Health-seeking behaviour and treatment preferences among women reporting symptoms suggestive of female genital schistosomiasis in KwaZulu-Natal, South Africa: A facility-based cross-sectional study
Figures
Abstract
Female Genital Schistosomiasis (FGS), a neglected manifestation of Schistosoma haematobium infection, affects 20–56 million women in sub-Saharan Africa and is associated with elevated HIV risk. In a Schistosoma haematobium-endemic area of KwaZulu-Natal, South Africa, 32% of girls aged 10–12 years reported a history of genital symptoms, yet awareness is low, and the proportion seeking appropriate care is undocumented. This study quantified healthcare-seeking behaviour and treatment preferences for FGS-suggestive symptoms across biomedical, traditional, and combined care pathways. A facility-based cross-sectional survey was conducted at eight primary healthcare clinics in iLembe District between June and August 2025, using REDCap. Of 713 women surveyed, 694 were included, and the analyses focused on 318 who reported at least one symptom suggestive of FGS, without microbiological, serological, or gynaecological confirmation. Multivariable logistic regression identified predictors of healthcare-seeking; among all symptomatic women, treatment preference was examined across biomedical-only, traditional/CAM-only, and combination pathways, with logistic models identifying predictors of preferring any traditional/CAM care. Analyses used Stata/SE 19. Mean age was 30.6 years (SD 8.0). Of the 318 women, 78.3% preferred only biomedical care. Women in informal employment had 4.12 times the odds of preferring traditional/CAM care compared with unemployed women (aOR = 4.12; 95% CI: 2.35–7.23; p < 0.001). In the multinomial analysis, informal employment was most strongly associated with preferring traditional/CAM care alone, carrying nearly nine times the relative risk relative to biomedical care (RRR = 8.89; 95% CI: 2.74–28.90; p < 0.001), whereas water-contact activity other than bathing was associated with a preference for combination care (RRR = 3.28; 95% CI: 1.44–7.47; p = 0.005). These findings support policies that integrate FGS screening and referral into primary healthcare, sexual and reproductive health, HIV, and antenatal services, as well as health worker training, community awareness, and coordinated referral pathways.
Citation: Shongwe S, Kalinda C, Hlongwana K, Ginindza T (2026) Health-seeking behaviour and treatment preferences among women reporting symptoms suggestive of female genital schistosomiasis in KwaZulu-Natal, South Africa: A facility-based cross-sectional study. PLOS Glob Public Health 6(11): e0007218. https://doi.org/10.1371/journal.pgph.0007218
Editor: Julia Robinson, PLOS: Public Library of Science, UNITED STATES OF AMERICA
Received: May 15, 2026; Accepted: November 6, 2026; Published: November 20, 2026
Copyright: © 2026 Shongwe et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files. The anonymised dataset and Stata do-file underlying the findings of this study are provided as S1 Data (S1_Data.zip). All direct participant identifiers were replaced with unique study identification numbers before analysis, in accordance with the confidentiality conditions of the study’s ethics approvals (see Ethics statement in the Methods section).
Funding: This research was funded by the European Union’s Horizon Europe Research and Innovation Programme under grant agreement No. 101057853 (DUALSAVE-FGS; https://cordis.europa.eu/project/id/101057853), awarded to K.H., T.G., and C.K. S.S. received salary support from this grant. Additional support was provided through the University of KwaZulu-Natal College of Health Sciences (UKZN CHS) PhD Scholarship, awarded to S.S., and the Union for African Population Studies (UAPS) AFRES-Data Fellowship, also awarded to S.S. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: aOR, Adjusted Odds Ratio; CAM, Complementary and Alternative Medicine; CI, Confidence Interval; FGS, Female Genital Schistosomiasis; HIV, Human Immunodeficiency Virus; KZN, KwaZulu-Natal; LMICs, Low- and Middle-Income Countries; LR, Likelihood Ratio; NTD, Neglected Tropical Disease; OR, Odds Ratio; Ref, Reference category; RRR, Relative Risk Ratio; SD, Standard Deviation; SE, Standard Error; WHO, World Health Organisation
Introduction
Schistosomiasis is an acute and chronic disease caused by trematode worms of the genus Schistosoma, with at least 93.9% of those requiring treatment living in Africa [1]. Female genital schistosomiasis (FGS), a complication of Schistosoma haematobium infection, is estimated to afflict between 20 and 120 million women of reproductive age globally [2]. Critically, FGS is also associated with an approximately three-fold increased risk of HIV acquisition, positioning it as both a reproductive health and HIV-prevention priority [3]. Despite this burden, FGS remains excluded from many gynaecological training programmes, reproductive health guidelines, and clinical diagnostic algorithms, leaving it underreported and misdiagnosed [2,4,5].
In sub-Saharan Africa (SSA), FGS is one of the most common gynaecological conditions, affecting about 56 million women and girls [2,5]. Across SSA, women frequently misinterpret FGS symptoms as sexually transmitted infections or routine reproductive complaints, thereby delaying care [6–8]. Low community and provider awareness, stigma, and poor integration of FGS into reproductive health services further restrict timely treatment [7,9–11]. Studies from other African countries highlight similar challenges [12–14]. In Ghana, for example, community awareness was only 38.9%, and just 48.8% of symptomatic women sought care, shaped by income, education, and trust networks [12]. Frontline workers in Liberia and Nigeria similarly show limited FGS knowledge, hindering early diagnosis and treatment [13,14].
In KwaZulu-Natal (KZN), FGS prevalence is estimated at 32–75% among women and girls in rural endemic areas, with approximately 89,000 cases province-wide [15,16]. Prevalence is highest in rural districts such as iLembe and lowest in the urban eThekwini district [17]. Despite this high prevalence, the specific behavioural determinants underlying healthcare-seeking and treatment choices that contribute to this persistent burden remain unexplored. Healthcare-seeking occurs within a pluralistic health system that recognises biomedical, traditional, and complementary and alternative medicine (CAM) as valid [18]. Traditional healers, izinyanga (Traditional Healers) and izangoma (diviners), are frequently consulted for reproductive concerns, and pregnant women in rural KZN commonly use preparations such as isihlambezo (a traditional remedy used to support pregnancy and childbirth) and umhlabelo (a traditional labour-preparation medicine) alongside clinic attendance [19–21]. This pattern is consistent with foundational work on health-seeking in low- and middle-income countries (LMICs), which has long described medical pluralism and parallel consultations with biomedical and traditional providers as the norm rather than the exception [22,23].
