Factors Associated with Low Drinking Water Intake among Adolescents

RESEARCH Research and Professional Briefs

Factors Associated with Low Drinking Water Intake among Adolescents: The Florida Youth Physical Activity and Nutrition Survey, 2007 SOHYUN PARK, PhD, MS; BETTYLOU SHERRY, PhD, RD; TERRENCE O’TOOLE, PhD; YOUJIE HUANG, MD, DrPH

ABSTRACT There is limited information on which characteristics are associated with water intake among adolescents. This cross-sectional study examined the association between demographic, dietary, and behavioral factors and low water intake as the outcome measure. Analyses were based on the 2007 Florida Youth Physical Activity and Nutrition Survey using a representative sample of 4,292 students in grades six through eight in 86 Florida public middle schools. Multivariable logistic regression was used to calculate adjusted odds ratios (ORs) and 95% confidence intervals for factors associated with low water intake (⬍3 glasses water per day). About 64% of students had low water intake. Factors significantly associated with low water intake were Hispanic ethnicity and nonHispanic other (vs non-Hispanic white; ORs 0.79 and 0.76, respectively), drinking no 100% juice, drinking it ⬍1 time/day, and drinking it 1 to 2 times/day (vs drinking it ⱖ3 times/day; ORs 1.83, 1.91, and 1.32, respectively), drinking no milk and drinking ⬍2 glasses of milk/day (vs drinking ⱖ2 glasses/day; ORs 1.42 and 1.41, respectively), drinking ⬍1 soda/day (vs drinking none; OR 1.40), drinking fruit-flavored drinks/sports drinks ⬍1 time/day and drinking it ⱖ1 time/day (vs drinking none; ORs 1.49 and 1.41, respectively), eating at a fast-food restaurant ⱖ3 days/week (vs none; OR 1.38, respectively), not participating on team sports or participating on 1 to 2 team

S. Park is an epidemiologist and B. Sherry is a lead epidemiologist, Division of Nutrition, Physical Activity, and Obesity, and T. O’Toole is a public health analyst, Division of Nutrition, Physical Activity, and Obesity, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, GA. Y. Huang is the state chronic disease epidemiologist, Bureau of Epidemiology, Florida Department of Health, Tallahassee. Address correspondence to: Sohyun Park, PhD, MS, Division of Nutrition, Physical Activity, and Obesity, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, 4770 Buford Hwy, NE, Atlanta, GA 30341. E-mail: [email protected] Manuscript accepted: February 7, 2011. Published by Elsevier Inc. on behalf of the American Dietetic Association. 0002-8223/$0.00 doi: 10.1016/j.jada.2011.05.006

sports in previous 12 months (vs participating on ⱖ3 teams; ORs 1.77 and 1.24, respectively), and consuming snack/soda while watching television/movies “sometimes” and “most/every time” (vs never; ORs 1.65 and 2.20, respectively). The strongest factor associated with low water intake was frequent consumption of snacks/sodas while watching television/movies. Although study findings should be corroborated in other states and in a nationally representative sample, they may be useful in targeting adolescents for increased water consumption. J Am Diet Assoc. 2011;111:1211-1217.

T

he prevalence of childhood obesity has increased substantially in the United States (1,2) and is a major public health concern because of morbidities associated with it (3-5). High consumption of sugar drinks has been associated with obesity among youth (6-10). Sugar drinks are the largest source of added sugar (11) and account for 7.1% of total energy in the average US diet (12). Given that the average energy intake among US youth has increased during the past few decades, particularly from sugar drinks (13,14), the substitution of drinking water for sugar drinks might help them to maintain a healthier weight (10,15). Adequate water consumption is important in preventing dehydration, which has been associated with adverse health outcomes, including certain cancers (16,17), coronary heart disease (16), kidney stones (16), and severe bronchopulmonary disorders (18). Although total water intake requirements can be met by consuming beverages and foods, drinking water is a zero-energy, thirst-quenching option (16) that obviously will not promote obesity. The adequate intake (AI) level for water in any form (foods, beverages, and drinking water) was established by the Institute of Medicine as a guide for preventing adverse effects of dehydration rather than for decreasing risks for chronic diseases (16). Specific AI levels vary by sex and age. The AI levels for total water are 2.4 L/day, including 1.8 L (⬇8 c) in the form of water or other beverages for boys 9 to 13 years of age and 3.3 L/day, including 2.6 L (⬇11 c) as beverages for boys 14 to 18 years of age. The AI levels for total water are 2.1 L/day, including 1.6 L (⬇7 c) in the form of water or other beverages for girls 9 to 13 years of age and 2.3 L/day, including 1.8 L (⬇8 c) as beverages for girls 14 to 18 years of age. People who are physically active or living in hot climates require even more water (16). However, based on the 2005-2006 National Health and Nutrition Examination Survey (NHANES), the average total water intake

