A bibliometric analysis of global research on drinking water and health in low- and lower-middle-income countries

Access to safe drinking water is a fundamental human right and has been acknowledged as such by the United Nations General Assembly through the human right to water (United Nations General Assembly 2010). The provision of safe, affordable, and reliable water is essential to human health and is indispensable to sustainable development. Nonetheless, in 2021, two billion people around the globe continue to lack access to safely managed drinking water services, which is defined as an improved water source that is located on premise, available when needed, and free from contamination (UN Water 2021). Most people who lack access to safe drinking water are situated in low- and lower-middle-income countries (LLMICs), which are consequently the regions disproportionately affected by the health burdens associated with the inadequate provision of water (WHO & UNICEF 2021). Some of these conditions include infectious diseases caused by waterborne pathogens, neglected tropical diseases, and adverse health outcomes such as stunting, wasting, and being underweight (Deshpande et al. 2020). It is estimated that universal access to drinking water, together with guaranteeing sanitation and hygiene services, would prevent 1.6 million deaths per year, accounting for 2.8% of global mortalities (Prüss-Ustün et al. 2019).

Lack of access to water also affects human health and development through several secondary mechanisms, making targeted health interventions complex. The inadequate provision of safe and reliable water is associated with food insecurity (Brewis et al. 2020) and reduced hygiene practices (Staddon et al. 2020), which can in turn lead to heightened rates of malnutrition and infectious diseases, respectively. Moreover, the ramifications of drinking water insecurity extend beyond the development of biological conditions and also involve potential consequences for mental health and well-being, including increased levels of emotional/perceived stress, depression, and anxiety (Wutich et al. 2020). As women and girls are primarily responsible for household water management in LLMICs, they bear the brunt of these mental health burdens (Hadley & Wutich 2009) and often experience heightened vulnerability related to water insecurity during periods of menstruation, pregnancy, and personal or household illness (Stevenson et al. 2012; Collins et al. 2019; Adams et al. 2021). Moreover, evidence suggests that the provision of on-site access to water is associated with an increase in educational and economic opportunities, with these benefits skewing heavily toward women (WaterAid 2021). Thus, obtaining universal access to drinking water is not only a prerequisite for health but is also necessary for achieving progress toward eliminating hunger, advancing women's empowerment, and enabling greater opportunities for economic development.

In recent decades, heightened interest in drinking water research that has been maintained by supportive funding sources has narrowed the gap in knowledge surrounding water and health. The volume of publications in the field has expanded as researchers investigate diverse perspectives on drinking water, such as those related to quality, availability, and accessibility and a multitude of themes, including major water-related diseases, contaminants of emerging concern, water and gender equality, and water governance. Yet it remains unclear exactly which interests dominate the field of drinking water research in LLMICs. Further, the globalization of research on global health topics has accelerated in recent decades, pointing to shifting collaborative efforts that have become increasingly diverse both geographically and across socioeconomic parameters (Addo-Atuah et al. 2020). However, research outputs from High-Income Countries (HICs) continue to dominate several fields of global health research with less representation from LLMICs where the disease burden is disproportionately higher (Ghani et al. 2021). The distribution of global research on drinking water and health in LLMICs needs to be better understood to identify whether these gaps exist with regard to drinking water research.

It is relevant to explore emerging trends in scientific activity to understand where research has been undertaken and by whom. Past bibliometric studies have examined specific topics on drinking water (Hu et al. 2010; Abejon & Garea 2015; Tang et al. 2020; Ekundayo et al. 2021) and concentrated on research out of specific geographical regions (Wambu & Ho 2016), or narrow time periods (Fu et al. 2013). Recent studies have also emerged to assess research on Sustainable Development Goal (SDG) 6 (Basu et al. 2021; Dibbern et al. 2022; Roy et al. 2022), which include a wide range of water, sanitation, and hygiene-related priorities and are again restricted to recent years. A comprehensive analysis examining research on drinking water and health that highlights changing trends over time and analyses data from the perspective of country-level income has yet to be conducted. Hence, the current study aims to assess research trends on water and health in LLMICs from 1980, focusing on progress made in closing research disparities over time. It aims to explore possible pathways forward in the global pursuit toward a more equitable scientific community, such as through initiatives that promote Open Access (OA) content. A comprehensive overview of the field may be used to inform future research priorities and decision-making processes, facilitate improved collaboration on research outputs, and identify where national research systems must be strengthened. In doing so, the following questions would be addressed: What are the overall volume, type, and growth rate of all published documents on drinking water and health in LLMICs? What countries, institutions, journals, or funding agencies have had the greatest influence on research related to drinking water and health in LLMICs? What relationships are present between influential groups? What percentage of global research on drinking water and health in LLMICs is universally accessible (i.e. OA)? What are the most frequently studied topics on drinking water and health in LLMICs?

