Two Google searches were performed following PRISMA guidelines on July 27, 2020, to assess both international (.com) and Australian (.com.au) websites (Multimedia Appendix 1). The first search was conducted on Google.com using the search terms “endometriosis” OR “adenomyosis,” OR “pelvic pain.” The second search was conducted on Google.com.au and “Australia” was added to the search string (Figure 1). Google accounts for 92.26% of the global market share compared to Bing (Microsoft Corp; 2.83%), and Yahoo Search provides results generated by Bing [20,21]. Hence, we reported results based on Google search only. To minimize the impact of any previous search history, the search was conducted in “incognito” mode. The first 30 eHealth websites listed were screened as most people do not investigate beyond this number [22]. The search was later updated on August 24, 2023, to include more recent websites. Duplicate results were removed, and the eHealth websites were screened for eligibility.

eHealth websites were included if they related to endometriosis or adenomyosis or pelvic pain, were written in English, and were free. eHealth websites that did not meet the inclusion criteria were excluded (Textbox 1).

Descriptive data were manually extracted by 1 researcher (DS) after reading the initial description and purpose of each eHealth website. These data were collated in an Excel spreadsheet (Microsoft Corp), under the following categories: (1) eHealth website name, (2) hyperlink, (3) developer, (4) funder, (5) intended purpose, (6) target audience, (7) category, (8) country of origin, and (9) last updated (Multimedia Appendix 2).

The ENLIGHT quality assessment tool [23] was used to evaluate all included eHealth websites. The ENLIGHT tool assesses seven criteria: (1) usability, (2) visual design, (3) user engagement, (4) content, (5) therapeutic persuasiveness, (6) therapeutic alliance, and (7) general subjective evaluation (Table 1). Each ENLIGHT quality assessment criterion is scored using a rating scale of 1-5 (very poor to very good and not applicable) [23].

The eHealth websites were reviewed in 2 stages. Initially, 1 researcher (DS) reviewed all included eHealth websites. Then, the eHealth websites were divided and independently reviewed by another member of the team (RO, MLH, NB, HS, MAP, and CHMN). The ENLIGHT scores of each eHealth website were collated in an Excel spreadsheet.

Discrepancies in ratings (any deviation greater than 1 rating unit) were resolved by discussion between pairs of reviewers. If evaluation differences were not resolved, a third independent assessor was consulted. After a detailed assessment, the average of the 2 reviewers’ (DS and RO or MLH or NB or HS or MAP or CHMN) ratings was used to calculate a score for each of the 7 domains (Multimedia Appendix 3).

A total score for each eHealth website was calculated according to the ENLIGHT formula [24] (Multimedia Appendix 3). eHealth websites with a total score of ≥3.5 are classified as “good” according to the ENLIGHT scoring system [24] and are recommended as high-quality eHealth websites for information on endometriosis. Interrater reliability was described using an intraclass correlation coefficient, which is estimated from a 2-way mixed effects model using an absolute definition of agreement [25].

An ethics approval was not required for this study because this was a systematic review of eHealth websites and did not involve the recruitment of participants.

A total of 117 eHealth websites were returned in the search, 58 from the first Google search (International) and 59 from the second Google search (Australian). Thirty-two eHealth websites were duplicates (duplicate websites identified in the international and Australian search), leaving 85 that were screened for this systematic review (Figure 1). Two eHealth websites were excluded, as they were not related to the topic. There was a conflict of interest in assessing 1 website (EndoZone) [26] since the authors were responsible for its development. Two eHealth websites—Royal Australian and New Zealand College of Obstetricians and Gynaecologists [27] and Cool Springs OBGYN [28]—could not be included in the final review, as the link was no longer available. A total of 80 eHealth websites were included in the final assessment and analysis (Multimedia Appendix 2).

There were discrepancies between the first and second reviewers (DS and RO or MLH or NB, HS, MAP or CHMN) in 7% of the ratings (316 variances across 4480 total ratings). All discrepancies were resolved without needing a third reviewer. The intraclass correlation coefficient was 0.61 (95% CI 0.45-0.73), indicating that interrater reliability was moderate [25].

