European Journal of Applied Physiology

, Volume 91, Issue 4, pp 493–498

The effect of drinking tea at high altitude on hydration status and mood


    • Colworth House Unilever R&D Colworth
  • Jane A. Rycroft
    • Colworth House Unilever R&D Colworth
  • Jennifer Aspen
    • Colworth House Unilever R&D Colworth
  • Clare Chapman
    • Colworth House Unilever R&D Colworth
  • Bryce Brown
    • Section of Emergency MedicineUniversity of Manitoba
Original Article

DOI: 10.1007/s00421-003-1015-z

Cite this article as:
Scott, D., Rycroft, J.A., Aspen, J. et al. Eur J Appl Physiol (2004) 91: 493. doi:10.1007/s00421-003-1015-z


The effect of drinking tea on hydration status and mood was studied in nine male and four female members of expeditions based at Mt. Everest base camp at an altitude of 5,345 m. Whilst exposed to altitude-cold diuresis, participants were subjected to a crossover experimental design comprising two 24-h dietary interventions. In the “tea” condition, hot brewed tea formed a major part of fluid intake, whereas in the “no-tea” condition tea was excluded from the diet. Subjects were prohibited in both cases from consuming other caffeinated beverages, caffeinated foods, and alcoholic drinks. Mean fluids ingested [mean (SE); tea=3,193 (259) ml versus no tea=3,108 (269) ml] and urine volume (tea=2,686 (276) ml versus no tea=2,625 (342) ml] were similar under both conditions. Statistical analysis found no difference in urine stimulated as a result of the tea intervention (P=0.81). Several markers of hydration status were also taken immediately pre and post each condition, including measures of urine specific gravity, urine electrolyte balance (K+, Na+), and urine colour. None of these measures indicated a difference in hydration status as a result of the dietary intervention in either the control or tea condition. A difference was, however, found in mood, with subjects reporting reduced fatigue when tea was included in the diet (P=0.005). The study shows therefore that even when drunk at high altitude where fluid balance is stressed, there is no evidence that tea acts as a diuretic when consumed through natural routes of ingestion by regular tea drinkers, but that it does have a positive effect on mood.


High altitudeHydrationMoodTea


At high altitude, humans experience a loss of body water via altitude-cold diuresis, high energy expenditures, and often poor access to water (Askew 1989; Consalazio et al. 1968). Many studies have shown that such dehydration is associated with impaired physiological performance (e.g. Cadwell et al. 1984; Sawka and Pandolf 1990). Effects of dehydration include decreased muscular strength (Craig and Cumming 1966), reduced lactate production (Houston et al. 1981), and a lowered capacity for aerobic-physical work (Saltin 1964). Under the demanding physical conditions endured by mountaineers and trekkers at high altitude, such decrements of performance are both undesirable and potentially dangerous. It is therefore important that the body is properly rehydrated whilst climbing, trekking, and acclimatizing to altitude.

In order to replenish bodily fluids at altitude several beverages are generally available. Water is widely consumed, typically boiled or chemically treated to remove bacteria, and often mixed with powdered flavours. But a second important source of fluids is tea. Tea is both popular with visiting mountaineers and a staple beverage of the indigenous population of many high altitude parts of the world, such as the Sherpas of the Nepal Himalaya. Despite the tradition and popularity of tea, concerns have been raised about its use in situations where maintenance of fluid balance is important. Tea provides valuable fluid, but it also contains a number of biologically active compounds, including the methylxanthines caffeine, theobromine, and theophylline. Tea usually contains caffeine in significant amounts. Caffeine content in a typical 200-ml tea infusion has been estimated by MAFF (1998) at 40 mg and by Ziebler and Filer (1996) at 59 mg. Pharmacologically, caffeine is known to have a stimulant action on the body and mind. It has been shown to enhance vigilance and improve mood (Lieberman et al. 1987), but also, under certain conditions, to stimulate the production of urine. It is therefore important to examine the question of the effects of tea at altitude, and to consider whether it is appropriate as a source of fluids in this context.