Journal of the AMERICAN DIETETIC ASSOCIATION

1211

among US youth (aged 12 to 19 years) was 2.4 L/day (1.6 L/day from food and beverages other than drinking water and 0.8 L/day in plain water) (19). Although many studies have examined the association of sugar drink consumption with various demographic and behavioral factors (14,20,21), few have examined the association between drinking water consumption and these factors. Some studies have shown possible associations between water intake and factors such as age, race/ ethnicity, education level, and physical activity among young children (22-24) or adults (25-27). However, there is limited information on the extent to which demographic and behavioral factors are associated with water intake among adolescents (19,26). Both societal norms/ culture/behaviors and individual knowledge/attitudes influence behaviors associated with weight status, so it is possible that low intake of drinking water may be associated with other less-healthful behaviors, such as high consumption of sugar drinks, excessive screen time, and low levels of physical activity. This is a potential link between low water consumption and obesity. The purpose of this cross-sectional study was to assess associations between low drinking water intake and demographic, dietary, and behavioral factors among Florida adolescents. METHODS This cross-sectional study was based on the Florida Youth Physical Activity and Nutrition Survey (YPANS), conducted in spring 2007. YPANS, developed by the Florida Department of Health, was a statewide, self-reported, school-based survey of public middle school students. YPANS monitored attitudes, health behaviors, and knowledge of physical activity and nutrition among students (28). YPANS used questions from the Centers for Disease Control and Prevention Youth Risk Behavior Surveillance System, which had previously been validated (29,30). The eligible population for the survey included all Florida regular public middle school students in grades six through eight. The population was stratified by region and sampled systematically proportional to enrollment in grades six through eight. Within schools, either all classes meeting during a particular period of day or all classes in a required subject were included in the sampling frame. Systematic equal probability sampling was used to select classes from each school that participated in the survey. All students in selected classes were included in the sample. Every fourth student in selected classes was asked to complete the survey. Consent forms from parents (passive consent in most cases) and assent from students were obtained before the survey was administered. Because these analyses involved existing deidentified data, this study was exempt from the Centers for Disease Control and Prevention Institutional Review Board process. A total of 4,669 students in grades six through eight from 86 Florida middle schools completed the 2007 survey. However, 377 respondents were excluded because of missing data for the outcome variable. The final analytic sample was 4,292 respondents. The estimated overall response rate for the Florida YPANS was 70%, based on

1212

August 2011 Volume 111 Number 8

the number of schools that participated and the number of eligible students. Outcome Variable The outcome of interest was water intake. Students were asked, “During the past 7 days, how many glasses or bottles of water did you drink?” For ␹2 tests, the outcome variable was categorized into drinking ⬍1 glass water per day, 1 to 2 glasses water per day, or ⱖ3 glasses water per day. For the logistic regression analyses, the outcome variable, low water intake, was defined as drinking ⬍3 glasses water per day, based on the findings from a previous study done in youth (19) and the data distribution from our study. Exposure Variables Student characteristics were examined. Demographic variables were age (12 years or younger, 13 years, and 14 years or older), sex, race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, and other). Other characteristics were self-reported academic grades (A, B, C, and D/F) and intention to control weight (trying to lose weight, trying to gain weight, trying to stay the same weight, and not trying to do anything about weight). Dietary variables (reported consumption during the previous 7 days) were frequency of 100% fruit juice consumed per day (none, ⬍1, 1 to 2, and ⱖ3), glasses milk consumed per day (none, ⬍2, and ⱖ2), sodas consumed per day (none, ⬍1, and ⱖ1), frequency of fruit-flavored drinks or sports drinks consumed per day (none, ⬍1, and ⱖ1), number of breakfasts eaten (none, 1 to 6, and 7), and days in which fast food was eaten (none, 1 to 2, and ⱖ3). Behavioral characteristics were team sports participation during the past year (none, 1 to 2 teams, ⱖ3 teams), individual sports participation during the past year (none, 1 to 2 sports, and ⱖ3 sports), time spent watching television on an average school day (none, ⱕ2 hours, and ⱖ3 hours), time spent watching television on an average weekend day (none, ⱕ2 hours, 3 to 4 hours, and ⬎4 hours), and frequency of snack or soda consumption while watching television or movies (never, sometimes, and most/every time). Unknown values or missing data regarding exposure variables ranged from 0.3% to 16% and were excluded from analysis when the variable was used. Statistical Analysis Data were weighted by sex, grade, and region to represent all Florida public middle school students in grades six through eight. Descriptive statistics were expressed as proportions. ␹2 tests were used to examine differences across categories. Odds ratios (ORs) and 95% confidence intervals for low drinking water intake were calculated using multivariable logistic regression. Because of significant correlation between team sport participation and individual sport participation, we used only team sport participation in the logistic regression model. Of note, there were no confounding effects of the various beverages included in the study on low water intake. Age, sex, plus all variables that were significantly associated with water intake based on ␹2 tests were included in the initial logistic regression model development. However, vari-

