Validity of the screening for early endometriosis detection (SEED) screening tool: protocol for a cross-sectional study

Referred to as a “missed disease” and “modern epidemic” by researchers, endometriosis remains largely unrecognised clinically. 1 Endometriosis is a chronic condition characterised by endometrial-like tissue growth outside of the uterus and is prevalent in approximately 10% of Canadian women and approximately 25% of transgender and gender diverse individuals. 2 3 Typical symptoms such as pelvic pain, infertility, heavy menstrual bleeding, as well as bowel and bladder symptoms contribute to poor mental and physical health and reduced quality of life. 4 5 Reports show the average time from symptom onset to receiving an endometriosis diagnosis is 11 years, with a recent survey in the UK suggesting the diagnostic delay is lengthening. 6 7 Despite attempts to improve diagnosis via non-invasive (eg, ultrasound and MRI) and symptom-focused methods, as well as recent calls to action to improve care for chronic pelvic pain, the diagnostic delay remains excessive. 8 – 12 Literature suggests the variability in the presenting symptoms and symptom overlap with other diseases, lack of a non-invasive diagnostic method, the stigmatisation and normalisation of menstrual pain, and lack of training/awareness from healthcare providers all contribute to the delay in diagnosis. 7 13 As such, streamlining the process from endometriosis identification to diagnosis represents a key area for endometriosis care refinement and improvement. The appropriate screening of people reporting symptoms related to endometriosis and subsequent referral to a gynaecologist to confirm diagnosis would streamline this process, support primary care providers and enhance access to timely care for patients. Developing a non-invasive screening tool to aid the diagnosis of endometriosis is the # 4 priority identified in an international priority setting with patients. 14 A recent scoping review identified eight endometriosis screening tools, with the ENDOPAIN-4D demonstrating the highest reliability, content validity and criterion validity, showing promise towards early detection of endometriosis with high accuracy and sensitivity. 15 16 However, the screening tool was validated in a 33–45-year-old French population, leaving its psychometric quality for a young adult (18–25-year-old) unknown. While validated in English, the ENDOPAIN-4D may feature non-Canadian spelling, uncommon phrases and concepts that require an above grade 8 reading level that may cause patient confusion and limit adoptability. With endometriosis symptom onset occurring as early as first menstruation, and symptomology differing as disease progresses, validating for young adults is especially important. 17 Additionally, the young adult population represents a unique patient population, as they are at a transitional point in biopsychosocial development, which necessitates specific healthcare needs. As an example, young adults have reported difficulty transitioning from paediatric to adult healthcare and struggle with expectations, as their goals may be impacted by physical and psychological barriers that differ from their healthy peers. 18 19 If pain management is not addressed in the early stages of symptom onset, often challenging without a clear diagnosis, it can become chronic and have devastating impacts on quality of life, education and vocation, relationships and mental health, including increased risk of opioid use disorder, that can persist into late adulthood. 20 – 23 Young adulthood is a window of opportunity for early identification and intervention that could be pivotal in altering the negative health trajectories associated with untreated endometriosis. Thus, our team conducted a Delphi study to refine the ENDOPAIN-4D for a young Canadian adult population. The result of the two-round Delphi study and working meeting was the development of a 22-item screening tool now called the SEED (Screening for Early Endometriosis Detection). The SEED is a 22-item self-reported clinical screening measure that reflects four primary domains associated with endometriosis and includes pelvic pain, pain during sexual intercourse, bowel and urinary complications, infertility and other relevant symptoms. The SEED is designed to be completed either alone or with the assistance of a healthcare provider. The SEED provides common definitions for words patients may not be familiar with and includes diagrams that refer to specific regions of the body to aid in reducing confusion for patients completing the measure on their own. The objective of this study is to evaluate the measurement properties, specifically criterion validity, specificity and sensitivity, of the SEED.

The study has received approval from the Women’s College Hospital Research Ethics Board (REB # 2024-0065-E). Patients who meet eligibility criteria will be approached by a member within their circle of care to discuss interest in participation. Patients who express interest will be forwarded to the research team, who will provide more detailed study information and assess eligibility. Written informed consent will be obtained electronically via REDCap prior to participation. The study findings will be disseminated through high-quality peer-reviewed journal publications, national and international conference presentations, and plain language English summaries for research participants. Working with key interest holders, our group will take a systematic approach to understanding barriers and facilitators associated with implementation and co-create an implementation toolkit and resources to support rapid adoption of endometriosis screening in young adults. The toolkit will be disseminated through our partner networks, reports, presentations and web-based platforms, facilitating information sharing among young people with endometriosis and clinicians.