A bibliometric analysis was conducted to produce an overview of publications and groups that have contributed to research on drinking water and health in LLMICs since 1980. The Web of Science Core Collection (WOSCC), a leading bibliographic source (Mongeon & Paul-Hus 2016), was used to extract data sets. The final search was performed on 28 June 2022.

The search strategy was developed to search title, abstract, author's keywords, and ‘keywords-plus’ (i.e. common words and phrases from article titles) to retrieve publications since 1980 in all languages. As defined in Table 1, search terms related to: (1) drinking or domestic water, including potable or household water; (2) health or disease; and (3) LLMICs.

LLMICs were determined based on the ‘list of ODA recipients effective for reporting on 2022 and 2023 flows,’ published by the Organisation for Economic Co-operation and Development (OECD) (DAC 2021). For inclusion in this study, only research on low-income countries (LICs) and lower-middle-income countries (LMICs) was searched, which are defined as nations with GNI per capita below $1,045 and between $1,046 and $4,095, respectively (Hamadeh et al. 2021). Literature on upper-middle-income countries (UMICs) and HICs was not explicitly searched as a part of this study due to the considerably higher rates of drinking water access in these regions. To retrieve relevant records, synonyms for LLMICs included now outdated terms such as ‘third world,’ as well as imprecise terms such as ‘global south’ (Khan et al. 2022; Lencucha & Neupane 2022). The use of this vocabulary, however, is not endorsed and henceforth will refer to LLMICs as a generic term. In recognizing the role that language plays in perpetuating or minimizing the marginalization of certain groups, this term was chosen for its comparatively objective connotation in this study and elsewhere. However, it should be acknowledged that any attempt to classify groups of people based on indicators, like national income status, will present inherent shortcomings, as reflected by the drastic income inequalities that persist within nations.

A subset of the search queries was reviewed to ensure the search strategy produced results that fit within the scope of this study. A total of 7,899 publications were retrieved from WOSCC and included in the analysis.

Analysis of the structure and dynamics of the research field was conducted using the functionality offered by Clarivate's InCites tool, in conjunction with the Bibliometrix R package. The leading institutions, countries, authors, and journals were identified using publication-related metrics (indicating productivity) and citation-related metrics (indicating research impact and academic influence). To minimize the likelihood that results were skewed by outliers, thresholds to limit the search were established. For example, during impact analysis (i.e. examining Category Normalized Citation Impact (CNCI)), publications were limited to those with up to 25 authors and from institutions/countries with a minimum of 25 publications that contributed to the field. This is done to ensure that publications with an abnormally high number of co-authors, as well as institutions/countries with minimal publications, do not have unduly influence on interpretation. To explore recent trends, the time period was adjusted to the past five complete years (2016–2021).

By examining the records retrieved, the author affiliations and citations that link documents were leveraged to identify relationships among researchers, institutions, and countries. Network analysis was performed to draw attention to the relative importance of research groups to the field and the strength of relational ties between institutions/countries was measured. The software tool VOSviewer 1.6.18 was used to generate, visualize, and analyze these bibliometric networks.

Bibliographic coupling was utilized to explore the topical research similarities at the institutional- and country-levels (Kessler, 1963). Additionally, co-authorship analysis was utilized to illustrate regional contributions to the field and inter-country collaboration patterns. Mapping such collaborations may be particularly relevant for highlighting countries or regions that typically partner to produce research outputs, which may have the potential to spark new collaborative networks that integrate underrepresented regions.

