Introduction

Fluorine is a naturally occurring element that is commonly found in nature in the form of fluoride. Fluoride minerals are naturally present in soil and aquifer sediments, which can result in the accumulation of fluoride in freshwater sources, particularly in groundwater1,2. High concentrations of fluoride in groundwater represent a significant global public health concern3,4, affecting approximately 200 million individuals across 25 countries3. The implementation of fluoride in public health programmes for the prevention of dental caries is well-documented; however, there is evidence of the detrimental consequences associated with excessive consumption5,6. Chronic exposure to fluoride can result in the development of dental and skeletal fluorosis, as well as impact various tissues and organs, including the thyroid7, liver, and kidney8, in addition to the immune system9 and the nervous system10. These effects often have a significant impact on the lives of those affected. Recent researches indicate that even lower levels of fluoride exposure can have deleterious effects on human health7,8,11. China is one of the countries facing a significant epidemic of drinking water fluorosis, affecting 28 provincial-level administrative divisions (PLADs) and over 70 million individuals3. In the 1980s, 812,000 individuals suffered from skeletal fluorosis and 15.06 million from dental fluorosis in China12. In order to address the health risks associated with endemic fluorosis from drinking water, China has implemented water quality improvement projects and defluoridation measures. Given that fluoride is colorless, odorless, and transparent, its presence in groundwater sources may remain undetected until tested13,14. China has conducted surveillance for endemic fluorosis to assess fluoride levels in drinking water and the effectiveness of control measures.

Health risk assessment is a fundamental method that establishes a link between environmental pollution and human health. By quantifying the impact of pollutants and predicting potential risks, it provides valuable insights into the potential health hazards associated with various environmental factors15. In the realm of drinking water safety, the method is particularly sensitive and widely applicable15, as it allows for the assessment of the health risks associated with contaminants in drinking water.

Snigdha et al.16 assessed water quality and health risks in the upper Brahmaputra floodplains of Assam, India. Their findings revealed that the hazard quotient for fluoride exceeds 1, surpassing the benchmark recommended by the U.S. Environmental Protection Agency (USEPA). Additionally, a study conducted in Tenerife indicated that dietary exposure to fluoride among children (9 to 17 years old) and adults in some municipalities approaches or exceeds the upper level (UL) by nearly 100%17. Previous studies have conducted health risk assessments of fluoride in drinking water among the Chinese population18,19,20,21,22,23. However, these studies have been limited in scope, focusing on specific regions and lacking comprehensive nationwide coverage. The health impact of fluoride in drinking water is regarded as a geochemical disease24, emphasising the necessity of analysing its spatial–temporal distribution to develop targeted control measures. It is of paramount importance to consider the vulnerability of young children to these health risks, given their higher substance consumption relative to body weight and underdeveloped nervous and immune systems25. This study conducted a national health risk assessment of drinking water fluoride in minors to gain a deeper understanding of the impact of controlling drinking water endemic fluorosis and to provide a foundation for adjusting control strategies in China.

Methods

Fluoride concentrations in drinking water

To ascertain the efficacy of the control of endemic fluorosis in drinking water, the Ministry of Health has conducted national surveillance of endemic fluorosis in drinking water in 28 PLADs, which were the original area selected for this study (Table S1). A simple random sampling method was employed to select surveillance points in villages affected by drinking water endemic fluorosis. The surveillance project adheres to a standardised programme, wherein all laboratories employ consistent methodologies for the collection, preservation and analysis of water samples. If a water improvement project is undertaken, a single tap water sample is collected during both the dry and wet seasons to measure the fluoride concentration (each sample is measured twice in parallel, and the resulting average value is calculated). In instances where the water supply remains unaltered, five water samples are collected at regular intervals from five different points (east, west, south, north, and center) to ascertain the fluoride content.

The collection and preservation, examination, and analysis quality control of drinking water samples were carried out by the requirements of the standard examination methods for drinking water-collection and preservation of water samples (GB/T 5750.2-2006)26, nonmetal parameters (GB/T 5750.5-2006)27, and water analysis quality control (GB/T 5750.3-2006)28 established by the Chinese national administrations, respectively. The water samples were collected in 5.0 L polyethylene bottles and stored at 4 ℃ prior to analysis. The fluoride content in the water was determined using the ion selective electrode method29. To ensure the reliability of the findings, both external and internal quality control measures are employed. The data for this study were obtained from the national surveillance report of the endemic disease control center of the China CDC, except Xizang, in 2014 and 2018 (Fig. 1). Furthermore, this study employed the fluoride concentration limit of 1.0 mg/L in drinking water as the exposure level, by the standards for drinking water quality (GB 5749-2022)30 established by the Chinese national administrations.

Fig. 1
figure 1

The distribution of the study regions (the map was generated using QGIS 3.30.0, https://qgis.org/).

