answer

sam1992

safwan.pdf

RE S E A R C H AR T I C L E

Fluid Intake and Cognitive Performance: Should Schoolchildren Drink During Lessons? TANJA FUCHS, MSca PETRA LÜHRMANN, Dr. oec. troph.b FAITH SIMPSON, MScc BIRTE DOHNKE, PhDd

ABSTRACT BACKGROUND: Evidence suggests that an insufficient fluid intake impairs cognitive performance. Drinking policies at schools—especially drinking during lessons—is a point of controversy. To provide a scientific base for this debate, more empirical evidence is needed on which aspects of fluid intake are crucial for cognitive performance. This study makes a contribution by investigating associations between quantitative and temporal aspects of fluid intake and cognitive performance in everyday school life.

METHODS: The study comprised 125 children (age: mean = 10.98 years, SD = 0.38). Amount of fluid intake and time span between fluid intake and completion of cognitive tests were determined on basis of self-reports. Cognitive performance was assessed by standardized tests.

RESULTS: Quantitative and temporal aspects were associated with cognitive performance: The more fluid the children consumed and the shorter the time span between their last fluid intake and test completion, the better they performed.

CONCLUSIONS: The amount of fluid intake should be adequate and moreover the time span between intake and cognitive efforts should be as short as possible. Schoolchildren are thus recommended to drink at regular intervals and also during lessons.

Keywords: fluid intake; cognitive performance; schoolchildren.

Citation: Fuchs T, Lührmann P, Simpson F, Dohnke B. Fluid intake and cognitive performance: should schoolchildren drink during lessons? J Sch Health. 2016; 86: 407-413.

Received on January 4, 2015 Accepted on January 9, 2016

Fluid intake among schoolchildren is an issue ofconcern in many countries. Studies from the United Kingdom, France, Italy, and the United States show that about 60% of the children do not consume enough fluid at breakfast, and thus, arrive at school with a hydration deficit.1-4 Starting the school day in this condition is all the more serious, because fluid intake at school also has been shown to be problematic. More than two thirds of schoolchildren do not consume the amount necessary to maintain an optimal hydration status.5 These results concur with others that found that 70% of the children who were dehydrated in the morning maintained this state in the afternoon.6 According to a German study,

aResearch Assistant, ([email protected]), University of Education Schwäbisch Gmünd, Oberbettringer Str. 200, 73525 Schwäbisch Gmünd, Germany. bProfessor, ([email protected]), University of Education Schwäbisch Gmünd, Oberbettringer Str. 200, 73525 Schwäbisch Gmünd, Germany. cResearch Assistant, ([email protected]), University of Education Schwäbisch Gmünd, Oberbettringer Str. 200, 73525 Schwäbisch Gmünd, Germany. dProfessor, ([email protected]), University of Education Schwäbisch Gmünd, Oberbettringer Str. 200, 73525 Schwäbisch Gmünd, Germany.

Address correspondence to: Tanja Fuchs, Research Assistant, ([email protected]), University of Education Schwäbisch Gmünd, Oberbettringer Str. 200, 73525 Schwäbisch Gmünd, Germany.

24% of schoolchildren never or only rarely drink at school.7

A sufficient fluid intake, however, is essential for health. Approximately 60% of the body is comprised of water; it is a major constituent of every cell and fulfills vital physiological functions.8 To maintain good health and proper body functions, the amount of body water should remain relatively constant. Because water cannot be stored, a constant supply is needed to replenish fluid losses—otherwise dehydration occurs.9

Among adults, a number of studies demonstrate that dehydration, induced by physical activity, heat exposure, or fluid restriction, impairs cognitive performance.10,11 These impairments were observable

at a dehydration-induced loss of just 2% body weight or more.

Children are particularly at risk of becoming dehydrated as they have immature thirst mechanisms, relatively high fluid loss (due to a large surface area to body mass ratio) and high activity levels.5

For ethical reasons, experiments involving induced dehydration are not conducted among children. Instead, experimental studies investigate effects of fluid intake on cognitive performance. They show that children who consumed a specific amount of water prior to a cognitive test performed better than those who did not. These effects were found for concentration performance12-15 and memory12,14 in particular.

