Abstract
Objectives. To evaluate whether differences in tap water and other beverage intake explain differences in inadequate hydration among US adults by race/ethnicity and income.
Methods. We estimated the prevalence of inadequate hydration (urine osmolality ≥ 800 mOsm/kg) by race/ethnicity and income of 8258 participants aged 20 to 74 years in the 2009 to 2012 National Health and Nutrition Examination Survey. Using multivariable regression models, we estimated associations between demographic variables, tap water intake, and inadequate hydration.
Results. The prevalence of inadequate hydration among US adults was 29.5%. Non-Hispanic Blacks (adjusted odds ratio [AOR] = 1.44; 95% confidence interval [CI] = 1.17, 1.76) and Hispanics (AOR = 1.42; 95% CI = 1.21, 1.67) had a higher risk of inadequate hydration than did non-Hispanic Whites. Lower-income adults had a higher risk of inadequate hydration than did higher-income adults (AOR = 1.23; 95% CI = 1.04, 1.45). Differences in tap water intake partially attenuated racial/ethnic differences in hydration status. Differences in total beverage and other fluid intake further attenuated sociodemographic disparities.
Conclusions. Racial/ethnic and socioeconomic disparities in inadequate hydration among US adults are related to differences in tap water and other beverage intake. Policy action is needed to ensure equitable access to healthy beverages.
Access to safe, clean drinking water is defined as a human right by the United Nations.1 Consuming adequate water, whether in the form of plain water, other beverages, or food, is essential for maintaining hydration status, which is in turn critical for proper physiological functioning. Drinking water instead of sugary drinks could help reduce the risk of obesity,2,3 and fluoridated water can promote dental health.4 Although severe dehydration can be life threatening and often requires urgent medical intervention, an emerging body of evidence suggests that mild dehydration or inadequate hydration (i.e., when one is beginning to feel thirsty) may increase the risk of (1) disruptions in cognitive function, (2) fatigue, and (3) lower endurance.5 Inadequate hydration has also been associated with worse mood and overall subjective feelings of poor health, such as headaches6; conversely, interventions to rehydrate via water intake can improve mood and fatigue.7,8
As the recent catastrophe with Flint, Michigan’s municipal water system has made plain, access to safe, clean drinking water is by no means universal in the United States and may be negatively affecting both perceptions of water quality and water intake (total water intake and water as a beverage). A national study of US adults found that non-Hispanic Black and Hispanic adults were substantially more likely to report that their local tap water was not safe to drink and that this was associated with drinking less water (either tap or bottled) and more sugar-sweetened beverages.9 Total water intake, that is, water from all food and beverage sources, is lower among non-Hispanic Black and Hispanic adults than among non-Hispanic White adults,10 and non-Hispanic White adults are more likely to specifically consume tap water.11
Despite growing research on inadequate hydration and health,12–16 disparities in hydration status have not yet been directly assessed in US adults. One recent study found significant disparities among US youths, with boys and non-Hispanic Black youths more likely to be inadequately hydrated.17 Two recent evaluations of urine osmolality18 and inadequate hydration as related to body mass index (BMI; defined as weight in kilograms divided by the square of height in meters)15 in the US population found it to vary by race/ethnicity; however, these studies’ aims did not include quantifying these disparities or exploring why these disparities exist. Understanding population disparities in hydration status among adults, and identifying whether differential consumption of tap water and other beverages influences disparities, may help inform public health efforts to promote well-being.
We sought to provide an assessment of disparities in hydration status among US adults by income and race/ethnicity using nationally representative cross-sectional data. We also sought to evaluate the extent to which racial/ethnic or socioeconomic differences in intake of tap water as well as other beverages could explain disparities in hydration status, adjusting for relevant confounders.
METHODS
We analyzed data from a nationally representative sample of participants in the National Health and Nutrition Examination Survey (NHANES). Urine samples were collected from study participants as well as demographic, body composition, and dietary intake data. We analyzed data collected during the 2009 to 2010 and 2011 to 2012 waves on US adults aged 20 to 74 years (n = 10 490 of 20 293 participants). We excluded individuals with missing information on urine osmolality, demographic indicators, dietary intake, physical activity, chronic conditions, blood pressure medication use, and body composition from the analyses, resulting in a final study sample of n = 8258.
Measures
The primary outcome was hydration status. NHANES laboratory staff collected urine samples from participants noting the time of examination (morning, afternoon, or evening). The laboratory staff analyzed urine samples for urine osmolality using the freezing point depression osmometry method. Measures of urine osmolality were not available for the NHANES 2013 to 2014 cycle, so we included only data from the 2009 to 2010 and 2011 to 2012 NHANES, because they provide consistent measures of urine osmolality. Details on specimen collection and processing can be found in the NHANES laboratory and medical technologists procedures manuals.19,20
Urine osmolality is a measure of urine concentration expressed as the amount of solute particles in milliosmoles per kilogram of urine. Although other measures of hydration status, including 24-hour urine volume and 24-hour urine osmolality, may be more precise indicators of an individual’s hydration status and are typically used in clinical studies,21 single time point measures of urine osmolality can be used for estimating population prevalences in a large epidemiological study such as this.12,15,17 To define inadequate hydration, we used a cutoff of 800 milliosmoles per kilogram, which has been used in similar cross-sectional studies.12–15,17
Primary predictors.
