Barriers to a Timely Diagnosis of Endometriosis: A Qualitative Systematic Review

23

24 The diagnosis of endometriosis, a common gynaecological condition, is characterised by delays of up 25 to eleven years. During this time, women may experience persistent symptoms that affect their 26 quality of life, productivity, and relationships, and the disease may progress. It is therefore a priority 27 for us to understand the factors that contribute towards this delay so that the diagnostic pathway 28 can be improved. We sought to examine the barriers to the timely diagnosis of endometriosis, from 29 the perspective of affected women and healthcare professionals, using a qualitative evidence 30 synthesis. 31 32 DATA SOURCES 33 We searched MEDLINE, PsycINFO and CINAHL using combinations of the synonyms of 34 “endometriosis”, “diagnosis”, and “qualitative research”. The search was restricted to English-35 language articles from database creation until May 2022. 36 37

OF STUDY SELECTION 38 The initial search yielded 899 articles. Studies were included if they explored the experiences or 39 views of affected women and/or healthcare professionals on the diagnosis of endometriosis, used 40 qualitative methods for data collection and data analysis. Risk of bias was assessed using the CASP 41 checklist. 42 43 TABULATION, INTEGRATION, AND RESULTS 44 3 Thirty seven articles were considered for full review, with thirteen eligible papers, identified for 45 inclusion. Thematic synthesis was used to identify four key themes; a) individual factors (n=6), b) 46 interpersonal influences (n=6), c) health system factors (n=13), and d) factors specific to 47 endometriosis (n=8). These encompassed twelve sub-themes which represented barriers to the 48 diagnosis of endometriosis; a) difficulty establishing pathological symptoms from normal 49 menstruation, b) menstrual stigma and the normalisation of menstrual pain, c) poor attitudes and 50 lack of training for healthcare professionals, delayed referrals to specialist services, poor explanation 51 of oral contraceptive pill use in the diagnostic process, and d) variability in presenting symptoms, 52 overlap with other conditions, lack of a non-invasive method of diagnosis, and concerns about the 53 value of a diagnosis. 54 55

56 This review identified barriers to the timely diagnosis of endometriosis from the perspective of 57 affected individuals and healthcare professionals. It highlights areas for improvement along the 58 diagnostic pathway, which may guide future strategies to reduce the delay. Further research to 59 identify the facilitators to diagnosis and would be beneficial. 60 61 62 63 64 65 66 67 4

68 Endometriosis is an inflammatory condition affecting women, in which endometrial tissue (the lining 69 of the uterus) is found at sites outside of the uterus1. Approximately 10% of women of reproductive 70 age are affected (190 million women worldwide)2,3. The symptoms associated with endometriosis 71 exert a substantial burden on the lives of individuals and on society as a whole 4. Endometriosis is 72 associated with reduced health-related quality of life (HR-QOL), reduced mental wellbeing, adverse 73 effects on intimate relationships, and reduced social activity4-6. It is also associated with reduced 74 productivity at work or school4, which contributes towards a significant loss of income5. 75 Endometriosis is estimated to cost the United Kingdom (UK) economy over £8.2 billion a year due to 76 lost productivity and healthcare costs, which is comparable to other chronic conditions, such as 77 diabetes mellitus and rheumatoid arthritis 7. 78 79 The burden of endometriosis is prolonged by a well-established diagnostic delay; there is an average 80 of between four and eleven years from the onset of the first symptom to receiving a diagnosis5. This 81 delay is seen worldwide, even in countries with access to universal healthcare 5. 82 There has been limited research to ascertain the reasons for the diagnostic delay in endometriosis. 83 This systematic review aimed to explore the barriers to the timely diagnosis of endometriosis from 84 the perspective of affected women and the healthcare professionals involved in the diagnosis. 85 86 SOURCES 87 A scoping search was undertaken to provide an overview of the literature and identify relevant 88 search terms for a subsequent systematic review. The SPIDER framework8 was used to identify key 89 concepts, including “endometriosis”, “diagnosis” and “qualitative research”, which provided free-90 text terms for initial searches of MEDLINE, The Web of Science, and Google Scholar. The first 100 91 5

from each database were screened. The titles, abstracts, and index terms of relevant studies 92 were reviewed, and a list of additional search terms was developed (Supplementary Tables 1-2). 93 The Cochrane Library and PROSPERO were searched to identify systematic reviews on the topic, 94 using the key terms “endometriosis” and “diagnosis” to ensure this review did not repeat any 95 previously published or ongoing reviews. 96 The research question was divided into key concepts based on the SPIDER framework 8. 97 Combinations of the free-text terms and Medical Subject Headings (MeSH) that had been identified 98 by the scoping search were then trialled for each concept until an acceptable balance between 99 sensitivity and specificity was reached. 100 Qualitative search filters, designed by the University of Texas were also added to the final search 101 strategy9. Filters to further narrow the search, for example, based on the characteristics of the study 102 population, date of publication, or study design were not used. 103 Three databases were used to identify relevant studies: 104 MEDLINE via Ovid (1946-May 2022) 105 PsycINFO via Ovid (1887- May 2022) 106 The Cumulative Index to Nursing and Allied Health Literature (CINAHL) via EBSCO (1937- May 2022). 107 Searches were initially carried out in July 2021 and then repeated in May 2022; all identified records 108 were exported to Rayyan10. A supplementary search was also performed using the CLUSTER 109 (Citations, Lead Authors, Unpublished materials, Scholar searches, Theories, Early examples, and 110 Related Projects)11 procedure. 111 Following the identification of key “pearl” papers, forward citation searches were carried out using 112 the Web of Science and the ‘Related Article’ functions on MEDLINE and CINHAL. The reference lists 113 of included studies and the name of the lead author were also searched. Authors were not 114 6 contacted directly, but efforts to identify their work included searching their publications lists and 115 institutional repositories. Finally, all volumes of the ‘British Journal of Obstetrics and Gynaecology’ 116 and the ‘Journal of Endometriosis and Pelvic Pain Disorders’ published between January 2011-May 117 2022 were searched for relevant articles using the key terms “endometriosis” and “diagnosis”. 118 Studies were selected using pre-determined inclusion and exclusion criteria. To be included in the 119 review, the study population must have included women with a diagnosis of endometriosis, and/or 120 healthcare professionals (regardless of speciality) that had experience of diagnosing the condition. 