This was a quantitative cross-sectional study conducted across two economically diverse settings in China and Rwanda from January to November 2025. In each country, two facilities were purposively selected based on the diagnostic capacity and availability of a gynaecology outpatient clinic where participants could be sampled. In China, these facilities included The First People’s Hospital of Jingzhou (FPHJ) and Jingzhou Central Hospital (JCH), while in Rwanda, University Teaching Hospital of Kigali (CHUK) and King Faisal Hospital, Rwanda (KFHR) were selected.

China is a fast-growing upper-middle-income country located in East Asia, with a total land area of approximately 9.6 million square kilometres and a population exceeding 1.4 billion people. The country has a tiered healthcare system comprising primary, secondary, and tertiary levels, with tertiary hospitals serving as regional or national referral centres offering specialised and technologically advanced care. The present study was conducted in Jingzhou City, located in Hubei Province, Central China. Jingzhou has a population of over 5.5 million people and is a rapidly developing urban area with expanding industrial and educational infrastructure. Both FPHJ and JCH are tertiary-level referral and teaching hospitals affiliated with Yangtze University. These facilities provide a wide range of specialist services, including advanced gynaecological and reproductive health care, and are equipped with modern diagnostic technologies such as laparoscopic surgery, magnetic resonance imaging (MRI), and assisted reproductive technologies (ART). China’s health system operates through a mixed model, combining government subsidies, social health insurance, and out-of-pocket payments.

In contrast, Rwanda is a landlocked low-income country located in East Africa. It is a small country covering 10,169 square miles on the ground with an estimated population size of 14 million people. The public health sector in Rwanda is organised into three levels (peripheral, intermediate and central) with each level having defined technical and administrative platforms. The peripheral level is at the district level, consisting of the district health offices with a district hospital, and primary health care facilities. These refer their patients to the national referral hospitals, including CHUK (public), and KFHR (private not-for-profit), supervised at the central level. Rwanda follows a universal healthcare model, which provides low-cost community-based health insurance through the mutuellesdesanté [18]. This scheme allows residents to pay premiums tailored to their socio-economic class to ensure that each citizen has access to medical care. Both CHUK and KFH offer advanced diagnostic services, including MRI, but advanced ART services are still limited.

Participants were recruited from outpatient gynaecology clinics at the selected hospitals. Women were eligible for recruitment if they were of reproductive age, 18 - 45 years, with a confirmed surgical diagnosis of endometriosis, and willing to offer consent. Women with a history of other chronic gynaecological conditions that may affect fertility (uterine fibroids, polycystic ovarian syndrome, primary ovarian insufficiency and chronic pelvic inflammatory disease) or any serious comorbidities that may adversely affect the quality of life (such as multiple sclerosis, schizophrenia, severe anxiety disorders, type 1 diabetes mellitus and conditions causing physical disabilities) were not included in this study.

Based on the primary objective of comparing the impact of endometriosis between the two countries, the sample size needed to detect a mean difference in scores of at least 5 with a standard deviation of 15 was calculated, with a power of 80% and a 5% margin of error. A sample size of 142 per country was obtained.

During the study period, at least two study contact persons were identified at each study site in China and Rwanda. These contact persons were stationed at their respective gynaecological clinics on clinic days and were responsible for screening, consenting and interviewing of study participants. Participants were screened consecutively against a set of eligibility criteria and, upon fulfilment, a questionnaire was administered.

Data was collected using a validated 63-item Endometriosis Impact Questionnaire (EIQ). In Rwanda, the study contact persons were trained to use the English version during interviews without the need for translation, while in China, the tool was first translated into Chinese, and it was self-administered. This tool captured data on the burdens experienced by women over three recall periods: the last 12 months, 1 to 5 years ago, and more than 5 years ago. The EIQ uses a 5-point Likert scale scored as follows: Not at all = 0, A little = 1, Somewhat = 2, Quite a lot = 3, Very much = 4 and Not applicable = 9. The original tool defines six distinct domains of quality of life, including physical-psychosocial (33 items), fertility (3 items), sexual (7 items), employment (11 items), educational (6 items), and lifestyle (3 items). In this study, the first domain (physical-psychosocial) was further divided into physical (13 items), psychological (16 items) and social (4 items) domains, giving rise to a total of eight domains for a comprehensive assessment. Each domain contributes equally to the maximum score, and the overall score is calculated by adding all applicable scores across the three recall periods. The total score for each domain is calculated as the mean of the three recall periods. The scores are then rescaled to 0 - 100, with 0 indicating the minimum possible impact and 100 the maximum possible impact [19].