The literature on the effects of tea is, however, not straightforward. Some studies recommend that caffeinated beverages such as tea should be avoided on account of their diuretic action (e.g. Robertson et al. 1978; Passmore et al. 1987; Neuhauser-Berthold et al. 1997). But others challenge these recommendations, showing that caffeine when consumed in beverages under more naturalistic conditions does not have a diuretic effect (e.g. Grandjean et al. 2000). Several factors complicate this issue and may explain the inconsistency. One is the dose size. Where the dose relationship has been systematically investigated it is only at a high dose of 360 mg that a diuretic action is found (Passmore et al. 1987). A recent systematic review of the accumulated evidence has shown that acute diuretic effects are observed more generally only in cases where at least 300 mg of caffeine is ingested (Maughan and Griffin 2004). This finding suggests that the effect is not linear, occurring only above a critical threshold equivalent to the caffeine contained in five to eight cups of tea. A second factor is habituation. Studies based on experimental models routinely impose extended washout periods, thus testing caffeine-withdrawn or -naïve individuals. It is known, however, that regular caffeine users become habituated to the effects of caffeine, thus diminishing its action. Hence, previously published results may exaggerate caffeine effects compared with consumption by regular users under everyday circumstances. Further issues arise from the use of caffeine itself in many published studies rather than caffeinated beverages like tea or coffee. The dose in the former is provided in the absence of the accompanying fluid that would be a necessary part of ingesting high doses of the substance in a caffeinated beverage. This method also ignores the many other active compounds present in beverages such as tea and coffee, and so it is not known what additional effects these may have. The question of whether tea does or does not act as a diuretic is therefore complex, and to answer it properly requires, in addition to controlled laboratory experiments, naturalistic studies employing realistic consumption conditions.

The psychological effects of caffeine and caffeinated beverages such as tea have also been widely studied, and a range of effects has consistently been shown. Compared with a placebo, caffeine has been found to increase self-ratings of alertness, to improve mood, and to enhance psychomotor and cognitive performance, including effects on reaction time, sustained attention, memory and logical reasoning (e.g. Lieberman et al. 1987; Fagan et al. 1988; Smith et al. 1993; Quinlan et al. 1997). Tea, moreover, has been shown to have additional social psychological effects. Ethnographic research has highlighted the important role tea plays in social rituals, particularly those in which one takes time out from stressful situations (Lee 1999). Research using diaries has found associations between tea drinking and feeling more relaxed over the course of the day, with the strength of these associations influenced by factors such as gender and coping resources (Steptoe and Wardle 1999). In addition to the possible physiological impact on hydration, therefore, drinking tea may for a number of reasons also have a psychological impact on mood. Although there has been some research on the effects of altitude on mood (e.g. Shukitt-Hale et al. 1990), showing a tendency for there to be adverse changes in some mood factors with increasing altitude, there has not as yet been any research on the effects of particular beverages on mood whilst at altitude.

The purpose of the present study is to investigate the effects of drinking tea at high altitude on hydration status and mood, using an experimental design that closely resembles the conditions in which the beverage is widely drunk. The objectives are twofold. First, the study aims to investigate whether there is any evidence to show that tea does lead to body water loss under these conditions, and hence should be avoided by mountaineers and trekkers at altitude. Second, the study investigates the effects of tea consumption on mood, to test whether it has any positive psychological effects at high altitude. The paper reports data collected over a 10-day period whilst at Mt. Everest base camp at an altitude of 5,345 m.



A total of 9 men and 4 women volunteered to take part in the study. Mean age was 33.5 years, the age range 18–51 years. Volunteers held supporting roles in expeditions attempting to climb either Mt. Everest or Mt. Lhotse. All were on rest days whilst participating in the study and remained at Mt. Everest base camp at an altitude of 5,345 m for the duration. Subjects had no significant past medical history of hepatic, renal, cardiac, pulmonary, digestive, haematological, neurological, thyroidal or psychiatric disease. Anyone taking medications that might influence weight or fluid and electrolyte balance (e.g. over-the-counter medications or supplements, and especially those containing methylxanthines) was excluded. All subjects were regular tea drinkers. The study was approved by the ethics committee of Unilever Research and Development Colworth and subjects provided written informed consent.