121 The ESHRE guidelines12 suggest a two-step approach in which a diagnosis can be made clinically 122 (based on the history, pelvic examination, and ultrasound imaging) before a trial of treatment, or by 123 direct visualisation during laparoscopic surgery13. Accordingly, laparoscopy is usually offered only to 124 patients with negative imaging results or where empirical treatment was unsuccessful or 125 inappropriate12. We therefore included patients that had been diagnosed in either way. This 126 definition was used successfully in a previous Cochrane review and ensured that the review captured 127 patients’ experiences of both methods14. 128 129 There were no limitations placed on the age of the participants, the severity of the endometriosis, 130 the period of time since diagnosis, or the sample size. As a range of healthcare professionals may be 131 involved in the diagnosis, studies involving any specialist with experience of the condition were 132 included. This review was limited to studies available in the English language, and published studies, 133 but there was no limit placed on the date of publication for searches via OVID. 134 135 As this study aimed to explore participants’ experiences and reflections on the existing factors 136 contributing to the delayed diagnosis of endometriosis, only non-interventional, qualitative studies 137 focusing on the participants’ views on the diagnostic pathway were included. Studies focusing on the 138 7 general experience of having the condition, treatment, second-hand accounts, or on a specific 139 intervention were therefore excluded. 140 Qualitative studies were sought as, by design, these approaches delve deeper into the experiences 141 of a patients’ condition; something quantitative surveys for example, would struggle to achieve. 142 Included studies must therefore have used an appropriate qualitative method for data collection and 143 data analysis such as focus groups or interviews analysed using thematic analysis, framework 144 analysis, content analysis or constant comparison. Data was also extracted from the qualitative 145 elements of relevant mixed-methods studies if there was adequate information to do so. 146 147 STUDY SELECTION 148 Records were filtered by a single reviewer (lead author). A title search was undertaken followed by a 149 review of the abstracts and those which did not meet the eligibility criteria were excluded. The full 150 texts of the remaining studies were retrieved using the research centre online search tool or by 151 contacting the research centre library, and those that met the inclusion criteria were chosen for the 152 data synthesis. 153 All decisions were recorded on Rayyan10 and any uncertainties about the eligibility of a study were 154 discussed with the corresponding author. Multiple reports from the same study were treated as a 155 single study, with reference made to all publications. 156 Relevant studies were appraised using the Critical Appraisal Skills Programme (CASP) checklist for 157 qualitative research15 but were included in the final review, regardless of methodological quality. 158 The CASP checklists are freely available, and easy to use materials to aid health professionals to 159 critique journal articles by highlighting the key areas to focus; different checklists are available for 160 varying study designs. Both Cochrane and CASP themselves recommend against rejecting studies 161 8 based on an arbitrary cut-off16 as including studies of lower quality may strengthen the evidence for 162 findings in the synthesis17 p.33. 163 Standardised data extraction forms were used to provide consistency, reduce the risk of bias, and 164 improve the validity of the data collection process17.The author, year of publication, country, setting, 165 research question and aim, study design, the participant characteristics, the method of data 166 collection, and the method of data analysis were recorded. Quotes from study participants (‘first 167 order constructs’), along with existing author interpretations (‘second-order constructs’) were then 168 collected. Thematic synthesis was used to analyse the findings of the included papers18. We 169 acknowledged the positionality of the lead author, a White female working as a junior doctor in the 170 UK, and reflected on how we could minimise its influence on the interpretation of results. 171 GRADE-CERQual19 was then used to evaluate the findings of this review and was recorded in a 172 GRADE Evidence Profile. This is an approach to judge how much confidence to hold in qualitative 173 evidence synthesis, assessing four key areas of methodological limitations, coherence, adequacy, 174 and relevance. A GRADE Evidence Profile is the name ascribed to the summary presentation of key 175 information from the articles in the synthesis. Due to the original nature of this project (a Masters 176 dissertation) the review protocol was not pre-registered. 177 178

179 995 records were identified, including 936 records from the electronic databases and 59 potentially 180 relevant records from the supplementary searches. Following the removal of duplicates, 899 unique 181 papers remained. 826 papers were excluded during the title review and 36 following the abstract 182 review. A further 24 papers were excluded following the full-text review: nine focused on an aspect 183 of endometriosis other than the diagnosis, six focused on a disease other than endometriosis, four 184 did not have an appropriate qualitative research design, three did not explore the diagnosis from the 185 9 perspective of affected women or healthcare professionals, and one was a conference abstract of an 186 included study. In addition, one paper20 could not be retrieved, despite assistance from the library 187 within the research centre and attempted contact with the author. 188 In total, 13 papers, reporting on 12 studies, were included in the qualitative synthesis (Figure 1); the 189 titles and characteristics of the included papers are provided in Table 1. All were published as journal 190 articles between 2003 and 2022. Of these, five were conducted in the United Kingdom (UK)21-25, 191 three in Australia26-29, two in the United States of America (USA)30,31, one in Sweden32, and one in the 192 Netherlands33. All studies were therefore carried out in high-income countries, with the majority 193 having universal health coverage. 194 Ten of the studies used a qualitative research design21-25, 27-29,31-33 and two used mixed methods with 195 adequate information provided to extract the qualitative data 26,30. Three studies used a narrative 196 approach21,24,27,28, two used grounded theory22,25, and one used phenomenology29. The remaining 197 studies did not specify their methodological background. 198 The samples of nine studies21-24,26-29,31,32 were made up exclusively of women with endometriosis and 199 one30 included women with endometriosis as part of their larger sample. The sample sizes ranged 200 from 9 to 222, with data being collected from 476 participants in total. The age of participants 201 ranged from 16-78 years, although one study did not report the age of their participants21. Few 202 papers reported further demographic information; those that did described a predominantly white, 203 working class population27/28,31. In most of the studies, participants were recruited using convenience 204 sampling methods from specialist endometriosis centres and non-profit organisations, with snowball 205 sampling used as an additional method in two studies21,30. 206 Six of the studies only included participants who had received a laparoscopic diagnosis of 207 endometriosis21-24,29,31, whereas three included those who had received a clinical diagnosis26-28. The 208 severity of endometriosis was only discussed in one paper 22, which stated 58.4% of the women had 209 mild endometriosis, 33.3% had moderate to severe endometriosis, and 8.3% had deep infiltrating 210 10 endometriosis. Most participants had experienced a delay from the onset of symptoms to diagnosis, 211 which ranged from an average of 3.8 to 8.1 years. 