The final data collection tool was designed to capture additional information on sociodemographic characteristics and appropriate reproductive health, such as a history of self-reported infertility. In this study, infertility was defined as history of delayed or difficulty in conception as reported by the participant, who was a sexually active woman of reproductive age. The data collection tools in English and Chinese were converted into electronic forms using Kobo Toolbox (Cambridge, MA, USA).

Upon completion of data collection, datasets were extracted from Kobo Toolbox and imported into Microsoft Excel for cleaning. The Chinese dataset was first back-translated with the help of the Google translation tool. After translation, both Chinese and Rwandan datasets were merged to complete the cleaning process. Before analysis, each participant’s responses to the 63-item EIQ were reviewed, and total scores were systematically calculated as described in the previous subsection. To facilitate a balanced comparison, a participant’s periodic score was calculated only if 50% of the items in the domain were answered, in line with the original developer’s recommendation [19]. The overall, periodic and domain scores were subsequently transformed to a 0 - 100 scale using the formula:

Additionally, the monthly income reported by participants was converted into US dollars (USD) to allow for comparison across participants from both countries. Conversion was completed using the prevailing dollar rate on November 1, 2025, on the Oanda Corporate Smarter Services platform [20]. The exchange rates were 1 Rwandan Franc (RWF) = 0.00069 USD and 1 Chinese Yuan Renminbi (CNY) = 0.14059 USD.

Descriptive statistics were used to summarise sociodemographic and clinical characteristics, using medians and interquartile ranges (IQRs), while group comparison was completed by use of the Mann-Whitney U test for continuous variables, and the Chi-square test or Fisher’s exact test (for variables consisting of frequencies less than 10) for categorical variables. Continuous variables were assessed for normality using visual inspection (histogram with normal overlay and kernel density plots) and statistical tests (Shapiro-Wilk and skewness-kurtosis tests). To identify predictors of EIQ scores, the multiple linear regression model was used, while logistic regression was employed to examine factors associated with infertility. Model assumptions were verified, multicollinearity assessed (mean VIF = 2.33), and significance set at p < 0.05. Results were presented with 95% confidence intervals. Data were analysed using Stata 15.0 (StataCorp, College Station, TX, USA).

Of the 153 women who participated in the study, 102 (66.7%) were from China (Table 1). The median age was 31 years (interquartile range (IQR): 25 - 39), with Chinese women significantly younger than their Rwandan counterparts (30 vs 35 years; p = 0.009).

Marked differences were observed in marital status; about two-thirds of Rwandan women (68.6%) were married compared with only 36.3% of Chinese participants (p < 0.001). Conversely, widowed status was more frequent among Chinese women (12.7%) and absent in Rwanda. Additionally, educational attainment differed significantly, China having more women with tertiary education compared to Rwanda (51.0% vs 41.2%, p = 0.005). Almost half of all participants (47.7%) had completed tertiary education. Employment rates were higher in China (85.3%) than in Rwanda (64.7%) (p = 0.004), despite a greater proportion of Chinese participants living in rural areas (62.7%) compared to Rwandans (47.1%). Household income varied markedly between countries (median = $674.8 in China vs $103.5 in Rwanda; p < 0.001). However, health insurance coverage was more common in Rwanda (84.3%) than in China (66.7%) (p = 0.022).

Regarding reproductive history, out of the 153 women in this study, 73.9% were sexually active, with no difference between countries (p = 0.603). However, self-reported infertility was more frequent among Chinese participants than Rwandans (74.3% vs 41.0%; p < 0.001). In both countries, about 60% of women who reported infertility, also reported having accessed treatment (p > 0.999) (Table 1).

Table 1. Characteristics of study participants, N = 153.

USD = United States Dollar. ^aThe p-values were estimated with Chi-square or Exact Fisher’s test for categorical variables and Mann-Whitney U test for continuous variables.