Children spend the vast majority of their waking day at school where they are constantly faced with tasks requiring high and long-lasting cognitive functioning. Creating conditions that facilitate cognitive perfor- mance should therefore be a priority—especially when considering that cognitive performance was found to be a predictor for children’s school achievements.16,17

Against this background, fluid intake should play a vital role in schools. Indeed, there is growing interest in promoting optimal drinking policies.5,18,19

There are demands for measures such as free access to drinking water and especially the permission to drink during lessons. Drinking freely during lessons, however, is often criticized, especially by teachers.5

Concerns include pupils being distracted and a general disturbance of the lessons. Therefore, drinking during lessons is still prohibited in many schools.19

Schoolchildren are recommended to drink 1-1.5 liters of fluid per day, depending on age, sex, weather, and physical activity.20 The question arises whether there is an actual need for children to drink regularly and at short intervals, and therefore, also during lessons, or whether it would be sufficient to consume this recommended amount at longer intervals or distributed throughout the day. To answer this, it is necessary to know which aspects of fluid intake are actually crucial for cognitive performance: quantitative aspects (ie, amount of fluid intake), temporal aspects (ie, time span between fluid intake and cognitive tasks), or possibly, the interaction of both.

Experimental studies investigating the effects of fluid intake on cognitive performance have so far not sufficiently addressed these aspects. Quantitative aspects of fluid intake were investigated by having schoolchildren consume a specified amount of water prior to the completion of cognitive tests. The amount was kept constant ranging from 250 mL14 to 500 mL.15

Moreover, it was not considered whether or how much fluid the children consumed prior to the experiment (at home or at school). A dose-response relationship between the overall fluid intake and the cognitive performance could not be tested. It remains unclear

whether children perform better the more fluid they consume. Finally, it must be stated that a water intake of 250 to 500 mL before a cognitive task is unlikely and not practical in everyday school life as children face cognitive tasks constantly across a school day. With respect to temporal aspects, similarly, no information can be derived from these studies as the time span between water intake and completion of the cognitive tests was kept constant with an average of 30 minutes.12-15 Thus, it remains unclear whether children perform better the shorter the time span between fluid intake and test completion.

The experimental design of these studies limits the generalizability of their results. It especially remains open which role quantitative and temporal aspects of fluid intake play in cognitive performance in general and most importantly in everyday school life. The aim of this study was to provide empirical evidence on these questions by testing the associations of (1) the overall amount of fluid intake prior to the completion of a cognitive task; (2) the time span between last fluid intake and completion of a cognitive task; and (3) the interaction of these 2 aspects with cognitive performance. Temporal aspects of fluid intake were examined further with respect to the current debates on drinking during school lessons. Explorative analyses were conducted investigating group differences in cognitive performance in dependence of last fluid intake. That is, children whose last fluid intake before the completion of a cognitive task was within a shorter time span were compared with those whose last fluid intake was within a longer time span.

METHODS

Participants The study took place in September 2012. Partic-

ipants were recruited from 2 secondary schools in southern Germany. The school type was Gymnasium, a secondary school which is designed to prepare pupils for university education. German secondary schools start at grade 5; this study took place among chil- dren from grade 6 (N = 125; age: mean = 10.98 years, SD = 0.38). The proportion of boys was 46.4%; the proportion of children with a migration background was 9.6%.

Procedure Data collection was conducted at midday during

a lesson and took approximately 45 minutes. Study assistants administered a questionnaire that assessed demographic data and fluid intake. Afterward, cogni- tive performance tests were conducted. Children were tested in the classroom but completed the tests indi- vidually. For each test, standardized instructions were given and practice items were presented.