Since 1980, the top 10 institutions with the greatest CNCI, a metric used to measure citation impact by adjusting for field or subject area, is illustrated in Table 2. Note that organizations in this table are unified, but data are displayed for both the parent organization (i.e. University of London) and the child institution (i.e. London School of Hygiene and Tropical Medicine). The leading institution is the University of California Berkley with a CNCI of 2.6, followed by the Swiss Federal Institute of Aquatic Science and Technology (EAWAG) (2.2), and the London School of Hygiene and Tropical Medicine (2.1). H-Index, a time-dependent measure of productivity and impact was also determined for organizations over the past 5 years (2016–2021). The following list presents the organizations with the highest H-index during this period; Tehran University of Medical Sciences (27), University of California System (22), COMSATS University, Islamabad (21), Indian Institute of Technology System (20), University of London (20), Chang‘an University (19), Chinese Academy of Sciences (19), the ICDDR (18), Stanford University (17), and Shahid Beheshti University (17). With only three of the top 10 leading organizations representing HICs, this list is suggestive of the growing impact of publications from LLMIC-affiliated institutions in recent years.

The countries with the highest citation impact by total citations are the United States (62,489 citations), India (37,709 citations), and the United Kingdom (29,764 citations). The past 5 years (2016–2021) mark the growing volume and impact of research output from LLMICs. While the United States and the United Kingdom remain top producers of highly cited research, only 3 of the top 10 countries constitute HICs (Table 3).

Since 1980, 9 of the top 10 most influential authors based on the H-index are affiliated with UMICs/HICs. On average, LLMIC-affiliated publications have an Impact Relative to World indicator of 0.84, in contrast to 1.23 from UMICs/HICs. This indicator is used to determine the relative research performance where the world average is equal to one.

For each article, InCites associates the Journal Impact Factor (JIF) of the journal in which it was published. As the ratio of incoming citations to outgoing references, the JIF provides an index of the relative importance of a journal in the field. A journal with a high JIF is widely read and contains articles that have attracted more citations than they themselves cite. While this journal-level metric does not confer any level of significance to a given article, it serves as a marker of the general quality of research found in that journal, as editors of a high JIF journal can be very selective of which manuscripts they accept. Results show that 47% of publications generated by researchers affiliated with UMICs/HICs are published in journals that have a JIF in the top quartile (i.e. attract more citations than 75% of all journals in the field), compared to 35% of publications produced by LLMIC-affiliated researchers. This indicates that as a percentage of the total published literature, documents produced by LLMIC-affiliated researchers are less likely to be published in highly cited journals.

The most relevant sources were Science of the Total Environment (236 articles), Environmental Monitoring and Assessment (172 articles), Environmental Science and Pollution Research (172 articles), International Journal of Environmental Research and Public Health (154 articles), American Journal of Tropical Medicine and Hygiene (146 articles), Journal of Water and Health (140 articles), Environmental Geochemistry and Health (133 articles), PLOS One (130 articles), Journal of Water Sanitation and Hygiene for Development (120 articles), and Environmental Earth Sciences (119 articles).

As a property of each node in the network, the sum of a country's collaborations is given as total link strength. The United States has the most collaborations (1,486 total link strength), followed by England (815), Bangladesh (745), and India (594), indicating that these countries have the strongest collaborative research ties. England had a total of 57 country links, followed by India (55), the United States (54), and Germany (49), which represent the countries with the most diverse range of collaborative ties. The strongest bi-lateral connections are observed between the United States and the following countries: Bangladesh (with link strength of 295), England (198), India (150), and Kenya (104). This was followed by connections between Bangladesh and Japan (94 link strengths) and between Pakistan and China (93).

Beyond comparing the volume of output and collaborations, this study also characterizes publications from the perspective of their accessibility to potential readers. Academic journals were traditionally available only via an often very expensive subscription. This model presents a barrier for researchers in LLMICs, whose institutional libraries often cannot afford the journal subscriptions (Matheka et al. 2014). More recently, the OA model, funded primarily by publication and article processing fees, has made an increasing proportion of academic literature free to read. OA refers to research that is accessible to anyone for free online and grants permission to copy, use, or distribute such research with proper recognition (Swan 2012). An OA journal can be further categorized by its chosen economic model, which is known as Gold, Hybrid, and Bronze. An OA article can be both Green as well as any of the aforementioned types. Definitions of these types are provided in Table 4.

In the past 5 years (2016–2021), the publications targeting the SDGs have primarily focused on SDG 3: good health and well-being (3,311 documents and 38,305 times cited), SDG 6: clean water and sanitation (2,171 documents and 27,089 times cited), and SDG 14: life below water (1,555 documents and 18,996 times cited).