Despite the documented burden of FGS in KwaZulu-Natal, little is known about the factors influencing healthcare-seeking among symptomatic women and their preferred healthcare pathways. Existing research has predominantly focused on parasitological prevalence [15,16,24] and clinical or histopathological characterisation [17,25] with limited attention to behavioural and health-systems determinants [26]. This evidence gap limits the capacity to translate epidemiological and clinical insights into focused, patient-centred interventions that can enhance diagnosis, treatment adherence, and health outcomes. Earlier foundational work by Danso-Appiah et al. in 2009 examined determinants of health-seeking behaviour for schistosomiasis-related symptoms in Ghana [27], while more recent qualitative and mixed-methods studies from Madagascar [28], Tanzania [29], and the Democratic Republic of Congo [30] have documented stigma, concealment, and delayed healthcare-seeking among women with FGS symptoms. South Africa’s formal recognition of traditional medicine requires a specific understanding of how it interacts with biomedical pathways in FGS care. This study, therefore, investigated the sociodemographic and behavioural determinants of health-seeking behaviour and treatment pathway preferences among women with FGS-suggestive symptoms in the KwaZulu-Natal province, with a particular focus on their navigation of pluralistic care options.
Methods and materials
Ethics statement
Study area and setting
The study was conducted in the iLembe district, which is situated on the east coast of KwaZulu-Natal, bordering the Indian Ocean [31]. Mandeni, KwaDukuza, Maphumulo, and Ndwedwe are the four local municipalities that make up the iLembe district, the smallest of the province's district municipalities [31]. The majority of the rural inland areas are tribal authority areas, characterised by subsistence farming [31]. Two of the four local municipalities in the iLembe District, Ndwedwe and Mandeni, both categorised as rural, were the study sites. This study is under a bigger umbrella project called the Dual Screening by Spectral Artificial Visual Examination (SAVE) for Female Genital Schistosomiasis (FGS) and Cervical Cancer (abbreviated as DUAL-SAVE-FGS), which is a clinical research project that aims to improve the detection and management of FGS in participating countries (South Africa, Eswatini and Mozambique). Evidence of Schistosoma haematobium intermediate host snails (Bulinus spp.) and documented water-contact activities within communities, which indicate past exposure risk in these areas, were considered when selecting the sites [15]. Based on their closeness to previously identified high-prevalence schools and communities, eight primary healthcare clinics, four in each municipality, were chosen as study sites.
Study design and population
Sample size
The sample size was determined based on the previously reported prevalence of schistosomiasis in the study area [15]. Calculations were performed at the 95% confidence level using the formula proposed by Lwanga et al. [32]. for estimating a population proportion with specified relative precision: n = Z2 (1 − p) / (ε2p). Where n is the required sample size, p is the estimated prevalence, Z is the standard normal deviate corresponding to a 95% confidence level, and ε represents relative precision.
A reported prevalence of 32% was used for the sample size calculation [33]. This estimate is derived from previous analyses of baseline genital symptom data among schoolgirls and young women in the iLembe district, KZN, which suggested that approximately 32% of female learners may present with genital symptoms not attributable to sexually transmitted infections and potentially associated with schistosomiasis 33. Although this estimate was derived from a schoolgirl population rather than adult women aged 18–49, it was considered the most contextually appropriate available proxy for three reasons: (1) The source study was conducted within the same district and in similar schistosomiasis-endemic communities (2) water-contact exposure patterns that drive FGS transmission are established during childhood and persist into adulthood, such that adult prevalence in endemic settings is expected to be comparable to or higher than adolescent prevalence [2,5] and (3) the absence of published adult women prevalence estimates for symptom-based FGS in this geographic context necessitated the use of a schoolgirl proxy, which, while contextually appropriate, underscores a critical data gap. With a relative precision of 5%-10% of the true prevalence (0.05 < ε < 0.10), a minimum sample size of 713 participants was deemed adequate for the study. The sample size was calculated for the parent prevalence study, not for the regression analyses presented here. We therefore describe the subgroup analyses as exploratory and report confidence intervals to indicate the precision of the estimates, rather than conducting a post hoc power calculation.
Sampling strategy
A two-stage probability sampling approach was used, with municipalities (Ndwedwe and Mandeni) as primary strata to ensure representation from both parent project implementation sites. In the first stage, probability-proportional-to-size (PPS) sampling was used to allocate the target sample size (n = 713) across eight selected primary healthcare clinics, based on outpatient headcount data obtained from the KwaZulu-Natal Provincial Department of Health District Health Information System (DHIS) (S1 Table). The clinic-specific PPS allocations are presented in S1 Table. In the second stage, consecutive enrolment was conducted within each clinic. All women attending routine clinic days during the study period were screened for eligibility; those meeting the inclusion criteria were invited consecutively until the target was reached. Because recruitment occurred on routine service days, women who attend clinics less frequently, including those in formal employment, may be under-represented. After excluding participants with incomplete data, 694 women were included in the overall analysis. Of the 318 women reporting FGS-suggestive symptoms, 202 (63.5%) had sought care for those symptoms and 116 (36.5%) had not; this distinction defined the healthcare-seeking outcome. Treatment preference, the type of care a woman would consider using, was recorded for all 318 women, irrespective of whether they had sought care. Both regression models were based on 317 women, as one participant had a missing age value. The full participant flow is shown in Fig 1. Both regression models were based on 317 women, as age was missing for one participant.
