Several public health guidelines concerning physical activity have been published in the last decades. Although results may vary depending on the definition used, most studies have only applied one criterion to assess the proportion of adults being sufficiently active. This study is one of the first to simultaneously analyze compliance with different physical activity guidelines, using an objective measure of physical activity. In addition, we investigated the relationship between socio-demographic characteristics and the odds of meeting the different guidelines in order to identify groups that are currently inactive and would thus derive substantial benefits from increasing their physical activity.
As expected, the prevalence of sufficient activity varied according to the physical activity guideline and decreased as the recommendations became more stringent . According to the ACSM/AHA guidelines, 73.8% of men and 55.7% of women were classified as moderately active and only 13.4% and 7.6% as vigorously active, when modified 10-min bouts were considered.
Previous studies have also shown that compliance rates were lower for vigorous physical activity as compared to moderate physical activity [14, 18, 23]. However, these studies used self-reports of physical activity and primarily focused on leisure time. Recently, accelerometers have been used more often to objectively quantify physical activity in daily life and evaluate the compliance with public health guidelines. However, most of these studies were restricted to recommendations for MVPA. The proportion of adults attaining at least 30 min MVPA per day from modified 10-min bouts varied from 1% in Swedish men and women , to 3.8% and 3.2% in men and women from the US  and 4.2-9.3% in Portuguese men and women .
Recent reports, such as the 2008 Physical Activity Guidelines for Americans and the Global Recommendations on Physical Activity for Health, focus on total volume of physical activity rather than a minimum number of sessions per week [5, 8]. Thus, individuals can achieve recommended amounts of activity in a number of different ways. For example, weekend warriors may participate in one or two bouts of exercise to meet these goals. Nonetheless, the US guidelines add that physical activity should preferably be spread throughout the week .
In the current sample, 94.2% of men and 86.5% of women accumulated ≥150 min MVPA per week, but only 73.8% and 55.7% did so in five or more sessions of ≥30 min/day. Thus, removing the frequency and duration requirements resulted in a considerable increase in the prevalence of sufficient MVPA. Furthermore, despite the upward shift from 60 to 75 min/week, the proportion of adults meeting the minimum levels of vigorous activity was higher with the current than with the ACSM/AHA guidelines.
The less restrictive nature of the recent guidelines is demonstrated in several previous studies. Carlson et al.  examined the prevalence of LTPA and reported that 34.8% and 30.5% of US men and women were considered active according to the ACSM/AHA guidelines, compared with 47.4% and 39.9% using the US 2008 guidelines. Reasons for the higher percentages were the removal of the frequency and duration requirements but also the possibility of combining moderate and vigorous physical activity. In addition, Rafferty et al.  demonstrated that 40.2% of men and 36.1% of women who reported walking in their leisure time, walked for a total of 150 min/week, but only 23.5% and 20.0% walked for at least 30 min/day five or more times per week.
The prevalence of objectively-measured physical activity is much lower, but also indicates the difference between both guidelines . In addition, Tucker et al.  noted that compliance with the US guidelines varied considerably according to the criteria used. When considering time spent in MVPA, 9.5% of men and 7.0% of women achieved the recommended minimum. In contrast, using the energy expenditure criteria (≥500 MET.min/week), 57.7% of men and 32.5% of women were classified as sufficiently active. However, energy expenditure was calculated by accumulating accelerometer MET-minutes individually, whereas only periods of 8–10 consecutive minutes were considered for time spent in MVPA.
It is important to note that most of the previous studies defined a 10-min bout as 10 or more consecutive minutes of at least moderate intensity, with allowance for 1–2 min below the threshold [15, 16, 21, 25]. In 1995 it was indeed noted that bouts of physical activity as short as 8–10 min provided beneficial health and fitness effects . However, since 2007, the reports state that activity should be accumulated from bouts lasting 10 or more minutes [4, 5, 8]. Therefore, it seems more appropriate to use a strict definition of a 10-min bout, i.e. without any interruptions. When we applied this criterion the prevalence of sufficient physical activity decreased, especially for moderate activity and to a smaller extent for vigorous activity. Moderate-intensity activities may represent activities of daily living, which are intermittent in nature, whereas vigorous activity is probably performed for exercise intentions and thus more continuous.