We will employ a cross-sectional study design with a clinical screening measure that is completed by patients in tertiary gynaecology and pain clinics as well as in chronic pain clinics. Participants will be recruited from Women’s College Hospital (WCH; Toronto, Ontario), Hamilton Health Sciences – McMaster (Hamilton, ON) and Foothills Medical Centre (Calgary, AB). All participants will be able to understand and read English. Participants will be excluded if they have significant cognitive impairment limiting their ability to read or comprehend the survey. Two groups will be recruited: (Group 1) people aged 18–25 years who were assigned female at birth, of any gender identity, and have endometriosis confirmed through laparoscopy and/or imaging interpreted by an endometriosis expert; and (Group 2) age-matched controls who were assigned female at birth, of any gender identity, and have pelvic pain according to the ICD-11 definition but do not have a confirmed endometriosis diagnosis. Endometriosis and chronic pelvic pain diagnoses will be confirmed through medical chart extraction, and relevant diagnostic details, including the method of diagnosis, will also be collected. Where possible, purposive recruitment strategies will be used to support the inclusion of gender-diverse participants and advance the study’s equity-oriented approach. Although the study may not be powered to establish measurement equivalence across gender groups, representation will allow for exploratory assessment of the tool’s performance and acceptability among gender-diverse young adults. Selecting a control sample with chronic pelvic pain will allow us to assess the ability of the SEED to distinguish endometriosis within a treatment-seeking population. Participants will be recruited from three Canadian pelvic pain and gynaecological clinics. Potential participants will be screened through clinic lists on the electronic medical record systems in real time for either query endometriosis or pelvic pain, and eligible patients will be invited to hear about this study from someone in their circle of care (eg, registered practical nurses, gynaecologists, medical secretaries, physicians). Interested patients will speak to a member of the research team who will provide detailed information on the study and conduct the informed consent process. Participants will complete questionnaires to assess convergent and discriminant validity. Following the completion of the questionnaire package, the participant’s medical chart will be reviewed for detailed extraction of information regarding endometriosis and other pelvic pain conditions. The questionnaire package will contain the following measures: Demographic characteristics: age, sex assigned at birth, gender identity, years lived with pain, ethnicity and socio-economic factors (education, employment, living status and income). This section is informed by the ADDRESSING framework to ensure that identity-related variables known to influence health outcomes will be captured. 24 25 PROMIS Emotional Distress – Anxiety Short Form 8a: 26 contains eight items rated on a 5-point Likert scale to assess symptoms of fear, worry and anxiety, with a total raw score ranging from 8 to 40. Responses can be converted to T-scores using conversion tables, where a higher score indicates more severe anxiety. Assessments of validity using Item Response Theory (IRT) found the 8a was more reliable than other PROMIS Short Forms (eg, 4a and 6a). 27 The 8a was also highly correlated with the original 29-item anxiety scale and supported the unidimensional structure of the PROMIS Anxiety item bank. 27 Measures of internal consistency and test-retest reliability were good (Cronbach’s alpha=0.93 (0.84, 1.0) and test-retest reliability=0.82 (0.75, 0.87) respectively). 28 PROMIS anxiety 8a has been validated in numerous countries, languages and clinical populations. PROMIS Emotional Distress – Depression Short Form 8a: 26 Contains eight items rated on a 5-point Likert scale to assess negative mood and depressive symptoms, with a total raw score ranging from 8 to 40. Responses can be converted to T-scores using conversion tables, where a higher score indicates more severe depression. Through assessment using IRT, the 8a was determined to be more reliable than other PROMIS Short Forms (eg, 4a and 6a). 27 The 8a was also highly correlated with the original 28-item depression scale. 27 Measures of internal consistency and test-retest reliability were good (Cronbach’s alpha=0.97 (0.88, 1.0) and test-retest reliability=0.88 (0.83, 0.92) respectively). 28 PROMIS depression 8a has been validated in numerous countries, languages and clinical populations. PROMIS Pain Intensity 3a: 29 contains three items rated on a 5-point Likert scale, two of which assess pain at its worst and on average in the past 7 days, and one item assesses current pain. Pelvic Pain Impact Questionnaire (PPIQ): 30 the questionnaire contains 8 mandatory items and two optional items rated on a 5-point Likert scale, with the total score ranging from 0 to 32. The original PPIQ showed good internal consistency with a Cronbach’s alpha of 0.92, and a study measuring the validity of the Hungarian version within a chronic pelvic pain and endometriosis population reported a Cronbach’s alpha of 0.88. 