Data collection
Data were collected between June and August 2025 by three trained enumerators using a structured, interviewer-administered questionnaire implemented in Research Electronic Data Capture (REDCap) on Android tablets [34]. The questionnaire was developed for this study, drawing on constructs from the Health Belief Model [35] and Kleinman's Explanatory Model [36], as well as on items used in prior FGS and reproductive health-seeking studies [12,27,37]. The instrument was administered in English by enumerators fluent in isiZulu, who translated items into isiZulu in real time during interviews to accommodate participants’ language preferences. Before data collection, the instrument was reviewed by the research team and supervisors for clarity and cultural appropriateness. A formal pilot study was not conducted; however, the enumerators were experienced isiZulu-speaking fieldworkers familiar with the study communities, and any items requiring clarification were resolved during enumerator training. The absence of formal pilot testing is acknowledged as a limitation. Data quality was supported by REDCap's built-in validation rules and required-field settings, which minimised missing and out-of-range entries. The questionnaire comprised closed-ended items and took approximately 20–30 minutes to complete. Given the sensitive nature of the questions, interviews were conducted privately in a separate consultation room within each clinic, out of the sight and hearing of other patients and staff. Because participants were already attending the clinic for routine services, travel reimbursements were not provided; however, airtime vouchers were offered at the end of each interview as a token of appreciation. Throughout this study, “FGS” refers to symptoms suggestive of FGS rather than microbiologically confirmed infection; no molecular, serological, or gynaecological examination-based diagnostic testing was performed.
Operationalisation of variables
The selection of independent variables was based on theoretical relevance derived from the Health Belief Model (HBM) [35] and Kleinman’s Explanatory Model of illness [36]. HBM constructs of perceived severity and perceived barriers guided the inclusion of symptom burden and symptom duration [35]. At the same time, Kleinman's framework, which emphasises how individuals interpret illness and navigate pluralistic care pathways, informed the inclusion of source of health information, prior awareness of schistosomiasis, and treatment preference [36]. Variable selection was further guided by empirical evidence from prior FGS and reproductive health-seeking studies in sub-Saharan Africa [12,27,37]. Sociodemographic variables were included as potential confounders, informed by the social determinants of health framework. Monthly income was recorded in South African Rand (ZAR) during interviews and converted to USD equivalents using the South African Reserve Bank exchange rate as of 1 June 2025 for international comparability. A full list of variables, their categorisation, and theoretical underpinnings is provided in S2 Table.
Two outcome variables were defined. First, healthcare-seeking behaviour was operationalised as a binary variable indicating whether a woman who reported at least one FGS-suggestive symptom had sought any form of care in the month preceding the interview (1 = Yes; 0 = No). Second, treatment preference, the type of care a woman would consider using for FGS-suggestive symptoms, was recorded for all symptomatic women as a categorical variable with three categories: biomedical care only, traditional/CAM care only, or a combination of biomedical and traditional/CAM care. Traditional and complementary/alternative medicine were grouped together because both represent care sought outside the formal biomedical system, which was the distinction of analytic interest; we acknowledge that this category is heterogeneous, encompassing consultation with traditional healers, herbal remedies, and self-care. This measure was analysed for all 318 symptomatic women, rather than only those who had sought care, consistent with the study objective of estimating the proportion who would consider each care pathway, either individually or in combination.
Data analysis
All analyses were conducted in Stata/SE 19.5 (StataCorp, College Station, Texas, USA). Descriptive statistics summarised participant characteristics. Categorical variables were presented as frequencies and proportions, and continuous variables were presented using appropriate measures of central tendency and dispersion. Bivariate analyses used Pearson's chi-square tests and univariable logistic regression to obtain crude odds ratios. Variables with p < 0.20 or of a priori theoretical significance were retained for multivariable modelling [38]. Employment status and income source were considered conceptually distinct predictors, capturing labour-market participation and primary income type. In South Africa, social grants and remittances mean that employment status does not always reflect the financial resources available for healthcare [39]. Where these measures overlapped, employment status was retained.
Two regression models were fitted for the two outcomes, each on all 318 symptomatic women. Binary logistic regression identified factors associated with healthcare-seeking behaviour (Yes vs No), with results reported as adjusted odds ratios (aORs) and 95% confidence intervals (CIs). Multinomial logistic regression identified factors associated with treatment preference, with biomedical-only preference as the reference category and traditional/CAM-only and combination preferences as comparison categories. Results are reported as relative risk ratios (RRRs) with 95% CIs to support interpretation of the practical significance of associations. To provide a pooled summary, a binary logistic model additionally contrasted any traditional/CAM preference with biomedical-only preference. Model fit was assessed using the likelihood-ratio chi-square test, pseudo-R2, and, for the healthcare-seeking model, the specification link test. For categorical predictors with more than two levels, the overall (joint) significance of the variable was assessed using adjusted Wald tests, rather than relying on individual category contrasts, with exact p-values reported
Clustering across the eight clinics was assessed using the intraclass correlation from null random-intercept models, and the robustness of associations was confirmed across cluster-robust standard errors and random-intercept specifications; estimates for the treatment-preference model were additionally checked against Firth penalised-likelihood estimation. As a further sensitivity analysis, self-reported perceived barriers of distance and financial constraint were added to the healthcare-seeking model to assess residual confounding by structural access factors. Robust (Huber–White) standard errors were used to account for potential misspecification. Observations with missing data for variables in a given model were excluded in a complete-case analysis; missingness ranged from 0% to 0.94% (S3 Table). Statistical significance was evaluated at a two-sided threshold of p < 0.05.RESULTS
Characteristics of the study population
The mean age of participants was 30.64 SD ± 8.00 years (range: 18–49 years), with the majority in the 18–29-year age group (47.6%). Most (90.9%) participants resided in rural areas, were single or never married (96.8%), and identified as Christian (86.2%). Regarding income, 31.8% relied on remittances, and 3.5% earned income through business or farming. In terms of educational attainment, the majority (55.7%) had completed secondary or high school, and 11.9% had tertiary education. Monthly wages were distributed across the lower-income brackets; 43.1% earned ≤R1000 (≤$55) per month. Regarding access to clean water, 80.8% of participants reported having access to it, and the most common water-contact activity was bathing (89.6%). Awareness of FGS was notably low in the study population, with only 21.7% of women reporting prior awareness of the condition. The most common source of health information was media (65.1%), followed by healthcare workers (31.1%), with family/friends and traditional healers each accounting for 1.9% (Table 1).
Missing data were minimal across all study variables (0%–0.94%). Age (0.31%) and marital status (0.94%) had missing values, whereas all other variables were complete. A detailed summary of missing data for all variables included in the descriptive and regression analyses is provided in S3 Table.
