Psychopharmacology (2002) 164:188–192 DOI 10.1007/s00213-002-1175-2
ORIGINAL INVESTIGATION
Carolyn F. Brice · Andrew P. Smith
Effects of caffeine on mood and performance: a study of realistic consumption Received: 23 January 2002 / Accepted: 17 June 2002 / Published online: 4 September 2002 Springer-Verlag 2002
Abstract Rationale: There is a vast literature on the behavioural effects of caffeine. Many of the studies have involved single administration of a large dose of caffeine that is not representative of the way in which caffeine is usually ingested. Further information is required, therefore, on the behavioural effects of realistic patterns of consumption. Objectives: The present study aimed to determine whether a realistic drinking regime (multiple small doses – 465 mg over a 5-h period) produced the same effects as a single large dose (200 mg). The smaller doses were selected so that the amount of caffeine present in the body after 5 h would be equivalent to that found with the single dose. Methods: A double-blind, placebocontrolled, within-subjects experiment was, therefore, carried out. The participants (n=24) attended for four sessions. Each session started with a baseline measurement of mood and performance at 0930 hours. On two of the sessions, coffee was then consumed at 1000, 1100, 1200 and 1300 hours. In one of these sessions 65 mg caffeine was added to the de-caffeinated coffee. In the other two sessions, the participants consumed coffee at 1300 hours and 200 mg caffeine was added in one of the sessions. The volunteers completed the battery of tests again at 1500 hours. Results: The results showed that in both consumption regimes caffeine led to increased alertness and anxiety and improved performance on simple and choice reactive tasks, a cognitive vigilance task, a task requiring sustained response and a dual task involving tracking and target detection. Conclusions: These results suggest that previous findings from studies using a large single dose may be applicable to normal patterns of caffeine consumption.
C.F. Brice · A.P. Smith ()) Centre for Occupational and Health Psychology, School of Psychology, Cardiff University, 63 Park Place, Cardiff CF10 1AS, UK e-mail:
[email protected] Tel.: +44-29-20874757 Fax: +44-29-20874658
Keywords Caffeine · Mood · Reaction time · Sustained attention · Dual tasks · Performance
Introduction There is now a very large literature on the behavioural effects of caffeine (see Lieberman 1992; Fredholm et al. 1999; Smith 2002, for reviews). Recent research has addressed the issue of whether results from previous laboratory studies are applicable to real life. For example, it has been demonstrated that effects of caffeine on artificial laboratory tasks may also be observed in simulations of real-life activities (Horne and Reyner 1996; Brice and Smith 2001a). Similarly it has been shown that effects of caffeine may be apparent when doses typically found in a single beverage are used (Durlach 1998; Smith A et al. 1999; Smit and Rogers 2000). In addition, effects of caffeine do not appear to be modified by the beverage in which it is given (Smith A et al. 1999). One problem with generalising from many laboratory studies to real life is that they often use a single large dose of caffeine (often equivalent to a person’s total daily intake). Caffeine is not usually ingested in this way, rather it is consumed in several small doses over the course of the day (Brice and Smith 2002). The aim of the present study was to compare the effects of several small doses of caffeine with those produced by a single large dose (the smaller doses leading to equivalent caffeine levels in the body to those produced by the single dose). The effects of these different forms of administering caffeine were assessed using a range of measures known to be sensitive to caffeine. It was predicted on the basis of our previous research, and the findings of other researchers, that caffeine would increase alertness (Smith et al. 1993; Warburton 1995; Brice and Smith 2001a), increase anxiety (Loke 1988; Smith A et al. 1999), reduce simple reaction time (Clubley et al. 1979; Smith et al. 1994; Smith et al. 1997) and improve sustained attention (Smith et al. 1992; Brice and Smith 2001b), dual-task
189
performance (Kerr et al. 1991) and the encoding of new information (Smith AP et al. 1999). The aim of the present study was to determine whether these effects would be observed with both administration of multiple small doses and a single large dose of caffeine.
coffee at 1300 hours, participants were allowed to go for lunch as normal and then returned to the lab at 1455 hours, when a second saliva sample was taken. Details of what they had eaten and drunk during the morning and at lunch were recorded at this time. At 1500 hours, participants began the second test session. On completion of the second test session, participants were required to answer a second questionnaire to ascertain their perceptions of whether they believed that they had previously received caffeinated or de-caffeinated coffee.