The study used an experimental design comprising two 24-h conditions. In both conditions participants were prohibited during the 12 h prior to the start of testing from consuming alcohol or caffeinated drinks or food, including tea, coffee, and chocolate. This washout period ran from 2000 hours to the start of the 24-h test period at 0800 hours. Since this was the time when most subjects usually slept, the standardized washout period represented a minimal disruption from their normal routine. In the “tea” condition participants were required during the course of the day to consume a minimum of two mugs of tea using vessels of known volume (400 ml) supplied by the investigators. Subjects used ‘Taj Mahal’ tea bags containing a blend of Indian black teas and principally Assam tea. The tea was analysed to test for the levels of methylxanthines present using an HPLC system (Dionex) with UV detection at 274 nm. Caffeine was present at 67.3 mg/g tea solids and theobromine at 4.8 mg/g. No trace of theophylline was found. To ensure the results of the study reflect the effects of naturalistic tea drinking practices, the tea was prepared and consumed according to subjects’ own taste. To remain further consistent with natural behaviour under these conditions, no further constraints were placed on tea consumption and no upper limit was imposed on participants.

The second, “no-tea” condition required participants to abstain from drinking tea in addition to the other prohibited foods and drinks (i.e. alcohol and caffeinated drinks or food, including coffee, and chocolate). A complete crossover design was used. Participants were assigned to one of the two conditions according to a randomized schedule, beginning the trial with either the tea or no-tea condition to balance for order effects. Beverages were consumed by participants individually in the expeditions from which they were recruited rather than as an experimental group; hence there was no influence of any systematic variation resulting from group behaviours promoted by the different experimental conditions.


Prior to the testing day, subjects were supplied with a beverage diary, and two graduated bottles, one for measuring fluids consumed and the other urine. Subjects were asked to record in the diary the type and quantity of every beverage ingested over the 24-h period as well as the volumes of every bladder void. Fluid intake was measured using the graduated bottles and the 400-ml cups. This provided a 24-h measure of total fluid volume ingested and urine excreted.

Subjects were also supplied with a separate bottle to retain the first void of the morning at both the start and end of the 24-h test period. These samples were tested within an hour of the void for specific gravity (Uspecific gravity)and pH (UpH), using reagent strips for urine analysis (Labstix SG; Bayer). Urine colour (Ucolour)was estimated from a colour chart with eight levels (Armstrong 2000). These measures have been shown to be both practical in a field setting and relatively sensitive to moderate levels of hypohydration (Armstrong et al. 1994; Shirreffs 2000). They are, moreover, reliably intercorrelated, with recent data showing correlations between Ucolour and Uspecific gravity of r2=0.88–0.98 (Ormerod et al. 2003). Aliquots (15 ml) were additionally taken from the morning void. All specimens were preserved in chilled conditions in insulated containers for the duration of the expedition and, on completion, were stored with dry ice and transported by air to the laboratory. They were subsequently analysed for sodium and potassium by Bedford General Hospital (Cobas Integra, Roche Diagnostics).

To measure mood, ratings were made at the end of each testing day on 12 adjectives derived from the Profile of Mood States (POMS; McNair et al. 1981). The highest loading items from each of the seven sub-scales of the POMS were selected. Participants rated these adjectives in response to the question ‘How have you been feeling generally today?’ using a 10-cm non-graduated line as a visual analogue scale with endpoints ‘Not at all’ and ‘Extremely’. The line was subsequently measured in millimetres to form a 101-point scale. Adjectives were rated representing the factors vigour (indicated by the adjectives alert and energetic), depression-dejection (miserable, hopeless), anger-hostility (bad-tempered, annoyed), relaxation (relaxed), fatigue (fatigued), confusion (confused), and friendliness (friendly, good-natured). Where the factor was represented by two items the mean was computed.

Subjects also completed a short questionnaire to measure any deleterious effects of altitude experienced whilst participating in the study, based on the Lake Louise symptoms of acute mountain sickness (Hackett and Oelz 1992). This includes items on gastrointestinal symptoms (anorexia, nausea or vomiting), fatigue or weakness, dizziness or light-headedness, and difficulty sleeping.


Statistical analysis was conducted using paired Student’s t-tests after appropriate manipulation of the data. To test the urinary measures of hydration status (Uspecific gravity, Ucolour, UpH) the difference was computed between ‘pre’ and ‘post’ time points as a measure of the effect of each condition, and then the difference between these measures for the two conditions was analysed. Fluid volumes were analysed by comparing the ratios of urine volume to total fluid intake for the tea and no-tea conditions. This data did not follow a normal distribution so the log of the values was taken and subjected to analysis. The difference in mood measures between conditions was compared directly. A probability level of 0.05 was set as statistical significance.


All 13 subjects recruited to the study successfully completed both conditions. None experienced any adverse effects as a result of the experimental intervention or any symptoms of acute mountain sickness during the study.