Table 5 illustrates the funding agencies dominating the field based on both productivity and impact. Funding agencies in UMICs/HICs comprise 17 of the top 20 with India (2) and Pakistan (1) being the only countries representing LLMIC-based agencies.

Of the global research on drinking water and health in LLMICs published since 1980, the United States and the United Kingdom have dominated the field based on the production and impact of scientific literature. India, Iran, Bangladesh, and Pakistan are emerging as key players, yet, except for Bangladesh, publications out of these countries remain disproportionately restricted behind paywalls. Contamination and phytoremediation, diarrheal disease, and water resources are the themes that characterize most of the research on water and health with the latter emerging as a key priority only in recent years.

The scientific literature on drinking water and health in LLMICs has grown, particularly since the start of the millennium, with publication volume increasing steeply since 2016. The sustained and robust production of research shown in the present study mirrors findings from similar investigations that evidence drinking water research as a developing field of study (Fu et al. 2013; Wambu & Ho 2016) and is relevant to broader conclusions that previous scholars have established: health and water, the combined focus of the present study, are among the most researched sustainable development priorities (Meschede 2020).

Following successful efforts to meet the water-related Millennium Development Goal (MDG) target, to reduce by half the proportion of the population without safe drinking water and sanitation, 2.6 billion people gained access to an improved source of drinking water, i.e. one that adequately protects from outside contamination, between 1990 and 2015 through supply programs and interventions (WHO & UNICEF 2015). In 2015, the United Nations extended far beyond the MDGs through the adoption of the 2030 Agenda for Sustainable Development, a new global blueprint toward a more sustainable future. Included in the agenda is SDG 6, which aims to ensure the availability and sustainable management of water and sanitation for all (UN DESA 2016). The universal clean water target (6.1) is particularly significant because it underscores several other SDGs (Bhaduri et al. 2016). For example, an adequate supply of water is indispensable for and intricately connected to achieving good health and well-being (SDG 3), no poverty (SDG 1), and gender equality (SDG 5) (UN-Water 2016). The growth of published drinking water research observed in the past 20 years is reflective of efforts to reduce drinking water inequalities through the MDGs and similarly, the acceleration of such research outputs since 2016 is likely a consequence of the international community's doubling down on drinking water priorities through the commitments to universal access outlined in the SDGs. The growth of the field can likely also be attributed to the expansion of internet access and telecommunications technologies, enhanced research capacity, and the improved ease of information sharing through digital channels over time. However, these conveniences have not been distributed equally across the globe, as some are better able to capitalize on technological tools and systems while others face barriers (Shumba & Lusambili 2021). As will be discussed in the following sections, a necessary step in accelerating the production and application of water research is to reduce research inequalities by addressing enduring barriers to participation that exist within underrepresented regions.

Furthermore, it is relevant to evaluate recent progress on drinking water targets and assess current global events to predict whether research in this field will continue to expand in the future. Worldwide, progress toward the use of safely managed drinking water, which was set as the new indicator, has progressed at a rate of 0.63% per year between 2015 and 2020 (WHO & UNICEF 2021). Yet, the world is not currently on track to achieve universal and equitable access to affordable drinking water by 2030 (UN Water 2021). LLMICs continue to confront challenges in closing the gap in drinking water disparities, such as reaching rural communities, transitioning existing facilities (i.e. from wells to piped), and ensuring consistent monitoring (WHO & UNICEF 2021). Likely to further challenge global progress on drinking water equality are emerging global threats associated with rapid population growth, viral pandemics, climate change, and conflict (Hannah et al. 2020). These pressures have the capacity to place additional stress on global water supplies and potentially interrupt efforts toward achieving universal access. Modern-day challenges, combined with the new evidence needed to advance SDG 6, are the likely catalysts for the marked expansion of research outputs highlighted by the present study and are predicted to drive the continued expansion of the research field in years to come.