Age, gender, and race/ethnicity (non-Hispanic White, non-Hispanic Black, Mexican American, Hispanic, and other race including multiracial) were collected from all participants via a questionnaire. We categorized those self-reporting as either Mexican American or other Hispanic as Hispanic. Those in the “other race” category included non-Hispanic individuals who reported more than 1 race or a race other than White or Black. Self-reported total household income was collected, and we analyzed this as the ratio of income to the poverty level, controlling for household size. We used the Department of Health and Human Services’ poverty guidelines for the respective years included. Those above 130% of the poverty level were categorized as higher income, and those at or below 130% of the poverty line were categorized as lower income.
Covariates.
Dietary intake of the day before urine collection was assessed by 24-hour recall. From this, NHANES staff calculated total moisture intake from all foods and beverages (in g). We categorized beverage intake into the following types:
plain water (further stratified into bottled and tap),
sugar-sweetened beverages,
milk,
100% juice,
diet beverages,
unsweetened coffee or tea, and
alcoholic beverages.22
We assessed whether individuals consumed any of each beverage type and estimated the number of 8-fluid-ounce servings of each beverage type consumed. For alcoholic beverages, serving sizes were 1.5 fluid ounces for spirits, 5 fluid ounces for wine, 8 fluid ounces for mixed drinks, and 12 fluid ounces for beer.23 We calculated total food moisture by subtracting all grams of moisture consumed from beverages from total moisture. NHANES staff calculated total intake, in grams, of sodium and protein. We have reported food moisture, sodium, and protein in quartiles for ease of interpretation.
NHANES staff calculated BMI from measured weight and height. We assessed physical activity levels from participants’ self-reported minutes spent per week in leisure-, work-, or biking- and walking-related moderate or vigorous activity. We then categorized participants as accumulating versus not accumulating at least 150 minutes of moderate to vigorous physical activity each week, consistent with a recent study of hydration status using NHANES.16 We assessed use of prescription medication for high blood pressure, which may influence urine concentration, from participants’ self-report of medication use. To account for general health status, which could affect urine concentration, we used an existing score of multiple chronic conditions, summing participants’ reports of having hypertension, heart disease, diabetes, cancer, stroke, chronic bronchitis, emphysema, current asthma, and kidney disease.24
Statistical Analysis
To account for the complex sampling design, we used the SAS PROC SURVEYMEANS for descriptive statistics on mean beverage intake and the SAS PROC SURVEYFREQ for estimates of the sociodemographic variables, the proportion of participants drinking any of each beverage type, and the proportion of participants with inadequate hydration (urine osmolality of ≥ 800 mOsm/kg). We estimated logistic regression models using the SAS PROC SURVEYLOGISTIC to account for the complex sampling design and to estimate crude differences in risk of inadequate hydration by race/ethnicity and income status. We also estimated adjusted logistic regression models to simultaneously quantify racial/ethnic and income differences in hydration status and additionally account for covariates that could influence urine concentration, including age, gender, time of day of urine examination, presence of chronic conditions, use of blood pressure medication, BMI, moderate to vigorous physical activity, sodium consumption, and protein consumption.
Next, to first explore whether tap water intake alone might contribute to disparities in hydration status, we examined whether any observed differences in hydration status by race/ethnicity or income were attenuated after including the variable “any tap water intake” in the adjusted logistic regression model. This approach of qualitatively assessing how much modeled estimates of disparities are attenuated after accounting for measures of potential mechanisms for those disparities has been used in similar studies investigating the drivers of sociodemographic disparities in health.25,26 In our final model, we included all other beverage types and food moisture. As a supplementary analysis, we also constructed linear regression models (adjusting for the sampling design) that estimated differences in intakes of 8 fluid ounce servings per day of the different beverages by race/ethnicity and income, adjusting for the same confounders (Table A [available as a supplement to the online version of this article at http://www.ajph.org]). We conducted all analyses using SAS version 9.4 (SAS Institute, Cary, NC).
RESULTS
Most of the sample identified as non-Hispanic White (68.3%), with fewer non-Hispanic Black (11.1%) and Hispanic (13.7%) participants (Table 1). A little more than one fifth (23.7%) of the sample was classified as being at or below 130% of the poverty level. Most of the sample was aged between 20 to 39 years (39.3%) or 40 to 59 years (41.5%). Nearly half of the sample had at least 1 chronic condition, and one fifth took blood pressure medication.