212 The samples of the remaining two studies were made up exclusively of healthcare professionals that 213 either had experience of diagnosing endometriosis25,33, or were expected to diagnose endometriosis 214 as part of their primary job role25. Zale30 also included healthcare professionals in their sample in 215 addition to patients. The number of healthcare professionals included in these samples ranged from 216 4-43, with data collected from 89 professionals in total; 85 general practitioners, two physicians that 217 had specialised in obstetrics and gynaecology (neither specialising in endometriosis), one radiologist, 218 and one internal medicine physician. Participants were recruited using convenience methods from 219 physician registers and from professional events. The number of years of experience in diagnosing 220 endometriosis ranged from 0 (doctors in training) to >20 years. 221 The phenomenon of interest in all studies was the diagnosis of endometriosis. Four studies gathered 222 information on affected women’s experiences of receiving a diagnosis of endometriosis 223 specifically23,24,26,32, five focused on women’s experiences of endometriosis in general, with the 224 diagnosis as a significant topic 21,22,27-29,31, and two aimed to explore healthcare professionals’ 225 experiences of diagnosing and managing endometriosis 25,33. Zale (2020)30 explored both patient’s 226 and physician’s experiences. 227 The primary outcome of all studies was the experiences, views, beliefs, or attitudes of the affected 228 women and/or the healthcare professionals. Seven studies gathered this information using semi-229 structured interviews21-25,30,32, one used unstructured interviews27/28, three used focus groups26,29,33 230 and one used an open-ended questionnaire31. All interviews and focus groups were audio-recorded 231 and were transcribed for analysis. Eight of the studies used inductive thematic analysis21,23-24,26-28,30,31 232 with grounded theory 22, the Modified Stevick-Colaizii-Keen method32, content analysis33, and 233 framework analysis25 all used in one study each. The methodological quality of the studies was 234 generally very good, including one22 which was of the best quality according to CASP15 (Table 2). 235 11 Thematic analysis of the studies highlighted multiple factors, that act as barriers to the timely 236 diagnosis of endometriosis. These were categorised into four overarching analytical themes focusing 237 on a) individual factors, b) interpersonal influences, c) health system factors, and d) factors specific 238 to our knowledge of endometriosis. Twelve sub-themes are presented within these categories. 239 Verbatim quotations, taken directly from the primary studies, are used throughout to illustrate the 240 themes and to provide evidence for the reviewer’s interpretations 34. 241 Individual factors 242 Individual factors contributed to the prolonged period between the onset of symptoms and the 243 seeking of medical advice, which accounted for up to half of the diagnostic delay27. This concept was 244 explored in six of the studies included in this review22,23,27-29,32, resulting in the identification of two 245 sub-themes. 246 As reported in all six studies, affected individuals struggled to distinguish their pathological 247 symptoms from the normal experience of menstruation. Having had painful periods from menarche, 248 they often assumed that their symptoms were within the spectrum of normal and something that all 249 women endured27. This assumption remained even when symptoms were severe and disruptive, 250 with many individuals reporting that they thought they were ‘unlucky’ to have ‘problematic periods’ 251 rather than unwell. 252 This was reinforced by their relatives. Participants in Markovic28 explained that they drew primarily 253 on the experiences of their mothers and grandmothers to learn about menstruation, and so, if they 254 too had gynaecological problems, they learned to regard severe period pain as normal. This is 255 particularly relevant given that endometriosis has a genetic component35. In these situations, 256 affected women were unlikely to recognise their own symptoms as atypical or in need of medical 257 attention. 258 259 12 260 261 These beliefs about the normality of the symptoms of endometriosis often delayed help-seeking 262 until a “circuit breaker” event forced the affected women to reconstruct their experience as 263 pathological27. For example, participants reported that they only sought medical advice when their 264 symptoms began to affect their ability to perform everyday tasks, when a significant other 265 encouraged them to seek help, or when they failed to conceive27. 266 267 268 Another common theme was the use of self-care techniques to manage symptoms. Participants in 269 Jones22, Ballard23 and Manderson27 reported that they developed a wide range of coping strategies, 270 including spending time in bed, using hot water bottles and heating pads, withdrawing from social 271 activities, and taking analgesia, which enabled them to mask their symptoms for some time. They 272 delayed seeking medical advice until self-management became unfeasible or an acute problem 273 arose. 274 Interpersonal Influences 275 Findings related to interpersonal factors were collated into two sub- themes; a) menstrual stigma 276 and b) the societal normalisation of menstrual pain. These concepts emerged from six of the studies 277 and discourage individuals from seeking advice for their symptoms22,23,25,29,30,33. 278 Firstly, affected individuals reported that they felt that talking about their menstrual problems was 279 socially unacceptable, seeing it as potentially embarrassing or shameful22,23,29. They often concealed 280 their symptoms from others, which exacerbated the difficulties they had in differentiating normal 281 from pathological menstruation. As menstruation was rarely discussed, they missed out on the 282 opportunity to gain comparative evidence to suggest that their experience was abnormal. 283 “I had witnessed my mum with so many problems in her life, she used to get a shot of pethidine every month, so when I started to get period pain in teenage years, I just thought that was fairly normal.” [Patient] (Manderson)27. “I remember one day someone [at school] saying, ‘You shouldn’t have any pain, that’s not normal’”. [Patient] (Manderson)27. 13 If they did discuss with peers, the social narrative around painful periods being a normal part of 284 womanhood was influential in deterring women from seeking help 22,24. Participants reported that 285 their symptoms were often dismissed by their relatives, peers, and colleagues as “just period pain”. 286 Their symptoms were often trivialised, and they were made to feel that there was a problem with 287 how they coped with pain, rather than an underlying disease. Interestingly, this phenomenon was 288 reported in studies as recent as 2021, suggesting that this persists 3,31,32,33. 289 290 291 From the perspective of healthcare professionals, the stigma around “women’s issues” and sense 292 that discussion could be embarrassing and uncomfortable made it difficult to gather enough 293 information to make a diagnosis 25. Patients rarely disclosed information about their symptoms in 294 the initial consultation, meaning that healthcare professionals often had to see a patient multiple 295 times before they could get an accurate impression of their experience. 