Full size image

Exposure parameters

The Exposure factors handbook of Chinese population (0–5 years)31 and (6–17 years)32 were used as a reference to obtain the recommended values of exposure parameters, including drinking water intake, bathing time, body weight and skin surface area, based on age and region. The handbook presents data from a nationwide survey conducted by the Ministry of Environmental Protection between 2013 and 2014. The survey concentrated on the environmental exposure behavioral patterns of the population aged 0–17 years. The handbooks serve as a repository of exposure parameters that reflect the distinctive characteristics observed in different regions of Chinese children. The minors were classified into three age groups: infants (0–2 years), children (3–11 years), and teens (12–17 years).

Health risk assessment

The technical guide for environmental health risk assessment of chemical exposure (WS/T 777-2021)33 was employed to evaluate the potential health risks of fluoride in water for minors. WS/T 777-2021 is a health industry standard issued by the National Health Commission of the People’s Republic of China in 2021. The health risk assessment model has been developed for the purpose of assessing the potential health risks associated with exposure to a single chemical substance via a range of pathways. The primary pathways of exposure to chemicals are ingestion and dermal absorption. This study employs the methodology delineated in WS/T 777-2021 to evaluate the non-carcinogenic risks associated with fluoride in drinking water via the two aforementioned pathways. The assessment model and detailed parameters are presented in Supplementary Tables S2 and S3, respectively.

The non-carcinogenic risk threshold established by WS/T 777-202133 is set at 1. Should the hazard quotient (HQ) or hazard index (HI) exceed this value, it indicates a significant non-carcinogenic risk that requires attention.

Standard protocol approvals, registrations, and consent statement

This study was approved by the Ethics Committee of Harbin Medical University, China. It adhered to the Guidelines for Accurate and Transparent Health Estimation Reporting for Population Health Research.

Since the exposure factors of the Chinese population are derived from a publicly available database and all participants’ data were anonymous, informed consent was not required.

Statistical analysis

The normality of fluoride concentration was evaluated using the Kolmogorov–Smirnov test (K-S test). Non-normally distributed data was summarized using median, minimum, and maximum values. Differential analysis was conducted using the Mann–Whitney U test. Categorical variables were compared using the Chi-square test. The health risks and spatio-temporal characteristics of fluoride exposure among minors in endemic fluorosis areas were depicted using risk probability and box plots. The data was organized, analyzed, and visually represented using R4.3.1 software. A statistically significant difference (two-tailed) was considered when P < 0.05.

Results

National surveillance situation

A total of 439 and 421 water samples from regions endemic to fluorosis were analyzed in 2014 and 2018, respectively. The fluoride concentrations reported are the averages for all areas in each year, with values of 0.84 mg/L (0.03 ~ 11.8 mg/L) in 2014 and 0.59 mg/L (0.05 ~ 5.94 mg/L) in 2018 (Table 1). In 2014, 42.8% of the samples exceeded the Chinese drinking water standard of 1.0 mg/L30, while in 2018, this percentage significantly decreased to 29.7%. The difference in the percentage of water samples with fluoride levels above the standard from 2014 to 2018 was statistically significant (χ2 = 16.013, P < 0.001).

Table 1 Fluoride concentrations in drinking water in different regions in 2014 and 2018.
Full size table

Spatial and temporal distribution of fluoride in drinking water

The spatial and temporal distribution of fluoride in drinking water across the six geographical regions of China is presented in Table 1. Except Southwest China, the median fluoride concentrations in water decreased in the other five regions in 2018 compared to 2014. Statistically significant reductions were observed in North China, Northwest China, and Northeast China (P = 0.024, P < 0.001, and P = 0.034, respectively). By 2018, the median fluoride concentration in drinking water across all regions was at or below 1.0 mg/L30, thus meeting the standard (Table 1).

Health risks of different exposure pathways

The dermal absorption health risks (HQderm) associated with fluoride exposure were found to be consistently below 1. Across different age groups, the fluoride intake through drinking water (HQoral) was significantly higher than dermal absorption, with three orders of magnitude difference (Table 2). In 2014, the average HQoral for infants and children in North China and East China exceeded 1, while it was below 1 in other regions. Conversely, in 2018, the average HQoral for infants, children, and teens in Southwest China was above 1, but below 1 in other regions (Table 2).

