Fluid intake of adults in four Chinese cities

Abstract

To investigate the fluid intake and knowledge on water and health in Chinese adults, and to provide a scientific basis for the development of adequate water intake values for people in China. A total of 1,483 adults aged 18–60 years were selected from four provincial capital cities using a multiple-stage random sampling method. The information on the amounts and types of daily fluid intake was collected using a 24-h fluid intake record for 7 consecutive days. Trained interviewers used a questionnaire to assess knowledge on water. The median of daily total fluid intake was 1,488 mL, with a wide range from 86 mL to 7,036 mL. Significant differences were found for fluid intake among the four cities. The medians of daily intakes of plain water, tea, and beverages were 786 mL, 109 mL, and 186 mL, respectively. Approximately 32% of the subjects drank less water than the amount recommended by the Chinese Nutrition Society (1,200 mL/day) and 71.6% of participants lacked knowledge of these recommendations. Moreover, 48.3% of them drank water only when they felt thirsty.

INTRODUCTION

Water is the major component of the human body and accounts for approximately 60–70% of body mass.^1,2 Water serves as the essential solvent for cellular biochemical reactions, and life will terminate without it. Nevertheless, overconsumption of water can impair physiological function and health. Therefore, an appropriate water intake that balances losses and thereby assures adequate hydration of body tissues plays a vital role in human health and life.^3,4 International organizations and some countries proposed the recommended dietary reference values for water based on the fluid intake studies.^3,5,6 For China, however, limited information on fluid intake is available. In order to determine the drinking water intake of the Chinese population and to provide scientific evidence for the development of adequate intake (AI) recommendations for water, a study on fluid intake was carried out among Chinese adults in and near the following four cities: Beijing, Shanghai, Chengdu, and Guangzhou, in 2010.

SUBJECTS AND METHODS

Sampling procedure

A multiple-stage random sampling method was used for subject selection. First, the cities of Beijing, Shanghai, Chengdu, and Guangzhou were selected. Second, one urban and one rural area were randomly selected from each city. Third, two streets/towns were randomly selected from each area. Fourth, two neighborhood committees/villages were randomly selected from each selected street/town. Finally, 12 subjects (6 men and 6 women) were randomly selected from each of four age groups (age ranges: 18–29 years, 30–39 years, 40–49 years, and 50–60 years). Participants were excluded if they hadbeen diagnosed in the past 5 years with liver, kidney, intestinal, heart, or endocrine disease that was not well controlled.

The Ethical Review Committee of the National Institute for Nutrition and Food Safety, Chinese Center for Disease Control and Prevention approved the study protocol. Written consent was obtained from all subjects.

Assessment of daily fluid intake

Fluid intake in our study consisted of the following three sources: 1) plain water, including tap water, mineral water, and purified water; 2) tea, including brewed green tea, semi-fermented tea, and fermented tea; and 3) beverages, including tea drinks, carbonated soft drinks, fruit and vegetable drinks, protein drinks (milk, yogurt, milk drinks), functional drinks, solid drinks, and coffee. The total amount of fluid intake was the sum of the amounts of these three sources.

A 24-h fluid record was used to collect the information on fluid intake for 7 consecutive days. Subjects recorded the type, place, and amount of fluid during eight occasions, including before breakfast, during breakfast, after breakfast, during lunch, after lunch, during dinner, after dinner, and during night.

On the first day, investigators introduced a standard container that was used to record the amounts of the water consumed during the following 7 days. The subjects then recorded daily amounts of water by themselves. To ensure record completeness, investigators interviewed each subject by telephone every 2 days.

In the Chinese Dietary Guidelines issued in 2007, the Chinese Nutrition Society recommended that Chinese adults who perform light physical activity at a moderate environmental temperature should drink at least 1,200 mL water a day,² which was used as a criterion in this study to classify whether the participants' daily water intake was adequate.

Assessment of knowledge of water and health

Participant knowledge of the effects of water on health was assessed using a questionnaire administered by trained interviewers.

Statistical analyses

The median and the quartile range were used to describe the average amount of daily fluid intake because the amount of fluid intake was found to have a positively skewed distribution. The Kruskal-Wallis rank-sum test was used to compare gender, city, and area differences in the amount of fluid intake. All statistical analyses were performed with SAS 9.1 software (SAS Institute Inc, Cary, NC). A two-tailed P value of <0.05 was regarded as statistically significant.

RESULTS

A total of 1,483 subjects (738 men and 745 women) completed the data collection. The sample sizes are shown in Table 1.