296 297 298 299 300 Health system factors and factors specific to endometriosis 301 Delays to diagnosis once the individual had sought medical help for their symptoms were a 302 recurring theme in all 13 studies.21-33. Barriers relating to the health system and endometriosis itself 303 were grouped into sub-themes which have been explored from the view of both patients and 304 healthcare professionals. They have been discussed together to allow deeper analysis of the way in 305 which the factors interact. 306 “They might not want to tell you the first time they meet you that they’re actually having really painful sex, and they might be quite worried about that. Or they might not want to open up to you about that, but maybe once you’ve seen them once or twice, they might say” [GP] (Dixon)25. “I am sure people thought she is making it up again. Oh god, everybody has a time of the month, why can’t she just snap out of it” [Patient] (Jones)22. 14 Firstly, participants described a sense of being dismissed or treated with scepticism by their general 307 practitioner (GP) when they sought help21,23,24,27,29-32. Indeed, most participants in Ballard23 cited their 308 doctors’ disbelief of the severity of their symptoms as the “least helpful aspect of their treatment”. 309 Patients felt that their symptoms were trivialised as “just period pain”, which were not worthy of 310 medical attention23. Some women, particularly in Grundstrom32, perceived healthcare professionals 311 as distant and nonchalant, getting the impression that they were not interested in hearing about 312 ‘women’s problems’. 313 314 315 316 When they continued to seek help after the initial dismissal, participants often reported that they 317 were accused of exaggerating or “imagining” their symptoms 21-24,28,30,31. The impact of this, alongside 318 the women’s lack of knowledge about ‘normal’ menstruation, led to some questioning their own 319 experiences, with participants in Ballard23 stating they felt they were “going mad”. This theme was 320 repeated, with many affected women discouraged from seeking further help as they felt they must 321 be wasting the doctor’s time. 322 323 324 325 Another issue was that patients perceived healthcare professionals to have poor knowledge of 326 endometriosis. Participants expressed frustration that their doctors appeared to lack even basic 327 awareness of the condition31,32 as “there is an expectation that clinicians have expertise in the 328 diseases that they treat”30. This was seen as a major barrier to diagnosis, as the healthcare 329 “He [the GP] brushes it off and says its period pain, and you can get on with it. That’s his entire attitude, have some ibuprofen and get on with it” [Patient] (Denny)24. “It’s a shame women have to endure years of physicians making light of their pain and being told ‘it’s part of being a woman’” [Patient] (Mikesell)31. “I was experiencing a lot of pain on penetration…I went to the doctors and they did an internal and said ‘Look, everything is perfectly normal’ and suggested that it might be a psychological problem, and I might just be anxious” [Patient] (Denny)21 15 professional must have known enough about the condition to ‘think of it in the first place’ and to 330 make an appropriate referral30. 331 332 333 This was also recognised by the healthcare professionals themselves 25,30,33. GPs in Van der Zanden33 334 reported that their knowledge of endometriosis was limited, admitting they found it difficult to 335 differentiate between normal menstruation and pathological conditions like endometriosis. Without 336 adequate knowledge of the disease, the doctors struggled to identify the ‘right’ questions to ask and 337 could not perform an appropriate examination. 338 339 340 This was attributed to a lack of clinical training and experience, with physicians stating that 341 endometriosis was not emphasised or covered comprehensively during their medical training25,30,33. 342 Multiple GPs and non-specialist gynaecologists also reported that they did not have access to clear 343 guidelines for diagnosing endometriosis,30,33. They suggested that clinical guidelines, developed by a 344 multidisciplinary team would be useful to advise them in which therapies they can start themselves, 345 when referral is advised, and what actions may be taken in secondary care 30,33. However, this was 346 not identified as a problem in Dixon 25, a study carried out with GPs in England. This suggests that this 347 problem is country-specific, with guidelines varying in quality and accessibility. 348 The variability in the symptoms of endometriosis was also identified as a barrier to diagnosis from 349 the perspective of healthcare professionals25,30,33. The symptoms of endometriosis can be vague, 350 ranging from chronic pelvic pain, to dyspareunia, bowel problems, urinary symptoms, and chronic 351 fatigue36. This made it difficult for healthcare professionals to differentiate it from other 352 gynaecological and gastrointestinal problems. Particularly, endometriosis was often masked by a 353 “My doctor did not have a great deal of knowledge in that area. With two of my doctors, I had to explain to them what endometriosis is and how to treat it. So I can’t exactly go to them and ask for help with my treatment” [Patient] (Moradi)29. “You have to think of it in the first place. You have to know the condition before it even occurs to you. And if you don’t think of it, you will not find the information you need” [GP] (Van der Zanden)33. 16 diagnosis of irritable bowel syndrome (IBS). These conditions may present with similar symptoms, 354 and indeed, often co-occur in the same individual, making them difficult to distinguish. 355 From the perspective of professionals, misdiagnosis was therefore an inevitable part of the process. 356 GPs are trained to work through a clinical hierarchy, where ‘red flags’ indicating serious disease must 357 be excluded before other diagnoses (including endometriosis) can be considered. As the symptoms 358 of endometriosis overlap with other conditions, and because endometriosis cannot be reliably 359 diagnosed in primary care, the tests offered were predominantly used to rule out other causes of the 360 symptoms. Once the life-threatening conditions were ruled out, a referral for specialist 361 endometriosis care could be made. 362 363 364 This is in line with clinical guidelines 34 but appeared to be poorly explained to the patients, who felt 365 that their concerns about endometriosis were being ignored. Participants also reported that there 366 was a delay before other diagnoses, such as endometriosis, were considered after the first line 367 treatment for alternative conditions (such as IBS) failed22. This could therefore act as a barrier to 368 diagnosis, as affected women could become disengaged and fail to return to their doctor when 369 symptoms persisted. 370 371 372 Another miscommunication appeared to occur with the use of the oral contraceptive pill (OCP). 373 From the perspective of the healthcare professionals, a trial of treatment with the OCP could be 374 informative to the diagnostic process. As recommended in the ESHRE guidelines12, the OCP was 375 often started early. If symptoms improved, it increased the likelihood that they were due to 376 endometriosis and therefore contributed to reaching the correct diagnosis25. However, if the 377 “Always as a GP, we’re terrified of missing something horrible, so you think about red flags” [GP] (Dixon)25. “They had to test for everything first and never once mentioned endometriosis as a possibility” [Patient] (Denny)24. 