bleeding disorders did not differ significantly compared with respondents from centers without such a clinic ( p = 0.402). This between-group comparison was assessed using a chi square test. Fig. 2 Estimated prevalence of underlying bleeding disorders in women suffering HMB reported by gynecologists and hematologists Estimated prevalence of underlying bleeding disorders in women suffering HMB reported by gynecologists and hematologists

Table  3 provides an overview of the pre-visit questionnaires used by gynecologist and hematologists (multiple options possible). The majority of respondents indicated that no questionnaires were send to patients prior to the consultation. Among gynecologists and hematologists, respectively 20% and 9% reported using any form of questionnaire. The ‘PBAC/menstruation calendar” and a ‘self-developed questionnaire’ were most commonly used by gynecolgists. A ‘self-developed questionnaire’ was not further defined. Table 3 Questionnaires before visit, multiple options possible Questionnaire Responding gynecologists n  = 149 Responding hematologists n  = 58 No questionnaire 120 (80%) 53 (91%) Questionnaire 29 (20%) 5 (9%) PBAC/menstruation calendar # 17 (11%) 2 (3%) ISTH-BAT # 0 (0%) 4 (7%) Self-BAT # 0 (0%) 0 Self-developed questionnaire # 15 (10%) 0 Different questionnaire* # 6 (4%) 2 (3%) Values are n (%) PBAC  Pictorial Blood Loss Assessment Chart, ISTH-BAT   International Society on Thrombosis and Hemostasis Bleeding Assessment Tool Self-BAT   self Bleeding Assessment Tool * intake questionnaire, decision aid, uterine fibroid symptom and quality of life questionnaire or general hospital questionnaire # Respondents could give multiple answers Questionnaires before visit, multiple options possible Values are n (%) PBAC  Pictorial Blood Loss Assessment Chart, ISTH-BAT   International Society on Thrombosis and Hemostasis Bleeding Assessment Tool Self-BAT   self Bleeding Assessment Tool * intake questionnaire, decision aid, uterine fibroid symptom and quality of life questionnaire or general hospital questionnaire # Respondents could give multiple answers While respondents working as gynecologists ( n  = 149) did not routinely inquire about bleeding history or a family history of bleeding disorders, more than 80% (52/62 and 49/62) of hematologist respondents reported that they consistently ask about these aspects (Fig.  3 ). A bleeding score (ISTH BAT) is never used by 75% (112/149) of gynecologist respondents. Only 6% of gynecologists reported frequently using laboratory tests to detect bleeding disorders. Fig. 3 Medical history and laboratory evaluation practices among gynecologists and hematologists Medical history and laboratory evaluation practices among gynecologists and hematologists

