Drinking water intake of adults in typical cities in the major river basins of China

Water is necessary to sustain all forms of life, and humans can only live a few days without it (WHO 2022). Drinking water can prevent dehydration, which might lead to unclear thinking, mood alteration, overheating, constipation, and kidney stones (Popkin et al. 2010; Karger 2018). For sedentary adult men and women under average circumstances, 2,900 and 2,200 mL of water per day are regarded as necessary for hydration; and for physically active men, women, and children in high temperatures, 4,500 mL of water per day is required (WHO 2005). EFSA suggested that the adequate total water intake for females should be 2.0 L/day and for males 2.5 L/day (EFSA 2010). In the United States, the recommended daily intake of total water for females is 2.7 L and for males it is 3.7 L (Patel et al. 2020).

Water ingestion is a pathway of exposure to environmental chemicals (ATSDR 2023). People may be exposed to contaminants in water when consuming water directly as plain water, or indirectly from foods and drinks made with water (Babuji et al. 2023). Estimating the magnitude of the potential dose of toxins from water ingestion requires information on the quantity of water consumed. In addition, when setting up the drinking water quality standard, it is also necessary to deduce the standard limit according to the relevant drinking water intake parameters.

For risk assessment related to drinking water contamination, a default water ingestion value of 2 L/day per person is suggested by the United States, the World Health Organization, and other countries or agencies (USEPA 2018; WHO 2022). However, several factors may affect the amount of water consumed, such as temperature (seasonal and/or regional effects), the aesthetic quality of drinking water, cultural differences, as well as age, gender, physical activity, and diet (Roche et al. 2012; Malisova et al. 2013; Armstrong & Johnson 2018). Water intake can also be affected by dietary habits and cooking methods. Asian people consume rice and noodles as staple foods, which contain a greater proportion of water (60% of weight) compared with bread (30–40%) (Tani et al. 2015). In China, two large-scale studies have been conducted, focusing on the total fluid intake and types of fluids consumed in urban China by age, gender, region, and city socioeconomic status, while fluid intake from foods was not assessed (Ma et al. 2012; Zhang et al. 2018).

In this study, we conducted a survey to collect information on the drinking water consumption of residents in major cities of China's major river basins. The water intake level of Chinese adults was evaluated, and direct plain water, indirect plain water, commercial beverages, total plain water, and total water intake were provided, and the effects of gender, season, region, age, and body mass index on water intake were explored. A recommended daily water ingestion rate for use in exposure assessments was also provided.

The participants are 18-year-olds and above adult residents recruited from the study areas until the quotas for age and gender in relation to the total country population are met.

Before the formal questionnaire survey was carried out, we chose Chaohu City to conduct a preliminary survey. In accordance with the results of the survey, we evaluated the selection of residents, the content of questionnaires, and survey methods, to enhance the validity and reliability of the survey; then, optimized and rectified any deficiencies. The formal survey is conducted in summer and winter, respectively. The survey method adopts the combination of a direct face-to-face questionnaire survey and a self-filling log. Each survey lasts for 7 days. On the first day, the investigator entered the house and conducted a face-to-face questionnaire survey with the residents. The investigator filled in the questionnaire according to the residents' answers to the questionnaire and collected the questionnaire. At the same time, the investigator told the residents how to complete a self-filling log. From the second day to the seventh day, the respondents filled the self-filling log themselves, and all self-filling logs were collected and stored by the investigator. To guarantee the quality of the self-filling log, the investigators conducted telephone interviews with residents from the second day to the seventh day and provided answers to various questions when they were filling in the questionnaires.

The survey information on drinking water intake mainly includes drinking habits, frequency, and intake of direct plain water, indirect plain water, and commercial beverages. Direct plain water refers to plain water consumed directly as cold water (e.g. tap water, spring water, purified water, groundwater, and mineral water) and boiled water such as barley tea. Indirect plain water was intended to include the tap water added for preparation in foods such as soup, porridge, rice, steamed bread, and noodles. Commercial beverages include many kinds of drinks which can be purchased in the market, such as bottled tea or coffee, sports drinks, milk products, juices, carbonated soft drinks, and alcoholic beverages. In China, the majority of the coffee that people consume is purchased from merchants, and a few people prepare it themselves. Hence, coffee is categorized as a kind of commercial beverage. In addition to water and tea, bottled tea also contains white granulated sugar, citric acid, sodium citrate, D-sodium erythorbate, vitamin C, honey, and other ingredients. As a result, it is also considered a commercial beverage.

The intake of all types of water was averaged based on 7-day survey results. Total plain water intake is the amount of indirect plain and direct plain water. Total water intake is the amount of indirect plain water, direct plain water, and commercial beverages. For calculating the amount of indirect plain water intake for each individual per day through diet food consumption, representative water contents of the soup, porridge, rice, steamed bread, and noodles were summed. The water content of each dish was computed by multiplying the volume of daily food intake with the water content of the food. The proportion of water content used to calculate is 95, 85,70.1, 40.3, and 72.3% for the soup, porridge, rice, steamed bread, and noodles, respectively (Yang 2019).