The present study assessed the key players in drinking water research, highlighting emerging influences at the institutional- and country-level. The results suggest that the United States and the United Kingdom have historically driven research, a finding that has been supported by similar literature analyzing the areas of drinking water and SDG 6 (Basu et al. 2021; Roy et al. 2022). Although the United States and the United Kingdom presently remain important contributors, heightened LLMIC-affiliated research in recent years out of India and to a lesser extent, Pakistan, Iran, and Bangladesh are shifting the landscape of research production on drinking water and health in LLMICs. The growing influence of LLMIC-based research highlighted by the present study is at variance with other studies on SDG 6, such as one conducted by Dibbern et al. (2022) that did not determine a significant contribution from these regions on the field. This may, in part, be due to the explicit use of ‘health’ and ‘LLMICs’ in the present study's search strategy instead of ‘SDG 6,’ the latter of which may be utilized less often by LLMIC-based researchers despite similar driving research agendas (i.e. regarding the complexities surrounding what is considered global health versus public health (Bozorgmehr 2010)). Moreover, the present study revealed that the impact of publications out of LLMICs relative to the world and the proportion of research published in Q1 journals were notably lower than UMICs/HICs, pointing to a possible undervaluing of this scientific literature.

The United States and the United Kingdom are consistently evidenced as frontrunners in the production of global health research (Cash-Gibson et al. 2018) and are among the top producers of research in several other scientific fields (Sumathipala et al. 2004; Sweileh 2020). It is therefore unsurprising that these countries have contributed significantly to the literature on drinking water and health in LLMICs. Part of this dominance may be due to supportive national commitments to reducing drinking water inequalities and connected funding priorities. For example, the United Kingdom's Department for International Development (DFID) – now known as the Foreign, Commonwealth and Development Office (FCDO) – has supported numerous research initiatives and interventions to target drinking water in LLMICs in recent decades, with evidence pointing to 80 million people who gained access to WASH services from DFID projects between 2011 and 2019 (Blaszczyk 2019). A push for greater assistance from governments and funding agencies in HICs to support global health research on drinking water, in addition to empowering LLMICs to participate in this research and drive agenda-setting, may facilitate the enhanced generation of impactful research in the coming years.

The growing representation of LLMICs in drinking water research in the past 5 years is promising, as India, Pakistan, Iran, and Bangladesh are now among the global leaders, while Nigeria and Ethiopia are also represented in the top 10 producers of literature. While other scholars have pointed India as the top LMIC contributor to health inequalities research in general (Cash-Gibson et al. 2018), the present study reveals that India is now the new global leader in research on drinking water and health in LLMICs, recently surpassing the United States on scientific production. Despite this, evidence suggests that significant disparities exist within countries, such as India, suggesting that research capacity may be developing unevenly (Kalita et al. 2015). For example, one study suggests that a heightened proportion of health research is produced by institutions in just four states in India, whereas the regions with the highest disease burdens are only minimally represented (Kalita et al. 2015). As such, it is crucial that priority-setting highlights opportunities for underserved populations to become involved in research so that inequalities within countries do not escalate as global inequalities in research are reduced.

Although the increased representation of LLMICs on the list of the top 10 producers of research is encouraging, this does not represent the minuscule production and underrepresentation of many other LLMICs that face barriers to establishing research capacity. In the present study, for example, LICs are significantly less represented as producers of drinking water research than LMICs, meaning that while new research systems are strengthening in some regions, countries that possess the least resources remain at the periphery of global drinking water research, at times with more problems and without a voice to drive the research agenda. This implies that greater effort is needed to ensure the field of drinking water research becomes more representative of the LICs that are likely the ‘subjects’ or ‘recipients’ of the implications of such research. This speculation is consistent with a study conducted by Dimitris et al. (2021) that evidences uneven progress in the representation of authors from LMICs in global health research, despite greater representation as a whole. With this understanding, capacity-building and improved authentic partnering to encourage research out of low-resource settings should be prioritized to strengthen national research systems.

The continued presence of water-related diseases in the modern era is indicative of the complexities that underpin efforts to achieve universal access to clean drinking water. As solutions to drinking water challenges are often based on complex systems of unique, contextual factors, ‘silver bullet’ solutions fail to consider the changing landscapes that surround people's relationships with water in local settings. This is why it is important for drinking water research to include regional investigations, for example through implementation science, that are driven by principles of co-creation, local decision-making, and strengthened local research capacity (Haque & Freeman 2021). Many scholars convey that barriers to reaching water and sanitation targets stem from a lack of local water governance, not simply a lack of funding, technology, or infrastructure (Herrera 2019). Local researchers from LMICs are needed to guide context-specific research investigations and enhance considerations of organizational, political, social, and infrastructural circumstances that define the specific WASH needs of the community under study. Local involvement also facilitates the production of health research that aligns with processes that facilitate the greatest impact. The field of global health is predicated on eliminating health disparities between people, particularly focusing on empowering low-resource and marginalized groups through access to fundamental health opportunities. Similarly, there is a need for the field of global health to encourage research out of low-resource settings, which will not only foster publication equity but also empower actors who are better positioned to identify unique factors that may disrupt interventions from actually improving health outcomes.