Table 1.  Urban, rural, and total sample sizes of participant groups in and around four cities in China.

Participants	Total			Beijing			Shanghai			Chengdu			Guangzhou
	Urban	Rural	Total	Urban	Rural	Total	Urban	Rural	Total	Urban	Rural	Total	Urban	Rural	Total
Total	747	736	1,483	187	171	359	191	189	380	183	190	373	186	185	371
Men	373	365	738	93	83	176	96	94	190	92	95	187	92	93	185
Women	374	371	745	94	89	183	95	95	190	91	95	186	94	92	186

The median of daily fluid intake was 1,488 mL, with a wide range from 86 mL to 7,036 mL. The daily fluid intake was significantly higher in men than women (1,679 mL versus 1,370 mL; Z = 8.34; P = 0.000). A significant difference in fluid intake was found among the four cities, with the highest in Shanghai (1,793 mL) and the lowest in Chengdu (1,150 mL). No significant difference was found between the urban and rural areas (1,514 mL versus 1,466 mL; Z = −0.81; P = 0.420). However, the differences between the two areas in Shanghai, Chengdu, and Guangzhou were significant; for example, the median daily water intake in the urban area of Shanghai was lower than in the rural area (Table 2).

Table 2.  Daily fluid intake (in mL) of participants in and around four cities in China.

Participant group	Beijing		Shanghai		Chengdu		Guangzhou		Total		X ^2 value	P value
	M	QR	M	QR	M	QR	M	QR	M	QR
Values are given in milliliters, unless otherwise indicated. *  Significant difference between urban and rural area (P < 0.05). ^†  Significant difference between men and women (P < 0.05). Abbreviations: M, median; QR, quartile range; B, Beijing; S, Shanghai; C, Chengdu, G, Guangzhou.
Total
Urban	1,579	757	1,748*	810	1,339*	897	1,326*	976	1,514	846	34.88	0.000	B = S > C = G
Rural	1,583	956	1,900	1,243	959	656	1,553	944	1,466	1,093	158.21	0.000	S > B = G > C
Total	1,579	829	1,793	1,005	1,150	772	1,467	988	1,488	950	154.31	0.000	S > B > G > C
Men
Urban	1,686	831	1,804	871	1,381	1,009	1,550	1,054	1,636	996	13.37	0.004	S > C = G
Rural	1,893	1,136	2,464	1,379	1,114	698	1,739	1,321	1,729	1,403	83.08	0.000	S > B = G > C
Total	1,744	1,016	1,994	1,171	1,257	847	1,671	1,100	1,679^†	1,167	82.38	0.000	S > B = G > C
Women
Urban	1,516	606	1,586	617	1,309	914	1,177	909	1,425	793	91.52	0.000	B = S = G > C
Rural	1,400	786	1,557	857	870	607	1,398	712	1,293	817	24.72	0.000	B = S > C = G
Total	1,474	694	1,563	676	1,043	727	1,316	810	1,370	812	80.58	0.000	S > B > G > C

The median of daily intakes of plain water, tea, and beverages of the subjects were 786 mL, 109 mL, and 186 mL, respectively. The amounts of tea intake were significantly higher among men compared with their female counterparts (229 mL versus 57 mL; Z = 7.52; P < 0.01), whereas no significant difference in the intakes of plain water and beverages was found between men and women. Among the four cities, subjects in Guangzhou had the highest intake of plain water (917 mL), followed by Beijing. Subjects in Shanghai had the highest tea and beverage consumption (257 mL and 323 mL, respectively). The intake of plain water, tea, and beverages in Chengdu was the lowest among these four cities. Subjects from urban areas had a lower consumption of plain water than their rural counterparts (693 mL versus 914 mL; Z = 6.88; P = 0.000), but significantly higher consumption of tea (186 mL versus 56 mL; Z = −4.56; P = 0.000) and beverages (286 mL versus 106 mL; Z = −13.00; P = 0.000) (Tables 3–5).

Table 3.  Daily plain water intake (in mL) of participants in and around four cities in China.