TABLE 1—
Sample Characteristics of Adults With Nonmissing Urine Osmolality, Demographic, and Dietary Intake Data, Stratified by Race/Ethnicity and Income Status: National Health and Nutrition Examination Survey, United States, 2009–2012
| Characteristic | Overall, No. (Weighted %) or Mean ±SE | Non-Hispanic White, No. (Weighted %) or Mean ±SE | Non-Hispanic Black, No. (Weighted %) or Mean ±SE | Hispanic, No. (Weighted %) or Mean ±SE | Other Race, No. (Weighted %) or Mean ±SE | Higher Income, No. (Weighted %) or Mean ±SE | Lower Income, No. (Weighted %) or Mean ±SE |
| Overall no. | 8258 | 3541 | 1805 | 2048 | 864 | 5396 | 2862 |
| Age, y | |||||||
| 20–39 | 3184 (39.3) | 1345 (35.0) | 620 (42.7) | 801 (53.8) | 418 (47.9) | 1902 (35.5) | 1282 (51.5) |
| 40–59 | 3080 (41.5) | 1319 (43.1) | 699 (41.6) | 750 (35.3) | 312 (37.6) | 2076 (43.6) | 1004 (34.9) |
| 60–74 | 1994 (19.2) | 877 (21.9) | 486 (15.7) | 497 (10.9) | 134 (14.5) | 1418 (20.9) | 576 (13.6) |
| Gender | |||||||
| Male | 4125 (49.8) | 1766 (50.2) | 888 (45.6) | 1032 (51.9) | 439 (50.0) | 2777 (50.8) | 1348 (47.1) |
| Female | 4133 (50.2) | 1775 (49.8) | 917 (54.4) | 1016 (48.1) | 425 (50.0) | 2619 (49.2) | 1514 (52.9) |
| Race/ethnicity | |||||||
| Non-Hispanic White | 3541 (68.3) | . . . | . . . | . . . | . . . | 2464 (74.1) | 1077 (49.6) |
| Non-Hispanic Black | 1805 (11.1) | . . . | . . . | . . . | . . . | 1165 (9.0) | 640 (17.7) |
| Hispanic | 2048 (13.7) | . . . | . . . | . . . | . . . | 1156 (9.9) | 892 (25.8) |
| Other race, including multiracial | 864 (6.9) | . . . | . . . | . . . | . . . | 611 (6.9) | 253 (6.9) |
| Income | |||||||
| Higher incomea | 5396 (76.3) | 2464 (82.8) | 1165 (62.2) | 1156 (55.3) | 611 (76.1) | . . . | . . . |
| Lower incomea | 2862 (23.7) | 1077 (17.2) | 640 (37.8) | 892 (44.7) | 253 (23.9) | . . . | . . . |
| Inadequate hydration | 2538 (29.5) | 928 (26.6) | 634 (36.4) | 727 (39.2) | 249 (27.4) | 1606 (27.7) | 932 (35.1) |
| Chronic conditions | |||||||
| 0 | 4612 (58.7) | 1938 (57.4) | 823 (51.2) | 1266 (66.9) | 585 (66.8) | 3064 (59.2) | 1548 (56.9) |
| 1 | 2113 (26.1) | 922 (27.2) | 536 (27.3) | 472 (22.6) | 183 (21.1) | 1428 (26.7) | 685 (24.3) |
| ≥ 2 | 1533 (15.2) | 681 (15.4) | 446 (21.4) | 310 (10.6) | 96 (12.1) | 904 (14.1) | 629 (18.8) |
| Using blood pressure medication | 1921 (20.4) | 785 (21.2) | 627 (29.6) | 386 (12.3) | 123 (14.0) | 1305 (21.2) | 616 (18.0) |
| Body mass index, kg/m2 | 28.9 ±0.13 | 28.6 ±0.16 | 31.1 ±0.25 | 29.6 ±0.22 | 26.5 ±0.39 | 28.7 ±0.16 | 29.4 ±0.19 |
| Physical activity ≥ 150 min/wk | 5096 (65.7) | 2312 (68.1) | 1040 (60.0) | 1199 (61.0) | 545 (61.2) | 3403 (66.8) | 1693 (62.3) |
Note. Sample size was n = 8258.
Defined by Department of Health and Human Services’ poverty guidelines for the respective years included. Those above 130% of the poverty level were categorized as higher income, and those at or below 130% of the poverty line were categorized as lower income.
In crude analyses of the population prevalence of inadequate hydration, we found that nearly one third of the overall sample (29.5%) was inadequately hydrated (Table 1). Whereas just more than one quarter of non-Hispanic White adults were inadequately hydrated (26.6%), more than one third of non-Hispanic Black (36.4%) and Hispanic (39.2%) adults were. Whereas more than one third of low-income adults were inadequately hydrated (35.1%), more than one quarter of high-income adults were (27.7%). Overall total moisture intake was 3198.6 grams, but we saw significant variation, with higher intake among non-Hispanic Whites (3341.1 g) than among non-Hispanic Blacks (2694.8 g), Hispanics (3004.7 g), and those identified as other race (2985.2 g; Tables A and B, available as a supplement to the online version of this article at http://www.ajph.org). We also found higher intake for higher-income adults (3247.8 g) than for lower-income adults (3040.6 g), although the difference was not statistically significant in adjusted regression models.
Non-Hispanic White adults reported consuming, on average, 3.5 servings of tap water each day, and adjusted regression models found that non-Hispanic Black adults reported consuming 1.38 fewer servings of tap water (95% confidence interval [CI] = −1.76, −1.00) and Hispanics reported consuming 1.20 fewer servings of tap water (95% CI = −1.86, −0.54) than did their non-Hispanic White counterparts. Conversely, non-Hispanic Black and Hispanic adults reported consuming significantly more servings of bottled water than did non-Hispanic Whites. Non-Hispanic Blacks and Hispanics also reported consuming significantly fewer servings of other noncaloric beverages such as diet drinks and unsweetened coffee or tea than did non-Hispanic Whites.
Lower-income adults also consumed significantly fewer servings of tap water (estimate: −0.47 servings; 95% CI = −0.86, −0.08) than did higher-income adults and reported consuming more bottled water servings than did higher-income adults although the difference was not statistically significant in adjusted models. Table 2 shows that only about one third of non-Hispanic Black (37.0%) and Hispanic (36.3%) adults reported consuming any tap water compared with more than half of non-Hispanic White adults (59.1%; Figure 1).
