Introduction An active lifestyle, characterized by a regular pattern of leisure time physical activity along the lifespan, improves cognitive functioning and mental health. To join in sports activities or exercise programs is associated with a better quality of life, and people who engage in these activities show not only a better physical fitness, but a better psychological and emotional wellbeing. An active lifestyle provides an apparent protective effect against the incidence of disorders such as anxiety or depression. However, in this regard, the dose-response in terms of what type, amount, duration and intensity of physical activity is necessary for achieving improvements has not been well established in studies with large samples. Objetives The aim of this study is to provide evidence that, from a descriptive point of view, shows the association between different indicators of mental health in the Spanish adult population and leisure time physical activity practice, also assessing the level of this leisure time physical activity. Methods Data were obtained from the last Spanish National Health Survey, a cross-sectional study conducted by the Ministry of Health and Social Policy Institute for Health Information-Spain, between June 2006 and June 2007. The sample was representative of the Spanish adult population. Persons over 16 years took part in this study, being the reference population available in the last official census (January 1, 2006). A total of 29 478 persons (11 645 men and 17 833 women, older than 16 years) were interviewed using the National Health Survey Questionnaire 2006. It was designed as multistage sampling, stratified cluster in which the first unit: were census tracts distributed by population size; the second stage units are the main family dwellings, and the last unit was selected as a person over 16 years from each dwelling. The census tracts studied were selected within each stratum with proportional probability to size with the goal of this whole procedure would lead to self-weighting samples in each stratum, and households and individuals were selected by a random procedure, taking into account sampling quotas for age and sex. The sampling error for a confidence interval was 95.5% +/- 0.6%. To minimize seasonal biases in the study, in terms of morbidity and lifestyle, the questionnaire was administered in four phases over the year: each stage involved 15 days, and the reference period of each variable explored comprised two weeks one year from the day of data collection. The gathering of information was conducted by trained interviewers. Leisure time physical activity pattern was obtained through this questionnaire by the question: "Usually, do you engage in any leisure time physical activity such as walking, sports, gymnastics ...?" This question could be answered "yes" or "no". Those who responded affirmatively were asked to describe the amount of physical activity in the last two weeks, in bouts of 20 or more minutes, in response to three levels of physical activity: light (<3.0 times resting energy expenditure), moderate (3.0 to 5.9 times resting energy expenditure) or vigorous (>= 6.0 times resting energy expenditure), providing the interviewer at least five examples of each level of activity. People who reported any amount of leisure time physical activity were classified into three groups based on the recommendations of physical activity practice for adults from the American Heart Association: insufficiently active (people who engage in some kind of physical activity, but do not meet the minimum recommendation to maintain health), moderately active (150 or more minutes of moderate physical activity, weekly) and vigorously active (60 or more minutes of vigorous physical activity, weekly). To assess mental health, we used a subjective mental health index obtained through a 12 items questionnaire (General Health Questionnaire, GHQ-12) designed for diagnosing psychiatric disorders, widely used in international literature and validated for Spanish adult population. The score ranges from 0 to 12 points, from the better mental health to the worse mental health. To facilitate statistical comparison, individuals are grouped into percentiles, containing the 95th percentile for people with worse mental health. Other variables included in this study were: age group, gender, perceived health status, history of depression or mental illness throughout life, depression or mental illness in the last 12 months, diagnosed depression, anxiety or other mental disorders at some time, psychiatrist attendance in the last 12 months, and finally, use of antidepressants, anxiolytics or other mental drugs for the past 12 months. We used a multinomial logistic regression to model the association between variables and calculating the odds ratio (OR) and confidence interval 95% (Cl) to determine the degree of association between leisure time physical activity and mental-health related factors, taking into account gender and age group. Also, the association between the level of practice and mental health-related factors as studied in both sexes. The statistical treatment of data was performed trough the SPSS statistical package software (15.0 version). Results Leisure time physical activity is associated with a higher prevalence of perceived good or very good health status, and with a better assessment on the GHQ scale of subjective mental health (50th percentile). In terms of levels of practice, it seems that any amount of leisure time physical activity was associated with a higher probability of showing better perceived health status (p<0.001) and good GHQ index (p<0.001) in both sexes, although the degree of association grows proportionally to the level of practice of physical activity achieved. In both sexes, vigorous activity level is strongly associated with the studied variables, showing that there are more positive indicators of mental well-being in more vigorously active subjects, being the only exception psychiatrist attendance for females. For men, there is an association at this level of vigorous physical activity with a smaller number of psychiatrist attendance (p < 0.05), but this association is less than the relationship with other variables, in which there is a higher significance level (p < 0.001). For the lower levels of physical activity, the relationship of practice with these variables is not significant on many occasions. Sometimes, there is no difference between those classified as insufficiently actives (not reaching the minimum) and those who have a pattern of physical activity considered as moderate, and even the differences are in favour of those individuals insufficiently active. Such is the case of the use of medication for mental health in men or psychiatrist attendance in the case of women, less probable in insufficiently active group than in the other two groups (moderate and vigorous leisure time physical activity). Conclusion Practicing a physical activity during leisure time is associated with a lower prevalence of negative mental health indicators.