Participant group	Beijing		Shanghai		Chengdu		Guangzhou		Total		χ ^2 value	P value
	M	QR	M	QR	M	QR	M	QR	M	QR
Values are given in milliliters, unless otherwise indicated. *  Significant difference between urban and rural areas (P < 0.05). Abbreviations: M, median; QR, quartile range; B, Beijing; S, Shanghai; C, Chengdu, G, Guangzhou.
Total
Urban	709	647	500	571	664	585	836	700	693*	643	44.74	0.000	G > B = C > S
Rural	1,069	818	943	971	678	614	1,007	838	914	832	42.72	0.000	B = S = G > C
Total	814	789	693	842	671	579	917	771	786	743	53.31	0.000	G > B > S = C
Men
Urban	693	659	471	519	630	645	861	989	671	700	28.9	0.000	G > B = C > S
Rural	1,200	1,193	679	1,107	800	640	971	1,143	871	936	16.37	0.001	B = G > S = C
Total	775	946	561	807	714	643	900	1,093	757	857	37.65	0.000	B = G > S = C
Women
Urban	726	629	543	593	679	550	814	486	710	586	60.08	0.000	B = S = G > C
Rural	1,036	700	1,114	871	543	518	1,043	614	943	736	16.13	0.001	B = S = G > C
Total	814	657	843	807	611	532	934	600	800	650	41.01	0.000	B = S = G > C

Table 4.  Daily tea intake (in mL) of participants in and around four cities in China.

Participant group	Beijing		Shanghai		Chengdu		Guangzhou		Total		χ ^2 value	P value
	M	QR	M	QR	M	QR	M	QR	M	QR
Values are given in milliliters, unless otherwise indicated. *  Significant difference between urban and rural area (P < 0.05). ^†  Significant difference between men and women (P < 0.05). Abbreviations: M, median; QR, quartile range; B, Beijing; S, Shanghai; C, Chengdu, G, Guangzhou.
Total
Urban	229	664	300	886	186	686	68	429	186*	657	18.8	0.000	B = S = C > G
Rural	29	404	114	1,171	0	171	107	500	56	491	24.75	0.000	S = G > B = C
Total	114	571	257	957	71	436	86	457	109	593	17.33	0.001	S > B = C = G
Men
Urban	300	671	457	1,068	364	789	129	564	286	764	12.26	0.007	S = C > G
Rural	100	571	986	2,157	0	366	279	986	186	1,014	48.09	0.000	S > G > B = C
Total	146	657	596	1,489	86	607	186	686	229^†	864	37.07	0.000	S > B = C = G
Women
Urban	225	629	200	686	157	507	46	343	143	557	6.93	0.074
Rural	11	257	0	116	0	143	57	161	0	164	12.1	0.007	G > B = S
Total	107	514	0	493	42	257	54	264	57	350	6.5	0.090

Table 5.  Daily beverage intake (in mL) of participants in and around four cities in China.

Participant group	Beijing		Shanghai		Chengdu		Guangzhou		Total		χ ^2 value	P value
	M	QR	M	QR	M	QR	M	QR	M	QR
Values are given in milliliters, unless otherwise indicated. *  Significant difference between urban and rural areas (P < 0.05). Abbreviations: M, median; QR, quartile range; B, Beijing; S, Shanghai; C, Chengdu, G, Guangzhou.
Total
Urban	363	371	500	441	221	235	118	214	286*	371	209.52	0.000	S > B > C > G
Rural	146	271	143	343	43	143	121	257	106	262	39.56	0.000	B = S = G > C
Total	264	369	323	514	107	262	121	236	186	348	137.08	0.000	B = S > C = G
Men
Urban	400	443	514	345	214	247	94	282	300	443	98.92	0.000	B = S > C > G
Rural	129	343	111	287	36	89	114	264	86	236	20.45	0.000	B = S = G > C
Total	290	500	319	543	86	250	110	264	161	365	70.69	0.000	B = S > C = G
Women
Urban	321	317	493	477	236	187	143	193	271	329	18.2	0.000	B = S > C > G
Rural	169	236	143	357	63	179	146	282	129	264	112.63	0.000	B > G > S > C
Total	249	303	332	436	143	257	143	229	193	315	64.68	0.000	B = S > C = G

Approximately 32.4% of the subjects drank < 1,200 mL of water per day, with a higher rate among women (χ² = 24.80; P < 0.01). No significant difference in the proportion was found between rural and urban areas (χ² = 3.07; P = 0.080). However, the proportion in the rural area of Chengdu was higher than its counterparts (χ² = 28.32; P < 0.01), and the proportion in the rural area of Guangzhou was lower than its counterparts (χ² = 9.9124; P < 0.01) (Table 6).

Table 6.  Proportion of participants with a daily fluid intake below 1,200 mL.