Instruments Fluid intake was assessed by self-report. Amount

of fluid intake was assessed by a 24-hour recall ques- tionnaire (paper-and-pencil), which was developed based on an instrument successfully used among pri- mary schoolchildren.21 It records the type and amount of beverages consumed at 6 different time intervals throughout the past 24 hours. Six beverage cate- gories were listed: ‘‘mineral water/sparkling water,’’ ‘‘tap water,’’ ‘‘softdrinks,’’ ‘‘fruit juice/fruit nectar,’’ ‘‘tea,’’ and ‘‘milk and milk drinks’’; additional bever- ages could be specified separately.22 For this study, the overall amount of fluid intake on the test day was considered from the time participants woke up in the morning until the cognitive tests were conducted. The amount of all fluids consumed within this interval was summed. No differences were made regarding com- position of the fluids (ie, osmolarity); all fluids were assumed to hydrate equally. The time span between last fluid intake and test completion was assessed asking par- ticipants to report the time of their last fluid intake. The time span was calculated in minutes (hereinafter referred to as last fluid intake).

Cognitive performance was assessed applying 4 standardized performance tests measuring concen- tration performance and memory. These tests were selected as previous research has shown that per- formance requiring these cognitive processes has been associated with fluid intake.23 Evidence sug- gests that concentration performance is best measured by combining tests representing different facets of concentration.24 Thus, 3 concentration tests which proved to load highly on the factor concentration were selected: Test d2, Zahlen-Verbindungs-Test (ZVT), Digit symbol substitution test (DSST). Furthermore, one memory test was selected: Memory Test. The tests were administered in a paper-pencil format and participants were instructed to complete each test as fast and as correctly as possible.

The Test d225 consists of 14 test lines each with 47 characters. The characters are the letter ‘‘d’’ or letter ‘‘p’’ marked with 1, 2, 3, or 4 small dashes. Participants are required to scan the lines and cross out all ds with 2 dashes while ignoring all other characters. The indicator for concentration performance is calculated from the number of ds crossed out correctly minus the number of errors (Cronbach alpha in this study = 0.935).

The ZVT26 is a variant of the widely used trail- making tests. The test comprises 4 parts. Each part consists of 90 circles distributed over a sheet of paper numbered 1-90. Participants are asked to connect the numbers in ascending order as quickly as possible. For each part, test duration is limited to 30 seconds. Concentration performance is indicated by the number of correctly connected numbers averaged across the 4 parts (Cronbach alpha in this study = 0.863).

The DSST was taken from the Berlin Intelligence Structure test battery.27 A coding table on top of the page provides 9 digit-symbol associations. Below, 4 rows of 17 double boxes are shown. Digits are in the upper boxes, whereas the boxes below are empty. The task is to add the corresponding symbols to the digits as fast as possible. Test duration is limited to 60 seconds. As indicator for concentration performance, the number of correctly added symbols is determined.

The Memory Test is a verbal span test which was also taken from the Berlin Intelligence Structure test battery.27 The test consists of 2 phases: a stimulus presentation phase and a recall phase. First, the stimuli (list of words) to be remembered is presented for 40 seconds via an overhead projector. Afterward, participants are asked to recall and write down as many words as possible in any order within 90 seconds. The number of words remembered is used as an indicator.

Data Analysis All analyses were conducted with SPSS 21 (IBM

Deutschland, Ehningen, Germany). To test the associ- ation between fluid intake and cognitive performance, hierarchical regression analyses were conducted. This was done in steps. In the first step, sex, age, migration status, and waking hours were included as possible control variables. To get the most parsimonious mod- els, only significant control variables remained in the final model. In the second step, the amount of fluid intake was entered, followed by last fluid intake in the third step. This way, the unique influence of both indicators of fluid intake can be determined. To test whether the interaction between the amount of fluid intake and the last fluid intake is associated with cognitive performance, an interaction term was calculated and entered in the final step. Significant interactions were further investigated using simple slope analyses.28

To investigate group differences in cognitive performance in dependence of last fluid intake, 1- way analyses of covariance were conducted. For this purpose, the sample was divided into 2 groups according to their last fluid intake. Group 1 comprised the children whose last fluid intake before test completion was within a shorter time span (ie, < 30 minutes), and Group 2 the children whose last fluid intake before test completion was within a longer time span (ie, > 30 minutes). To determine the unique influence of the last fluid intake, the amount of fluid consumed was included as control variable. Furthermore, sex, age, migration status, and waking hours were considered if significant.