17 purpose of the trial of the OCP was not adequately explained to the patient, they often became 378 disengaged, believing that their symptoms were being “masked” rather than treated22,23,25,28,33. 379 Patients reported that they did not want to take a medication without knowing the underlying cause 380 of their symptoms21. They also worried that the diagnostic pathway would need to begin again as 381 soon as they could no longer use the OCP, for example, if they wanted to conceive or if they 382 experienced side effects22,23,25. 383 384 385 386 387 When further investigations were carried out, the lack of an accurate, non-invasive method of 388 diagnosis became an issue 23-25,29,30,32,33. This delayed diagnosis in two ways; unspecific, non-389 discriminatory investigations were often used to inappropriately rule out endometriosis as a 390 diagnosis, or diagnosis was delayed due to concerns about the risks of a laparoscopic procedure. 391 In both Ballard 23 and Mikesell31 , patients reported that a diagnosis of endometriosis was discounted 392 following a normal transvaginal ultrasound, which falsely reassured their doctors that further 393 investigation or treatment was not necessary. This is an inappropriate use of the imaging modality 394 as, although the ESHRE guidelines recommend the use of ultrasound in the diagnostic work-up for 395 endometriosis, a negative finding cannot be used to exclude endometriosis, particularly in superficial 396 peritoneal disease or in cases of high clinical suspicion12. 397 398 399 400 “You have to really try and positively engage the patient to come back and really convince them that you want them to come back. Because you know if you give them a 3-month trial of treatment, it doesn’t help, and they think ‘Well nothing really happened last time, I’m not bothered, I’m going to leave it 6 months, and then I’m going to come back’, the years slip so quickly in that kind of timescale” [GP] (Dixon25). “It was awful just going for these flipping internals all the time and being told there’s nothing there. To actually keep going backwards and forwards and having it, and there’s nothing showing up. And when I’ve mentioned about having the… is it the laparoscopy – having that done, they’re ‘Well, no, it won’t be done because there’s nothing showing up on these [ultrasounds]’!” [Patient] (Ballard23). 18 The misconception amongst healthcare professionals that a diagnosis can only be made by 401 laparoscopic surgery also acted as a barrier. Individuals with endometriosis in six of the papers 402 reported that there was a delay from the first consultation in primary care to receiving a referral to a 403 specialist endometriosis service21-24,26,27,31,32. Participants in Denny24 perceived that their GP was 404 reluctant to refer them, with most affected women reporting that they didn’t think they were 405 referred to secondary care soon enough. 406 From the GPs’ perspective in Dixon25 and Van der Zanden33, this was due to a difficulty in 407 determining when a referral was appropriate. In some cases, for example, if there were concerns 408 about fertility, referral was straightforward, but for patients with mild or vague symptoms, GPs 409 reported that they struggled to determine whether a referral would be in the best interest of the 410 patient33. They were mindful that, traditionally, diagnosis required an invasive procedure with 411 potential risks, and some therefore saw it as a last resort33. 412 413 414 This was exacerbated by doubts about the added value of establishing a definitive diagnosis, since 415 the treatment options often appeared to be the same whether the diagnosis was confirmed or not. 416 Several GPs described experiences where specialists had prescribed medications that they could 417 have given themselves33. This led to some GPs opting to trial treatment before making a referral to 418 secondary care. 419 Others expressed doubt about the value of a diagnostic label at all. In Dixon25, healthcare 420 professionals cautioned that having a diagnosis might not address all the patients’ concerns, and 421 could instead lead to further anxiety and distress. This led to a low sense of urgency for some of the 422 GPs, with participants stating that they would prefer to see if a trial of treatment with the OCP 423 worked before even raising endometriosis as a possible diagnosis. 424 “The only way you can really make it is with a laparoscopy and you don’t really want to put a woman through that unless you’ve reached the absolute end of the road” [GP] (Dixon)25. 19 425 426 427 Notably, participants did not share this sentiment. The majority of affected women in seven of the 428 studies reported that getting a diagnosis for their symptoms was positive, allowing them to process 429 their experience and access specialist support21-24, 26,29,32. 430 Finally, participants in Zale (2020)30, a study carried out in the USA, which does not have universal 431 health coverage, also identified financial barriers to seeking a diagnosis. Even patients with access to 432 health insurance reported constraints on the providers that they were able to visit, the tests that 433 they could access, and ultimately their treatment options due to high co-pays and deductibles. This 434 meant diagnosis was often delayed due to financial restraints. 435 436

437 Thirteen papers, reporting twelve studies, were included identifying twelve barriers that delayed 438 diagnosis of endometriosis. Affected individuals reported that they struggled to differentiate 439 pathological symptoms from normal menstruation, which was reinforced by the societal 440 normalisation of menstrual pain and the lack of opportunity to discuss menstrual problems with 441 others. This prevented them from seeking medical advice until an acute event caused them to 442 reinterpret their experience as abnormal, prolonging the delay to seek medical advice. During this 443 time, many patients with endometriosis used self-management strategies to cope with symptoms, 444 further delaying their presentation to a medical professional. This suggests that there is a need to 445 provide education about what constitutes ‘normal’ menstruation, potentially at secondary school as 446 the symptoms of endometriosis often start in adolescence, and to raise awareness of menstrual 447 conditions. This would have multiple potential benefits; affected women may be encouraged to seek 448 “Moderator: For example a young girl, 16 years old, who comes to your office with complaints about her menstruation, would you tell her, you consider endometriosis when you prescribe her the pill? GP3: I wouldn’t even think of it. GP2: No, me neither”. (Van der Zanden)33. 20 advice earlier, and their family members, peers, and employers may be more supportive if they have 449 better understanding of the condition. 450 Once affected individuals sought medical advice, they experienced delays before being referred to 451 secondary care. From the patients’ perspective, this was caused by dismissal of their symptoms by 452 GPs, and the perception that doctors did not have adequate knowledge of the condition, leading to 453 misdiagnosis and the use of non-discriminatory tests. 