Table 1. Drinking water intake among public middle school students by students’ characteristics, Florida Youth Physical Activity and Nutrition Survey, 2007a Drinking Water Intake During the Previous 7 d Student characteristic

All respondents

Total Age (y) ⱕ12 13 ⱖ14 Sex Boy Girl Race/ethnicity White, non-Hispanic Black, non-Hispanic Hispanic or Latino Other, non-Hispanic Self-reported academic grades Mostly A’s Mostly B’s Mostly C’s Mostly D’s and F’s Intention to control weight Trying to lose weight Trying to gain weight Trying to stay the same weight Not trying to do anything about weight

n 4,292 4,279 1,584 1,422 1,273 4,274 2,061 2,213 4,114 1,737 840 1,097 440 3,800 1,266 1,449 840 245 3,882 1,790 541 954 597

3 glasses/d

P valueb

4™™™™™™™™™™™™™™™™™™ %⫾standard error ™™™™™™™™™™™™™™™™™3 100 38.4⫾0.9 25.2⫾0.7 36.3⫾0.8 0.75 35.7⫾1.6 33.7⫾0.9 30.5⫾1.2

38.1⫾1.3 39.3⫾1.4 37.7⫾1.6

25.2⫾1.1 24.2⫾1.2 26.5⫾1.4

36.7⫾1.2 36.6⫾1.5 35.8⫾1.3

51.1⫾0.9 48.9⫾0.9

38.5⫾1.4 38.3⫾1.1

24.0⫾1.0 26.6⫾1.1

37.5⫾1.2 35.1⫾1.0

45.3⫾2.8 20.0⫾2.1 23.5⫾2.2 11.1⫾0.6

37.5⫾1.2 45.2⫾1.8 37.8⫾1.6 32.6⫾2.6

27.7⫾1.2 21.8⫾1.7 22.9⫾1.1 26.1⫾2.3

34.8⫾1.0 33.0⫾1.6 39.3⫾1.5 41.3⫾2.3

33.5⫾1.5 38.1⫾1.0 22.0⫾1.1 6.5⫾0.6

32.9⫾1.6 37.8⫾1.5 43.5⫾1.5 46.4⫾3.9

29.5⫾1.3 24.7⫾1.2 21.9⫾1.5 19.4⫾2.6

37.6⫾1.5 37.5⫾1.4 34.6⫾1.6 34.1⫾4.1

45.4⫾0.8 14.3⫾0.6 24.5⫾0.6 15.9⫾0.6

37.2⫾1.4 42.3⫾2.3 37.9⫾1.7 36.4⫾2.0

23.8⫾1.1 25.1⫾2.0 27.3⫾1.6 26.9⫾1.9

39.0⫾1.3 32.6⫾2.4 34.9⫾1.5 36.7⫾1.9

0.18 ⬍0.0001

⬍0.0001

0.08

a

Because of rounding, weighted percentages may not add up to 100%. ␹ tests were used for examining differences across categories.

