Physical activity (PA) is defined as any bodily movement produced by the skeletal muscles, resulting in increased energy expenditure 1 . Regular PA is associated with a lower risk of cardiovascular and overall morbidity and mortality 2 . Indeed, physical inactivity was recently ranked as the fourth most important risk factor of mortality worldwide, by the World Health Organization (WHO) 3 . The recommended level of medium intensity aerobic activity being 150 minutes/week, only half of the adult population in the USA reach this level of PA according to self-reported data 4 , while the Eurobarometer study estimated that only 23 % of the Swedish population were sufficiently physically active 5 .

Both cardiorespiratory fitness (CRF), defined as the maximum oxygen uptake on treadmill or by bicycle ergometer 1 , and the level of physical activity have been found to be independently associated with cardiovascular risk factors, such as blood lipids, body mass index (BMI), blood pressure and insulin resistance 6 7 , as well as predicting morbidity and mortality 8 9 . In addition, sedentary behaviour and few breaks in sitting time, is also independently associated to cardiovascular risk factors 10 . In addition to the direct effects on the cardiovascular system, physical inactivity also increases future risk of developing mental disorders and stress 11 .

The definitions of sedentary behaviour has varied in the literature, initially including a low physical activity level in addition to being still, making comparisons between studies difficult 12 . The health risk related to sedentary behaviour seems to be independent of the time spent performing light and moderate to vigorous PA 13 14 . In these studies, sedentary behaviour was defined as an accelerometer count of <100/min, but has also been defined as all activities that consume not over 1.0-1.5 metabolic equivalent units (METs; equivalent to energy expenditure from oxygen uptake of 3.5 ml·kg^-1·min^-1) 15 . Recently the Sedentary Behaviour Research Network has suggested a standardized description 16 , defining inactive individuals are those not meeting recommended level of PA, while sedentary behaviour is defined as any awake activities while sitting or reclining posture with an energy expenditure <1.5 METs. Moreover the actual sitting time has recently also been independently related to increased risk for all-cause and cardiovascular disease (CVD) mortality 17 18 .

Indeed, the greatest net-effect for health may be gained by any positive change from sedentary behavior 19 . As a result, identifying and targeting physically inactive, including sedentary, or insufficiently active individuals, is one of the major tasks in primary and secondary prevention in everyday health care.

The Framingham Risk Score and the European Society of Cardiology’s SCORE, based on cholesterol, gender, age, blood pressure and smoking, are used to identify individuals with increased risk of cardiovascular events 20 21 . However, both risk score charts lack information about the physical activity level, which would be a valuable contribution to identify patients at risk, as both physical activity levels and CRF are important independent risk factors for CVD 9 22 . While direct measurement of the individual CRF level, using e.g. bicycle ergometers or treadmill tests may not be feasible in everyday health care, a simple assessment of the physical activity level, identifying patients at potentially increased cardiovascular risk, by being physically inactive, would be useful.

In a previous longitudinal study, we used a four level scale to assess the self-reported physical activity level in a working population, showing an association with both perceived stress and self-reported mental disorders 11 . This scale was originally developed by Saltin and Grimby (here termed Saltin-Grimby Physical Activity Level Scale, SGPALS) 23 .

In the present study, we thus aimed to study the relationship between the self-reported level of PA according to the modernized version of SGPALS and other traditional cardiovascular risk factors in a large Swedish population. We also wanted to characterise the subjects having the lowest level of PA, according to SGPALS, with regard to their risk of having pre-defined abnormal levels of the classical risk factors.

This study is a part of the Intergene study, which was conducted between April 2001 and December 2004, and included a random sample of inhabitants of Region Västra Götaland from central population registry, of both sexes, aged between 25 and 74 years. In total, 8820 individuals were initially selected to participate of whom 194 were not eligible, due to either having an unknown address, change of address or having died prior to contact. The survey response-rate was 44 % (women) and 39 % (men), and a total of 3610 individuals (1908 women/1702 men), with a mean age of 52 (women) and 51 years (men), were finally included in the study. The question regarding PA level was answered by 1903 women and 1685 men.

The aim of the Intergene study was to investigate the INTERplay between GENEtic susceptibility and environmental factors predisposing to chronic disease. The Intergene study included a questionnaire, a clinical examination and blood measurements. Sample individuals received a letter of information, including an invitation to participate in the study. Written informed consent was required before study inclusion. Screening was centralised to the most inhabited areas, using a mobile test centre (bus) outside the city of Gothenburg, as earlier described by Berg 24 .