Heavy menstrual bleeding (HMB) is common in the general population and affects at least one-quarter of women in reproductive age [ 1 ]. According to the Federation of Gynecology and Obstetrics (FIGO), HMB is characterized as menstrual blood loss that negatively affects a woman’s physical, emotional, social, or material quality of life, and is associated with substantial economic costs [ 2 – 4 ]. HMB may result from a variety of underlying pathologies. The PALM-COEIN classification system was developed as a framework for understanding the causes of abnormal uterine bleeding [ 5 ]: Polyps, adenomyosis, leiomyoma, malignancy and hyperplasia represent structural causes, whereas coagulopathy, ovulatory dysfunction, endometrial disorders, iatrogenic and unclassified entities represent causes that cannot be visibly defined by ultrasound or MRI. Systemic disorders of hemostasis are included under the category of coagulopathy. HMB is a leading cause of gynecological consultations in both primary and secondary healthcare settings [ 6 ]. A bleeding disorder is, however, rarely considered in the differential diagnosis of HMB [ 7 ]. A recent systematic review and meta-analysis reported an overall prevalence of bleeding disorders of 30% in hospital-based outpatient population of adults presenting with HMB [ 8 ]. In adolescents with HMB, a prevalence of 19.4% for underlying bleeding disorders has been demonstrated [ 9 ]. The most common bleeding disorder associated with HMB is von Willebrand disease, with prevalence estimates ranging from 5% to 24% [ 10 ]. Platelet disorders, factor deficiencies, and hemophilia carriership are also associated with HMB. Recognizing the presence and severity of bleeding disorders is a fundamental step in the management of patients with HMB. An accurate diagnosis is essential to mitigate bleeding risks during surgery or childbirth and to prevent unnecessary interventions such as endometrial ablation or hysterectomy [ 11 , 12 ]. Moreover, establishing a diagnosis supports family awareness in hereditary conditions and validates patients’ experiences. Nevertheless, in European hemophilia treatment centers, standardized algorithms for managing HMB are lacking, and effective multidisciplinary management is needed [ 13 ]. Diagnosing a bleeding disorder can be challanging and requires specialized expertise. The Dutch Society of Obstetrics and Gynecology (NVOG) recommends against routine screening for von Willebrand disease in cases of HMB and provides no guidance on how to screen for bleeding disorders [ 14 ]. We designed this study to gain insight into current awareness and diagnostic practices regarding underlying bleeding disorders in patients with HMB. The primary objective is to evaluate the awareness and diagnostic work up related to bleeding disorders among gynecologists and hematologists in the Netherlands. Specifically, we aimed to determine whether gynecologists and hematologists are aware of the prevalence of bleeding disorders in patients with HMB, whether gynecologists use screening tools to identify women suspected of having a bleeding disorder, whether laboratory testing is performed in this context, and whether patients with HMB are referred to hematologists for further evaluation.

Our study demonstrates that awareness of underlying bleeding disorders in women with HMB has remained largely unchanged over the past two decades in the Netherlands. The prevalence of bleeding disorders is underestimated-particularly by gynecologists, but also by hematologists- and referral rates remain low. Diagnostic work-up, including the use of screening tools and laboratory testing, is not routinely performed by gynecologists, and structured multidisciplinary discussions are lacking in both specialities. These findings highlights the need for clearer guidelines, targeted education, practical screening tools, and stronger collaboration between gynecology and hematology.

In order to assess which characteristics hematologists consider associated with HMB, and to explore the potential for alignment with gynecologists’ definitions, hematologists were asked to select all relevant characteristics. Almost all symptoms were chosen in over 60% of the cases except for “passage of menstrual clots” and “PBAC score > 100” (Table  2 ). Table 2 Characteristics that hematologists associate with HMB N = 61 Menstruation during > 7 days 43 (71%) Changing pads more often than every two hours 54 (89%) Passage of menstrual clots 26 (43%) PBAC score > 100 24 (39%) Absence from work/school 46 (75%) Hormonal therapy related to HMB 39 (64%) Iron deficiency anemia 39 (64%) Need for iron supplementation 45 (72%) Need for blood transfusion 57 (93%) Values are n (%), characteristics were based on HMB ACOG and NICE guidelines, as well as the ISTH-BAT questionnaire [ 15 , 16 ] Characteristics that hematologists associate with HMB Values are n (%), characteristics were based on HMB ACOG and NICE guidelines, as well as the ISTH-BAT questionnaire [ 15 , 16 ]