All calculations and statistical procedures were performed using SPSS software version 18.0 for Windows (IBM SPSS Statistics, Chicago, IL, USA). Wilcoxon rank test was used to compare the means of water intake rate by gender, season, age groups, region, and BMI. Categorical regression analysis was performed to evaluate the impact of various influence factors on the water intake rate. All statistical tests were two-tailed and the significance level was set at p < 0.05.

The demographic characteristics of the participants are presented in Table 1. A total of 13,106 participants aged from 18 to ≥65 were included, with 6,741 in summer and 6,365 in winter. The proportions of different age groups (18–21, 25–34, 35–44, 45–54, 55–64, and ≥65) were 12, 19, 18, 17, 17, and 17%, respectively. The male-to-female ratio was 1.00:1.08.

Some socio-demographic factors, particularly income, affect the priority of needs, higher income earners are more likely to pay to improve drinking water quality and use bottled/barreled water and water purifiers (Huang et al. 2015; Van Houtven et al. 2017). A study in Iran found that about 1.04 billion liters on average of bottled water was consumed annually between 2000 and 2015. Bottled water consumption increased from 0.41 to 48.9 L/capita-year (Aslani et al. 2021). In a survey of drinking water consumption patterns among private well users in Ontario, 45.5% of survey respondents selected bottled water as their primary household drinking water supply (Lavallee et al. 2021). The use of bottled/barreled water and filtered water from household water purifiers also reflects the rapid economic development in some parts of China and the relatively high income of residents. The low utilization rate of well water and filtered water of water purifiers in the community reflected the high popularity rate of using tap water in China, and the quality of drinking water supplied to residents is fully guaranteed.

As shown in Table 2, the average daily total plain water intake was 1,666 ml/day, which is higher than the reported values of 1,493 ml/day for Koreans (Ji et al. 2010), 1,466 ml/day for Americans (Kahn & Stralka 2009), and 1,380 ml/day for Canadians (Roche et al. 2012). The total water intake was 1,805 ml/day, which is higher than the total fluid intake reported by Zhang et al. (2018) was 1,387 ml/day, and exceeds the adequate intakes set by the Chinese Nutrition Society. But, it is lower than the recommended value of total water intake for the USA (Patel et al. 2020), EFSA (2010), and WHO (2005). Direct plain water was the main contributor to the total water intake (approximately 67% of the total water). It was also the main contributor to the total plain water, accounting for approximately 73%. The intake of indirect plain water was 451 ml/day, which is close to the value of 477 ml/day for Koreans (Ji et al. 2010) and is lower than the intake of moisture in foods (664 g/day) of the adult US population (Yang & Chun 2014) and much lower than the findings in Japan with the value of 1,130 ml/day (Tani et al. 2015). Compared with people in the United States, Korea, and China, it is considered that people consume more water from food moisture due to the inclusion of boiled white rice, vegetables, and soup in Japanese meals. Commercial beverages contributed the least to the total water intake. Commercial beverages accounted for 7.7% of total water intake. This value was much lower than that contribution (including coffee, sweetened beverages, tea, fruit beverages, milk, and diet beverages) in the USA which accounted for 48.8% of total nonalcoholic beverage consumption (Martin et al. 2020). These commercial beverages on the market may constitute an important proportion of the daily water intake; however, they would not reflect the local water contamination, and hence would not be appropriate to be considered in the exposure assessment. The value of 1,666 ml/day is the average obtained through extensive surveys, which can represent the daily water intake of most Chinese adults. This value is proposed as the recommended value of the daily total plain water intake rate for use in exposure assessments in China.

The water consumption rates by gender are shown in Table 2. The total plain water intake and total water intake were 1,777 and 1,942 ml/day for males, and 1,564 and 1,678 ml/day for females, respectively. The water consumption (including total plain water, direct plain water, indirect plain water, commercial beverages, and total water) was significantly higher in male adults than female adults (p < 0.001). This observation is similar to that of Ershow & Cantor (1989), Ji et al. (2010), Tani et al. (2015), and Zhang et al. (2018) who reported higher water consumption in males. This may be attributed to men's higher weight, body water content, and metabolism, men have greater average sweat rates than women, and women generally have smaller body sizes and lower metabolic rates when performing a given task than men (Buono & Sjoholm 1988).

The water consumption rates in summer and winter are also shown in Table 2. The total plain water intake and total water intake were 1,794 and 1,944 ml/day in summer, and 1,531 and 1,657 ml/day in winter, respectively. The water consumption (including total plain water, direct plain water, indirect plain water, and total water intake) was significantly higher in summer than in winter. There was no significant difference in commercial beverage intake in summer and winter (p = 0.068). We found greater water consumption in summer than in winter. High temperature and humidity also might have exacerbated dehydration (Westerterp et al. 2005; Malisova et al. 2013). There are some relevant studies on the effect of hot weather on the water consumption rate. Tani et al. (2015) found that the mean total water intake was 2,230 ml/day (high in summer: 2,331 ml/day; low in winter: 2,134 ml/day). Ji et al. (2010) also reported higher tap water consumption during the summer. Regnier et al. (2015) found when the outdoor air temperature exceeded 90°F, water consumption increased by 28%. However, in England and Wales (DWI 2008), people do not decrease total tap water consumption in the winter. The reason is that tea and coffee accounted for 32 and 18% of all tap water consumption. Nearly 80% of the sample say they drink tea or coffee the same all year, and 16% tend to drink more tea or coffee in winter, while 4 and 1% said they drink more tea and coffee in summer.