Urinary indices were available as an indication of the hydration status of participants at the start of the trials. In the tea condition, Uspecific gravity was 1.022 (0.002) [mean (SE)] and Ucolour was 2.04 (0.30). For the no-tea condition Uspecific gravity was 1.021 (0.002) and Ucolour was 1.88 (0.29). Neither measure indicated a significant difference in the starting hydration status of subjects for the two conditions.

The total volume of fluid consumed in 24 h in the no-tea condition was 3,108 (269) ml and in the tea condition was 3,193 (259) ml. In the latter, the quantity of tea consumed was 1,735 (181) ml. This represents 56.5% of the total fluids ingested during the 24 h of the study. Urine volume in the no-tea condition was 2,625 (342) ml and in the tea condition was 2,686 (276) ml. The percentage of fluid retained by the body after the no-tea condition was therefore 15.5%, and after the tea condition was 15.9% (Armstrong 2002). Statistical analysis of the ratio of total urine to total fluid intake was non-significant (P=0.81). Total urine was also modelled using a mixed modelling technique using total fluid intake as a covariate. This form of analysis also concluded no difference between treatment means. Data derived from the urinary analysis are given in Table 1.
Table 1

Effect of experimental conditions on urinary measures of hydration after 24 h. Values are mean (SE). No differences were significant, P<0.05

Urinary measure

No-tea condition

Tea condition







Specific gravity

1.021 (0.002)

1.024 (0.001)


1.022 (002)

1.025 (0.001)


Urine colour

1.9 (0.3)

2.0 (0.2)


2.0 (0.3)

2.2 (0.4)


Sodium (mmol/l)

94.1 (11.4)

110.8 (14.8)


86.0 (12.3)

102.4 (10.9)


Potassium (mmol/l)

30.1 (5.8)

29.2 (4.1)


22.3 (3.7)

29.7 (5.0)


No difference was found in any of the urinary measures of hydration status in the differences resulting from the tea compared with the no-tea condition. Likewise, there was no statistically significant difference found in six of the seven mood factors (see Fig. 1). There was, however, a difference in fatigue, which was lower in the tea condition than the no-tea condition (P=0.005).
Fig. 1

Difference in the seven mood factors in the tea compared with the no-tea condition


The purpose of this study was to test the effects on hydration and mood of consuming tea at high altitude. The study was designed to investigate possible diuretic effects of tea in a context where the beverage is commonly drunk and where the fluid balance of the body is stressed. The results suggest that tea consumed under the quasi-naturalistic conditions of the present study does not promote fluid loss and dehydration. Using two different types of hydration measure, one based on 24-h urine volume and the other on analysis of early-morning voids, no differences were found as a consequence of drinking tea. An effect was, however, found on mood, with subjects reporting feeling less fatigued during the day on which they drank tea.

Research that has looked at the diuretic effects of caffeine has shown that caffeine exerts a diuretic action only above a threshold of 300 mg (Maughan and Griffin 2004), the equivalent of five to eight cups for a typical infusion of tea. In the present study participants consumed a mean of eight to nine cups, over the course of 24 h. Hence the results appear broadly consistent with previous findings based on data at sea level. However, since the focus of the study was on tea as naturally infused and drunk at altitude rather than on the effects of caffeine per se, caffeine levels were not controlled or measured. The results relate therefore to the consumption of tea rather than to the ingestion of caffeine. It is only possible to speculate about the actual dose of caffeine ingested, and further research is required to determine the precise level of caffeine in tea needed to induce diuresis under these conditions.

Several factors other than the caffeine levels may have also contributed to the present findings, especially in comparison with previous laboratory experimental designs. Since the study aimed at studying an ecologically valid scenario, regular drinkers of caffeinated beverages were used, and they were subjected only to minimal, overnight washout conditions. Hence, subjects were only minimally, if at all, caffeine withdrawn at the beginning of the test periods. It is well known that the effects of caffeine are subject to habituation and with regular consumption tolerance develops to its effects (e.g. Denaro et al. 1991). Caffeine tolerance may well have contributed, therefore, to the lack of any diuretic effects in the study. Though this is more representative of the typical situation for regular tea drinkers, it does not address the question of the effects of tea on non-tea drinkers. The object of the investigation was moreover the complete tea beverage rather than isolated components of tea such as caffeine. Any caffeine ingested was therefore consumed along with the necessary accompanying fluid and other active components in tea consistent with that dose. Hence, the effect of tea was tested under conditions in which it would be experienced in natural circumstances and this may have had an effect in comparison with studies using pure caffeine.