b 2

ables that were not significantly associated with low water intake were deleted from the final model. For the modeling analyses, our sample size was reduced to ensure that only individuals with complete data on all of the variables examined were included. Our analytic sample for the modeling was 2,790 students. All statistical tests were two-sided and assumed significance at P⬍0.05. All statistical analyses were done using Statistical Analysis Software (version 9.2, 2009, SAS Institute Inc, Cary, NC) and accounted for the two-stage cluster sample (students selected within classrooms within schools) and selection probability of the survey design. RESULTS AND DISCUSSION About 64% of respondents reported drinking ⬍3 glasses water per day during the previous 7 days. Based on ␹2 tests, drinking water intake varied significantly by race/ ethnicity and self-reported academic grades (Table 1) as well as dietary (juice, milk, soda, fruit-flavored drinks/ sports drinks, breakfast, and fast-food consumption) and behavioral factors (sports participation, television viewing, and consumption of a snack/soda while watching television/movies) (Table 2). Results of multivariable logistic regression showed the following factors to be significantly associated with low drinking water intake: Hispanic ethnicity and non-Hispanic other (vs non-Hispanic white; ORs 0.79 and 0.76,

respectively), drinking no 100% juice, drinking it ⬍1 time/day, and drinking it 1 to 2 times/day (vs drinking it ⱖ3 times/day; ORs 1.83, 1.91, and 1.32, respectively), drinking no milk and drinking ⬍2 glasses milk/day (vs drinking ⱖ2 glasses/day; ORs 1.42 and 1.41, respectively), drinking ⬍1 soda/day (vs drinking none; OR 1.40), drinking fruit-flavored drinks/sports drinks ⬍1 time/day and drinking it ⱖ1 time/day (vs drinking none; ORs 1.49 and 1.41, respectively), eating at a fast-food restaurant ⱖ3 days/week (vs none; OR 1.38, respectively), not participating on team sports or participating on one to two team sports in previous 12 months (vs participating on ⱖ3 teams; ORs 1.77 and 1.24, respectively), and consuming snack/soda while watching television/movies “sometimes” and “most/every time” (vs never; ORs 1.65 and 2.20, respectively) (Table 3). AI levels for beverages (including drinking water) among adolescents range from 1.6 L/day to 2.6 L/day depending on sex and age, and those living in hot climates should consume even more (16). However, average water intake among Florida adolescents was somewhat lower than that in other youth populations. For example, a mean water intake (tap and bottled) was 1.31 L/day among 138 Canadian adolescents (aged 13 to 17 years) (31). In another study, mean plain water intake was 0.80 L/day among 1,997 US youth (aged 12 to 19 years) (19).

August 2011 ● Journal of the AMERICAN DIETETIC ASSOCIATION

1213

Table 2. Drinking water intake among public middle school students by students’ dietary and behavioral factors, Florida Youth Physical Activity and Nutrition Survey, 2007a Drinking Water Intake During the Previous 7 d Dietary or behavioral factor 100% fruit juice intake during the previous 7 d None ⬍1 time/d 1-2 times/d ⱖ3 times/d Milk intake during the previous 7 d None ⬍2 glasses/d ⱖ2 glasses/d Soda intake during the previous 7 d None ⬍1 soda/d ⱖ1 soda/d Fruit-flavored drink or sports drink intake during the previous 7 d None ⬍1 time/d ⱖ1 time/d Eating breakfast during the previous 7 d None 1-6 mornings/wk Every morning Fast-food consumption during the previous 7 d None 1-2 days/wk ⱖ3 days/wk Team sports played during the past year None 1-2 teams ⱖ3 teams Individual sports played during the past year None 1-2 sports ⱖ3 sports Television viewing—weekday None ⱕ2 h/d ⱖ3 h/d Television viewing—weekend None ⱕ2 h/d 3-4 h/d ⬎4 h/d Eating a snack or drinking a soda while watching television or movies Never Sometimes Most or every time a

Because of rounding, weighted percentages may not add up to 100%. ␹ tests were used for examining differences across categories.

b 2

1214


All respondents

3 glasses/d

n 3,741 747 1,788 748 458 4,107 475 2,079 1,553 4,110 784 2,056 1,270 3,875

4™™™™™™™™™™™™™™™™ %⫾standard error ™™™™™™™™™™™™™™3

580 1,794 1,501 4,188 591 1,667 1,930 4,107 1,066 1,978 1,063 3,953 1,549 1,605 799 3,934 732 2,090 1,112 3,672 288 1,890 1,494 3,599 224 1,279 979 1,117 3,926