The present study analyzed co-authorship to determine which countries have the strongest collaborative research ties and used bibliographic coupling to determine whether similar thematic areas emerge among organizations/countries. The visualizations in the present study suggest that the United States, England, Bangladesh, and India had the strongest international research collaborations. Partnerships were also strong between Bangladesh and Japan, and between Pakistan and China; however, these links were not as strong as connections between the United States and the following countries: Bangladesh, England, India, and Kenya. Heightened collaborative ties observed for the United States and England are consistent with previous findings by Basu et al. (2021); however, the strong international networks shown for Bangladesh and India are perhaps under-recognized in the literature. Moreover, findings suggest that similar research themes are pursued among European and American organizations, whereas Asian and Middle Eastern organizations tend to cluster together more.

In general, the network analysis in the present study depicts a rather robust map of international partnerships with several collaborative hubs. In particular, the well-established research partnerships between the United States and England and several other countries illustrate the continued role these two countries play in driving research in this field. The comparatively stronger ties between the United States/United Kingdom and certain LLMICs may have developed based on linguistic and historical happenings. As a large portion of global health research is driven by English-speaking HICs (Cash-Gibson et al. 2018), it is likely that partnerships that were previously established based on the similar language spoken may have fostered collaborative research on drinking water as well. For example, strong ties between the United States/United Kingdom and English-speaking East African countries are particularly high. Bangladesh and India, also English-speaking LLMICs, are strong centers for international collaboration in their own right, which may further be explained by the heightened production of research out of these countries in general.

The strong collaborative ties between Japan and Bangladesh as well as China and Pakistan are notable. This is mirrored in the bibliographic coupling networks, which indicate Asian/Middle Eastern countries produce literature on different thematic areas than the rest of the world. This separation may be indicative of differing regional needs for drinking water research, apart from advancements in basic WASH that drive a significant amount of research on SDG 6. For example, China, India, Bangladesh, and Pakistan are among the leading producers of research on heavy metals, such as arsenic, in drinking water (Abejon & Garea 2015; Han et al. 2020), which aligns with heightened concerns over contaminated waterways in populated Asian nations. Countries in the Middle East have differing agendas for advancing drinking water development, which requires research related to water scarcity and security (World Bank 2018). Thus, it is speculated that clusters of co-authorship, in addition to bibliometric coupling networks are reflective of the pervasive research needs unique to each region.

Despite the well-established networks displayed in the present study, partnerships between LICs appear to be much fewer, with weaker link strengths. This highlights possible entry points for enhanced collaborative partnerships. International collaborations play a pivotal role in facilitating information sharing, enabling the transfer of technology, and widening the reach of research to affect global health outcomes (Addo-Atuah et al. 2020). Such partnerships, however, must involve authentic partnering, inclusion, and shared benefits to ensure that power imbalances are not propagated in the process (Plamondon & Bisung 2019).

Strengthening meaningful international partnerships across income groups and also between LLMICs can facilitate the advancement of drinking water targets through improved knowledge generation. The 2030 Agenda for Sustainable Development explicitly calls for expanded international co-operation and capacity-building to support low-resource settings with water and sanitation-related activities and programs (6.a) as well as supporting the participation of local communities in water management (6.b) (UN DESA 2016). Not only is there a need to improve inter-country partnerships, but also, there is a growing call for water researchers to collaborate with actors in related sectors and partner with those working on connected development goals (Bhaduri et al. 2016). In the face of contemporary global issues tied to rapid population growth and climate change, it is likely that inter-regional and inter-sectoral collaborations on water research will grow to explore the myriad of water-related challenges that flow through a range of development priorities. In the coming decade, collaborative research partnerships that transcend national borders will be crucial to attaining drinking water equity and minimizing the elevated water-related disease burden that weighs heavily on LLMICs.