TABLE 2—
Number (Weighted %) of Participants Who Drank Any of Each Beverage Type in the Past 24 Hours, by Race/Ethnicity and Socioeconomic Status: National Health and Nutrition Examination Survey, United States, 2009–2012
| Beverage | Overall, No. (Weighted %) | Non-Hispanic White, No. (Weighted %) | Non-Hispanic Black, No. (Weighted %) | Hispanic, No. (Weighted %) | Other Race, No. (Weighted %) | Higher Income,a No. (Weighted %) | Lower Income,a No. (Weighted %) |
| Plain water, any type | 6625 (81.9) | 2786 (82.2) | 1413 (77.6) | 1661 (81.7) | 765 (86.0) | 4470 (83.4) | 2155 (77.0) |
| Tap water | 3833 (53.1) | 1942 (59.1) | 680 (37.0) | 719 (36.3) | 492 (53.0) | 2712 (56.4) | 1121 (42.5) |
| Bottled water | 3224 (33.6) | 1018 (28.1) | 815 (44.2) | 1050 (49.3) | 341 (39.7) | 2103 (32.5) | 1121 (37.0) |
| SSB | 4727 (53.8) | 1830 (49.7) | 1216 (68.7) | 1283 (64.1) | 398 (50.1) | 2871 (50.2) | 1856 (65.2) |
| Milk | 3382 (43.3) | 1647 (46.4) | 466 (25.8) | 902 (42.3) | 367 (42.5) | 2229 (44.2) | 1153 (40.2) |
| 100% juice | 1858 (21.8) | 707 (20.6) | 435 (24.6) | 546 (26.6) | 170 (19.6) | 1248 (21.7) | 610 (22.2) |
| Diet beverages | 1397 (20.8) | 840 (24.6) | 179 (9.5) | 270 (13.0) | 108 (17.1) | 1052 (23.4) | 345 (12.4) |
| Coffee or tea, unsweetened | 4572 (58.4) | 2204 (63.7) | 720 (35.4) | 1145 (50.1) | 503 (59.8) | 3210 (62.3) | 1362 (45.9) |
| Alcoholic beverages | 1679 (21.1) | 858 (23.5) | 345 (18.6) | 374 (16.0) | 102 (11.0) | 1196 (22.7) | 483 (15.8) |
Note. SSB = sugar-sweetened beverage.
Those above 130% of the poverty level were categorized as higher income, and those at or below 130% of the poverty line were categorized as lower income.
FIGURE 1—
Weighted % of Participants Who Drank Any of Each Beverage Type in the Past 24 Hours, by Race/Ethnicity: National Health and Nutrition Examination Survey, United States, 2009–2012
Note. Sample size was n = 8258.
*Significantly different from non-Hispanic White persons (P ≤ .001). P value is from logistic regression models comparing likelihood of consuming any of each beverage type for non-Hispanic Black, Hispanic, and other race groups to non-Hispanic White persons, accounting for complex sampling design and adjusted for age category (20–39, 40–59, 60–74, and ≥ 75 y), gender (male or female), number of chronic conditions, blood pressure medication use (yes or no), time of day of urine sample (morning vs afternoon or evening), body mass index (kg/m2), whether participant met moderate to vigorous physical activity recommendations of ≥ 150 min/wk (yes or no), sodium consumption (in quartiles), and protein consumption (in quartiles).
In crude models testing for statistically significant racial/ethnic and income differences in the population prevalence of inadequate hydration, Hispanic adults had 1.78 times higher odds and non-Hispanic Black adults had 1.58 times higher odds of inadequate hydration than did non-Hispanic White adults, and lower-income adults had 1.41 times the odds of inadequate hydration as did higher-income adults (Table 3). After adjusting simultaneously for race/ethnicity, income status, age, gender, chronic conditions, medication use, the time of day of the urine osmolality assessment, BMI, physical activity, and sodium and protein consumption, these disparities persisted, although they were attenuated. Both non-Hispanic Blacks and Hispanics had about 1.4 times the odds of inadequate hydration as did non-Hispanic Whites (non-Hispanic Blacks: adjusted odds ratio [AOR] = 1.44; 95% CI = 1.17, 1.76; Hispanics: AOR = 1.42; 95% CI = 1.21, 1.67). Lower-income adults had 1.23 times higher odds of inadequate hydration (95% CI = 1.04, 1.45) than did higher-income adults after adjustment for covariates.