454 GPs and non-specialist gynaecologists themselves reported that they had insufficient knowledge 455 about the symptoms, diagnosis and treatment of endometriosis, and some struggled to know when 456 to refer patients to secondary care. Additional mandatory training about menstrual conditions is 457 required during medical school, as well as comprehensive in-service education, and clear, evidence-458 based guidelines for reference. 459 As participants in the studies consistently reported feeling dismissed by healthcare professionals, 460 there may also be an opportunity to reflect on the ‘gender health gap’, a term used to describe the 461 disparities in the way that women access and experience healthcare 37. Opportunities for healthcare 462 professionals to enhance their knowledge of women’s health and challenge their unconscious biases 463 may be helpful, as may workshops to improve communication skills given that misunderstandings 464 were frequently reported. 465 466 Strengths and Limitations 467 This review was the first to specifically explore barriers to the diagnosis of endometriosis from a 468 qualitative perspective. Previous reviews by Young (2015)38 and Culley (2013)6 have explored 469 diagnostic delay as part of their wider scope. Consistent with our findings, Young (2015) 38 found that 470 the normalisation of symptoms by the patient, their peers, and their GPs was a significant problem, 471 as was the misdiagnosis of symptoms as psychological or as irritable bowel syndrome. Similarly, 472 21 Culley (2013)6 reported that diagnostic delays occurred at the patient level due to poor awareness of 473 the condition, and at the medical professional level due to the dismissal of symptoms. 474 This review was also the first to explore the barriers to diagnosis from the perspective of both 475 affected women and healthcare professionals, giving more balanced impression. The findings are 476 based on good quality data and provide valuable information, especially as the improvement of the 477 diagnosis delay is one of the top ten research priorities for endometriosis39. 478 However, all of the primary studies included were conducted in high-income countries, most of 479 which had universal healthcare, potentially limiting generalisability to other contexts (it is likely that 480 this review underestimates the effect of financial barriers to seeking a diagnosis). Additionally, whilst 481 not all of the studies reported on the demographics of the participants, of those that did, the 482 majority of participants were White women. The review may therefore be unrepresentative of the 483 experiences of people of Colour, especially as the prevalence and diagnosis of endometriosis is 484 suspected to be influenced by ethnicity40. 485 All studies focusing on the views of affected women recruited their participants from specialist 486 gynaecology clinics or from support organisations, which generally support those with severe 487 disease. Therefore, the samples may not be skewed towards those with more complex, debilitating 488 disease. Finally, the papers focusing on the views of healthcare professionals exclusively recruited 489 doctors, the majority of which were GPs, so the views of other healthcare professionals, including 490 endometriosis specialists, could not be analysed. 491 The review was also conducted by a single reviewer (the lead author) which may have increased the 492 risk bias or methodological error17. Future reviews could include additional databases and journals, 493 unpublished material and non-English papers to identify more eligible studies and reduce the risk of 494 publication and language biases. 495 Future research priorities 496 22 Identification of factors that facilitate diagnosis and demonstrations of good practice are necessary 497 to guide the production of evidence-based, decision-making resources. Additionally, the 2022 ESHRE 498 guidelines suggest a two-step approach in which empirical treatment is started based on clinical 499 suspicion and imaging (rather than laparoscopic findings)12, and so further research to analyse the 500 effect of this on the time to diagnosis and patients’ experiences is warranted. 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 23

518 1. Zondervan KT, Becker CM, Koga K, Missmer SA, Taylor RN, Viganò P. Endometriosis. Nat Rev Dis 519 Primers 2018;4(1):9. doi: 10.1038/s41572-018-0008-5 520 2. Shafir AL, Farlan LV, Shah DK, Harris HR, Kvaskoff M, Zondervan KT et al. Risk for and 521 consequences of endometriosis: A critical epidemiological review. Best Pract Res Clin Obstet 522 Gynaecol 2018;51:1-15. doi: 10.1016/j.bpobgyn.2018.06.001 523 3. Zondervan KT, Becker CM, Missmer SA. Endometriosis. N Engl J Med 2020;382(13):1244-56. doi: 524 10.1056/NEJMra1810764 525 4. Nnoaham K, Hummelshoj L, Webster P, D’Hooghe T, de Cicco Nardone F, de Cicco Nardone C et al. 526 Impact of endometriosis on quality of life and work productivity: a multicentre study across ten 527 countries. Fertil Steril 2011;96(2):366-78. doi: 10.1016.j.fertstert.2011.05.090 528 5. Agarwal SK, Chapron C, Giudice L, Laufer M, Leyland N, Missmer SA et al. Clinical diagnosis of 529 endometriosis: a call to action. Am J Obstet Gynecol 2019;220(4):354.e1-354.e12. doi: 530 10.1016/j.ajog.2018.12.039 531 6. Culley L, Law C, Hudson N, Denny E, Mitchell H, Baumgarten M et al. The social and psychological 532 impact of endometriosis on women’s lives: a critical narrative review. Hum Reprod Update 533 2013;19(6):625-39. doi: 10.1093/humupd/dmt027 534 7. Simoens S, Dunselman G, Dirksen C, Hummelshoj L, Bokor A, Brandes I et al. The burden of 535 endometriosis: costs and quality of life of women with endometriosis and treated in referral centres. 536 Hum Reprod 2012;27(5):1292-3. doi: 10.1093/humrep/des073 537 8. Cooke A, Smith D, Booth A. Beyond PICO: the SPIDER tool for qualitative evidence synthesis. Qual 538 Health Res 2012;22(10):1435-43. doi: 10.1177/1049732312452938 539 9. Wagner M, Rosumeck S, Küffmeier C, Döring K, Euler U. A validation study revealed differences in 540 design and performance of MEDLINE search filters for qualitative research. J Clin Epidemiol 541 2010;120:17-24. doi: 10.1016/j.jclinepi.2019.12.008 542 10. Ouzzani M, Hammady H, Fedorowicz Z, Elmagarmid A. Rayyan – a web and mobile app for 543 systematic reviews. Syst Rev 2016;5:210. doi: 10.1186/s13643-016-0384-4 544 11. Booth A, Harris J, Croot E, Springett J, Campbell F, Wilkins E. Towards a methodology for cluster 545 searching to provide conceptual and contextual “richness” for systematic reviews of complex 546 interventions: case study (CLUSTER). BMC Med Res Methodol 2013;13:118. doi: 10.1186/1471-2288-547 13-118 548 12. European Society of Human Reproduction and Embryology (ESHRE) Guidelines: Endometriosis. 549 Accessed 19 April 2023. https://www.eshre.eu/Guideline/Endometriosis 550 13. Hsu AL, Khachikyan I, Stratton P. Invasive and non-invasive methods for the diagnosis of 551 endometriosis. Clin Obstet Gynecol 2010;53(2):413-9. doi: 10.1097/GRF.0b013e3181db7ce8 552 14. Brown J, Crawford T, Allen C, Hopewell S, Prentice A. Nonsteroidal anti-inflammatory drugs for 553 pain in women with endometriosis. Cochrane Database Syst Rev 2017;23(1). doi: 10.1002/14651858 554 15. Critical Appraisal Skills Programme. CASP Qualitative Checklist: CASP. Accessed 31 December 555 2022. https://casp-uk.net/casp-tools-checklists/ 556 16. Long H, French D, Brooks J. Optimising the value of the critical appraisal skills programme (CASP) 557 tool for quality appraisal in qualitative evidence synthesis. BMC Med Res Methodology 2020;1:31-558 42. doi: 10.1177/2632084320947559 559 24 17. Centre for Reviews and Dissemination. Systematic Review; 3rd Edition, University of York. 560 Accessed 31 December 2022. https://www.york.ac.uk/crd/SysRev/!SSL!