RESULTS

Descriptive information and intercorrelations of the study variables are presented in Table 1. Participants

Table 1. Means and Intercorrelations of the Study Variables

Mean SD 2 3 4 5 6

1 Amount of fluid intake†

0.71 0.45 −.414*** .079 .319*** −.013 .181*

2 Last fluid intake‡ 88.68 116.80 −.236** −.224* −.132 −.182* 3 Test d2 141.96 29.89 .239* .269* .144 4 ZVT 35.31 5.80 .327*** .218* 5 DSST 25.45 6.16 .203 6 Memory Test 7.60 2.24

*p < .05; **p < .01; ***p < .001. ZVT, Zahlen-Verbindungs-Test; DSST, Digit symbol substitution test. †Amount of overall fluid intake at test day indicated in liters. ‡Time span between last fluid intake and cognitive tasks indicated in minutes.

consumed an average of 0.7 liters of fluid prior to test completion. Their last fluid intake was on average 89 minutes before. The amount of fluid intake and last fluid intake correlated significantly. The more fluid the children consumed, the shorter the period since their last fluid intake. The mean scores of all cognitive tests were mainly in the respective midrange. Participants were neither in the upper or lower range of any test score, indicating an adequate level of difficulty. The 3 concentration performance tests showed moderate correlations reflecting the fact that these tests assess different facets of concentration performance. Memory was related to the concentration test ZVT only.

Association Between Fluid Intake and Cognitive Performance

The results of the regression analyses are displayed in Table 2. The amount of fluid intake significantly predicted the scores of the concentration test ZVT and the Memory Test, accounting for 4.9% and 3.6% variance, respectively. That is, the more fluid participants consumed, the better they performed in the respective tests. No influence could be found regarding the concentration tests Test d2 and DSST.

Adding last fluid intake in the next step, a significant influence could be found for the concentration test Test d2. The shorter the time span between the last fluid intake and test completion, the better the performance.

The inclusion of the interaction term in the last step increased the explained variance of the ZVT by another 5%. To further investigate this interaction, conditional effects at different levels of the moderator were estimated. Figure 1 shows that the amount of fluid intake was not a significant predictor for the ZVT test score when the last fluid intake was within a longer time span (+1 SD; B = −3.42; p > .05). Among the children, whose last fluid intake was within a shorter time span, the amount of fluid intake significantly predicted test scores (−1 SD; B = 6.24; p < .001): The more fluid participants consumed, the better they performed in this test. Taken together, the interaction was in line with our prediction; participants performed

better when the amount of fluid intake was high and at the same time the last fluid intake was within a short time span.

Group Differences in Cognitive Performance in Dependence of Last Fluid Intake

The mean time span between last fluid intake and test completion in Group 1 (last fluid intake < 30 minutes) was 23.55 minutes (SD = 8.34), in Group 2 (last fluid intake > 30 minutes) 109.21 minutes (SD = 127.24). Thus, in both groups the last fluid intake was on average less than two hours before test completion. Results of the analyses of covariance are shown in Table 3. Across all cognitive tests, there was a trend that participants whose last fluid intake was within 30 minutes before test completion, performed better than those whose last intake was within a longer time span. Significant differences were found for the concentration test ZVT and the Memory Test. These temporal effects were independent of quantitative aspects, because the amount of fluid intake was controlled for in all analyses.

DISCUSSION

This study contributes to the research on fluid intake and cognitive performance among schoolchildren. The findings confirm that both quantitative and temporal aspects of fluid intake in particular are associated with cognitive performance.