The daily water intake of different age groups is also shown in Table 2. The total plain water intake ranged from 1,541 ml/day for the 18–24 age group adults to 1,764 ml/day for the 35–44 age group adults, and total water intake ranged from 1,718ml/day for the >65 age group adults to 1,902 ml/day for 35–44 age group adults. The water consumption (including total plain water, direct plain water, indirect plain water, and total water) was significantly higher in the 35–64 age groups than 18–34 and ≥65 age groups. However, commercial beverage consumption was the highest in the 18–24 age groups, and there is a significant downward trend with the increasing age. This is similar to Ershow & Cantor's (1989) observation of an increase in water consumption among subjects 45–64 years and a gradual decrease in older populations. The total tap water consumption rate of Koreans was greatest among the age group of 40–49 years, and the consumption rate decreased in age groups of >50 years (Ji et al. 2010). The reason is many older adults deliberately avoid drinking water because they fear nighttime incontinence (Asplund & Aberg 1991). Older adults may not recognize that they are thirsty as the sensation of thirst decreases with age (Kenney & Chiu 2001).

The water consumption among different regions is summarized in Table 3. The cities in the major river basin in the central and southern regions of China had the highest intake of total plain water and total water. The cities in the major river basin in the northeast and west regions of China had the lowest intake of total plain water and total water. The direct plain water intake rate was the lowest, but the consumption of indirect plain water and commercial beverages was the highest in Beihai city. The region's effect on water consumption could be found in some other studies (Ershow et al. 1991; Zhang et al. 2018). The different water consumption in different regions may be caused by differences in climate conditions and differences in lifestyle (Ershow et al. 1991). People living in different places take nutrients from different kinds of food. The highest intake of indirect plain water and commercial beverages in Beihai city could be explained by the fact that local people are used to drinking soup and drink commercial beverages such as tea with milk.

BMI was assessed as an index of obesity. BMI = Weight (kg)/(Height (m))². The normal range is 18.5–22.9, slightly overweight is 23.0–24.9, overweight is 25.0–29.9, and obese is ≥30.0 (according to the standard of Asian BMI). Daily water consumption variation by different BMI is summarized in Table 4. The intake of total plain water, direct plain water, and total water significantly increased with BMI (p < 0.001). There was no significant association between the intake of beverages and indirect plain water and BMI (p = 0.675 and p = 0.218). The intake of commercial beverages was the highest in the group with a BMI ≥ 30, followed by a BMI < 18.5 groups. The intake of indirect plain water was the highest in the group of 25 ≤ BMI < 30. The association between water consumption and body weight outcomes is generally well-documented elsewhere. Fulgoni (2007) found that obese adults consumed more plain water than normal-weight adults. Kant et al. (2009) reported that plain water consumption was higher in adults of higher BMI groups. In this study, no significant correlation was found between BMI and the intake of commercial beverages and indirect plain water. This further illustrated that the relation between commercial beverage and indirect plain water consumption and BMI was not strong in the study population (Forshee et al. 2005).

To reveal what season, gender, age, region, and BMI were associated with water consumption, we performed categorical regression analysis using these influence factors as independent factors (Table 5). Except BMI had no significant correlation with the intake of indirect plain water and commercial beverages, other factors were found to be associated with the intake of total plain water, direct plain water, indirect plain water, commercial beverages, and total water. BMI had a significant correlation with the intake of total plain water, direct plain water, and total water. From Table 6, we could see the region was the most important factor influencing the intake of total plain water, direct plain water, and total water, followed by season and gender. The intake of indirect plain water was mainly related to the region, and other factors had little influence on the indirect plain water intake. Age was the most important factor of the impact on the intake of commercial beverages, followed by region and gender.

The safety and quality of drinking water are of crucial significance to the health of both community and individuals. Data on the intake of plain water are utilized for multiple purposes. The application of appropriate water intake rates is important in assessing the risk of waterborne illness caused by chemical and microbiological hazards or in determining the limit value of a drinking water quality indicator. This study explored the drinking water intake of adults in typical cities within major river basins of China. The daily volume of water consumed daily was associated with age, gender, season, region, and BMI. The mean volume of total plain water consumed was 1,666 ml/day. The water intake rates reported in this study could be applied in China and other countries with similar geographical locations, temperatures, climates, and living customs.

This study was supported by grants from the Major Science and Technology Project of Water Pollution Control and Management in China (No. 2018ZX07502001). The authors are grateful to all participants and local staff of study areas for their participation in this study.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.