Table 2 Non-carcinogenic health risks of fluoride in drinking water by different exposure pathways, 2014 and 2018 (median (min, max)).
Full size table

Health risks of different groups of minors

The HI was specifically calculated for different groups of minors, utilizing the average values obtained from all areas, to assess the risk of fluoride exposure from drinking water. Risk probability plots illustrated the distribution of non-carcinogenic risks related to fluoride in drinking water, limited by HI = 1 (Fig. 2). In 2014, the non-carcinogenic risks of fluoride in drinking water for infants, children, and teens surpassed the recommended value by 39.2%, 25.5%, and 10.7%, respectively. Similarly, in 2018, these risks exceeded the recommended levels by 27.1%, 15.0%, and 5.23%, respectively. Furthermore, the non-carcinogenic risks for infants, children, and teens decreased in 2018 compared to 2014, showing statistically significant differences (χ2 = 13.604, P < 0.001; χ2 = 14.602, P < 0.001; and χ2 = 8.492, P = 0.004, respectively).

Fig. 2
figure 2

Probability distribution of non-carcinogenic risk of fluoride in drinking water for infants, children and teens, in 2014 and 2018. HI hazard index.

Full size image

Spatial and temporal distribution of health risks

The non-carcinogenic health risks of fluoride in drinking water for infants, children, and teens in regions other than the Southwest were lower in 2018 compared to 2014 (Fig. 3). Specifically, the non-carcinogenic health risks of fluoride in drinking water for infants in East China, Northeast China, and Northwest China were significantly lower in 2018 than in 2014 (χ2 = 4.194, P = 0.041; χ2 = 5.655, P = 0.017; χ2 = 7.010, P = 0.008, respectively). Conversely, there was no statistically significant increase in health risk in Southwest China (χ2 = 1.466, P = 0.226).

Fig. 3
figure 3

Non-carcinogenic risk of fluoride in drinking water for infants, children and teens, in different regions, in 2014 and 2018.

Full size image

For children, the non-carcinogenic health risks of fluoride in drinking water in North China, Northeast China, and Northwest China were significantly decreased in 2018 compared to 2014 (χ2 = 5.453, P = 0.020; χ2 = 4.412, P = 0.036; χ2 = 10.575, P = 0.001, respectively). However, there was no statistically significant increase in health risk in Southwest China (χ2 = 3.523, P = 0.061).

The non-carcinogenic health risks of fluoride in drinking water for teens were also significantly lower in East China and Northwest China in 2018 compared to 2014 (χ2 = 4.191, P = 0.041; χ2 = 5.809, P = 0.016, respectively). In Southwest China, there was no statistically significant increase in health risk (χ2 = 3.523, P = 0.061).

Effects of water improvement projects on fluoride in drinking water

The rate of water improvement in areas endemic to fluorosis reached 92.4% in 2018, representing a significant increase from 83.4% in 2014 (χ2 = 16.341, P < 0.001). Moreover, the fluoride concentration in drinking water in areas where water improvement had been implemented was notably lower than in non-improvement areas (Z = − 10.583, P < 0.001).

Effects of water improvement projects on health risks of fluoride exposure in drinking water

Water improvement projects implemented during the period from 2014 to 2018 were found to decrease the non-carcinogenic risks of fluoride exposure in minors, with odds ratios and 95% confidence intervals (OR and 95% CI) of 0.369 (0.268, 0.509) for infants, 0.556 (0.452, 0.683) for children, and 0.823 (0.740, 0.914) for teens (Fig. 4).

Fig. 4
figure 4

Effects of water improvement projects implemented during the period from 2014 to 2018 on the non-carcinogenic risk of fluoride in drinking water for infants, children, and teens.

Full size image

Health risk assessment based on the Chinese national standards for drinking water quality

China is a vast country with diverse geographical characteristics, and the factors related to health risk assessment—such as body size, diet, and lifestyle habits—exhibit significant variation across different regions (Table S3). The study used 1.0 mg/L30 as the fluoride exposure concentration for minors to evaluate the potential health risks associated with the current Chinese national standards for drinking water quality. In regions of North, East, South, and Southwest China, the HIs for non-carcinogenic health risks associated with fluoride in drinking water exceeded 1 for infants, with the national average also surpassing this threshold. However, the HIs for non-carcinogenic risks of fluoride in drinking water for children and teens were below 1 (Fig. 5 and Supplementary Table S4).

Fig. 5
figure 5

Non-carcinogenic health risks for minors at an exposure concentration of fluoride in drinking water of 1.0 mg/L.