Our study evaluated the awareness and diagnostic work up regarding underlying bleeding disorders in patients with HMB among gynecologists and hematologists in the Netherlands. Both groups substantially underestimated the evidence-based prevalence of approximately 30% [ 8 ], although hematologists provided estimates that were closer to the reported prevalence. Pre-visit questionnaires were rarely used, and gynecologists did not apply screening tools to identify an increased risk of an underlying bleeding disorder in patients with HMB– such as obtaining a detailed bleeding and family history, conducting laboratory evaluations, or using the ISTH BAT. During or after consultation, gynecologists infrequently performed structured medical history assessments (e.g. ISTH BAT) or laboratory testing aimed at diagnosing bleeding disorders. Moreover, 94% of responding gynecologists seldom or never refer patients with HMB to a hematologist for further analysis, which is consistent with their perception of a low prevalence of bleeding disorders in this population. The underestimation of the prevalence, the low referral rate, and infrequent use of medical history and diagnostic evaluation among gynecologists demonstrate limited awareness or prioritization of bleeding disorders in the assessment of HMB. Despite the need for effective multidisciplinary management of HMB [ 13 , 17 ], only 13% of respondents participating in multidisciplinary team discussions on bleeding disorders and HMB. Together, these findings suggest missed opportunities for timely diagnosis and treatment. Limited awareness may be partly attributable to the current Dutch HMB guideline, which, according to 49% of responding gynecologists, provides insufficient guidance on how to screen for bleeding disorders. Its primary focus is on identifying anatomical or structural causes of HMB [ 14 ]. The lack of clear recommendations may reflect the limited availability of high-quality evidence on effective screening strategies for gynecologists. Additionally, multidisciplinary team meetings focused on bleeding disorders and HMB appear uncommon in the Netherlands, which may further contribute to low awareness, limited referrals, and infrequent diagnostic evaluation. This is notable given that Curry et al. showed that 90% of women with inherited bleeding disorders consider a multidisciplinary model as beneficial [ 17 ], and Galen et al. similarly emphasize the need for effective multidisciplinary management of HMB in the context of bleeding disorders [ 13 ]. Hematologists also underestimated the prevalence of bleeding disorders in women with HMB and showed gaps in recognizing symptoms indicative for HMB. This may partly explain earlier findings that bleeding disorders are underestimated in hemophilia treatment centers [ 13 ]. Aligning definitions and terminology between gynecologists and hematologists may support more consistent interpretation and communication in multidisciplinary settings, although our findings indicate that such discussions are not routinely implemented. Pre-visit questionnaires are rarely used by hematologists, and the self-BAT is never applied despite available validation studies [ 18 ]. Hematologists do, however,, frequently use the ISTH-BAT during consultations. Notably, 7% of hematologist respondents reported sending the ISTH BAT to patients before consultation, even though it is designed to be administered by a clinician. Our study demonstrates that awareness has remained largely unchanged over the past two decades. Surveys conducted among gynecologists in the US (2002) and UK (2006) showed that bleeding disorders were considered rare causes of HMB: only 4% of US respondents considering von Willebrand disease [ 7 ], and most UK respondents estimated its prevalence to be < 1% [ 19 ]. Women with bleeding disorders continue to be underrecognized and underdiagnosed worldwide, reflecting longstanding gender bias in healthcare [ 20 ]. Underdiagnosis may be partly due to the lack of consensus on which patients with HMB should be screened for bleeding disorders. Even when a gynecological cause is identified, clinicians should consider the possibility of a coexisting bleeding disorder, as these conditions are not mutually exclusive [ 21 ]. Several international guidelines recommend screening individuals presenting with HMB for potential bleeding disorders. The American College of Obstetricians and Gynecologists (ACOG) provides a committee opinion (2019) for screening adolescents with HMB [ 15 ], recommending that clinicians identify risk factors for bleeding during medical history taking and consider using the adapted Philipp screening tool (2011) [ 22 ]. The National Institute for Health and Care Excellence (NICE) similarly recommends considering testing for bleeding disorders in women who present with HMB since menarche and who have a personal or family history suggestive for a bleeding disorder [ 6 ]. In our study, 30% of responding gynecologists reported routinely assessing personal bleeding history, and 40% indicated that they rarely inquire about a family history suggestive for a bleeding disorder. The NVOG guideline currently advises against routine screening for von Willebrand disease in cases of HMB and provides no further guidance on how screening for bleeding disorders in this context [ 14 ]. Consequently, it is not surprising that laboratory testing to detect bleeding disorders is rarely used by gynecologists in treating patients with HMB. Comprehensive coagulation tests are also not routinely available in general hospitals in the Netherlands. Combined with limited awareness and limited collaboration with hematology departments, these factors likely contribute the low rates of laboratory evaluation observed. Previous studies corroborate this pattern. An Australian retrospective audit reported that 40–54% of women with HMB attending a tertiary women’s hospital who were at risk of having a mild bleeding disorder were not tested for [ 23 ]. Among post-pubertal girls and adolescents with HMB, screening for von Willebrand disease occurred in only 8% of cases, and in 16% of severe cases [ 24 ]. Similarly, a UK survey found that only 12% of gynecologists would test an 18-year-old with HMB for von Willebrand disease, and only 2% would test a 35-year-old [16]. In our study, 94% of responding gynecologists rarely or never referred patients to a hematologist for further evaluation of the coagulation system. This aligns with the findings of Dilley et al. in 2002, who reported that only 3% of surveyed American gynecologists referred HMB patients to other specialists [ 7 ]. The persistence of such low referral rates over decades suggests systemic gaps in interdisciplinary collaboration. A major strength of this study is its nationwide scope, including both gynecologists and hematologists, which provides complementary perspectives on the diagnostic approach to bleeding disorders in patients with HMB. Moreover, the survey addressed not only perceived prevalence but also evaluation and referral practices, thereby highlighting key gaps in awareness and clinical practice that may serve as targets for improving care pathways. An additional strength is that the survey was developed and refined in collaboration with an expert group, ensuring clear phrasing of questions and response options that closely reflect daily practice. Several limitations should be acknowledged. First, due to anonymity, center-level comparisons could not be made, and findings are based on self-reported practices rather than observed behavior. Because multiple respondents may have originated from the same center, responses may not be fully independent. This potential clustering effect was not accounted for in the analyses and should be considered when interpreting the results. Second, volunteer bias may have influenced the findings. Physicians with a particular interest or involvement in this topic are likely overrepresented in the sample, which may limit the representativeness of Dutch gynecologists and hematologists managing HMB. These respondents may have been more knowledgeable or attentive when completing the questionnaire, potentially leading to an overestimation of best practices. Additionally, variation in denominators due to item non‑response may have introduced minor differences in sample composition across analyses. Finally, only 9.8% of gynecologists and 9.5% of hematologists notified via national distribution were ultimately included. Although the survey was sent to all gynecologists and hematologists in the Netherlands, information on the age groups of the patients they treated was not collected. Consequently, it is unknown whether pediatric hematologists, adolescent medicine specialists, or pediatric/adolescent gynecologists participated. The inability to distinguish between pediatric/adolescent provides from adult-focused providers may have influenced the observed levels of awareness and limits the interpretation of our findings. Future studies should focus on strategies to increase awareness of bleeding disorders as an underlying cause of HMB, as this is essential for appropriate referral and timely diagnosis. These studies should explicitly include the patient population (pediatric, adolescent and adult) to better capture potential differences in knowledge, diagnostic approaches, and clinical practice patterns. In addition, further research is needed to develop and validate efficient, practical screening tools that can be implemented in gynecology practice. Such tools should aim to identify women with HMB who are at risk for bleeding disorders early in the diagnostic process. Finally, the development of a uniform international guideline for the assessment of bleeding disorders in HMB is warranted to reduce diagnostic delays and improve diagnostic workup and overall patient care.

A total of 145 gynecologists responded to the question ‘which laboratory tests do you use in the first step? (multiple options possible)’. When performing laboratory testing, most gynecologists initially requested hemoglobin/hematocrit (132/145), ferritin (78/145) and von Willebrand activity (75/145). Common laboratory tests requested by hematologists are platelet count (62/62), hemoglobin/hematocrit (61/62), prothrombin time, activated partial thromboplastin time (60/62) and von Willebrand activity (51/62). Figure  4 shows an overview of the laboratory tests performed. Fig. 4 Initial laboratory testing by gynecologists ( n =145) and hematologists ( n =62) Initial laboratory testing by gynecologists ( n =145) and hematologists ( n =62)

Supplementary Material 1. Supplementary Material 1.