The overall median EIQ impact score was 43.3 (IQR 14.2 - 58.5), with no significant difference between countries (p = 0.549). Likewise, there was no statistical difference across the three recall periods, that is, median = 47.3 (24.6 - 63.2), p = 0.328 for less than 12 months period, median = 42.5 (15.9 - 60.9), p = 0.411 for 1 - 5 years period and median = 0 (0.0 - 50.9), p = 0.215 for more than 5 years period (Table 2).

Table 2. Endometriosis Impact Scores for the different recall periods among study participants in China and Rwanda, N = 153.

IQR = Interquartile range.

However, as shown in Figure 2, significant differences in the impact of endometriosis on some domains were observed across countries, including sexual (median score 23.8 vs 45.3, p = 0.016), fertility (median score 19.4 vs 61.8, p = 0.005) and lifestyle (median score 0 vs 0, p = 0.001). There was no statistical difference in the impact on the physical, psychological, social, educational and employment domains.

After adjusting for demographic and reproductive factors, country of residence was not significantly associated with EIQ scores (β = 4.5, 95% CI: –8.2,17.1; p = 0.484). Age, marital status, employment, income, and insurance coverage were also non-significant (Table 3). However, education level emerged as a significant independent predictor. Compared with women with primary level education, those with secondary education (β = –13.6; 95% CI: –27.1, –0.1; p = 0.049) and tertiary education (β = –17.0; 95% CI: –31.5, –2.5; p = 0.022) reported significantly lower impact scores, indicating less impairment of quality of life.

Figure 2. Median scores by participants in the different domains assessed. p-values indicate between-country differences.

Table 3. Multiple linear regression model of the predictors of the overall endometriosis impact scores among participants.

SE = Standard error.

Out of the 113 sexually active women, 71 (62.8%) reported history of infertility. In the unadjusted analysis, country of residence and income were significantly associated with infertility. Chinese participants had over fourfold higher odds of reporting infertility compared with Rwandan women (cOR = 4.16; 95% CI 1.83 - 9.49; p = 0.001).

However, after adjusting for age, marital status, income, and insurance status, the association with country was attenuated and no longer significant (aOR = 0.83; 95% CI 0.20 - 3.51; p = 0.797).

Monthly income remained a significant independent predictor (aOR = 1.002; 95% CI: 1.000, 1.003; p = 0.034), suggesting that women with higher income levels were more likely to report infertility. Health insurance showed a borderline protective association (aOR = 0.35; 95% CI 0.12 - 1.03; p = 0.056), whereas age and marital status were not significantly related to infertility (Table 4).

Table 4. Factors associated with a history of infertility among participants.

USD = United States Dollar; cOR = Crude Odds Ratio; aOR = Adjusted Odds Ratio.

Among participants who experienced infertility (n = 71), 43 (60.5%) reported having received at least one fertility treatment modality, 33 (76.7%) of whom were in China. The most commonly reported modality across both countries was surgical interventions reported by 12/33 women in China and 8/10 women in Rwanda (Figure 3). This was followed by In vitro fertilization (IVF) which was reported exclusively by Chinese women (11/33 reported having used it) and medical therapy (e.g., clomiphene) reported by 9/33 Chinese and 2/10 Rwandan women and lastly intrauterine insemination (IUI) reported by 3/33 Chinese women.

The median satisfaction ratings across treatment modalities were 4 (IQR: 2 - 5) among Chinese women and 2 (IQR: 0 - 3) among Rwandan women.

Figure 3. Treatment modalities received by participants for fertility. Participants also rated the treatment received on a scale of 0 - 5. Provided are medians with interquartile ranges in parentheses. IVF = In Vitro Fertilization, IUI = Intrauterine Insemination.

This study has some limitations that should be considered when interpreting the findings. First, because this was a hospital-based study, participants may have been more health-conscious than the general population, which may limit representativeness. Nevertheless, the hospital setting provided the most reliable means of identifying women with a confirmed diagnosis of endometriosis, ensuring that the questionnaire was administered to an appropriate target population. Second, the EIQ captures experiences across three recall periods extending up to more than five years, which may introduce recall bias. However, because recall-period-specific analyses yielded similar results, the impact of recall bias was likely minimal. Lastly, the target sample size was not fully achieved, particularly in Rwanda, where only 51 women participated, reducing statistical power and could have impacted the statistical reliability of the non-significant findings between countries.