TABLE 3—
Logistic Regression Models Estimating Associations Between Race/Ethnicity, Socioeconomic Status, and Risk of Inadequate Hydration, With and Without Adjustment for Differences in Any Beverage Category Intake and Food Moisture: National Health and Nutrition Examination Survey, United States, 2009–2012
| Characteristic | Model 1,a OR (95% CI) | Model 2,b AOR (95% CI) | Model 3,b,c AOR (95% CI) | Model 4,b,d AOR (95% CI) |
| Race/ethnicity | ||||
| Non-Hispanic White (Ref) | 1 | 1 | 1 | 1 |
| Non-Hispanic Black | 1.58 (1.30, 1.92) | 1.44 (1.17, 1.76) | 1.39 (1.12, 1.71) | 1.22 (0.98, 1.52) |
| Hispanic | 1.78 (1.56, 2.03) | 1.42 (1.21, 1.67) | 1.37 (1.18, 1.59) | 1.45 (1.24, 1.70) |
| Other race | 1.04 (0.84, 1.30) | 1.01 (0.81, 1.26) | 1.00 (0.80, 1.25) | 1.08 (0.87, 1.35) |
| Incomee | ||||
| Higher income (Ref) | 1 | 1 | 1 | 1 |
| Lower income | 1.41 (1.18, 1.70) | 1.23 (1.04, 1.45) | 1.21 (1.01, 1.44) | 1.09 (0.91, 1.31) |
| Gender | ||||
| Female (Ref) | . . . | 1 | 1 | 1 |
| Male | . . . | 1.61 (1.38, 1.88) | 1.60 (1.38, 1.87) | 1.52 (1.31, 1.76) |
| Age, y | ||||
| 20–39 | . . . | 1 | 1 | 1 |
| 40–59 | . . . | 0.73 (0.64, 0.83) | 0.73 (0.64, 0.82) | 0.83 (0.72, 0.95) |
| 60–74 | . . . | 0.43 (0.35, 0.51) | 0.43 (0.36, 0.52) | 0.53 (0.43, 0.66) |
| Chronic conditions | . . . | 0.88 (0.79, 0.98) | 0.88 (0.79, 0.98) | 0.89 (0.79, 0.99) |
| Blood pressure medication | . . . | 0.77 (0.59, 1.01) | 0.77 (0.60, 1.01) | 0.78 (0.59, 1.02) |
| Time of day | . . . | 1.13 (0.97, 1.32) | 1.13 (0.97, 1.32) | 1.16 (0.99, 1.36) |
| Body mass index, kg/m2 | . . . | 1.04 (1.03, 1.06) | 1.04 (1.03, 1.06) | 1.04 (1.02, 1.05) |
| Physical activity, min/wk | ||||
| < 150 (Ref) | . . . | 1 | 1 | 1 |
| ≥ 150 | . . . | 1.07 (0.91, 1.26) | 1.09 (0.92, 1.28) | 1.12 (0.95, 1.31) |
| Sodium (quartiles) | . . . | 1.05 (0.95, 1.16) | 1.05 (0.95, 1.16) | 1.10 (1.00, 1.21) |
| Protein (quartiles) | . . . | 1.05 (0.94, 1.18) | 1.06 (0.95, 1.18) | 1.14 (1.03, 1.27) |
| Beverages | ||||
| Any tap water | . . . | . . . | 0.83 (0.70, 0.98) | 0.77 (0.64, 0.94) |
| Any bottled water | . . . | . . . | . . . | 0.80 (0.67, 0.94) |
| Any SSBs | . . . | . . . | . . . | 1.00 (0.85, 1.16) |
| Any milk | . . . | . . . | . . . | 0.92 (0.80, 1.06) |
| Any juice | . . . | . . . | . . . | 0.81 (0.68, 0.97) |
| Any diet beverages | . . . | . . . | . . . | 0.75 (0.59, 0.96) |
| Any coffee or tea | . . . | . . . | . . . | 0.70 (0.58, 0.86) |
| Any alcoholic beverages | . . . | . . . | . . . | 0.83 (0.71, 0.97) |
| Food moisture (quartiles) | . . . | . . . | . . . | 0.83 (0.78, 0.89) |
Note. AOR = adjusted odds ratio; CI = confidence interval; OR = odds ratio; SSB = sugar-sweetened beverage. Sample size was n = 8258.
Risk of inadequate hydration, crude.
Risk of inadequate hydration, adjusted for age category (20–39, 40–59, 60–74, and ≥ 75 years), gender (male or female), number of chronic conditions, blood pressure medication use (yes or no), time of day of urine sample (morning vs afternoon or evening), body mass index (kg/m2), whether participant met moderate to vigorous physical activity recommendations of ≥ 150 min/wk (yes or no), sodium consumption (in quartiles), and protein (in quartiles).
Risk of inadequate hydration, additionally adjusted for any tap water intake.
Risk of inadequate hydration, additionally adjusted for beverage and moisture from food intake.
Those above 130% of the poverty level were categorized as higher income, and those at or below 130% of the poverty line were categorized as lower income.
Consuming any tap water was associated with lower risk of inadequate hydration (AOR = 0.83; 95% CI = 0.70, 0.98). Differences in any tap water intake further attenuated the likelihood of inadequate hydration from an AOR of 1.44 to 1.39 (95% CI = 1.12, 1.71) for non-Hispanic Black adults and to AOR = 1.37 (95% CI = 1.18, 1.59) for Hispanic adults when compared with non-Hispanic White adults (Table 3). There was little attenuation in the AOR for lower-income adults compared with higher-income adults when accounting for any tap water intake (AOR = 1.21; 95% CI = 1.01, 1.44). When we included intake of moisture from all other sources, the AOR for non-Hispanic Blacks was no longer statistically significant (AOR = 1.22; 95% CI = 0.98, 1.52), suggesting that both lower tap water and lower intake of other beverages among non-Hispanic Black adults may fully explain disparities in hydration status. We did not observe a similar attenuation after additional adjustment for intake of moisture from all other sources for Hispanic adults, although we did observe it for lower-income adults.