14.8⫾0.6 47.1⫾0.8 38.0⫾0.9

36.5⫾2.5 39.8⫾1.1 37.0⫾1.4

21.0⫾1.8 26.9⫾1.0 26.3⫾1.1

42.6⫾2.2 33.3⫾1.1 36.7⫾1.2

14.0⫾0.8 39.9⫾0.8 46.1⫾1.2

46.6⫾2.4 41.7⫾1.3 33.2⫾1.3

21.1⫾1.9 24.7⫾1.1 27.0⫾1.1

32.3⫾2.3 33.6⫾1.2 39.9⫾1.3

26.2⫾0.9 48.1⫾0.8 25.7⫾1.0

36.7⫾1.9 37.5⫾1.2 41.6⫾1.4

22.3⫾1.6 26.8⫾1.0 25.9⫾1.4

41.0⫾1.8 35.7⫾1.0 32.6⫾1.5

38.9⫾1.1 40.8⫾0.8 20.3⫾0.9

44.6⫾1.3 36.2⫾1.5 30.3⫾1.7

23.4⫾1.1 26.6⫾1.3 24.8⫾1.5

32.0⫾1.1 37.3⫾1.3 44.9⫾1.8

18.5⫾0.8 53.1⫾0.9 28.5⫾0.7

49.3⫾2.0 39.3⫾1.5 29.3⫾1.6

23.0⫾1.6 26.9⫾1.2 22.9⫾1.1

27.7⫾1.7 33.8⫾1.1 47.8⫾1.6

8.0⫾0.5 51.8⫾1.3 40.2⫾1.3

38.0⫾3.3 34.8⫾1.3 41.5⫾1.4

19.9⫾2.4 27.4⫾1.1 23.9⫾1.1

42.1⫾3.2 37.8⫾1.1 34.6⫾1.3

6.2⫾0.4 36.0⫾1.2 27.2⫾0.8 30.5⫾1.3

43.9⫾3.4 32.7⫾1.7 37.7⫾1.6 43.2⫾1.5

15.9⫾2.6 29.1⫾1.4 25.8⫾1.4 22.6⫾1.5

40.3⫾3.8 38.2⫾1.5 36.5⫾1.6 34.2⫾1.4

250 2,241 1,435

6.6⫾0.4 57.0⫾1.0 36.4⫾1.0

30.0⫾3.1 35.6⫾1.2 43.5⫾1.4

19.9⫾2.5 26.1⫾1.0 25.0⫾1.2

50.1⫾3.0 38.3⫾1.0 31.5⫾1.1

20.2⫾0.9 48.1⫾0.9 19.6⫾0.7 12.2⫾0.7

44.1⫾2.0 41.5⫾1.3 26.5⫾1.6 28.8⫾2.1

21.3⫾1.7 26.3⫾1.0 31.7⫾1.6 22.6⫾1.8

34.6⫾1.8 32.2⫾1.2 41.8⫾1.7 48.6⫾2.6

11.2⫾0.6 50.3⫾0.8 38.5⫾0.8

46.5⫾2.3 41.6⫾1.2 32.0⫾1.2

20.5⫾1.9 26.7⫾1.0 24.6⫾0.9

33.1⫾1.9 31.7⫾0.9 43.4⫾1.4

18.9⫾0.7 50.1⫾1.0 31.0⫾1.0

32.0⫾2.1 42.5⫾1.2 35.4⫾1.5

22.8⫾1.6 25.3⫾0.9 27.1⫾1.4

45.2⫾2.0 32.1⫾1.0 37.4⫾1.4

P valueb ⬍0.0001

⬍0.0001

⬍0.0001

0.001

⬍0.0001

0.003

⬍0.0001

⬍0.0001

⬍0.0001

⬍0.0001

Table 3. Factors associated with low water intake (drinking ⬍3 glasses water/d) among middle school students, Florida Youth Physical Activity and Nutrition Survey, 2007 (N⫽2,790)a