Bivariate factors associated with health-seeking behaviour
In the bivariate analysis, 81.8% (n = 108) of women with symptoms lasting 1 week or longer sought care, compared with 50.5% (n = 94) of those with symptoms lasting less than 1 week (p < 0.001). A higher number of reported symptoms was also associated with seeking care (p < 0.001). A previous schistosomiasis diagnosis was associated with healthcare-seeking (p = 0.003). 78.6% of women with a previous diagnosis sought care, compared with 59.3% among those without a previous diagnosis. Water contact activity (p = 0.058), source of health information (p = 0.465) and age group (p = 0.562) were not significantly associated with whether women sought care (p = 0.562) (Table 2).
Factors associated with seeking care among women with FGS-suggestive symptoms
Among the sociodemographic and exposure characteristics included in the adjusted binary logistic regression model, symptom duration (p < 0.001) and number of symptoms (p = 0.002) were associated with seeking care. Women who experienced symptoms for one week or longer were 3.19-fold more likely to seek care than those with symptoms of less than one week (aOR = 3.19; 95%CI: 1.82–5.61), and each additional symptom reported was associated with a 2.52-fold higher odds of seeking care (aOR = 2.52; 95% CI: 1.39–4.57).
Water-contact activity (aOR = 0.49; 95% CI: 0.21–1.12), previous schistosomiasis diagnosis (aOR = 1.26; 95% CI: 0.61–2.62), source of health information (adjusted Wald test p = 0.249), and age group (30–39 years p = 0.530; 40–49 years p = 0.281) were not significantly associated with healthcare-seeking in the adjusted model (Table 3).
Treatment preference among women with FGS-suggestive symptoms
Among all 318 symptomatic women, most preferred biomedical care only (78.3%; 95% CI: 73.4–82.5), followed by a combination of biomedical and traditional/CAM care (14.8%; 95% CI: 11.3–19.1), while a smaller proportion preferred traditional or complementary and alternative medicine alone (6.9%; 95% CI: 4.6–10.3) (Table 4). Preference varied by employment status (p < 0.001); women in informal employment were far more likely to prefer traditional/CAM care alone (29.2%) than unemployed (4.4%) or formally employed women (6.6%). Schistosomiasis awareness was also associated with preference (p = 0.007), with women unaware of schistosomiasis more likely to prefer traditional/CAM care alone (15.5%) than those who were aware (5.0%). Water-contact activity was associated with preference for combination care (p = 0.006); women reporting water-contact activities other than bathing more often preferred combination care (33.3%) than those whose primary activity was bathing (12.6%). Preference did not differ significantly by age group (p = 0.746), marital status (p = 0.220), or symptom duration (p = 0.094) (Table 4).
Factors associated with treatment preference: Multinomial logistic regression
Multinomial logistic regression was used to examine factors associated with treatment preference, with biomedical-only preference as the reference category (Table 5). The overall model was statistically significant (LR χ2(8) = 29.61; p = 0.0002).
In the traditional/CAM-only versus biomedical-only comparison, women in informal employment had nearly nine times the relative risk of preferring traditional/CAM care alone rather than biomedical care, compared with unemployed women (RRR = 8.89; 95% CI: 2.74–28.90; p < 0.001), while women aware of schistosomiasis had a 70% lower relative risk of this preference than those unaware (RRR = 0.30; 95% CI: 0.12–0.80; p = 0.016). In the combination-versus-biomedical-only comparison, women who engaged in water-contact activities other than bathing had a 3.28-fold higher relative risk of preferring combination care over biomedical care (RRR = 3.28; 95% CI: 1.44–7.47; p = 0.005). Employment status and schistosomiasis awareness were not significant in this comparison.
The robustness of these associations was confirmed in sensitivity analyses that accounted for clustering across the eight clinics (intraclass correlation = 0.18) and used Firth penalised-likelihood estimation (Table 5).
Discussion
This study investigated the determinants of health-seeking behaviour and preferred treatment pathways among women with FGS-suggestive symptoms in KwaZulu-Natal, contributing to the emerging evidence on the health-systems dimensions of FGS, a neglected tropical disease affecting millions of women in sub-Saharan Africa [40]. In this setting, longer symptom duration and greater symptom burden were associated with seeking care, and women's treatment preferences spanned biomedical, traditional, and combined pathways, shaped by socioeconomic and informational factors rather than the availability of clinical services alone. Although these patterns were observed in South Africa, they are relevant across sub-Saharan Africa, where FGS is endemic in many countries and pluralistic health-seeking, low disease awareness, and socioeconomic barriers to biomedical care are widely documented.
Women with longer symptom duration were more likely to seek care. Likewise, women with multiple symptoms may have interpreted their illness as more serious and so sought treatment. These data highlight the importance of symptom intensity and persistence as drivers of health-seeking behaviour among women reporting symptoms suggestive of FGS. Similar patterns have been reported in studies of urogenital schistosomiasis from sub-Saharan Africa, where women commonly normalise genital symptoms such as genital itching, contact bleeding, and abnormal discharge, attributing them to menstruation, sexually transmitted infections (STIs), or other reproductive conditions, often leading to delayed healthcare-seeking until symptoms become persistent or functionally disruptive [6,10,41]. This normalisation of FGS-related gynaecological symptoms may act as a barrier to early case identification and may accelerate the progression of FGS sequelae, including chronic pelvic pain, infertility, and increased vulnerability to HIV, a link documented in several southern African cohorts [8,40,42]. The finding that women tolerate single symptoms but respond to the accumulation of symptoms mirrors observations in other gynaecological conditions, such as pelvic inflammatory disease and cervical pathology in LMICs [43,44]. Interventions that enhance early recognition of FGS symptoms and challenge the assumption that gynaecological discomfort is normal are therefore essential and should be embedded within community-based health education in endemic areas.