Of the 80 eHealth websites, 44 (55%) belonged to Australian organizations, while 36 (45%) belonged to international organizations. The majority of the eHealth websites (n=49, 61%) provided education to the community on natal female pain (eg, period pain or conditions that cause pelvic pain in women), 25 (31%) were business pages of private organizations, 5 (6%) were related to endometriosis research, and 1 (1%) was a media release of a study at a university. Of the 49 eHealth websites that provided community education, 12 also provided support features to the endometriosis community. These included resources (eg, booklets, webinars), links to support group networks, and social media platforms for digital engagement with the endometriosis community (Multimedia Appendix 2). Of the 80 eHealth websites included in this study, none required payment or a subscription fee to access content.

The majority of the eHealth websites (n=70, 88%) were designed for use by the endometriosis community. Five (6%) eHealth websites provided information or education to health care providers and 5 (6%) provided information for researchers.

We conducted a comprehensive, multidimensional quality assessment of endometriosis eHealth websites using the ENLIGHT tool that captures quality constructs like persuasive design and therapeutic alliance, which are considered central to the successful uptake of eHealth websites among end users [22]. Our systematic review identified 4 high-quality endometriosis eHealth websites that can be used as educational resources for the community and health care providers. This is the first systematic review to use the ENLIGHT tool and comprehensively assess endometriosis eHealth websites.

Our systematic review has several strengths. Novel findings are presented on high-quality eHealth websites assessed using the ENLIGHT quality assessment tool and guide the community and health care providers toward quality, credible, and supportive health information. Health care providers can use these eHealth websites as educational resources and recommend high-quality eHealth websites to their patients. Finally, this review presents recommendations (ie, good design features and use of SEO tools) when designing or updating endometriosis eHealth websites as evidence suggests that good design features can help improve a website’s ranking, reach, and discoverability [52].

Our study has the following limitations. The ENLIGHT quality assessment criteria [22] were challenging to apply to websites. Most eHealth websites did not provide a health or behavior-related intervention for natal female pain. Hence, it was difficult to evaluate therapeutic persuasiveness and therapeutic alliance criteria in their entirety. We did not evaluate the entire eHealth website. We only evaluated the landing page or article the Google search engine result produced to mimic real-world circumstances. However, in some cases, this led to an uploaded document (the National Action Plan for Endometriosis on the EndoActive website) rather than the actual website, so it was not truly assessing an eHealth website. Google accounts for 92.26% of the global market share as compared to Microsoft Bing (2.83%) [20]. Additionally, Yahoo Search provides results generated by Microsoft Bing [21]. Hence, we reported results based on Google search only. Although, we searched in “incognito mode,” the Google algorithm may have automatically incorporated our location when searching, which may have influenced the results. Furthermore, the digital world is changing rapidly. The time and geographical location of the search may influence results conducted today versus the results presented above. Due to the lack of translation services, we did not include eHealth websites in languages other than English. We did not assess if the eHealth websites included in this study were developed using a genuine cocreation process hence, we do not know whether they are representative of all ethnicities or races. Finally, while we recommend the top 4 high-quality eHealth websites based on the ENLIGHT scoring system as good sources of information on endometriosis, we believe that the word “good” may well be influenced in context, that is, the generalizability of interpretation of good may vary for users in low income versus high-income countries. Nevertheless, this study has shown how eHealth websites can be assessed using the ENLIGHT checklist, which is a validated tool with its 7 quality assessment criteria.

Our systematic review presents novel findings on the quality assessment of eHealth websites using the ENLIGHT checklist to obtain endometriosis-related information. The findings of our study are (1) suggestive of high-quality eHealth websites for community use and (2) can be used by health care providers for educational purposes and recommendations. We recommend the development of (1) accurate and evidence-based endometriosis eHealth websites for the community; (2) accessible endometriosis eHealth websites for health care providers supporting daily practice and professional development; (3) interactive eHealth websites that promote user engagement, therapeutic persuasiveness, and therapeutic alliance; and (4) leveraging SEO tools to improve Google search ranking.