Some studies already have provided evidence for the importance of fluid intake among schoolchildren. They showed that children who consumed a specific amount of water prior to a cognitive test performed better than those who did not with respect to cognitive performance.12-15 However, this was tested in experi- mental designs with no differentiated investigation of quantitative and temporal aspects of fluid intake.

This study considerably extends the existing litera- ture by specifically investigating these different aspects of fluid intake in everyday school life. With respect to quantitative aspects of fluid intake, it shows a dose- response relationship between the overall amount of fluid intake and cognitive performance; the more fluid the schoolchildren consumed, the better they performed. Similar to previous studies, this applied to concentration performance12-15 and memory.12,14

Moreover, this study is the first to examine tempo- ral aspects of fluid intake, namely the time span between the last fluid intake and test completion. Results showed that a shorter time span was associ- ated with better performance. Finally, we found that quantitative and temporal aspects of fluid intake have an interactive effect—children performed at their best the more fluid they consumed and, at the same time,

Table 2. Hierarchical Regression Analyses Predicting Cognitive Performance by Fluid Intake

Test d2 ZVT DSST Memory Test

Step Predictor Beta !R2 Beta !R2 Beta !R2 Beta !R2

1 Waking hours/Sex -/- — .254**/- .064 .226*/−1.97* .085 −.188*/- .035 2 Amount of intake† .015 .000 .256* .049 −.036 .001 .191* .036 3 Last intake‡ −.268** .062 .083 .005 −.146 .017 −.103 .009 4 Interaction (Amount of intake * Last intake) .150 .016 −.263** .050 −.141 .014 .022 .000

*p < .05; ** p < .01. ZVT, Zahlen-Verbindungs-Test; DSST, Digit symbol substitution test. Betas and !R2s represent unique effects from the step the predictor was first entered into the model. Migration status and age were tested as control variables but had no effects. †Amount of overall fluid intake at test day indicated in liters. ‡Time span between last fluid intake and cognitive tasks indicated in minutes. Sex: female = 0, male = 1.

Figure 1. Interaction Between Amount of Fluid Intake and Last Fluid Intake on Cognitive Performance (Test ZVT)

low (-1SD) high (+1SD)

Amount of fluid intake

Last fluid intake low (-1SD) Last fluid intake high (+1SD)

Table 3. Mean Scores in Cognitive Performance Tasks in Dependence of Last Fluid Intake

Last Fluid Intake

Test <30 Min. >30 Min. F p η2

Test d2 146.82 (21.53) 140.44 (32.00) 0.675 .413 0.006 ZVT 37.77 (5.47) 34.53 (5.72) 4.280 .041 0.035 DSST∗ 27.17 (5.00) 24.91 (6.41) 2.285 .133 0.019 Memory test 8.93 (1.83) 7.17 (2.20) 12.634 .001 0.097

ZVT, Zahlen-Verbindungs-Test; DSST, Digit symbol substitution test. Analyses controlled for amount of fluid intake. ∗Controlled for migration status.

the shorter the time span between the last fluid intake and test completion.

To conclude, care should be taken that schoolchil- dren drink adequate amounts of fluid and that not too much time passes between fluid intake and cog- nitive tasks. However, everyday school life is not

characterized by singular cognitive tasks but requires long-lasting cognitive functioning. Thus, to ensure a constant and sufficient fluid supply, schoolchildren should drink at regular intervals. The question, how- ever, is whether children should also drink during lessons. With respect to the current debates about drinking during school lessons, more detailed analy- ses of temporal aspects were conducted in this study. Results showed that children whose last fluid intake was within 30 minutes before completion of a cogni- tive task achieved better results than those whose last fluid intake was longer ago. This was found for con- centration performance as well as for memory. Given that lessons often last up to 90 minutes, it can be rea- sonably assumed that children benefit from drinking during lessons.