Most women in the current study preferred biomedical treatment, consistent with studies of gynaecological health-seeking in peri-urban and rural South Africa [45,46]. This preference may partly reflect the comparatively extensive public-sector healthcare infrastructure in KwaZulu-Natal and the geographical accessibility of clinics to the study population [46]. Nevertheless, the preference for traditional/CAM care alone (6.9%) or in combination with biomedical care (14.8%) indicates that pluralistic healthcare remains relevant. This finding reflects the wider sub-Saharan African context, where biomedical, traditional and herbal therapies often coexist, particularly in rural and resource-constrained communities [18,22]. Across the region, limited access to diagnostic services and trained healthcare providers contributes to the underdiagnosis or misdiagnosis of FGS as an STI, potentially prompting women with persistent symptoms to move between different care systems [6,47]. Traditional and herbal treatments may also be used because they are culturally familiar, locally accessible or considered responsive to social and spiritual understandings of illness, although evidence regarding their effectiveness and safety for schistosomiasis remains limited [48]. In South Africa, traditional medicine is legally protected and culturally significant, formalised by the Traditional Health Practitioners Act of 2007 [49]. Data regarding traditional healer usage varies, with studies indicating that approximately 27% (or lower, with some studies reporting 2%–10% in certain populations) of South Africans consult traditional healers, with higher rates among rural residents and those groups with less formal education, closely aligned with our study population [50–52]. These findings therefore suggest that strengthening biomedical diagnosis and treatment for FGS should be accompanied by culturally responsive health education, community engagement and appropriate collaboration with traditional health practitioners, as part of the integrated and participatory approaches recommended for schistosomiasis control in sub-Saharan Africa [48, 53].
Informal employment was associated with preferring traditional/CAM care. In Lesotho, a study on the use of traditional, complementary, and alternative medicine among patients receiving healthcare found that self-employed participants and students were more likely to use TCAM than unemployed participants [54]. Direct costs (transport, user fees) and indirect costs (lost income from time away from informal or agricultural work) often make repeated clinic attendance unaffordable for women in informal economic roles [55]. In contrast, traditional healers typically operate closer to home and may accept flexible or in-kind payment arrangements [21]. Furthermore, in agrarian communities of KZN, illness beliefs often link gynaecological symptoms to social or spiritual causes rather than to biological infection, sustaining a preference for traditional remedies even when biomedical facilities are nearby [21]. Qualitative studies from rural KZN show that women frequently turn to traditional healers first for gynaecological complaints, resorting to biomedical care only after traditional treatment is perceived as ineffective, a sequential pattern that delays diagnosis and may exacerbate FGS pathology [9, 21, 41]. However, factors such as socioeconomic support may shape these preferences; for example, the relatively high preference for biomedical care among unemployed women (81.0%) may reflect the protective role of South Africa's social-grant system, which can afford greater time and financial flexibility for clinic visits than is available to women whose livelihoods depend on daily informal or agricultural work [56].
Women engaged in water-contact activities beyond bathing, such as laundry, agriculture, swimming, and cooking, were more likely to use a combination of biomedical treatment and traditional/CAM care pathways, suggesting that recurrent or continuing exposure to contaminated water sources may lead women to pursue different forms of treatment in an effort to manage ongoing symptoms. While evidence specific to FGS healthcare-seeking and water exposure remains limited, this finding aligns with studies indicating that regular exposure to freshwater in endemic environments increases the risk of infection and symptom recurrence, potentially shaping subsequent healthcare-seeking behaviour [2,57]. Evidence from Tanzania illustrates that gendered water-contact roles affect treatment pathways, with women using pluralistic healthcare systems, especially for persistent or poorly understood reproductive symptoms [29]. Given the well-established link between prolonged or repeated freshwater contact and Schistosoma haematobium transmission, this association merits further investigation in adequately powered studies.
In the multinomial model, awareness of schistosomiasis was associated with a lower relative risk of preferring traditional/CAM care alone over biomedical care, but not with a preference for combination care, suggesting that women who understood the disease's biomedical aetiology were more likely to seek biomedical care. Similar associations have been observed for other NTDs, such as lymphatic filariasis and soil-transmitted helminthiases, where disease-specific knowledge consistently predicts higher uptake of preventive chemotherapy and biomedical treatment following targeted awareness interventions [58]. As this was a cross-sectional analysis, the direction of this association cannot be established; women aware of schistosomiasis may prefer biomedical care, or those already oriented toward biomedical services may be more likely to have encountered such information. Nonetheless, the finding implies that well-designed, culturally grounded awareness campaigns explaining transmission via cercarial exposure in freshwater, the role of praziquantel in treatment, and the benefits of early care could increase both healthcare-seeking and uptake of appropriate treatment. Such campaigns should be tailored to low-literacy contexts and explicitly address local beliefs about gynaecological symptoms that sustain traditional medicine as a first-line option.
Collectively, these findings highlight the importance of integrating FGS into routine sexual and reproductive health services, strengthening community health workers’ roles in early symptom recognition and referral, and reorienting health communication strategies away from unidirectional mass media campaigns toward participatory, interpersonal models that respect pluralistic care practices while guiding women toward timely biomedical care.
Strengths and limitations
This study has several limitations. Foremost, FGS was not confirmed by microbiological, molecular, serological, or gynaecological examination. FGS-suggestive symptoms overlap considerably with sexually transmitted infections, pelvic inflammatory disease, cervical pathology, candidiasis, vaginitis, and urinary tract infections, and the symptom-defined population, therefore, almost certainly includes women without FGS. Our findings consequently describe health-seeking for FGS-suggestive gynaecological symptoms rather than FGS-specific behaviour, and should not be read as characterising healthcare-seeking among women with confirmed disease. This reflects the diagnostic reality of primary care in endemic settings, where no point-of-care FGS test is available; it nonetheless limits direct inference for FGS-specific interventions, and future work linking symptom-based screening to confirmatory diagnosis is needed to establish which of these healthcare-seeking patterns are specific to FGS. As this was a facility-based study, our sample comprised women who had reached a clinic, and the findings may not represent women who never access care. The cross-sectional design does not establish cause and effect; therefore, the association between schistosomiasis awareness and treatment preference should not be interpreted as causal. Because participants were recruited during routine clinic visits, women who attend clinics less often, including those in formal employment, may be under-represented. We were able to examine this; partially self-reported perceived barriers of distance and financial constraint were not associated with healthcare-seeking (distance p = 0.94; financial constraint p = 0.93), and adjusting for them changed the estimates for symptom duration and symptom count by less than 0.5%. This absence of an access-barrier gradient among the women in our sample is itself consistent with under-representation of the most severely obstructed women, for whom distance or cost may be a decisive barrier to reaching a clinic at all. We did not capture objective measures of geographic access, such as travel distance or time, and residual confounding by structural access factors cannot be excluded. Future research should aim to differentiate health-seeking behaviours and treatment preferences specifically among women with microbiologically confirmed FGS to understand nuances not captured by symptom-based studies. The single-province scope further constrains applicability to other sub-Saharan African settings with different health systems and disease burdens, and small subgroup sizes (e.g., married women, informally employed women) produced wide confidence intervals and reduced power within strata. Future studies should consider targeted sampling strategies for specific, hard-to-reach populations, such as exclusive traditional medicine users, or utilise multi-site studies to increase overall sample size and substantially improve statistical power. The instrument was not formally piloted and was administered in English with real-time isiZulu translation rather than from a validated translated version; both may have introduced some inconsistency in item delivery, which we acknowledge as a limitation.