Nevertheless, even when using the strict definition of a 10-min bout, the proportion of adults meeting the different guidelines was relatively high compared to previous studies. However, making comparisons between studies is difficult and several factors might explain the observed differences. First, various instruments were used to measure physical activity. Accelerometers accurately assess ambulatory activities, but may underestimate overall physical activity, due to the inability of detecting cycling, upper body movement, carrying loads or walking on an incline [26, 27]. In contrast, by combining accelerometry with physiological sensors, the SenseWear may capture the additional energy-cost of these lifestyle activities [28, 29]. In addition, the use of the electronic diary allowed us to impute missing data for swimming. Secondly, methodological differences in collecting and analyzing data make results difficult to compare. For example, estimates of time spent in MVPA may differ substantially according to the cut-points used. Hagströmer et al.  showed that, when using cut-points derived from both ambulatory and non-ambulatory activities, 95% of the sample achieved 30 min/day of MVPA, compared to 52% when applying cut-points based on ambulatory activities only. Moreover, estimates of adherence in previous studies were frequently based on data from participants with 1–4 valid monitoring days [15–17, 25]. In contrast, the current study required seven valid days to be included in analyses. In addition, physical activity was assessed over a 24-hour period, whereas the minimum wear-time in previous studies was limited to ≥10 hours/day. Thus, during some of the waking hours, activities were not registered. Another factor that limits the comparison of results is the difference in the interpretation of compliance with the physical activity guidelines. Finally, most studies used random sampling techniques, while the present study consisted of a group of healthy volunteers. Accordingly, our sample may present a more active group of adults, compared to the general population.
It should be noted that the recommended amounts of physical activity in the current guidelines are based on associations between self-reported physical activity and health outcomes . Perhaps, these guidelines are not directly translatable to objective measures of physical activity. Troiano et al.  suggested that less than 30 min of objectively-measured physical activity may be needed to achieve substantial health benefits. However, other studies noticed that, since these self-reports primarily captured LTPA, the current recommendations should be viewed as the minimum level of physical activity over and above the routine activities of daily living [22, 35]. Thus, when evaluating physical activity across the day, a higher cut-point for sufficient activity would be more suited. Several authors have proposed a cut-point for health-enhancing physical activity (HEPA) of 3000 MET.min of MVPA accumulated over 7 days or 1500 MET.min of vigorous activity accumulated over 3 days or more [22, 35, 36].
When we applied this cut-point, 43.0% of men and 17.8% of women were considered active (results not shown). These numbers are comparable to previous studies that used the IPAQ. Bergman et al.  reported that in a sample of Swedish adults, 33.5% of men and 19.1% of women reached the high physical activity category. Other studies investigated the prevalence of physical activity across countries and reported that, in Belgium, 29.6-37.2% of men and 20.5-21.9% of women met the recommended amounts of HEPA [35, 36].
Another issue that needs to be addressed is whether the requirement of continuous bouts is similar for objectively measured physical activity. In a previous accelerometer study it was shown that although 52% of adults accumulated at least 30 min/day of MVPA, only 1% achieved those minutes from three or more continuous bouts of at least 10 minutes .
Clearly, previous recommendations should be reconsidered based on the associations between objectively measured physical activity and health outcomes.
One recommendation that was specifically created for objective measures of physical activity is the guideline of 10000 steps/day. This guideline represents 30 min of MVPA in addition to a minimum level of baseline physical activity. It has been suggested that 30 min of moderate activity translate to 3000–4000 steps, at a stepping rate of 100 steps/min [12, 13, 37]. Adding this amount to an estimated minimum of 6000–7000 steps, taken during the routine activities of daily living, approximates the proposed 10000 steps/day .
The current study showed that 16.3% men and 14.1% of women met the guidelines of ≥10000 steps/day on seven consecutive days. However, when the frequency requirement was decreased to 5 days/week, 45.4% of men and 55.1% of women were compliant. Most studies did not use the criterion of 10000 steps/day for a specific number of days, but simply looked at the average daily value, calculated from all valid days. The prevalence of adherence among adults ranged from 13.9-16% in samples from the US [38, 39], to 34.5% in Canada  and 41.6% in Belgium . In addition, Chastin et al.  reported that 53% of a group of UK-based postal workers achieved the recommended minimum of 10000 steps/day on at least 5 days/week.
Furthermore, the present findings show that, despite the high compliance with the ACSM/AHA and US guidelines, only 34.9% of men and 21.6% of women attained an average PAL of 1.75. To increase PAL, a high level of physical activity throughout the day would be required. This explains why individuals who accumulate ≥30 min/day of MVPA but are otherwise sedentary may meet the guidelines for cardiovascular health, without achieving the minimum levels to avoid excessive weight gain. Thus, in the light of the current obesity epidemic continued interventions to increase physical activity are needed.
This study also identified socio-demographic characteristics that are associated with meeting the guidelines and should be considered for the planning of future interventions. Similar to what is typically reported, the prevalence of sufficient physical activity was higher among men than women, irrespective of the guideline used [20, 21, 23–25]. Furthermore, the findings that the likelihood of meeting the physical activity recommendations was lower among women and decreased with age, are consistent with several previous studies. However, the results are not directly comparable because of discrepancies in the definition of sufficient physical activity.
Bryan et al.  investigated the prevalence of LTPA among Canadian adults and reported that the probability of achieving ≥30 min MVPA or ≥20 min vigorous physical activity on 4 days/week was higher for men than women in 1994–1995 and 1998–1999. But, this difference disappeared around 2001. In addition, it was shown that women were less likely to meet the HEPA-criterion of 3000 MET.min MVPA or 1500 MET.min vigorous activity per week [22, 35].