Limitations The investigation of fluid intake was based on self-

report (24-recall). Especially for children, it is not easy to remember the beverages and to estimate the quantity consumed. Therefore, inaccuracies are possible, but should be modest because only the time intervals of the same day were considered. A tendency to overestimate fluid intake can also be assumed and the absolute amounts have to be interpreted with care. However, slightly inaccurate or overestimated amounts do not affect our results or its interpretation because the aim of the study was to test associations. Furthermore, associations between fluid intake and concentration performance were not consistently found across the 3 concentration tests; they were found for the Test d2 and the ZVT only. This might reflect the fact that most concentration measures include additional cognitive abilities besides concentration performance. The Test d2 and the ZVT also load highly on figural abilities, whereas the DSST does not.25 Future research should investigate which aspects of concentration performance are sensitive for variations in fluid intake and whether there are further cognitive abilities, which are influenced by fluid intake. Because this study focused on cognitive abilities which have shown to be associated with

children’s school achievements,16,17 studies might test associations between fluid intake and school grades.

Our study provides first evidence about the importance of drinking sufficient amounts at short time intervals. These results should be replicated and further investigated in comprehensive studies. To provide more detailed recommendations regarding fluid intake at school, it would be necessary to identify the optimal pattern of fluid intake, that is, how much fluid children should consume at school and at which time intervals. Moreover, external influences such as sports lessons or climatic conditions should be considered.

IMPLICATIONS FOR SCHOOL HEALTH

Our results have implications for drinking policies at schools. To facilitate cognitive functioning, an adequate fluid intake across the school day should be promoted. The introduction of health-promoting drinking policies is of key significance here. Drinking policies have a major impact on children’s drinking behavior. Evidence shows that children who drink water during lessons are more likely to reach the recommended daily level of fluid intake.5 This finding in combination with results of this study leads to the conclusion that schoolchildren should be allowed to drink during lessons in order to improve cognitive performance. However, efforts should go further still. Easy access to drinking water at school should be guaranteed. This could be realized by allowing children to decant water from taps or water fountains. A superior approach would be to provide free drinking water directly in the class room. An intervention study found that children who were provided free water bottles in the class room improved their drinking behavior compared to children from a control group who did not receive any water.22 Such structural measures should be combined with measures on the behavioral level.

Adequate water drinking behavior at school should be taught and developed. Quantitative and temporal aspects should be addressed. From a health perspective, the type of beverages should additionally be addressed to promote the consumption of energy-free beverages. Additionally, because thirst proved to be insufficient for regulating fluid intake in schoolchildren,23 it would be advisable for teachers to encourage pupils to drink across the school day.

Human Subjects Approval Statement The study was approved by the local school

authority (State Education Authority of the city of Göppingen) and the Institutional Review Board of the university. Moreover, parents provided written informed consent for their children to participate.

REFERENCES

1. Barker M, Benefer M, Russell J, Lepicard E, Constant F. Hydration deficit after breakfast intake among British schoolchildren. FASEB J. 2012;26(Suppl; Ib395).

2. Bonnet F, Lepicard EM, Cathrin L, et al. French children start their school day with a hydration deficit. Ann Nutr Metab. 2012; 60(4):257-263.

3. Assael BM, Cipolli M, Meneghelli I, et al. Italian children go to school with a hydration deficit. J Nutr Disorders Ther. 2012; 2(3):1-6.

4. Stookey JD, Brass B, Holliday A, Arieff A. What is the cell hydra- tion status of healthy children in the USA? Preliminary data on urine osmolality and water intake. Public Health Nutr. 2012; 15(11):2148-2156.

5. Kaushik A, Mullee MA, Bryant TN, Hill CM. A study of the association between children’s access to drinking water in primary schools and their fluid intake: can water be cool in school? Child Care Health Dev. 2007;33(4):409-415.

6. 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.

7. Heseker H. Untersuchungen zum Trinkverhalten von Schüler/innen und Studenten/innen. Available at: http://www. forumtrinkwasser.de/fileadmin/user_upload/Dokumente/Doku mente_Studien/Studie_Viertel_aller_Sch%C3%BCler_trinkt_n ichts.pdf. Accessed November 19, 2014.