30 31 Interstitial Cystitis Symptom Index and Problem Index: 32 composed of two indices to measure Interstitial Cystitis symptoms (ICSI) containing four items with a total score from 0 to 20 (Cronbach’s alpha=0.90) and interstitial cystitis problems containing four items with a total score from 0 to 16 (Cronbach’s alpha=0.85). 32 An ICSI score of 5 or above received a sensitivity of 94% and specificity of 50% in a chronic pelvic pain population with 38% of participants also being diagnosed with interstitial cystitis. 33 Fibromyalgia Rapid Screening Tool (FiRST): 34 comprised of six yes-or-no items with a cut-off score of ≥5, the FiRST has received acceptable psychometric properties (sensitivity, 90.5% and specificity, 85.7%) in various languages (eg, Greek, Spanish and Turkish). 34 – 37 The FiRST was chosen due to the ease of completion (estimated completion time of 3 min) and has yet to be used in a chronic pelvic pain or endometriosis population. Sample size calculations are based on sensitivity estimates, using the ideal profile for sample size as the alternative hypothesis (90%) and the null hypothesis at 60%. As we are creating a screening rather than a diagnostic measure, we are aiming for a sufficiently high degree of sensitivity as our primary outcome, with a lower degree of specificity being tolerated. 38 Assuming the prevalence of endometriosis is 20% in the population 39 and 50% in the recruiting clinics, 4 and a p value of 0.05, a minimum of 100 patients with confirmed endometriosis are needed (Group 1) and 200 age-matched controls are needed (Group 2; ie, those assigned female at birth with chronic pelvic pain). 38 Further, we assume that there will be some missing data and, as such, we will inflate the recruitment by 10%. A total of 300 participants are needed to participate in this study. Criterion validity will help determine whether the screening tool performs well enough to be useful in practice. Specifically, it assesses whether the measure can accurately identify individuals whose symptoms are consistent with endometriosis and support appropriate next steps in care, such as referral to a gynaecologist for further assessment. This phase of the project will measure criterion validity using three approaches: (1) the area under the receiver operating characteristic curve (ROC), (2) sensitivity and (3) specificity. An ROC curve will be produced to express the trade-off between sensitivity and specificity over a range of cut-off scores on the SEED. 40 The area under the ROC curve will be calculated to interpret the ability of the SEED to discriminate between participants with suspected or diagnosed endometriosis versus controls. Area under the curve values will be interpreted as: 0.5 no discrimination, 0.7–0.8 acceptable, 0.8–0.9 excellent and >0.9 outstanding. 41 A cut-off score will be chosen on the ROC curve that achieves (as close as possible) the hypothesised profile for sensitivity and specificity. The shoulder of the ROC curve is typically chosen as the cut-off score with a balanced profile of sensitivity and specificity; 40 however, as the SEED is for screening, particular attention will be paid to sensitivity (>85%) over specificity (>80%) with the input of clinical rationale (ie, consultation with the gynaecologists on our study team). Convergent validity will be evaluated through assessing the relationship between the SEED and other measures that are conceptually related. Correlations will be used to assess the association between the SEED and measures of pain intensity (PROMIS Pain Intensity 3a) and pelvic pain interference (PPIQ). It is hypothesised that the SEED will be moderately positively correlated with measures of pain and pain interference (r>0.60). Discriminant validity will be evaluated through assessing the relationship of the SEED with other measures that are conceptually different. Correlations will be used to assess the association between the SEED and measures of depression (PROMIS Emotional Distress – Depression Short Form 8a), anxiety (PROMIS Emotional Distress – Anxiety Short Form 8a), fibromyalgia (FiRST) and interstitial cystitis (Interstitial Cystitis Symptom Index and Problem Index). We expect the SEED to be weakly positively correlated with measures of depression, anxiety, fibromyalgia and interstitial cystitis (r<0.40). Internal consistency, or the degree of interrelatedness between items in the SEED, will be evaluated through McDonald’s omega. We expect there to be a high degree of internal consistency. We have engaged people with lived and living experience throughout every stage of this study, including project inception, proposal, protocol and early research stages. Specifically, we have engaged both interest holder organisations with representatives on our study team, as well as individuals with lived experience and expertise in sex and gender-based research and women’s health. These knowledge users are part of our project’s advisory team and have participated in co-design of the SEED screening tool. We continue to engage patient partners, primary care leads and pelvic pain specialists throughout the project lifecycle, ensuring relevance, applicability and appropriate dissemination. Participant recruitment began at Women’s College Hospital on 25 February 2026. Recruitment and site activation are ongoing, with study completion anticipated by 1 March 2028.