/WebHelp/SysRev3.htm 561 18. Thomas J, Harden A. Methods for the thematic synthesis of qualitative research in systematic 562 reviews. BMC Med Res Methodol 2008;8:45. doi: 10.1186/1471-2288-8-45 563 19. Munthe-Kaas H, Bohren M, Glenton C, Lewin S, Noyes J, Tunçalp Ö. Applying GRADE-CERQual to 564 qualitative evidence synthesis findings – paper 3: how to assess methodological limitations. 565 Implement Sci 2018;13(Suppl 1):9. doi: 10.1186/s13012-017-0690-9 566 20. Denny E. ‘You are one of the unlucky ones’: Delay in the diagnosis of endometriosis. Divers 567 Health Social Care 2004;1(1):39-44 568 21. Denny E. Women’s experiences of endometriosis. J Adv Nurs 2004;46(6):641-8. doi: 569 10.1111/j.1365-2648.2004.03055.x 570 22. Jones G, Jenkinson C, Kennedy S. The impact of endometriosis upon quality of life: a qualitative 571 analysis. J Psychosom Obstet Gynaecol 2004;25(2):123-33. doi: 10.1080/01674820400002279 572 23. Ballard K, Lowton K, Wright J. What’s the delay? A qualitative study of women’s experiences of 573 reaching a diagnosis of endometriosis. Fertil Steril 2006;86(5):1296-1301. doi: 574 10.1016/j.fertnstert.2006.04.054 575 24. Denny E, Mann C. Endometriosis and the primary care consultation. Eur J Obstet Gynecol Reprod 576 Biol 2008;139(1):111-5. doi: 10.1016/j.ejogrb.2007.10.006 577 25. Dixon S, McNiven A, Talbot A, Hinton L. Navigating possible endometriosis in primary care: a 578 qualitative study of GP perspectives. Br J Gen Pract 2021;71(710):e668-76. doi: 579 0.3399/BJGP.2021.0030 580 26. Cox H, Henderson L, Andersen N, Cagliarini G, Ski C. Focus group study of endometriosis: 581 struggle, loss and the medical merry-go-round. Int J Nurs Pract 2003;9(1):2-9. doi: 10.1046/j.1440-582 172x.2003.00396.x 583 27. Manderson L, Warren N, Markovic M. Circuit breaking: pathways of treatment seeking for 584 women with endometriosis in Australia. Qual Health Res 2008;18(4):522-34. doi: 585 10.1177/1049732308315432 586 28. Markovic M, Manderson L, Warren N. Endurance and contest: women’s narratives of 587 endometriosis. Health 2008; 12(3):349-67. doi: 10.1177/1363459308090053 588 29. Moradi M, Parker M, Sneddon A, Lopez V, Ellwood D. Impact of endometriosis on women’s lives: 589 a qualitative study. BMC Womens Health 2014;4:123. doi: 10.1186/1472-6874-14-123 590 30. Zale M, Lambert E, LaNoue MD, Leader AE. Shedding light on endometriosis: Patient and 591 provider perspectives on a challenging disease. J Endometr Pelvic Pain Disord 2020;12(2):69-76. doi: 592 10.1177/2284026520905239 593 31. Grundström H, Alehagen S, Kjølhede P, Berterö C. The double-edged experience of healthcare 594 encounters among women with endometriosis: A qualitative study. J Clin Nurs 2018;27(1):205-11. 595 doi: 10.1111/jocn.13872 596 32. Van der Zanden M, Tenuissen DA, van der Woord IW, Braat DD, Nelen WL, Nap AW. Barriers and 597 facilitators to the timely diagnosis of endometriosis in primary care in the Netherlands. Fam Pract 598 2020;37(1):131-6. doi: 10.1093/fampra/cmz041 599 25 33. Mikesell L, Bontempo AC. Healthcare provides’ impact on the care experiences of patients with 600 endometriosis: The value of trust. Health Commun 2022;14:1-13/ doi: 601 10.1080/10410236.2022.2048468 602 34. Corden A, Sainsbury R. Using Verbatim Quotations in Reporting Qualitative Social Research: 603 Researcher’s views. University of York Social Policy Research Unit 2006. Accessed 31 December 604 2022. https://www.york.ac.uk/inst/spru/pubs/pdf/verbquotresearch.pdf 605 35. Vercellini P, Viganò P, Somigliana E, Fedele L. Endometriosis: pathogenesis and treatment. Nat 606 Rev Endocrinol 2014;10(5):261-75. doi: 10.1038/nrendo.2013.255 607 36. National Institute for Health and Care Excellence. Endometriosis: diagnosis and management 608 2017. NICE guideline NG73. Accessed 31 December 2022. 609 https://www.nice.org.uk/guidance/ng73/evidence/full-guideline-pdf-4550371315 610 37. Women’s Health Strategy for England; Education and training for health and care professionals. 611 Department of Health and Social Care. Accessed April 19, 2023. 612 https://www.gov.uk/government/publications/womens-health-strategy-for-england/womens-613 health-strategy-for-england#education-and-training-for-health-and-care-professionals 614 38. Young K, Fisher J, Kirkman M. Women’s experiences of endometriosis: a systematic review and 615 synthesis of qualitative research. J Fam Plann Reprod Health Care 2015;41(3):225-34. doi: 616 10.1136/jfprhc-2013-100853 617 39. Horne A, Saunders P, Abokharis I, Hogg L. Top ten endometriosis research priorities in the UK and 618 Ireland. Lancet 2017;3(10085):2191-2. doi: 10.1016/S0140-6736(17)31344-2 619 40. Bougie O, Yap MI, Sikora L, Flaxman T, Singh S. Influence of race/ethnicity on prevalence and 620 presentation of endometriosis: a systematic review and meta-analysis. BJOG: An International 621 Journal of Obstetrics & Gynaecology 2019;126(9):1104-1115. doi: 10.1111/1471-0528.15692 622 26 27 Table 1: The Characteristics of the Included Studies. Author(s) (Year) Title Setting of Research Study Aim Sampling Method Sample size (age range of participants in years) Data Collection

Data Analysis

Denny E (2004)19 Women’s experiences of endometriosis. Birmingham, United Kingdom. To explore women’s experiences of living with endometriosis. Convenience sampling via the message board of a self-help group, and the gynaecology department of a local hospital, and snowball sampling. 15 patients (not given). Semi-structured interviews. Inductive thematic analysis. Jones G, Jenkinson C, Kennedy S (2004)20 The impact of endometriosis upon quality of life: a qualitative analysis. Oxford, United Kingdom. To identify and understand, from the patient’s perspective, the areas of HRQoL that are affected by endometriosis. Convenience sampling from the gynaecology outpatient clinic at John Radcliffe Hospital. 24 patients (21- 44). Semi-structured interviews. Grounded theory. Ballard K, Lowton K, Wright J (2006)21 What’s the delay? A qualitative study of women’s experiences of reaching a Southeast England, United Kingdom. To investigate the reasons that women experience delays in the diagnosis of endometriosis and the impact of this on women’s Purposive sampling of women attending a pelvic pain clinic due to suspected endometriosis. 28 patients (16- 47). Semi-structured interviews. Inductive thematic analysis. 28 diagnosis of endometriosis. experiences of the condition. Denny E, Mann C (2008) 22 Endometriosis and the primary care consultation. Birmingham, United Kingdom. To explore the experience of women with endometriosis in the primary care setting. Purposive sampling from a dedicated endometriosis clinic. 30 patients (19- 44). Semi-structured interviews. Inductive thematic analysis. Dixon S, McNiven A, Talbot A, Hinton L (2021)23 Navigating possible endometriosis in primary care: a qualitative study of GP perspectives. Oxford, United Kingdom. To explore what GPs identify as important considerations when caring for women with symptoms that raise the possibility of endometriosis. -Purposive sampling from Local Clinical Research Network teams, followed by snowball sampling. 42 general practitioners (not given). Semi-structured interviews. Inductive framework analysis. Cox H, Henderson L, Andersen N, Cagliarini G, Ski C (2003)24 Focus group study of endometriosis: Struggle, loss and the medical merry-go- round. Victoria, Australia To increase awareness of women’s experiences of endometriosis and the lack of support for women sufferers. Convenience sampling from a specialist endometriosis centre and a charity organisation. 61 patients (20- 64). Semi-structured focus groups. Inductive thematic analysis. Manderson L, Warren N, Circuit breaking: pathways of treatment Victoria, Australia. To we explore the circumstances under which the Purposive sampling from community newspapers, and 40 patients (20- 78). In-depth 29 Markovic M (2008)25 seeking for women with endometriosis in Australia. cycle of personal- or professional- induced delay in endometriosis was interrupted. notice boards and from snowballing. unstructured interviews. Inductive thematic analysis. Markovic M, Manderson L, Warren N (2008)26 Endurance and contest: women’s narratives of endometriosis. To enrich our understanding of the relationship between the patient’s socio- demographic

