Physical activity has emerged as an important preventive action against chronic diseases. For this reason, various consensuses have established the necessity to persuade the general population to increase their physical activity patterns. In Greece, data on the prevalence of physical activity during leisure time are scarce, especially on the elderly population. Therefore, the aim of the present survey was to estimate the prevalence of leisure-time physical activity (LTPA) in a sample of elderly Greeks and the relationship with selected sociodemographic variables.

For the purposes of the study, a total of 156 subjects (101 men, 55 women, mean age 69.6±8.1 years) volunteered. Calculations of LTPA were based on the average self-reported time spent on various physical activities typically undertaken by older adults. Metabolic equivalents (METs) assigned to each activity were used to quantify the amount of LTPA by converting it into MET-hrs∙wk-1. Participants were classified according to gender, marital status, and the highest level of education achieved. In addition, for each subject the Body Mass Index (BMI) was calculated as the weight in kilograms divided by the squared height in meters. The WHO criteria were applied for the definition of overweight and obesity. One-way ANCOVA and chi-square test were used for statistical analyses.

Men showed a slightly higher prevalence of LTPA (57.8%) compared with women (49.1%). Likewise, men presented a notably higher mean of MET-hours a week than women (20.2 MET-hrs∙wk-1 vs. 8.7 MET-hrs∙wk-1) (P=0.003). No significant associations were found between MET-hrs∙wk-1 and marital status. No relationship was observed between prevalence or amount of physical activity during leisure time and educational or smoking status. The previously reported inverse association of BMI and physical activity was not confirmed in the present study (s. Table 1). Similar results were observed when the analyses were carried out separately for men and women, except for women grouped in the category "widowed/divorced", which exhibited higher LTPA levels than married women (15.5 vs. 4.7 MET-hrs∙wk-1, p=0.025

Discussion/ Conclusions
Our data indicate that Greek elderly men had significantly higher levels of physical activity when compared with women. On the other hand, a high percentage of Greek people older than 50 yr. remain physically inactive. This finding underlines the need for defining intervention strategies attempting to improve the health of older Greeks through increased physical activity. The finding of no association between physical activity and BMI is consistent with findings of other researchers.

[1]. Dipietro L et al (1993). Med Sci Sports Exerc 25, 628-42.
[2]. Washburn RA et al (1993). J Clin Epidemiol 46, 153-62
[3]. Martinez-Gonzalez MA et al (2001). Med Sci Sports Exerc 7, 1142-6.

NOTA: O texto com a iconografia está no anexo.