Despite these limitations, the study has notable strengths. It is among the first to empirically examine FGS-related health-seeking behaviour and treatment preferences in KwaZulu-Natal, addressing a clear gap in the NTD literature. The use of multivariable and multinomial regression with robust standard errors enabled adjustment for multiple confounders, and the inclusion of biomedical, traditional/CAM, and combination care as distinct outcomes reflects the pluralistic realities of health-seeking in this setting. The sample was sufficiently large and geographically representative of a high-transmission endemic region, strengthening internal validity.
Conclusion
Health-seeking behaviour among women with FGS-suggestive symptoms in KwaZulu-Natal is predominantly driven by symptom duration and burden, with prolonged or multiple symptoms markedly increasing care utilisation. Although biomedical care is the dominant approach, socioeconomically disadvantaged women frequently use traditional and complementary medicine, alone or alongside biomedical services. Awareness of FGS remains strikingly low, even among more educated women, pointing to gaps in both community health literacy and formal clinical training. These findings support the World Health Organisation's call to prioritise FGS in national NTD strategies and highlight the need to integrate FGS into routine sexual and reproductive health services. Specifically, standardised FGS screening protocols and structured training modules for frontline healthcare workers should be developed to enable consistent adoption, alongside community-pre-tested awareness campaigns delivered through interpersonal channels. Policy responses should recognise the pluralistic use of biomedical and traditional health systems, foster collaboration between them, and address structural determinants, such as unsafe water access, that sustain transmission and shape treatment pathways.
Supporting information
S1 Table. Probability-proportional-to-size (PPS) sample allocation per clinic.
https://doi.org/10.1371/journal.pgph.0007218.s001
(DOCX)
S2 Table. Study variables, categorisation, and theoretical framework.
https://doi.org/10.1371/journal.pgph.0007218.s002
(DOCX)
S1 Data. Anonymised dataset and Stata analysis code.
https://doi.org/10.1371/journal.pgph.0007218.s004
(ZIP)
Acknowledgments
The authors would like to sincerely thank all the women who participated in this study and generously shared their time and experiences. We are also grateful to the clinic staff who facilitated data collection and supported research activities at the participating facilities. We extend our appreciation to the enumerators for their dedication and commitment during fieldwork. We also acknowledge the support provided by the iLembe District Health Office and the KwaZulu-Natal Department of Health for granting permission to conduct the study and facilitating access to the study sites. The authors further acknowledge financial support from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 101000365, as well as from the University of KwaZulu-Natal College of Health Sciences PhD Scholarship.
References
- 1.
World Health Organization. Schistosomiasis. 2023. https://www.who.int/news-room/fact-sheets/detail/schistosomiasis Accessed 2026 April 8.
- 2.
World Health Organization Regional Office for Africa E. Female genital schistosomiasis: A silent threat to women’s health and gender equality. 2024. https://espen.afro.who.int/female-genital-schistosomiasis-silent-threat-womens-health-and-gender-equality Accessed 2026 April 8.
- 3. Kjetland EF, Ndhlovu PD, Gomo E, Mduluza T, Midzi N, Gwanzura L, et al. Association between genital schistosomiasis and HIV in rural Zimbabwean women. AIDS. 2006;20(4):593–600. pmid:16470124
- 4. Tetteh CD, Ncayiyana JR, Makhunga SE, Manyeh AK, Asiamah EA, Ginindza TG. Knowledge and management of female genital schistosomiasis in sub-Saharan Africa: A scoping review protocol. S Afr J Infect Dis. 2024;39(1):553. pmid:38962372
- 5. Umbelino-Walker I, Wong F, Cassolato M, Pantelias A, Jacobson J, Kalume C. Integration of female genital schistosomiasis into HIV/sexual and reproductive health and rights and neglected tropical diseases programmes and services: a scoping review. Sex Reprod Health Matters. 2023;31(1):2262882. pmid:37850814
- 6. Kayuni SA, Cunningham LJ, Kumwenda D, Mainga B, Lally D, Chammudzi P, et al. Challenges in the diagnosis and control of female genital schistosomiasis in sub-Saharan Africa: an exemplar case report associated with mixed and putative hybrid schistosome infection in Nsanje District, Southern Malawi. Front Trop Dis. 2024;5.
- 7. Mazigo HD, Samson A, Lambert VJ, Kosia AL, Ngoma DD, Murphy R, et al. “Female genital schistosomiasis is a sexually transmitted disease”: Gaps in healthcare workers’ knowledge about female genital schistosomiasis in Tanzania. PLOS Glob Public Health. 2022;2(3):e0000059. pmid:36962298
- 8. Shukla JD, Kleppa E, Holmen S, Ndhlovu PD, Mtshali A, Sebitloane M, et al. The association between female genital schistosomiasis and other infections of the lower genital tract in adolescent girls and young women: a cross-sectional study in South Africa. J Low Genit Tract Dis. 2023;27(3):291–6. pmid:37379442
- 9. Nemungadi TG, Furumele TE, Gugerty MK, Djirmay AG, Naidoo S, Kjetland EF. Establishing and integrating a female genital Schistosomiasis Control Programme into the existing health care system. Trop Med Infect Dis. 2022;7(11):382. pmid:36422933
- 10. Mazigo HD, Samson A, Lambert VJ, Kosia AL, Ngoma DD, Murphy R, et al. Healthcare workers’ low knowledge of female genital schistosomiasis and proposed interventions to prevent, control, and manage the disease in zanzibar. Int J Public Health. 2022;67:1604767. pmid:36188750
- 11. Pillay LN, Umbelino-Walker I, Schlosser D, Kalume C, Karuga R. Minimum Service Package for the integration of female genital schistosomiasis into sexual and reproductive health and rights interventions. Front Trop Dis. 2024;5.