Characteristics Age (y) ⱕ12 13 ⱖ14 Sex Girl Boy Race/ethnicity White, non-Hispanic Black, non-Hispanic Hispanic or Latino Other, non-Hispanic 100% fruit juice intake during the previous 7 d None ⬍1 time/d 1-2 times/d ⱖ3 times/d Milk intake during the previous 7 d None ⬍2 glasses/d ⱖ2 glasses/d Soda intake during the previous 7 d None ⬍1 soda/d ⱖ1 soda/d Fruit-flavored drink or sports drink intake during the previous 7 d None ⬍1 time/d ⱖ1 time/d Fast-food consumption during the previous 7 d None 1-2 d/wk ⱖ3 d/wk Team sports played during the previous 12 mo None 1-2 teams ⱖ3 teams Frequency of snack or soda consumption while watching television or movies Never Sometimes Most or every time

Adjusted odds ratio

95% confidence interval

Referent group 0.87 1.04

— 0.71-1.08 0.86-1.24

Referent group 0.97

0.82-1.15

Referent group 0.97 0.79 0.76

— 0.79-1.20 0.65-0.97 0.59-0.98

1.83 1.91 1.32 Referent group

1.39-2.42 1.46-2.49 1.01-1.72 —

1.42 1.41 Referent group

1.10-1.83 1.20-1.66 —


— 1.14-1.71 0.86-1.33


— 1.17-1.91 1.10-1.80


— 1.00-1.50 1.04-1.83

1.77 1.24 Referent group

1.45-2.16 1.02-1.52 —


— 1.21-2.25 1.54-3.14

a To have a complete data set without any missing data, this logistic regression model included a sample of 2,790 students.

The finding that Hispanics were significantly less likely to have a low drinking water intake than non-Hispanic whites contrasted with results of a study of 3,867 US children and youth (aged 2 to 19 years) by Kant and Graubard (19), who found no significant difference in mean plain water intake levels among non-Hispanic whites (0.63 L/day), Mexican Americans (0.51 L/day), and non-Hispanic blacks (0.48 L/day) (19). There was no association between water intake and sex. Findings from previous studies of the association between water intake and sex have been inconsistent. Based on 2005-2006 NHANES data, mean plain water intake was significantly higher among boys than girls (aged 2 to 19 years) (19) and also higher among men than women (aged ⱖ19 years) (26); results of a study based on 2005-2006 NHANES data showed no significant different by sex among adults (aged ⱖ20 years) (27); and results of a Canadian study involving subjects aged 0 to 60 years showed a significantly higher mean water intake among women (31). In a study among US adults, water intake among those who tried to lose weight during the previous year was not significantly different from that among those who did not try to lose weight (27). Results of other studies addressing whether water intake might facilitate weight management have been mixed, and the concept remains controversial (32). A few studies examining possible associations between water intake and dietary or behavioral factors have been conducted among children (19) and adults (25,27). Consistent with previous findings, our study showed that students with low water intake were less likely to be 100% juice and milk consumers and more likely to be soda and fruit-flavored drink/sports drink consumers. Our findings suggest that low water intake might be associated with a higher consumption of a cluster of less healthful beverage intake behaviors. In a study of 4,755 US adults (aged ⱖ18 years), regular soft drinks were consumed daily by 45% of water consumers and 63% of non–water consumers. Furthermore, after controlling for demographics, water consumers were 25 times less likely to drink sugar drinks than non–water consumers. In contrast, milk was consumed daily by 49% of water consumers and 38% of nonconsumers; and fruit juices were consumed daily by 29% of water consumers and 17% of nonconsumers (25). The finding that Florida adolescents who participated in team or individual sports during the previous year were likely to drink more water than those who did not was similar to findings of Kant and Graubard (19) that participation in physical activity was associated with higher water intake among 3,867 US children and youth (aged 2 to 19 years) (19). Television viewing was significantly associated with low water intake in our study, whereas Kant and Graubard (19) found no significant difference in plain water intake by television and computer use. Furthermore, consumption of a snack/soda while watching television/movies was associated with low water intake. Our findings may be attributable in part to extensive television advertisements for high-energy and low-nutrient beverages and foods that are targeted to adolescents (33). Television viewing has been associated with sugar drink consumption (34). Another explanation might be an association between television viewing and