Furthermore, several studies showed that the odds of engaging in sufficient physical activity decreased with increasing age [18, 23, 35]. Bergman et al.  demonstrated that those being younger than 55 were 1.5-1.7 times more likely to be classified as active according to the ACSM/AHA guidelines in crude analyses. However, after adjustment for other socio-demographic characteristics, these associations were no longer significant. Nevertheless, 18–34 year-olds had a 1.8-fold higher odds of reaching the HEPA cut-points than 55–74 year-olds, after adjustment for all socio-demographic correlates.
The present findings also demonstrated that physical activity patterns differed between educational groups. Subjects with the lowest educational level had a significantly lower probability of obtaining 75 min of vigorous activity per week compared to those with a college or university degree. This may reflect the lower probability of participating in LTPA, as indicated by several previous studies [41, 42]. In contrast, lower-educated individuals may have more physically demanding jobs that require a large amount of ambulatory activity. This might explain why the lower-educated were almost 4 times as likely to take 10000 steps/day.
Our results confirm those of earlier studies. Macera et al.  showed that the odds of accumulating 30 min of moderate-intensity activity on 5 days or 20 min of vigorous activity on 3 days/week during non-working hours increased with educational level. Additionally, the proportion of adults meeting the ACSM/AHA or US guidelines was higher with greater educational attainment [20, 24]. In contrast, Bergman et al.  showed that education was not related to meeting the ACSM/AHA guidelines, but that subjects with a college/university degree were less likely to meet the HEPA-criterion than those with basic education.
With regard to steps, Chastin et al.  demonstrated that 77% of delivery postal workers met the minimum of 10000 steps/day on 5 days/week, compared to only 28% of the office-based postal workers. Furthermore, it was shown that the number of steps/day on weekdays differed between occupational groups with professionals and managers recording the lowest (7883 steps) and blue collar workers the highest number of steps (11784 steps). The difference of almost 4000 steps/day suggest that those with the highest occupational status would have to walk for an additional 30 min during non-working hours to reach the steps of those with a physically active job .
In addition, it has been suggested that factors of an unhealthy lifestyle such as smoking and being physical inactive tend to cluster, which is consistent with the present observations [41, 44, 45]. Smokers had a 60% reduced likelihood of participating weekly in 150 min MVPA compared to non-smokers. Similarly, Bertrais et al.  demonstrated that current smokers were 24-27% less likely to achieve 150 min MVPA or 60 min vigorous physical activity per week.
Finally, in accordance with the present results, Bergman et al.  showed that after adjustment for other socio-demographic correlates, marital status was not related to meeting the ASCM/AHA or HEPA guidelines. However, in gender-specific analyses, being single was positively associated with achieving the HEPA guideline with single women being twice as likely to meet this goal compared to women who were married or co-habited.
A major strength of this study was the use of a valid activity monitor to objectively assess physical activity across the day. Furthermore, participants were asked to wear the monitor 24 hours a day and only participants with at least 22 hours and 48 min (95% of 24 hours) of data for seven consecutive days were included in the analyses. Less than seven days may be enough to assess habitual physical activity , but when evaluating compliance with recommended amounts of physical activity per week, it is stronger to use seven days rather than estimating the prevalence of compliance based on data from participants with one or more valid monitoring days . On the other hand, it has been suggested that compliance rates could be overestimated by excluding those who did not wear the monitor for seven days, because the least active tend to be less compliant with the study protocol . However, 81% of the current sample wore the monitor for seven consecutive days and physical activity levels did not differ between those with and without seven valid monitoring days.
However, some limitations should be recognized. As previously stated, participants volunteered to engage in the study. This may have led to a selection bias because subjects who agreed to participate may have been more active than the general Flemish population. Accordingly, the generalizability of these findings may be restricted. In addition, subjects knew they participated in a physical activity study and were monitored for their activity. Thus, because of a possible Hawthorne effect, participants could have performed more physical activity than usual. However, it was not our aim to describe habitual physical activity in the general population, but rather to compare different activity guidelines within the same sample. Secondly, similar to other activity monitors, the SenseWear is known to overestimate energy expenditure of moderate-intensity activities and underestimate that of very vigorous activities, mainly due to a ceiling effect at 10 MET [31, 32]. However, this would not affect the estimate of time spent in vigorous activity, since the threshold was set at 6 MET. Nevertheless, Berntsen et al.  showed that time spent in MVPA was overestimated by both the Actigraph and SenseWear compared to indirect calorimetry (2.5% and 2.9%, respectively). In addition, Dwyer et al.  reported that during treadmill walking, the average SenseWear step count was 5% less than the manual count, which is similar to other motion sensors and pedometers, designed specifically to measure steps .