8. Chumlea WC, Schubert CM, Reo NV, Sun SS, Siervogel RM. Total body water volume for white children and adolescents and anthropometric prediction equations: the Fels Longitudinal Study. Kidney Int. 2005;68(5):2317-2322.

9. Kleiner S. Water: an essential but overlooked nutrient. J Am Diet Assoc. 1999;99(4):200-206.

10. Grandjean A, Grandjean N. Dehydration and cognitive performance. J Am Coll Nutr. 2007;26(5 Suppl):549S-554S.

11. Ritz P, Berrut G. The importance of good hydration for day-to- day health. Nutr Rev. 2005;63(6):6-13.

12. Benton D, Burgess N. The effect of the consumption of water on the memory and attention of children. Appetite. 2009;53(1): 143-146.

13. Booth P, Taylor B, Edmonds C. Water supplementation improves visual attention and fine motor skills in schoolchil- dren. Educ Health. 2012;30(3):75-79.

14. Edmonds C, Burford D. Should children drink more water? Appetite. 2009;52(3):776-779.

15. Edmonds C, Jeffes B. Does having a drink help you think? 6-7- year-old children show improvements in cognitive performance from baseline to test after having a drink of water. Appetite. 2009;53(3):469-472.

16. Freudenthaler HH, Spinath B, Neubauer AC. Predicting school achievement in boys and girls. Eur J Pers. 2008;22(3):231-245.

17. Steinmayr R, Ziegler M, Träuble B. Do intelligence and sustained attention interact in predicting academic achievement? Learn Individ Differ. 2010;20(1):14-18.

18. Patel A, Hampton K. Encouraging consumption of water in school and child care settings: access, challenges, and strategies for improvement. Am J Public Health. 2011;101(8): 1370-1379.

19. Stillman P. Drinking in schools - a report on best practice for hydration in schools: expert group on hydration. Avail- able at: http://www.montgomeryspring.co.uk/pdf-downloads/ Drinking-in.Schools.pdf. Accessed November 8, 2014.

20. European Food Safety Authority. EFSA Panel on Dietetic Products, Nutrition, and Allergies (NDA). Scientific opinion on dietary reference values for water. EFSA J. 2010;8(3):1-48.

21. Muckelbauer R, Libuda L, Kersting M. Relative validity of a self-completion 24h recall questionnaire to assess beverage consumption among schoolchildren aged 7 to 9 years. Public Health Nutr. 2010;13(2):187-195.

22. Lührmann P, Simpson F, Sickinger S, Fuchs T, Dohnke B. The effects of the provision of mineral water in schools on pupils’ beverage consumption. Ernährungs Umschau. 2014;61(6):82-89.

23. Benton D. Dehydration influences mood and cognition: a plausible hypothesis? Nutrients. 2011;3(12):555-573.

24. Schmidt-Atzert L, Bühner M, Enders P. Messen Konzen- trationstests Konzentration? Eine Analyse der Komponenten von Konzentrationsleistungen. [Do concentration tests assess concentration? Analyzing components of concentration test performances]. Diagnostica. 2006;52(1):33-44.

25. Brickenkamp R. Test d2 Aufmerksamkeits-Belastungs-Test. Han- danweisung. 9., erweiterte und neu gestaltete Auflage. Göttingen, Germany: Hogrefe; 2002.

26. Oswald W, Roth E. Der Zahlen-Verbindungs-Test. 2. überarbeitete und erweiterte Auflage. Göttingen, Germany: Hogrefe; 1987.

27. Jäger A, Süß H-M, Beauducel A. Berliner Intelligenzstruktur-Test: BIS-Test, Form 4 [Test for the Berlin Intelligenzelligence Structure model (BIS)]. Göttingen, Germany: Hogrefe; 1997.

28. Aiken LS, West SG. Multiple Regression Testing and Interpretation. Newbury Park, CA: Sage Publications; 1996.