¹The Rowett Research Institute, Bucksburn, Aberdeen, UK

²Biomathematics and Statistics, The Rowett Research Institute, Bucksburn, Aberdeen, UK

³BioPsychology Group, Psychology Department, University of Leeds, Leeds, UK

Correspondence to: R J Stubbs, The Rowett Research Institute, Greenburn Road, Bucksburn, Aberdeen AB21 9SB, UK. E-mail: J.Stubbs@rri.sari.ac.uk

^†Part of this work was presented as an abstract at the Nutrition Society Summer Symposium, University of Surrey, Guildford, June 1998.

Aim: We assessed the effect of graded increases in exercised-induced energy expenditure (EE) on appetite, daily energy intake (EI), total daily EE and body weight in six lean women using a within-subject, repeated measures design.

Method: Subjects were each studied three times during 7 day treatments, corresponding to no-exercise (control; Nex; 0 MJ/day), medium exercise level (Mex; ~1.9 MJ/day) and high exercise level (Hex; ~3.4 MJ/day), with 2 day maintenance beforehand. Subjects self-weighed ad libitum food intake. EE was assessed by continual heart rate monitoring. During waking hours subjects recorded hourly sensations of hunger and appetite.

Results: EE amounted to 9.2, 11.0 and 12.1 MJ/day (F (2, 10)=5.67; P=0.023 (s.e.d.=0.87)) on the Nex, Mex and Hex treatments, respectively. The corresponding values for EI were 8.9, 9.2 and 10.0 MJ/day (F (2, 10)=4.80; P=0.035 (s.e.d.=0.36)). There were very weak treatment effects on hunger. Weight loss was significantly different from zero on the Mex and Hex treatments.

Conclusion: Markedly increasing EE through exercise produced significant but partial compensations in EI (~33% of EE due to exercise). Accurate adjustments of El to acute increases in EE are likely to take weeks rather than days.

International Journal of Obesity (2002) 26, 866-869. doi:10.1038/sj.ijo.0801874

exercise; appetite; human; energy balance; feeding behaviour