- 12. Patel K. Understanding the socio-cultural determinants of health-seeking behaviour and health information trust among women at-risk for Female Genital Schistosomiasis in Ghana. Université d’Ottawa/University of Ottawa. 2022.
- 13. Oluwole AS, Bettee AK, Nganda MM, Piotrowski HL, Fapohunda VO, Adejobi JB, et al. A quality improvement approach in co-developing a primary healthcare package for raising awareness and managing female genital schistosomiasis in Nigeria and Liberia. Int Health. 2023;15(Suppl 1):i30–42. pmid:36960805
- 14. Piotrowski H, Oluwole A, Fapohunda VO, Adejobi JB, Nebe OJ, Soneye I, et al. Mixed-methods evaluation of integrating female genital schistosomiasis management within primary healthcare: a pilot intervention in Ogun State, Nigeria. Int Health. 2023;15(Suppl 1):i18–29. pmid:36960808
- 15. Livingston M, Pillay P, Zulu SG, Sandvik L, Kvalsvig JD, Gagai S, et al. Mapping schistosoma haematobium for novel interventions against female genital schistosomiasis and associated HIV Risk in KwaZulu-Natal, South Africa. Am J Trop Med Hyg. 2021;104(6):2055–64. pmid:33939629
- 16. Mwazha A, Nhlonzi GB, Kjetland EF. Presence of tissue schistosomiasis in KwaZulu-Natal, South Africa: a retrospective histopathologic review. Front Trop Dis. 2023;4.
- 17. Banhela N, Taylor M, Zulu SG, Strabo LS, Kjetland EF, Gunnar S. Environmental factors influencing the distribution and prevalence of Schistosoma haematobium in school attenders of ILembe and uThungulu Health Districts, KwaZulu-Natal Province, South Africa. Southern African Journal of Infectious Diseases. 2017;32(4):132–7.
- 18. Jama NA, Nyembezi A, Ngcobo S, Lehmann U. Collaboration between traditional health practitioners and biomedical health practitioners: Scoping review. Afr J Prim Health Care Fam Med. 2024;16(1):e1–11. pmid:39099280
- 19. Ramulondi M, de Wet H, Ntuli NR. The use of African traditional medicines amongst Zulu women during childbearing in northern KwaZulu-Natal. Afr J Reprod Health. 2022;26(1):66–75. pmid:37585018
- 20. Semenya S, Maroyi A, Potgieter M, Erasmus L. Herbal medicines used by Bapedi traditional healers to treat reproductive ailments in the Limpopo Province, South Africa. Afr J Tradit Complement Altern Med. 2012;10(2):331–9. pmid:24146458
- 21. Zuma T, Wight D, Rochat T, Moshabela M. The role of traditional health practitioners in Rural KwaZulu-Natal, South Africa: generic or mode specific?. BMC Complement Altern Med. 2016;16(1):304. pmid:27549895
- 22. Moshabela M, Bukenya D, Darong G, Wamoyi J, McLean E, Skovdal M, et al. Traditional healers, faith healers and medical practitioners: the contribution of medical pluralism to bottlenecks along the cascade of care for HIV/AIDS in Eastern and Southern Africa. Sex Transm Infect. 2017;93(Suppl 3):e052974. pmid:28736393
- 23. Sundararajan R, Mwanga-Amumpaire J, King R, Ware NC. Conceptual model for pluralistic healthcare behaviour: results from a qualitative study in southwestern Uganda. BMJ Open. 2020;10(4):e033410. pmid:32317259
- 24. Pillay P, Taylor M, Zulu SG, Gundersen SG, Verweij JJ, Hoekstra P, et al. Real-time polymerase chain reaction for detection of Schistosoma DNA in small-volume urine samples reflects focal distribution of urogenital Schistosomiasis in primary school girls in KwaZulu Natal, South Africa. Am J Trop Med Hyg. 2014;90(3):546–52. pmid:24470560
- 25. Mwazha A, Nhlonzi GB, Kjetland EF. Presence of tissue schistosomiasis in KwaZulu-Natal, South Africa: a retrospective histopathologic review. Front Trop Dis. 2023;4.
- 26. Jacobson J, Pantelias A, Williamson M, Kjetland EF, Krentel A, Gyapong M, et al. Addressing a silent and neglected scourge in sexual and reproductive health in Sub-Saharan Africa by development of training competencies to improve prevention, diagnosis, and treatment of female genital schistosomiasis (FGS) for health workers. Reprod Health. 2022;19(1):20. pmid:35073965
- 27. Danso-Appiah A, De Vlas SJ, Bosompem KM, Habbema JDF. Determinants of health-seeking behaviour for schistosomiasis-related symptoms in the context of integrating schistosomiasis control within the regular health services in Ghana. Trop Med Int Health. 2004;9(7):784–94. pmid:15228488
- 28. Schuster A, Randrianasolo BS, Rabozakandraina OO, Ramarokoto CE, Brønnum D, Feldmeier H. Knowledge, experiences, and practices of women affected by female genital schistosomiasis in rural Madagascar: A qualitative study on disease perception, health impairment and social impact. PLoS Negl Trop Dis. 2022;16(11):e0010901. pmid:36342912
- 29. Mazigo HD, Samson A, Lambert VJ, Kosia AL, Ngoma DD, Murphy R, et al. “We know about schistosomiasis but we know nothing about FGS”: A qualitative assessment of knowledge gaps about female genital schistosomiasis among communities living in Schistosoma haematobium endemic districts of Zanzibar and Northwestern Tanzania. PLoS Negl Trop Dis. 2021;15(9):e0009789. pmid:34591872
- 30. Wambui CW, Madinga J, Ashepet MG, Anyolitho MK, Mitashi P, Huyse T. Knowledge, attitudes and practices toward female genital schistosomiasis among community women and healthcare professionals in Kimpese region, Democratic Republic of Congo. PLoS Negl Trop Dis. 2024;18(7):e0011530. pmid:38995976
- 31.
iLembe District Municipality. Municipal profile. https://ilembe.gov.za/municipal-profile/ Accessed 2026 April 8.