Participant group	No. (%) of participants					χ ^2 value	P value
	Beijing	Shanghai	Chengdu	Guangzhou	Total
*  Significant difference between men and women (P < 0.05). ^†  Significant difference between urban and rural area (P < 0.05).
Total	93 (25.9)	63 (16.6)	199 (53.4)	125 (33.7)	480 (32.4)	125.44	0.000
Gender
Men	38 (21.6)	20 (10.5)	88 (47.1)	48 (26.0)	194 (26.3)	0.000	0.000
Women	55 (30.1)	43 (22.6)*	111 (59.7)*	77 (41.4)*	286 (38.4)*	0.000	0.000
Region
Urban	48 (25.7)	29 (15.2)	72 (39.3)	77 (41.4)	226 (30.3)	40.54	0.000
Rural	45 (26.2)	34 (18.0)	127 (66.8)^†	48 (26.0)^†	254 (34.5)	122.01	0.000

The proportion of subjects who were unaware of the recommended daily water intake proposed by the Chinese Nutrition Society was 28.4%, with higher values in women (32.4% versus 24.4%; χ² = 11.55; P < 0.01) and the highest value in Chengdu (41.8%). The proportion of subjects unaware of the deleterious effects of dehydration was 14.4%, and was higher in Chengdu (18.5%) and Guangzhou (18.3%) and higher in rural subjects than their urban counterparts (18.6% versus 10.2%; χ² = 21.47; P < 0.01). Moreover, nearly half (48.3%) of all subjects drank water when they felt thirsty. The proportion in Chengdu was the highest (59%). Approximately 20% of subjects did not know that plain water is the best source for water.⁴ In contrast, 17.9% of them considered beverages other than water as the best source.

DISCUSSION

On the basis of fluid intake survey data, Europe and the United States have developed adequate intake (AI) values for total water intake and fluid intake.⁵ A 24-h fluid intake recall for 1 day, combined with dietary recall methods, was used in the United States to estimate the daily drinking water intake and water from food,⁶ whereas a 7-day dietary record was used in France and in the United Kingdom.^7,8 The method of assessing fluid intake in the dietary survey requires the subjects to record or recall a large amount of data, which will reduce compliance and may underestimate the fluid intake.⁵ The method used in the Drinking Water Consumption Survey⁹ in the United States collects information on how often, when, and how much water is consumed at specific time periods during the day. This approach was specifically designed to assess fluid intake by reducing burdens on the subjects, improving accuracy, and reducing omissions. However, this method can only obtain information about daily fluid intake, not the types of drinking water. The protocol in our study was designed for adults to specifically have the advantages as the Drinking Water Consumption Survey approach. Moreover, information on the types of drinking water was also obtained.

Britain, France, and Mexico have recommended a daily fluid intake of 1,500–2,000 mL, based on the national data of actual drinking water intake among individuals.^4,5 The United States and Canada proposed AIs for fluid of 3,700 mL for men and 2,700 mL for women aged ≥19 years based on the results from the Third National Health and Nutrition Examination Survey.⁶ The World Health Organization proposed a daily water intake recommendation of 1,500 mL, taking into account the differences in national groups.⁵ The Chinese Nutrition Society proposed a daily intake of 1,200 mL of water in the 2007 Chinese Dietary Guideline, based on the relationship between energy metabolism and water requirements.² In the current study, the median daily fluid intake of subjects from the four cities studied was 1,488 mL. Both the recommended water intake and the actual daily fluid intake in China are lower than in other countries. This difference may be due to the different eating behaviors and dietary patterns. Plant foods are the main foods in China, and steaming, stewing, and stir-frying are the most commonly used cooking methods, which can retain most of the water in the food. Moreover, some foods contain water that is added during cooking. Therefore, a larger proportion of the total water intake comes from solid food, which may reduce the amount of fluid consumed. In addition, there are ethnic differences in body size and resting metabolic rate that will result in different water requirements.

CONCLUSION

The average intake of fluid in Chinese adults, as measured in and around four cities, is higher than the present recommended dietary intake of water for the Chinese population. However, approximately one-third of the participants in the present study did not meet the recommended level of intake. The current study provides useful information for developing an AI for water for the Chinese population.

Acknowledgments

The authors acknowledge support from the Xicheng District Center for Disease Control and Prevention, Beijing; Huairou District Center for Disease Control and Prevention, Beijing; Shanghai Municipal Center for Disease Control and Prevention; Chengdu Center for Disease Control and Prevention; and the Center for Disease Control and Prevention of Guangdong Province. The authors also acknowledge support from all of the team members and the participants. This study was funded by the Danone Waters Research and Development Center in China.

Declaration of interest.  There authors have no relevant interests to declare.