1215

physical inactivity (35,36), which would lead to a decreased need for fluid intake. Leisure-time physical activity and higher average physical activity have been associated with higher water intakes among youth (aged 2 to 19 years) (19) and adults (aged ⱖ20 years) (27). Our study is one of the first to examine associations between low water intake and demographic, dietary, and behavioral factors among adolescents. However, there are a few limitations. First, YPANS is self-reported and is subject to potential reporting bias. Second, because the terms “glass” and “bottle” were not defined in YPANS and water intake from water fountains was not included in estimates of water intake, the actual amount of water consumed by respondents is uncertain. Third, due to the cross-sectional nature of the study design, no causal inferences can be made in these analyses. Our findings represent only factors associated with the outcome of low water consumption. Longitudinal studies are needed to identify predictors for low water intake and causal relationships and mechanisms would have to be examined in experimental laboratory studies. Finally, the response rate was 70%, and only 68% of survey participants were included in the logistic regression, which further limited the generalizability of findings. CONCLUSIONS Nearly two thirds of respondents reported drinking ⬍3 glasses water daily in the previous week, and the percentage who did so varied by demographic, dietary, and behavioral characteristics. The factor most strongly associated with low drinking water intake was frequent consumption of a snack/soda while watching television/movies. Other factors significantly associated with low water intake were non-Hispanic ethnicity, low intake of 100% fruit juice and milk, high soda and fruit-flavored drink/ sports drink intake, frequent consumption of fast foods, and nonparticipation in team sports. These findings suggest that efforts to increase water consumption among adolescents may need to consider low daily water intake as part of a cluster of other less-healthful behaviors and should focus on those with these associated factors; however, findings should be corroborated in other states and in a nationally representative sample. STATEMENT OF POTENTIAL CONFLICT OF INTEREST: No potential conflict of interest was reported by the authors. ACKNOWLEDGEMENTS: The authors thank the Florida Department of Health for providing the Youth Physical Activity and Nutrition Survey data. The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. References 1. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 19992004. JAMA. 2006;295:1549-1555. 2. Ogden CL, Carroll MD, Curtin LR, Lamb MM, Flegal KM. Prevalence of high body mass index in US children and adolescents, 2007-2008. JAMA. 2010;303:242-249. 3. Dietz WH. Health consequences of obesity in youth: Childhood predictors of adult disease. Pediatrics. 1998;101:518-525.