The main finding of the present study was that self-reported physical activity level, according to the SGPALS, was negatively associated with traditional cardiovascular risk factors. The most inactive group (scoring 1 out of a possible 4) has an increased prevalence (odds-ratio) for having pre-defined abnormal levels of several cardiovascular risk factors. While an association between increasing levels of PA and lower levels of most traditional cardiovascular risk factors is well known, our findings may serve as an easy and inexpensive way to identify physically inactive and high risk individuals. Our study also confirms the validity of the modernized SGPALS in the setting of an unselected Swedish population.

While CVD related mortality has, in the last decades, declined in the Western world, CVD still remains a major cause of mortality 3 . Today, the health care system increasingly uses different risk score methods to identify cardiovascular risk patients, for example SCORE as advocated by the European Society of Cardiology 20 . Many existing risk score methods are predictive of future cardiovascular events and/or cardiovascular death 20 21 . However, they do not regularly take into consideration the level of physical activity and/or the maximal oxygen consumption (Vo_2max), which independently predicts cardiovascular and overall mortality 9 . Indeed, the true CRF in combination with heart rate recovery adds significantly to the predictive power of the Framingham score 9 . Individuals classified by SCORE as low risk, may become medium risk by having a low CRF 35 . Therefore, there is a need for complementary information regarding the CRF level and/or habitual physical activity levels, to increase the yield of the risk stratification in any given individual 35 .

In this study we found that individuals reporting being physically inactive during leisure-time, defined as scoring 1 out of 4 on the SGPALS, had elevated cardiovascular risk factors. Physically inactive individuals were four times more likely to be smokers, three times more likely to have an increased waist circumference, a high serum-triglyceride level, and a low serum-HDL level. In addition, individuals reporting being physically inactive, were more likely to have an elevated resting heart rate, having perceived high levels of stress, and elevated plasma-glucose level, and to be overweight (Table 3).

These findings are in line with earlier studies, showing a linear relationship between BMI and CRF 36 , as well as the level of PA. It has previously been reported that a higher level of physical activity is associated with lower levels of several cardiovascular risk factors 37 , while the lowest levels of PA are being associated with the highest risk 19 . The present definition of SGPALS 1 includes patients being sedentary as well as those reporting a very low leisure-time PA level. This perhaps explains the strong relationships with a multitude of cardiovascular risk factors, as both a low level of PA and sedentary behaviour, independently, are associated with higher cardiovascular risk 38 . The SGPALS indeed seems to capture “the worst of the worst” regarding activity behaviour.

The results of the present study have clinical implications, as the modernized SGPALS may be used as an sole and simple indicator of inactive behaviour (including sedentary and very low level of leisure-time PA), and thus indirectly of cardiovascular risk. The SGPALS has previously been found to be associated with the CRF level of the individual 25 26 27 . Thus, this scale can be a feasible tool to identify individuals in need of further risk assessment. Physically inactive men also reported increased levels of perceived stress. This is important, since high levels of stress have been related to cardiovascular morbidity, independent of other lifestyle factors 39 , supporting the cardiovascular risk assessment potential of SGPALS.

Furthermore, the effectiveness in changing longitudinal outcome by using risk score methods has been reviewed 40 , with two studies comparing risk score assessment with placebo showed no benefit, while another study showed an increased use of lipid lowering and antihypertensive drugs, and one study did show a small reduction in systolic BP, using risk assessment. Present risk score tools have probably affected smoking habits 41 , while cholesterol levels have been mainly changed by food habits 42 . Whether the use of the SGPALS will affect future health outcome, is of course beyond the scope of this paper. However, the use of a reliable tool for assessment of the PA-level is clinically important, while still underutilized. Therefore, it is also important that the present study validates the modernized SGPALS in the setting of a large Swedish population.

The present study in a large Swedish cohort, shows that the modernized Saltin-Grimby Physical Activity Level Scale (SGPALS), identifies individuals reporting being predominately sedentary or having a very low leisure-time PA level. Importantly, this physically inactive group was associated with a multitude of elevated cardiovascular risk factors. Accordingly, SGPALS may be useful as a tool in clinical risk assessment, alone or in combination with other risk score methods in everyday health care, to identify physically inactive including sedentary individuals in possible need of PA counselling and further risk assessment.

Hence, we suggest that using the SGPALS in risk assessment, may be suitable as a supportive instrument for health care providers using physical activity on prescription as a preventive and treatment method 44 , to improve health care.

The authors declare that they have no competing interests.

LR, IJ, AR, LB, DT and MB have made contributions to the conception and design of the study. LR and GL have conducted the statistical analysis. LR drafted the manuscript. IJ and MB have been actively involved in drafting the manuscript. IJ, AR, LB, DT, GG and MB have critically revised the manuscript. All authors read and approved the final manuscript.