and health-related phenomena, by identifying distinctive differences among women’s narratives. 30 patients (20- 78). Moradi M, Parker M, Sneddon A, Lopez V, Ellwood D (2014)27 Impact of endometriosis on women’s lives: a qualitative study. Canberra, Australia. To explore women’s experience of endometriosis and its impact. Convenience sampling from a dedicated endometriosis centre, and from a general practice in the community. 35 patients (17- 53). Semi-structured focus groups. Inductive thematic analysis. Zale M, Lambert E, LaNoue MD, Shedding light on endometriosis: Patient and Multiple cities, United States of America. 1) To assess the perceptions and awareness of endometriosis -Convenience sampling of providers from conferences and 4 providers (not given). Semi-structured interviews. Inductive thematic analysis. 30 Leader AE (2020)28 provider perspectives on a challenging disease. amongst providers who diagnose and treat endometriosis and 2) to assess the endometriosis patient experience in the United States. seminars for physicians in gastroenterology, obstetrics and gynaecology, internal medicine and primary care. -Convenience sampling of patients from the social media page of a national endometriosis charity. 12 patients (20- 43). Grundström H, Alehagen S, Kjølhede P, Berterö C (2016)29 The double- edged experience of healthcare encounters among women with endometriosis: a qualitative study. Östergötland County, Sweden. To identify and describe the experiences of healthcare encounters among women with endometriosis. Purposive sampling from an outpatient gynaecology clinic. 9 patients (25- 55). Semi-structured interviews. Modified Stevick- Colaizii-Keen

of analysis. van der Zander M, Teunissen DA, van der Woord I, Braat D, Nelen WL, Barriers and facilitators to the timely diagnosis of endometriosis The Netherlands. To identify barriers and facilitators to the timely diagnosis of endometriosis Purposive sampling of group practices from the College of General Practitioners based 43 providers (not given). Semi-structured focus groups. Content analysis. 31 Nap AW (2020)30 in primary care in the Netherlands. from the GPs’ perspective. on geographical spread. Mikesell L, Bontempo AC (2022). 31 Healthcare Providers’ Impact on the Care Experiences of Patients with Endometriosis: The Value of Trust Multiple cities, United States of America. 1) To explore how patients with endometriosis perceive healthcare professionals to serve as barriers and facilitators to the their and 2) to assess the role that trust in healthcare professionals plays in shaping care experiences. Convenience sampling of patients from the social media pages of endometriosis organisations. 222 patients (18-64) Open-ended questionnaire. Inductive thematic analysis. 32 Table 2: The findings of the CASP Qualitative Checklist13 for each study. ✔ represents a ‘Yes’ decision and ? represents a ‘Cannot Tell’ decision. X represents a ‘No’ decision. Study Clear Aims Appropriate Methodology Appropriate design Recruitme nt Data Collectio n Researcher Relationshi p Ethica l Issues Data Analysi s Clear Finding s Valuabl e Overall Assessme nt Denny (2004)19 ✔ ✔ ✔ ✔ ✔ ? ✔ ✔ ✔ ✔ Minor concerns Jones (2004)20 ✔ ✔ ✔ ✔ ✔ ✔ ✔ ✔ ✔ ✔ No or very minor concerns Ballard (2006)21 ✔ ✔ ✔ ✔ ✔ ? ✔ ✔ ✔ ✔ Minor concerns Denny (2008)22 ✔ ✔ ✔ ✔ ✔ ? ✔ ✔ ✔ ✔ Minor concerns Dixon (2021)23 ✔ ✔ ✔ ✔ ✔ ? ✔ ✔ ✔ ✔ Minor concerns Cox (2003)24 ✔ ✔ ✔ ? ✔ ? ? ✔ ✔ ✔ Moderate Concerns Manderso n (2008)25 and ✔ ✔ ✔ ✔ ✔ ? ? ✔ ✔ ✔ Minor concerns 33 Markovic (2008)26 Moradi (2014)27 ✔ ✔ ✔ ✔ ✔ ? ✔ ✔ ✔ ✔ Minor concerns Zale (2020)28 ✔ ✔ ✔ ✔ ✔ ? ✔ ✔ ✔ ✔ Minor concerns Grundströ m (2016)29 ✔ ✔ ✔ ✔ ✔ ? ✔ ✔ ✔ ✔ Minor concerns van der Zander (2020)30 ✔ ✔ ✔ ✔ ✔ ? ? ✔ ✔ ✔ Minor concerns Mikesell (2022)31 ✔ ✔ ✔ ? ? ? ? ✔ ✔ ✔ Moderate concerns 34 Supplementary Table 1: Search Strategy for MEDLINE via Ovid Database: Ovid MEDLINE(R) and Epub Ahead of Print, In-Process, In-Data Review & Other Non- Indexed Citations and Daily Search Term 1 Exp endometriosis/ 2 Endometrio*.mp. 3 “Chronic pelvic pain”.mp. OR Pelvic Pain/ 4 Dysmenorrhea/ 5 Menstruation Disturbances/ 6 Dyspareunia/ 7 Dyschezia.mp. 8 1 or 2 or 3 or 4 or 5 or 6 or 7 9 exp Diagnosis/ 10 exp Delayed Diagnosis/ 11 exp Missed Diagnosis/ 12 “Late Diagnosis”.mp. 13 Time Factors/ and diagnos*.mp. 14 9 or 10 or 11 or 12 or 13 15 8 and 14 35 16 exp qualitative research/ 17 Interviews as topic/ or Interview/ 18 Focus groups/ 19 Narration/ 20 “Semi-structured” or semistructured or unstructured or informal or “in-depth” or indepth or “face-to-face” or structured or guide.mp. 21 Discussion or questionnaire*.ti,ab 22 Qualitative or ethnograph* or fieldwork or “field work”.mp. 23 16 or 17 or 18 or 19 or 20 or 21 or 22 24 Patient Satisfaction/ 25 Health Knowledge, Attitudes, Practice/ 26 Attitude of Health Personnel/ 27 Attitude* or experience* or view* or reflect* or perception* 28 24 or 25 or 26 or 27 29 23 and 28 30 15 and 29 36 Supplementary Table 2: Search Strategy for APA PsycINFO Database: APA PsycInfo via Ovid Search Term 1 Endometrio*.mp. 2 (“Chronic Pelvic Pain” or Pelvic Pain).mp. 3 Exp Dysmenorrhea/ 4 Exp Menstrual disorders/ 5 Exp Dyspareunia/ 6 Dyschezia.mp. 7 1 or 2 or 3 or 4 or 5 or 6 8 Exp Diagnosis/ 9 “Delayed diagnosis”.mp. 10 “Missed diagnosis”.mp. 11 “Late diagnosis”.mp. 12 Time factors and diagnos*.mp. 13 8 or 9 or 10 or 11 or 12 14 7 and 13 15 Exp Qualitative Methods/ 16 qualitative.tw. 37 17 Exp Interviews/ or interview.tw. 18 Exp Focus Group/ 19 (“Semi-structured” or semistructured or unstructured or informal or “in-depth” or indepth or “face-to-face” or structured or guide or guides).ti,ab,id. 20 Ethnograph* or fieldwork or “field work” or discussion*.ti,ab,id. 21 15 or 16 or 17 or 18 or 19 or 20 22 Exp Client satisfaction/ 23 Exp Health Attitudes/ or Exp Health Knowledge/ 24 Exp Health Personnel Attitudes/ 25 Attitude* or experience* or view* or reflect* or perception*.mp. 26 22 or 23 or 24 or 25 27 21 or 26 28 14 and 27 38 Supplementary Table 3: Search Strategy for EBSCO CINAHL Database Searched: EBSCO CINAHL Search Term 1 (MH “Endometriosis”) or “chronic pelvic pain” or (MH “Pelvic Pain”) or (MH Menstruation Disorders+) or (MH “Dysmenorrhoea”) or (MH “Dyspareunia”) or “dyschezia” 2 (MH “Diagnosis”) or (MH “Diagnosis, Delayed”) or (MH “Failure to Diagnose”) or (MH “Time Factors) 3 (MH “Qualitative Studies+”) or MH (“Interviews+”) or (MH “Semi-structured interview”) or (MH “Structured interview”) or (MH “Unstructured interview”) (MH “Focus Groups”) or (MH "Ethnographic Research") or (MH "Discussion") or (MH "Fieldwork") OR "Qualitative" 4 (MH “Patient Satisfaction+”) or (MH “Attitude to Health”) or (MH “Attitude of Health Personnel”+) or “attitude” or “experience” or “view” or “reflection” or “perception” 5 3 or 4 6 1 and 2 and 5