The findings on OA trends in the present study are consistent with other studies, which point to publications becoming increasingly OA over time (Kurata et al. 2013; Bosman & Kramer 2018). In the present study, UMICs/HICs exhibit a higher percentage of OA publications, with the United Kingdom demonstrating the highest rates across all years, likely due to funder mandates. Interestingly, in the past 5 years, East African countries had among the highest levels of OA content globally. Despite the progress realized in this region, several other LLMICs continue to lag behind the global average on OA indicators, particularly those emerging as leaders in the field, namely India, Iran, and Pakistan. These findings are consistent with conclusions by Iyandemye & Thomas (2019) that point to sub-Saharan Africa as having the highest rates of OA biomedical publications, with the Middle East and South Asia displaying among the lowest. Importantly, several countries with limited research capacity were left out of the analysis as they did not meet the minimum threshold of 100 publications in the field to be evaluated.

Since 2003, when the OA movement was declared a global priority at the UN's World Summit on Information Technology (King & Tamber 2004), OA has been advocated as a tool to support the expansion of an equitable scientific knowledge base for social and economic development. The present study reveals a marked increase in OA content in the years following the announcement of the OA movement, which may illustrate the influence of international commitments to equitable information sharing on changing the nature of how research is accessed. Moreover, there are several possible explanations for the heightened rates of OA content in East Africa, and other low-income regions that the present study highlights. It is speculated that the role of OA fee waivers for low-income countries may be one possible reason. Further, heightened collaboration with countries with strong OA policies (i.e. UK) may have enabled a greater percentage of OA-published research to come out of LLMICs since publications with international and inter-regional collaboration tend to have considerably higher OA rates compared to single-country outputs (Chan 2019). High OA rates in East Africa may also be explained by the heightened production of research into topics that are overrepresented in OA literature, such as poverty-related diseases (Breugelmans et al. 2018). Finally, the expansion of OA repositories in countries such as Kenya, Tanzania, and Uganda in recent years, may have also contributed to heightened rates of OA publications (Otando 2015).

In the present study, it is notable that India, Pakistan, and Iran have among the lowest percentages of OA publications, despite producing a large volume of literature on water and health. These findings may provide evidence of regions that can be better supported in the transition to OA research. There are numerous reasons why some countries and institutions are slow to adopt OA frameworks and practices. As Bosman & Kramer (2018) suggest, funder/publisher policies, institutional mandates, and repository infrastructure play influential roles in determining whether research will shift in the direction of OA. Further research into barriers that are restricting OA progress may be key to supporting regions slow to adopt OA priorities. Advocates of the OA movement suggest that it is a necessary mechanism for leveling the playing field between those who can afford to produce and access fee-based scientific literature and those who cannot (Tennant et al. 2016).

The sustained production of research on both contamination and diarrheal disease is an indication of the global challenges that have driven long-term priorities in global water research. Chemical pollution from industrial discharge and untreated sewage remains a heightened area of concern in the modern, industrialized era since evidence points to a range of chronic diseases from exposure to contaminants found in water (Villanueva et al. 2014; Hutton & Chase 2016). Moreover, diarrheal disease is consistently targeted as a research priority as it accounts for 829,000 deaths per year, representing one of the leading causes of death in children under five (Prüss-Ustün et al. 2019). Scientific monitoring of contaminants in water and improved knowledge generation on diarrheal disease have the capacity to greatly reduce the associated health burdens, hence the continued effort to develop scientific literature on these topics.

In the present study, it is unsurprising that the greatest number of publications and most cited literature are associated with SDG 3 (good health and well-being) and SDG 6 (clean water and sanitation), since the search strategy explicitly included ‘health’ and ‘drinking water,’ both of which respectively represent these goals. Notably, SDG 14 (life below water) is the third most-published SDG in this study, aligning also with the increase in literature on the topic of ‘water resources’ in recent years. Such evidence may point to planetary health, a field dedicated to the relationships between human health and the earth's natural systems, as a growing global priority. The extension of the SDGs to incorporate efficient water resource allocation alongside environmental protection, rather than as a separate agenda item, is a likely explanation for the popular keywords ‘groundwater’ and ‘contamination’ highlighted in the searched health literature. The rise in publications related to water resources in the wake of announcements for the 2030 Agenda for Sustainable Development may therefore be reflective of a broader movement in the drinking water sector toward a more holistic, cross-sectoral integration of priorities on social and environmental development.