- 32.
Manual AP. Sample size determination in health studies. Geneva. 1991.
- 33. Hegertun IEA, Sulheim Gundersen KM, Kleppa E, Zulu SG, Gundersen SG, Taylor M, et al. S. haematobium as a common cause of genital morbidity in girls: a cross-sectional study of children in South Africa. PLoS Negl Trop Dis. 2013;7(3):e2104. pmid:23556009
- 34. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–81. pmid:18929686
- 35. Strecher VJ, Rosenstock IM. The health belief model. Cambridge handbook of psychology, health and medicine. 1997. p. 117.
- 36. Kleinman A, Eisenberg L, Good B. Culture, illness, and care: clinical lessons from anthropologic and cross-cultural research. Ann Intern Med. 1978;88(2):251–8. pmid:626456
- 37. Anyolitho MK, Nyakato VN, Huyse T, Poels K, Masquillier C. Health-seeking behaviour regarding schistosomiasis treatment in the absence of a mass drug administration (MDA) program: the case of endemic communities along Lake Albert in Western Uganda. BMC Public Health. 2023;23(1):1072. pmid:37277773
- 38.
Hosmer DW, Lemeshow S, Sturdivant RX. Applied logistic regression. New York: Wiley. 2000.
- 39.
Statistics South Africa. General Household Survey 2024. Pretoria: Statistics South Africa. 2025.
- 40.
Joint United Nations Programme on HIV/AIDS. No more neglect: female genital schistosomiasis and HIV: integrating sexual and reproductive health interventions to improve women’s lives. 2019.
- 41. Gadoth A, Mvumbi G, Hoff NA, Musene K, Mukadi P, Ashbaugh HR, et al. Urogenital schistosomiasis and sexually transmitted coinfections among pregnant women in a schistosome-endemic region of the democratic Republic of Congo. Am J Trop Med Hyg. 2019;101(4):828–36. pmid:31392943
- 42. Søfteland S, Sebitloane M, Galappaththi-Arachchige HN, Kleppa E, Holmen S, Pillay P, et al. Diagnosis of female genital schistosomiasis and other genital infections in young South African women: challenges in the syndromic approach. Front Trop Dis. 2024;5.
- 43. Soper DE, Wiesenfeld HC. The continued challenges in the diagnosis of acute pelvic inflammatory disease: focus on clinically mild disease. J Infect Dis. 2021;224(12 Suppl 2):S75–9. pmid:34396404
- 44. Low EL, Whitaker KL, Simon AE, Sekhon M, Waller J. Women’s interpretation of and responses to potential gynaecological cancer symptoms: a qualitative interview study. BMJ Open. 2015;5(7):e008082. pmid:26150145
- 45. Ebonwu J, Mumbauer A, Uys M, Wainberg ML, Medina-Marino A. Determinants of late antenatal care presentation in rural and peri-urban communities in South Africa: A cross-sectional study. PLoS One. 2018;13(3):e0191903. pmid:29518082
- 46.
Statistics South Africa. The status of women’s health in South Africa: Evidence from selected indicators. Pretoria: Statistics South Africa. 2022.
- 47. Sturt AS, Randrianasolo B, Downs JA, Jøker K, Mazigo HD, van Lieshout L, et al. Diagnosis and treatment of female genital schistosomiasis. Lancet Microbe. 2026;7(6):101356. pmid:42127951
- 48. Mwangi VI, Mumo RM, Nyachieo A, Onkoba N. Herbal medicine in the treatment of poverty associated parasitic diseases: A case of sub-Saharan Africa. Journal of Herbal Medicine. 2017;10:1–7.
- 49.
Republic of South Africa. Traditional Health Practitioners Act 22 of 2007. 2007.
- 50.
Do KW. Do 80% of South Africans regularly use traditional healers? 2013. https://groundup.org.za/article/do-80-south-africans-regularly-use-traditional-healers/ Accessed 2026 April 8.
- 51. Nattrass N. Who ConsultsSangomasin Khayelitsha? an exploratory quantitative analysis. Social Dynamics. 2005;31(2):161–82.
- 52. Louw G, Duvenhage A. Do the majority of South Africans regularly consult traditional healers?. Australasian Medical Journal (Online). 2016;9:506.
- 53. Manjang B, Ochola EA, Elliott SJ. The use of non-pharmaceutical interventions for the prevention and control of schistosomiasis in sub-Saharan Africa: A systematic review. Glob Public Health. 2022;17(3):469–82. pmid:33460349
- 54. Mokhesi T, Modjadji P. Usage of Traditional, Complementary and Alternative Medicine and Related Factors among Patients Receiving Healthcare in Lesotho. TOPHJ. 2022;15(1).
- 55. Debie A, Wilson A, Geremew AB, Roberts CT, Wassie MM, Stephens JH. Facilitators and barriers of healthcare financing modalities for universal maternal healthcare services in East Africa: a qualitative systematic review. BMC Health Serv Res. 2025;25(1):897. pmid:40604774
- 56. Nhede NT, Mazenda A, Gondwe D. Beyond poverty alleviation: The impact of child support grants on healthcare access and contraception use in South Africa. Dialogues Health. 2025;7:100228. pmid:40809838
- 57. Galappaththi-Arachchige HN, Amlie Hegertun IE, Holmen S, Qvigstad E, Kleppa E, Sebitloane M, et al. Association of Urogenital Symptoms with History of Water Contact in Young Women in Areas Endemic for S. haematobium. A Cross-Sectional Study in Rural South Africa. Int J Environ Res Public Health. 2016;13(11):1135. pmid:27854250
- 58. Montresor A, Gabrielli AF, Chitsulo L, Ichimori K, Mariotti S, Engels D, et al. Preventive chemotherapy and the fight against neglected tropical diseases. Expert Rev Anti Infect Ther. 2012;10(2):237–42. pmid:22339196