A further potential influence on the results was the initial hydration status of participants. It has been shown that in the case of the potent diuretic alcohol, when subjects are even mildly fluid deprived a significant proportion of fluid ingested is retained when the levels of alcohol are low (Eggleton 1941). It was expected therefore that under conditions of high altitude, hypohydration might have mediated any diuretic effects. However, at the start of the two conditions, participants were found not to be dehydrated. Uspecific gravity was 1.021 and 1.022, which is within the typical range of 1.013–1.029 for normal specimens for healthy adults, and well below 1.030, the level that indicates dehydration or hypohydration (Armstrong et al. 1994). This finding is furthermore consistent with the Ucolour results of 1.88 and 2.04 compared with scores of 1–3 on the Ucolour scale that correspond to adequate hydration (Armstrong 2000). The total fluid intake of participants was relatively high with a mean of 3,150 ml across the two conditions and this appears to have been sufficient to counter the diuresis typically experienced at altitude (Askew 1989). The results therefore relate to high-altitude conditions in which hydration status is stressed, but not where participants are already dehydrated. It is not clear from the results of the present study whether tea has a similar effect to alcohol on increased fluid retention under these circumstances.

The effect of tea on mood was only partly consistent with previous findings in the literature. Subjects felt less fatigued when drinking tea, but there were no differences in any of the other mood factors between the two conditions. Typically, caffeinated beverages increase objective measures of alertness under acute conditions (e.g. Lieberman et al. 1987) and prevent a steady decline of alertness when drunk over the course of the day (Hindmarch et al. 1998). A subjective measure of feelings of alertness was included in the vigour factor of the POMS and hence a difference in this factor was expected. Though there was a trend in the predicted direction, this failed to reach statistical significance. Likewise tea has been found to enhance relaxation (e.g. Steptoe and Wardle 1999), but this was not reflected in any difference in the tension-anxiety factor in the present case. Any effects may have been moderated, however, by factors such as gender and social support, which were not explored in the present study. More work is required before further conclusions can be drawn about the effect of tea on mood at altitude.

The results have implications for both theoretical debates on the diuretic and mood effects of caffeinated beverages and practical applications for dietary recommendations for those attempting to maintain their hydration status whilst climbing or trekking at altitude. The recommendations that follow from the present study extend to high altitude those proposed by Maughan and Griffin (2004). Since no evidence was found to support the view that tea causes a significant increase in the water lost in the urine, the study suggests that tea makes a useful contribution to the hydration status of the body. For regular tea drinkers there is therefore no justification for recommending that tea consumption be restricted at altitude. Moreover, in the context of conditions typically experienced at high altitude where dietary choice is limited and appetite may be suppressed, it is especially important that a varied, desirable, and nutritious diet is available. Tea is practical and generally obtainable, and so forms a valuable part of the limited sources of fluid on offer. Moreover, tea has additional health benefits since it is a rich source of flavonoids, which contribute to the daily intake of antioxidants (Wiseman et al. 1997). Since the other major dietary sources of antioxidants are fruit and vegetables, and these are often in short supply in remote regions of the world, tea is of particular value.

In conclusion, the data reported in the present study are consistent with findings discussed previously in the literature that suggest the caffeine found in tea when consumed under typical conditions by regular tea drinkers is insufficient to promote dehydration. The present study provides evidence to support the extension of this conclusion to a situation where fluid balance is stressed, and additionally showed a positive effect of tea in the reduction of fatigue. The study shows therefore that even when drunk at high altitude, where fluid balance might be compromised, there is no evidence that tea acts as a diuretic, when consumed through natural routes of ingestion by regular tea drinkers, but that it does have a positive effect on mood.


The authors wish to thank Matt Rowson, Alex Faupel, Caroline Booker, Joan Lane, Jacqui Coverly, Karen Lovett and Sian James from Unilever R&D Colworth; Todd Burleson, Gordon Janow, Willie Prittie, Vern Tejas, Jibhan Gimire and Bharat Karki from Alpine Ascents International; and the participants in the study for their efforts beyond the call of duty. The experiment completed for this research complies with the current laws of Nepal, where it was conducted.

Copyright information

© Springer-Verlag 2004