Full size image

Discussion

Fluoride is naturally abundant and readily dissolves in groundwater, making drinking water a significant source of daily fluoride exposure18. Fluoride distribution in the natural environment is highly variable due to the geochemical behaviour of the element, which is further influenced by human activities34. An analysis of surveillance data from areas endemic to fluorosis revealed a considerable range in water fluoride concentrations. Following the implementation of water improvement projects, fluoride levels in drinking water have generally shown improvement. However, despite these efforts, fluoride concentrations in drinking water in certain areas remained above the standard. In particular, in Southwest China, the median concentration of fluoride in drinking water was observed to be higher in 2018 compared to 2014. The studies conducted by Xu35 and Deng36 in Southwest China have highlighted fluoride as a significant contaminant in local drinking water, with local soil conditions and industrial production identified as key influencing factors. Furthermore, Deng’s study demonstrated an increase in groundwater fluoride concentrations from 2005 to 2010–2012, with drinking water fluoride levels exceeding the national standard in some regions. These findings, along with the results of our study, indicate that water fluoride concentrations are unstable, emphasizing the importance of monitoring the fluoride levels in water to mitigate health risks. In 2018, all average concentrations of fluoride in drinking water met the limit value (≤ 1.0 mg/L); however, some specific areas exceeded this limit significantly. The implementation of water improvement projects lacks fully effective control of the fluoride concentrations in drinking water possibly due to factors such as aging infrastructure, inadequate maintenance, or suboptimal technology and site selection, which have led to elevated fluoride levels. The study suggests that additional investment and science-based water improvement projects are essential to protect public health.

The safety of drinking water is of paramount importance for human health, forming the foundation for the management of water resources and the establishment of standard limits. Previous research has concentrated on the evaluation of the health risks associated with fluoride in drinking water in specific regions of China18,20,21,22,23. Nevertheless, the paucity of research in this area has limited our understanding of the health risks associated with fluoride in drinking water in regions where fluorosis is prevalent. A further national study provided an overview based on a secondary analysis of existing literature19, but the results were influenced by the original research, leading to uncertainties. Furthermore, previous studies have not considered specific population exposure parameters. In consideration of the vast expanse of China and the considerable variations in physiological indicators and water usage behaviours across regions, the utilisation of generic exposure parameters for health risk assessment may potentially yield inaccurate conclusions. This study was conducted with the specific objective of evaluating the health risks associated with fluoride in drinking water for minors residing in regions with endemic fluorosis in China. The study employed exposure parameters for minors of varying ages across different regions, thereby ensuring that the results accurately reflected the situation in China. The findings indicated that fluoride in drinking water in regions with endemic fluorosis posed a non-carcinogenic risk to minors, with infants being more susceptible than children and teens due to their higher intake of the substance relative to their body weight37. Furthermore, the research demonstrated that the non-carcinogenic health risk associated with fluoride in drinking water was predominantly linked to oral ingestion rather than skin contact, which is consistent with previous studies19.

In the context of managing drinking water safety, the environmental health risk assessment method has been demonstrated to be a more sensitive and applicable approach than the traditional standard limit-setting approach15,38. Several countries have implemented restrictions on the concentration of fluoride in drinking water to mitigate potential health risks. By the Guidelines for Drinking Water Quality39, the World Health Organization has established an acceptable level of 1.5 mg/L of fluoride in drinking water. In the United States, a concentration limit of 0.7 mg/L40 has been established, whereas in China30, the limit is set at 1.0 mg/L. This study employed China’s current standard limit of 1.0 mg/L for fluoride in drinking water30 to assess the non-carcinogenic risk to minors. The study’s findings indicated that the existing standard limit may present potential hidden risks to infants in terms of water quality safety and health management. Similarly, health risk assessment studies on arsenic in drinking water have also suggested that the current standard limits for drinking water may not be the most effective basis for water safety and health management25,41. Consequently, further research is needed to establish appropriate standard limit concentrations for fluoride and other pollutants in drinking water.

This study had some limitations. The surveillance employed a simple random sampling method to select the points, which may have resulted in an insufficiently representative sample. Moreover, alterations in a few sampling locations for tap water over the two-year study period may have influenced the study’s stability. The assessments focus on domestic drinking water without including other sources like bottled or purified water. This may not provide a comprehensive view of the situation. The study lacked a classification and evaluation of the various methods of improving water quality. Further research is required to provide a more robust foundation for the implementation of water improvement measures.

Conclusions

The implementation of water improvement projects has significantly reduced fluoride concentrations in areas affected by endemic fluorosis, demonstrating the efficacy of these initiatives from 2014 to 2018. However, a resurgence of fluoride concentrations has been observed in some regions, particularly in southwest China, underscoring the necessity for ongoing vigilance. Consequently, it is essential to maintain continuous surveillance of drinking water quality and to enhance water improvement projects. Although the risk of non-carcinogenic effects from fluoride exposure in drinking water for minors has diminished due to controlled fluoride levels, this risk remains, particularly for infants aged 0–2 years, who are especially vulnerable to the adverse impacts of fluoride exposure. Despite existing limits on fluoride in drinking water, a non-carcinogenic risk persists for infants in this age group. Therefore, while progress has been made in controlling fluoride levels in drinking water in fluorosis areas, continued efforts are required to maintain and improve water quality. Additionally, reevaluating the fluoride limit in drinking water is crucial to mitigate the health risks associated with fluoride exposure.