DISCUSSION
In this nationally representative cross-sectional study, we found substantial racial/ethnic and income-based disparities in inadequate hydration status. Compared with non-Hispanic White adults, and adjusting for other sociodemographic factors, non-Hispanic Black adults and Hispanic adults had more than 40% higher odds of inadequate hydration. We also observed income disparities in hydration status: those with lower incomes had 20% higher odds of inadequate hydration than did higher-income adults, after adjustment for covariates. Differences in tap water intake only partially accounted for racial/ethnic disparities in inadequate hydration, whereas differences in other beverage and total fluid intake appeared to account for disparities by income and by race/ethnicity for non-Hispanic Black compared with non-Hispanic White adults. Because poor hydration status may negatively affect cognitive functioning, mood, and well-being,5–7 disparities in hydration status have concerning implications for disparities in overall well-being and daily functioning.
Sociodemographic differences in total fluid intake,10 particularly tap water,11 are a challenge for promoting health equity. Although intake of beverages other than tap water can clearly promote adequate hydration, these beverages do not have the potential of tap water, when fluoridated, for positive effects on dental health.4 Additionally, consumption of sugar-sweetened beverages in particular raises obesity risk.27 Previous research has identified higher consumption of sugar-sweetened beverages in non-Hispanic Blacks,28 and in our own analysis we found that non-Hispanic Blacks and Hispanics were significantly less likely to consume noncaloric beverages such as tap water, diet drinks, and unsweetened coffee or tea. In addition to concerns about how disparities in the intake of sugary alternatives to tap water may influence disparities in health and obesity risk, there is an economic concern. Purchasing beverages, including bottled water, rather than using tap water is an economic burden that our study and previous studies show falls disproportionately on lower-income individuals and people of color.11
Tap water is the least expensive way to maintain hydration status while also having additional health benefits. However, as recent events in Flint, Michigan, and in California’s Central Valley have highlighted,29,30 there are substantial social inequities in access to safe drinking water in the United States and in perceptions of water quality.31 Many communities in the United States, particularly communities of color, perceive tap water to be unsafe, either because of poor access to tap water that is safe for drinking or because of historically poor access.9,32,33 Perception of poor tap water safety is, in turn, linked with higher consumption of both bottled water and sugary drinks among non-Hispanic Black and Hispanic adults.9,34 Policies addressing the renovation of deteriorating drinking water infrastructure throughout the country and public health efforts to address widespread perceptions of poor tap water quality among communities of color may be needed, as is research to more comprehensively document access to safe tap water across the United States and how that access may differ by regional and sociodemographic factors. Research is also needed on how to more successfully promote the intake of healthy beverages in general.
In this study, we found similar disparities in inadequate hydration status by race/ethnicity and gender as found in a previous study of inadequate hydration among US children and adolescents.17 However, the previous study also found inadequate hydration status to be more prevalent overall (54.5%) than what we observed among adults (29.5%), as did another recent study focusing on children.12 The reason for this stark difference in hydration status between adults and children is unclear. It may in part be because even though there are racial/ethnic and socioeconomic differences in beverage access, most adults have far fewer restrictions in their daily lives than do children on when, where, and how they can access a beverage if they are thirsty. Other age- and illness-related issues regarding urine concentration may also contribute to the lower prevalence of inadequate hydration in the adult sample.
Strengths and Limitations
Strengths of this study include the large, representative population and the laboratory-based measurements of hydration status.
There are also several important limitations. Urine osmolality measures from a single point in time can be subject to substantial variation and may be influenced by factors beyond adequate water intake, including dietary factors and physiology.21 Therefore, a random sample of urine osmolality is not used as a precise diagnostic tool for clinical assessment of the hydration status of individuals, and the hydration cutoff of 800 milliosmoles per kilogram is a crude measure. Although single urine osmolality measurements include random error, measures from a single time point can still serve as indicators of population prevalence of hydration status. The estimated proportion of adults with inadequate hydration was not substantially different from an estimate in a recent study of inadequate hydration among adults across 3 European countries that collected 24-hour urine osmolality, a more precise measure of hydration status.35
The single 24-hour dietary recall used to assess beverage intake is also subject to measurement error and may not be representative of usual intake; however, this error is likely to be random, biasing results toward the null. One concern could be that the time lag between the previous day’s dietary recall and an afternoon measure of urine osmolality could obscure true relationships between beverage intake and urine osmolality, because a urine osmolality measure from a single time point is strongly linked with recent fluid intake. However, although dietary intake can vary substantially from day to day, beverage intake is remarkably stable,36 suggesting that the 24-hour recall may be an adequate measure of usual beverage intake. Additionally, upon restricting the sample to only those with a urine sample collected in the morning, which theoretically better reflects the previous day’s diet, we observed results similar to that of the full sample.
The cross-sectional nature of our study precludes us from making firm conclusions about the causal links between race/ethnicity, beverage intake, and hydration status. However, race/ethnicity necessarily precedes both beverage intake and hydration status (i.e., beverage intake cannot “cause” race/ethnicity). Although we adjusted for many factors that could influence urine concentration and hydration status, we were not able to account for possible regional and seasonal variations that can affect urine osmolality.
Finally, we were unable to assess whether structural or environmental differences in tap water and other beverage access, which may be patterned by race/ethnicity, could help explain disparities in both beverage intake and hydration status. Future studies should continue to evaluate the influence of inadequate hydration on well-being and assess whether disparities in inadequate hydration may partly explain disparities in health and well-being. Further examination of age and gender differences in hydration status are warranted.
