ORIGINAL ARTICLE Trends in resting pulse rate among students attending Glasgow University between 1948 and 1968: analyses of cross sectional surveys G Davey Smith 1 , B Galobardes 1 , M Jeffreys 2 and P McCarron 3 1 Department of Social Medicine, University of Bristol, Canynge Hall, Whiteladies Road, Bristol, BS8 2PR, UK; 2 Centre for Public Health Research, Massey University (WGTN Campus), Private Box 756, Wellington, New Zealand and 3 Department of Epidemiology and Public Health, Mulhouse Building, Grosvenor Road, Belfast, BT12 6BJ, UK Evidence on long-term trends in physical activity is limited. We report that resting pulse rates – a proxy indicator of physical activity and fitness – increased among young adults attending Glasgow University between 1948 and 1968. International Journal of Obesity (2006) 30, 380–381. doi:10.1038/sj.ijo.0803139; published online 25 October 2005 Keywords: pulse rate; secular trends; physical activity; physical fitness Introduction The epidemic of obesity has focussed attention on changes in energy intake and energy expenditure, but no consensus has been reached regarding their relative contribution. 1 Evidence on long-term trends in physical activity is limited and contradictory, particularly when overall physical activity – combining leisure-time and occupational activity – is considered. Resting pulse rate is inversely associated with amount of physical activity 2 and thus long-term trends in pulse rate can provide evidence of changes in physical activity otherwise difficult to obtain. We have therefore examined this issue in the Glasgow Alumni Study. Methods and results Students attending Glasgow University between 1948 and 1968 were invited to participate in a health survey consisting of a questionnaire and clinical examination. 3 The 15 322 students who participated in these examinations represents almost 50 percent of the Glasgow student intake over this period. The gender and faculty distribution of the partici- pants is similar to that of the entire student population, with a slight excess of medical students in the participating group. The excess of medical students is because while the examinations were voluntary for other students, for medical students they were, for a time, compulsory, since medical students had to have chest x-rays for tuberculosis. Data collected by physicians included socio-demographic details, health behaviours, and medical history. Height, weight, blood pressure and resting pulse rate were also recorded. A total of 15 322 students participated (11 756 men and 3566 women). We excluded 2 with date of birth missing, 1640 who were older than 25 at the time of examination and 414 for whom pulse rate was not known. Extreme values of pulse rate, 4165 and o45 (n ¼ 13), were also excluded as they might reflect data recording errors. We assessed trends in pulse rate with linear regression analysis with year of birth as a continuous variable. The mean pulse rate was also estimated with year of birth categorized in 6 groups and continuous variables centred at their mean values. Fully adjusted models included age, height, body mass index (BMI), smoking, and father’s social class, with the addition of age at menarche in women, measured at University. The table presents pulse rate by year of birth group (Table 1). A clear increase is seen, and the results are little influenced by adjustment for potential confounders. For each successive birth year the change in resting pulse rate in fully adjusted models was 0.17 (95% confidence interval (CI) 0.14–0.21) for men and 0.08 (95% CI 0.03–0.13) for women. Received 21 February 2005; revised 30 August 2005; accepted 9 September 2005; published online 25 October 2005 Correspondence: Professor G Davey Smith, Department of Social Medicine, University of Bristol, Canynge Hall, Whiteladies Road, Bristol, BS8 2PR, UK. E-mail: zetkin@bristol.ac.uk International Journal of Obesity (2006) 30, 380–381 & 2006 Nature Publishing Group All rights reserved 0307-0565/06 $30.00 www.nature.com/ijo