Materials and methods Design
Matching of the caffeine doses
A within-subjects design was employed with each participant attending four test sessions on separate days. On 2 days, the participant received a single dose of either 200 mg caffeine or placebo, and on the remaining 2 days participants received four smaller doses of 65 mg caffeine or placebo. The order of caffeine condition presentation was counterbalanced with 50% of participants performing first under the caffeine condition and 50% under the placebo condition. Order of single/ multiple dose was also counterbalanced, as was the order in which the tests were performed. Administration of the caffeine was double blind.
A pharmacokinetic analysis to ascertain equivalent dosage for the two conditions was conducted [utilising a linear two-compartment model (parametric population PK–PD model) employing the computer programme NONMEM version IV, level 1.0 (Shi et al. 1993)]. The amount of caffeine assigned to the two dose conditions was calculated to ensure that the same levels of caffeine were present in the body at the final test session. From this analysis, it was ascertained that over a 5-h consumption period the caffeine dosage for each condition would be: a single dose of 200 mg (consumed at 1300 hours) and four doses of 65 mg (consumed at 1000, 1100, 1200 and 1300 hours).
Participants
Nature of the drinks
Twenty-four male participants were recruited from the student population of the University of Bristol. The mean age of the sample was 20.96 years (range 19–23 years); the mean weight was 71.91 kg (range 60–86 kg); and the mean daily caffeine consumption, assessed by self-report, 210 mg (range 100–375 mg). In order to directly compare the two methods of administering the caffeine conditions, it was essential to control for factors known to influence caffeine metabolism (Brice and Smith 2001b). Participants were not eligible for the study if they were female, taking medication or if they smoked. All included participants were required to sign a consent form that outlined the experiment, explained that they were free to withdraw at any time and confirmed the confidentiality of all information. The study was approved by the local ethics committee. Participants were paid £25.00 (pounds sterling) on completion of the study.
For both conditions all drinks were made with one rounded teaspoon of decaffeinated coffee in a 150-ml mug of boiling water. Milk and sugar were added in accordance with the participants’ usual preferences and this was recorded. Saliva samples Throughout this study caffeine levels were monitored via saliva samples collected through salivettes (Newton et al. 1981). An Emit caffeine assay (a homogenous enzyme immunoassay) was used to determine caffeine levels. Mood and performance All of the tasks are described in detail in Brice and Smith (2001b).
Procedure Prior to the experimental sessions, all participants were familiarised on the laboratory tests to ensure a full understanding of each of the performance tasks. On the evening prior to each testing day, participants were required to limit their alcohol intake to a maximum of four units, and on each test day participants were required to abstain from alcohol consumption, performing any strenuous physical exercise and consuming any caffeinated products. Testing took place on four separate days, at consistent times of day, for all participants and conditions. The test-day schedule for each dose condition initially assumed the same format. Upon arrival at 0920 hours, participants provided a baseline saliva sample. At 0930 hours, participants performed a 30-min battery of the computer tasks. Between 1000 hours and 1300 hours, the testing schedule varied for the multiple-/single-dose conditions. In the multiple-dose condition, participants consumed their first cup of coffee immediately after completing the performance tasks at 1000 hours. When they had consumed all of the drink, participants were allowed to go about their normal everyday activities. Participants were reminded not to consume any caffeinated products. They were then required to return to the lab at 1100, 1200 and 1300 hours for a cup of coffee. For the single-dose condition, participants were allowed to go about their normal everyday activities at 1000 hours, and then returned at 1300 hours for a cup of coffee. From 1300 hours onwards, the testing schedule for both dose conditions resumed the same format. After drinking
Mood Mood was assessed both pre- and post-performance using 18 computerised visual analogue mood rating scales. Three main factors were derived from these scales; alertness, hedonic tone and anxiety. Performance Dual tracking/detection task. This task measured hand–eye coordination and dual task attention and lasted for approximately 3 min. For the target detection component of the task, the variables measured were mean reaction time for correct hits, number of correct and late responses (>1200 ms but