Public Health Implications
Significant racial/ethnic and income disparities in hydration status exist in the US adult population. Differences in tap water intake partially explain the racial/ethnic disparities, and additional differences in total beverage and fluid intake appear to fully attenuate income and some racial/ethnic disparities. Public health action should focus on systematically researching access to safe drinking water nationwide and promoting adequate access to and intake of healthy beverages and fluids among lower-income adults and adults of color. This could include improving drinking water infrastructure in areas with poor access to make safe tap water access universal and improving perceptions of tap water and increasing intake to advance health equity.
ACKNOWLEDGMENTS
C. J. Brooks was supported by the Harvard T. H. Chan School of Public Health, Cardiovascular Disease Epidemiology Training Program in Behavior, the Environment, and Global Health (grant T32 HL098048); S. L. Gortmaker and E. L. Kenney were supported by a donation made in memory of Melvin R. Seiden. M. W. Long was supported by the Milken Institute School of Public Health. A. L. Cradock was supported by the Harvard T. H. Chan School of Public Health Prevention Research Center.
HUMAN PARTICIPANT PROTECTION
No protocol approval was necessary for this study because data were obtained from secondary sources.
Footnotes
See also Patel and Schmidt, p. 1354.
REFERENCES
- 1.UN General Assembly resolution 64/292. The Human Right to Water and Sanitation. July 28, 2010. [Google Scholar]
- 2.Tate DF, Turner-McGrievy G, Lyons E et al. Replacing caloric beverages with water or diet beverages for weight loss in adults: main results of the Choose Healthy Options Consciously Everyday (CHOICE) randomized clinical trial. Am J Clin Nutr. 2012;95(3):555–563. doi: 10.3945/ajcn.111.026278. [Erratum in Am J Clin Nutr. 2013;98(6):1599] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Pan A, Malik VS, Hao T, Willett WC, Mozaffarian D, Hu FB. Changes in water and beverage intake and long-term weight changes: results from three prospective cohort studies. Int J Obes (Lond) 2013;37(10):1378–1385. doi: 10.1038/ijo.2012.225. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Armfield JM, Spencer AJ, Roberts-Thomson KF, Plastow K. Water fluoridation and the association of sugar-sweetened beverage consumption and dental caries in Australian children. Am J Public Health. 2013;103(3):494–500. doi: 10.2105/AJPH.2012.300889. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Popkin BM, D’Anci KE, Rosenberg IH. Water, hydration, and health. Nutr Rev. 2010;68(8):439–458. doi: 10.1111/j.1753-4887.2010.00304.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Shirreffs SM, Merson SJ, Fraser SM, Archer DT. The effects of fluid restriction on hydration status and subjective feelings in man. Br J Nutr. 2004;91(6):951–958. doi: 10.1079/BJN20041149. [DOI] [PubMed] [Google Scholar]
- 7.Pross N, Demazieres A, Barnouin R, Klein A, Le Bellego L. Hydration and cognitive performance. Cell Physiol Biochem. 2011;28(6):1329. [Google Scholar]
- 8.Paik IY, Jeong MH, Jin HE et al. Fluid replacement following dehydration reduces oxidative stress during recovery. Biochem Biophys Res Commun. 2009;383(1):103–107. doi: 10.1016/j.bbrc.2009.03.135. [DOI] [PubMed] [Google Scholar]
- 9.Onufrak SJ, Park S, Sharkey JR, Sherry B. The relationship of perceptions of tap water safety with intake of sugar-sweetened beverages and plain water among US adults. Public Health Nutr. 2014;17(1):179–185. doi: 10.1017/S1368980012004600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Rosinger A, Herrick K. Daily Water Intake Among U.S. Men and Women, 2009–2012. Hyattsville, MD: National Center for Health Statistics; 2016. NCHS data brief no. 242. [Google Scholar]
- 11.Drewnowski A, Rehm CD, Constant F. Water and beverage consumption among adults in the United States: cross-sectional study using data from NHANES 2005–2010. BMC Public Health. 2013;13:1068. doi: 10.1186/1471-2458-13-1068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Stookey JD, Brass B, Holliday A, Arieff A. What is the cell hydration status of healthy children in the USA? Preliminary data on urine osmolality and water intake. Public Health Nutr. 2012;15(11):2148–2156. doi: 10.1017/S1368980011003648. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Fadda R, Rapinett G, Grathwohl D et al. Effects of drinking supplementary water at school on cognitive performance in children. Appetite. 2012;59(3):730–737. doi: 10.1016/j.appet.2012.07.005. [DOI] [PubMed] [Google Scholar]
- 14.Bar-David Y, Urkin J, Kozminsky E. The effect of voluntary dehydration on cognitive functions of elementary school children. Acta Paediatr. 2005;94(11):1667–1673. doi: 10.1080/08035250500254670. [DOI] [PubMed] [Google Scholar]
- 15.Chang T, Ravi N, Plegue MA, Sonneville KR, Davis MM. Inadequate hydration, BMI, and obesity among US adults: NHANES 2009–2012. Ann Fam Med. 2016;14(4):320–324. doi: 10.1370/afm.1951. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Rosinger AY, Lawman HG, Akinbami LJ, Ogden CL. The role of obesity in the relation between total water intake and urine osmolality in US adults, 2009–2012. Am J Clin Nutr. 2016;104(6):1554–1561. doi: 10.3945/ajcn.116.137414. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kenney EL, Long MW, Cradock AL, Gortmaker SL. Prevalence of inadequate hydration among US children and disparities by gender and race/ethnicity: National Health and Nutrition Examination Survey, 2009–2012. Am J Public Health. 2015;105(8):e113–e118. doi: 10.2105/AJPH.2015.302572. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Yeh HC, Lin YS, Kuo CC et al. Urine osmolality in the US population: implications for environmental biomonitoring. Environ Res. 2015;136:482–490. doi: 10.1016/j.envres.2014.09.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Centers for Disease Control and Prevention. National Health and Nutrition and Examination Survey (NHANES) laboratory procedures manual. 2011. Available at: https://www.cdc.gov/nchs/data/nhanes/nhanes_11_12/2011-12_laboratory_procedures_manual.pdf. Accessed July 1, 2017.