1216


4. Deckelbaum RJ, Williams CL. Childhood obesity: The health issue. Obes Res. 2001;9(suppl 4):239S-243S. 5. Li Y, Yang X, Zhai F. Childhood obesity and its health consequence in China. Obes Rev. 2008;9(suppl 1):82-86. 6. Ludwig DS, Peterson KE, Gortmaker SL. Relation between consumption of sugar-sweetened drinks and childhood obesity: A prospective, observational analysis. Lancet. 2001;357:505-508. 7. Ebbeling CB, Feldman HA, Osganian SK, Chomitz VR, Ellenbogen SJ, Ludwig DS. Effects of decreasing sugar-sweetened beverage consumption on body weight in adolescents: A randomized, controlled pilot study. Pediatrics. 2006;117:673-680. 8. Malik VS, Schulze MB, Hu FB. Intake of sugar-sweetened beverages and weight gain: A systematic review. Am J Clin Nutr. 2006;84:274288. 9. Fiorito LM, Marini M, Francis LA, Smiciklas-Wright H, Birch LL. Beverage intake of girls at age 5 y predicts adiposity and weight status in childhood and adolescence. Am J Clin Nutr. 2009;90:935942. 10. Muckelbauer R, Libuda L, Clausen K, Toschke AM, Reinehr T, Kersting M. Promotion and provision of drinking water in schools for overweight prevention: Randomized, controlled cluster trial. Pediatrics. 2009;123:E661-E667. 11. Guthrie JF, Morton JF. Food sources of added sweeteners in the diets of Americans. J Am Diet Assoc. 2000;100:43-51. 12. Block G. Foods contributing to energy intake in the US: Data from NHANES III and NHANES 1999-2000. J Food Comp Anal. 2004;17: 439-447. 13. Duffey KJ, Popkin BM. Shifts in patterns and consumption of beverages between 1965 and 2002. Obesity. 2007;15:2739-2747. 14. Wang YC, Bleich SN, Gortmaker SL. Increasing caloric contribution from sugar-sweetened beverages and 100% fruit juices among US children and adolescents, 1988-2004. Pediatrics. 2008;121:e1604e1614. 15. Wang YC, Ludwig DS, Sonneville K, Gortmaker SL. Impact of change in sweetened caloric beverage consumption on energy intake among children and adolescents. Arch pediatr Adolesc Med. 2009;163:336343. 16. Institute of Medicine, Food and Nutrition Board. Dietary Reference Intakes for Water, Potassium, Sodium, Chloride, and Sulfate. Institute of Medicine Panel on Dietary Reference Intakes for Electrolytes and Water, Standing Committee on the Scientific Evaluation of Dietary Reference Intakes. Washington, DC: National Academies Press; 2005. 17. Jiang XJ, Castelao JE, Groshen S, Cortessis VK, Shibata DK, Conti DV, Gago-Dominguez M. Water intake and bladder cancer risk in Los Angeles County. Int J Cancer. 2008;123:1649-1656. 18. Manz F. Hydration and disease. J Am Coll Nutr. 2007;26(suppl 5): 535s-541s. 19. Kant AK, Graubard BI. Contributors of water intake in US children and adolescents: Associations with dietary and meal characteristics— National Health and Nutrition Examination Survey 2005-2006. Am J Clin Nutr. 2010;92:887-896. 20. Grimm GC, Harnack L, Story M. Factors associated with soft drink consumption in school-aged children. J Am Diet Assoc. 2004;104:12441249. 21. Storey ML, Forshee RA, Anderson PA. Beverage consumption in the US population. J Am Diet Assoc. 2006;106:1992-2000. 22. Heller KE, Sohn W, Burt BA, Feigal RJ. Water consumption and nursing characteristics of infants by race and ethnicity. J Public Health Dent. 2000;60:140-146. 23. Sohn W, Heller KE, Burt BA. Fluid consumption related to climate among children in the United States. J Public Health Dent. 2001;61: 99-106. 24. Campos R, Montenegro-Bethancourt G, Vossenaar M, Doak C, Solomons N. Volume, frequency and participation in plain drinking water consumption by third and fourth-grade schoolchildren in Quetzaltenango, Guatemala. Asia Pac J Clin Nutr. 2009;18:164-170. 25. Popkin BM, Barclay DV, Nielsen SJ. Water and food consumption patterns of U.S. adults from 1999 to 2001. Obes Res. 2005;13:21462152. 26. Fulgoni VL. Limitations of data on fluid intake. J Am Coll Nutr. 2007;26(suppl 5):588s-591s. 27. Kant AK, Graubard BI, Atchison EA. Intakes of plain water, moisture in foods and beverages, and total water in the adult US populationnutritional, meal pattern, and body weight correlates: National Health and Nutrition Examination Surveys 1999-2006. Am J Clin Nutr. 2009;90:655-663. 28. Zapata LB, Bryant CA, McDermott RJ, Hefelfinger JA. Dietary and

29. 30. 31. 32.

physical activity behaviors of middle school youth: The Youth Physical Activity and Nutrition Survey. J Sch Health. 2008;78:9-18. Brener ND, Collins JL, Kann L, Warren CW, Williams BI. Reliability of the Youth Risk Behavior Survey questionnaire. Am J Epidemiol. 1995;141:575-580. Brener ND, Kann L, McManus T, Kinchen SA, Sundberg EC, Ross JG. Reliability of the 1999 Youth Risk Behavior Survey questionnaire. J Adolesc Health. 2002;31:336-342. Pintar KDM, Waltner-Toews D, Charron D. Water consumption habits of a south-western Ontario community. J Water Health. 2009;7:276-292. Dennis EA, Flack KD, Davy BM. Beverage consumption and adult weight management: A review. Eat Behav. 2009;10:237-246.

33. Institute of Medicine. Food Marketing to Children and Youth: Threat or Opportunity? Washington, DC: National Academies Press; 2006. 34. Barr-Anderson D, Larson N, Nelson M, Neumark-Sztainer D, Story M. Does television viewing predict dietary intake five years later in high school students and young adults? Int J Behav Nutr Phys Act. 2009;6:1-8. 35. Lowry R, Wechsler H, Galuska D, Fulton J, Kann L. Television viewing and its associations with overweight, sedentary lifestyle, and insufficient consumption of fruits and vegetables among US high school students: differences by race, ethnicity, and gender. J Sch Health. 2002;72:413-421. 36. Hager RL. Television viewing and physical activity in children. J Adolesc Health. 2006;39:656-661.


1217