Interventions, policies, and research agendas are becoming increasingly multifaceted because of the recognition of how fundamental and interdependent water resources are for so many sectors. In 2016, water insecurity was named the greatest long-term risk facing society by the World Economic Forum (WEF 2018). It is also evidenced that among other interconnected consequences, the growing strain on water supply would put global GDP and grain production at serious risk by 2050 (WWAP 2016). As such, drinking water research has expanded to examine the interconnections between development goals and across targets listed within goals. Evidence from surveys of WASH professionals across 36 countries suggests that water research must further engage in multisectoral coordination to meet contemporary global needs (Setty et al. 2020). It is predicted that collaborative approaches to water-related sustainable development, rather than fragmented ones, will continue to facilitate partnerships between actors across research fields and levels of society, which may lead to topics that reflect interdisciplinary priorities.

The patterns in publications that address social determinants of health, implementation science, or qualitative methods remain unexplored by the present study. There is growing recognition by health researchers, including those studying drinking water, of the relevance of social science perspectives to better understand how thoughts, attitudes, beliefs, and social norms shape behavior (Lilje & Mosler 2017). For example, critical WASH technologies are rendered ineffective if psychosocial factors impede the rate of adoption by the receiving communities, such as beliefs among women that using latrines is strange or scary, compared to practicing open defecation (Aiemjoy et al. 2017). Understanding how and why drinking water programs benefit communities or fail to reach intended outcomes is critical to reaching vulnerable populations most burdened by inadequate drinking water services (Loevinsohn et al. 2015). Considerations of social determinants in drinking water research may therefore result in a more comprehensive understanding of the barriers associated with universal drinking water and lead to more effective interventions (Dreibelbis et al. 2013). The possible movement toward the integration of social science perspectives in drinking water research is not captured by the present study; however, such investigations may make an interesting extension as it would further define the nature of research on drinking water and health in LLMICs.

Although most of the literature on drinking water and health in LLMICs was captured in this review, this study has several limitations. First, only articles published in WOSCC were included, thereby excluding potentially relevant publications from local journals that are not indexed in this database. Moreover, papers from predatory journals may inevitably be indexed in reputable databases and due to the large number of papers included in this study, it was not feasible to screen out such papers. Thus, countries/institutions with a heightened volume of papers published in predatory journals may be inflated. Second, the search strategy does not include research on UMICs or low-resource settings within HICs, which may also contain disadvantaged groups that lack access to adequate drinking water but were not captured here. The search was also conducted in English, thereby missing records that were cataloged in different languages. This may have resulted in the underrepresentation of certain regions, such as China, that likely have contributed more to the field. Moreover, the search strategy included imprecise terms related to the ‘developing world,’ whose membership has changed throughout history or can be redefined based on different classifications. Third, there are numerous bibliometric indicators that can be utilized to evaluate a given field of research, and this study is limited by the indicators and visualization tools selected by the authors. For instance, VosViewer references ‘England’, without having the functionality to analyze the ‘United Kingdom,’ whereas data generated from Clarivate InCites are generally analyzed using ‘United Kingdom.’ Thus, international partnerships out of this particular region were likely underrepresented in network maps, as the ‘United Kingdom’ would have yielded greater international collaborative links. Finally, this study used citation-based metrics to examine the impact of publications in the field of drinking water research. However, a strong bibliometric impact may not always align with publications that yield the greatest societal impact, the latter of which may be better understood by evaluating policies and programs that were established as a result of published research articles. Despite these limitations, the current study provides a useful overview of global research on water and health in LLMICs.

As sustainable development priorities drive the continued expansion and globalization of research on drinking water and health, special efforts are needed to support the participation of low-resource and underrepresented researchers, particularly from LICs. Local voices and perspectives may be particularly significant for facilitating the translation of research into context-specific solutions and producing evidence-informed decisions that advance both policy and implementation efforts. In recognizing that information sharing is critically important to building research capacity, the OA movement, and policies that support low-resource researchers to access and publish accessible drinking water research may also play an important role in developing a more equitable scientific community. The present study may be used to inform future research priorities and decision-making processes, facilitate improved collaboration on research outputs, and identify where national research systems must be strengthened. This will enable gaps in global drinking water inequalities to be filled, while at the same time accelerating equitable, inclusive research in the realm of water and health.

All relevant data are available from an online repository or repositories (10.6084/m9.figshare.21944102).

The authors declare there is no conflict.