- 20.National Health and Nutrition and Examination Survey (NHANES) Laboratory Procedures Manual. Atlanta, GA: Centers for Disease Control and Prevention; 2009. Available at: https://www.cdc.gov/nchs/data/nhanes/nhanes_09_10/lab.pdf. Accessed October 1, 2016. [Google Scholar]
- 21.Armstrong LE. Assessing hydration status: the elusive gold standard. J Am Coll Nutr. 2007;26(5 suppl):575S–584S. doi: 10.1080/07315724.2007.10719661. [DOI] [PubMed] [Google Scholar]
- 22.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(6):e1604–e1614. doi: 10.1542/peds.2007-2834. [DOI] [PubMed] [Google Scholar]
- 23.National Institute on Alcohol Abuse and Alcoholism. What is a standard drink? Available at: https://www.niaaa.nih.gov/alcohol-health/overview-alcohol-consumption/what-standard-drink. Accessed October 1, 2016.
- 24.Freid VM, Bernstein AB, Bush MA. Multiple Chronic Conditions Among Adults Aged 45 and Over: Trends Over the Past 10 Years. Hyattsville, MD: National Center for Health Statistics; 2012. NCHS data brief no. 100. [PubMed] [Google Scholar]
- 25.Sharifi M, Sequist TD, Rifas-Shiman SL et al. The role of neighborhood characteristics and the built environment in understanding racial/ethnic disparities in childhood obesity. Prev Med. 2016;91:103–109. doi: 10.1016/j.ypmed.2016.07.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Wang Y, Chen X. How much of racial/ethnic disparities in dietary intakes, exercise, and weight status can be explained by nutrition- and health-related psychosocial factors and socioeconomic status among US adults? J Am Diet Assoc. 2011;111(12):1904–1911. doi: 10.1016/j.jada.2011.09.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Malik VS, Pan A, Willett WC, Hu FB. Sugar-sweetened beverages and weight gain in children and adults: a systematic review and meta-analysis. Am J Clin Nutr. 2013;98(4):1084–1102. doi: 10.3945/ajcn.113.058362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Rosinger A, Herrick K, Gahche J, Park S. Sugar-Sweetened Beverage Consumption Among US Adults, 2011–2014. Hyattsville, MD: National Center for Health Statistics; 2017. NCHS data brief no. 270. [Google Scholar]
- 29.Balazs CL, Morello-Frosch R, Hubbard AE, Ray I. Environmental justice implications of arsenic contamination in California’s San Joaquin Valley: a cross-sectional, cluster-design examining exposure and compliance in community drinking water systems. Environ Health. 2012;11:84. doi: 10.1186/1476-069X-11-84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Hanna-Attisha M, LaChance J, Sadler RC, Champney Schnepp A. Elevated blood lead levels in children associated with the Flint drinking water crisis: a spatial analysis of risk and public health response. Am J Public Health. 2016;106(2):283–290. doi: 10.2105/AJPH.2015.303003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Balazs CL, Ray I. The drinking water disparities framework: on the origins and persistence of inequities in exposure. Am J Public Health. 2014;104(4):603–611. doi: 10.2105/AJPH.2013.301664. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Hobson WL, Knochel ML, Byington CL, Young PC, Hoff CJ, Buchi KF. Bottled, filtered, and tap water use in Latino and non-Latino children. Arch Pediatr Adolesc Med. 2007;161(5):457–461. doi: 10.1001/archpedi.161.5.457. [DOI] [PubMed] [Google Scholar]
- 33.Gorelick MH, Gould L, Nimmer M et al. Perceptions about water and increased use of bottled water in minority children. Arch Pediatr Adolesc Med. 2011;165(10):928–932. doi: 10.1001/archpediatrics.2011.83. [DOI] [PubMed] [Google Scholar]
- 34.Hu Z, Morton LW, Mahler RL. Bottled water: United States consumers and their perceptions of water quality. Int J Environ Res Public Health. 2011;8(2):565–578. doi: 10.3390/ijerph8020565. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Malisova O, Athanasatou A, Pepa A et al. Water intake and hydration indices in healthy European adults: the European Hydration Research Study (EHRS) Nutrients. 2016;8(4):204. doi: 10.3390/nu8040204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Lundin R. Distribution of Sugar Sweetened Beverage Consumption in the United States and Its Associations With Blood Lipids and Food Stamp Eligibility [dissertation]. Boston, MA: Harvard T. H. Chan School of Public Health; 2011. [Google Scholar]