Obesity has become a significant epidemic in both developed and developing countries due to its association with increased morbidity and mortality [1
]. Overweight and obesity have also reached epidemic proportions in Portugal [2
]. The consequences of obesity include physical, psychological and social aspects, which impact one’s quality of life. Although genetics is a strong component of obesity [4
], lifestyle and environmental changes typical of industrialized societies are more likely to explain the recent obesity epidemic [5
]. Indeed, obesity occurs when energy intake exceeds energy expenditure over time.
Physical inactivity has been viewed as one of the most important risk factors for coronary heart disease and other chronic diseases [6
]. For instance, physical activity (PA) levels, as well as sedentary behavior appear to play an important role in long-term weight regulation [8
]. Thus, the need to increase PA is a public health priority [9
]. Current public health campaigns to reduce obesity and type 2 diabetes have largely focused on increasing PA and/or exercise, but have paid little attention to the reduction of sedentary behaviors [10
]. Moreover, some studies have shown that non-exercise activity thermogenesis (NEAT) plays a key role in differences of energy expenditure in special populations such as the obese [11
]. In fact, altering one’s postural allocation from a seated to standing position or engaging in light ambulation has been shown to significantly increase energy expenditure [11
]. Conversely, it has been shown that workers performing their job functions in their usual fashion (seated) might expend more than 200 extra calories daily using a walking workstation [12
]. Therefore, sitting is strongly and inversely associated with caloric expenditure that is likely an important cause of the obesity epidemic. Indeed, in our daily contemporary life, sitting is a predominant behavior for many hours per day [13
Several studies have shown positive associations between BMI and different measures of sedentary behavior, such as TV viewing [14
] motorized transportation [17
] occupational sitting time [19
] or sedentary behaviors during leisure time [20
]. However, these studies fail to address total sedentary behavior and/or adjust the analysis for total PA.
In this context, the purpose of this study was to examine cross sectional relationship between body mass index (BMI) and total sitting time (i.e., sitting across all three main domains: transportation, occupational and leisure time), adjusting the analysis for total PA time in Azorean adults.
2.1. Study Design and Sampling
Data for the present study are derived from the Azorean Physical Activity and Health Study. The study methods are reported elsewhere [3
]. Briefly, data were collected in 2004 by mailing questionnaires to the adult residents of all the Azorean Islands and municipalities, a Portuguese archipelago. For the present study, only men who were employed (n = 3,939) or studying (n = 152) full-time and for whom questionnaires contained complete information on the variables of interest (i.e.
, weight, height, PA, sitting, education level, meal frequency, sleep duration, smoking status and alcohol consumption) were included (n = 4,091, corresponding to 95.8% of the total sample of men). The decision to exclude unemployed or retired men was based on a previous study with this sample that showed that these groups of men were less likely to achieve higher levels of PA [23
] and showed significantly higher total sitting time compared with men employed or studying full time (data not shown).
BMI [weight (kg) / height (m)2] was calculated from self-reported weight and height.
PA and sitting were assessed with the International Physical Activity Questionnaire (short last week version) and data was handled according to the IPAQ scoring protocol [24
Total PA time was computed by multiplying the reported minutes of moderate and vigorous PA by the number of PA days of each type of PA. Subjects were categorized according to the ACSM/AHA PA guidelines: insufficiently active (participants who reported fewer than 150 min/week of at least moderate-intensity PA or less than 20 min/week of vigorous-intensity PA) and sufficiently active (participants who reported 150 min/week or more of at least moderate-intensity PA or 20 min/week or more of vigorous-intensity) [7
The time spent sitting in an ordinary week day was considered a proxy measure of sedentary behavior. Subjects were categorized as having low or high total sitting time based on the median value for total sitting time found in this sample (180 min/day) [24
Participants were also categorized into the following four groups: low total sitting time/sufficient PA time; low total sitting time/insufficient PA time; high total sitting time/sufficient PA time and high total sitting time/Insufficient PA time.
Other variables included in this analysis were:
Education level: four years of education; 5–9 years of education; 10–12 of years education and higher education.
Smoking: non smokers, former smokers, occasional smokers and current smokers.
Alcohol consumption: non drinkers, former-drinkers, occasional drinkers, regular drinkers and heavy drinkers.
Sleep duration: number of sleeping hours per day.
Meal frequency: daily meal frequency was assessed by the question: “How many meals per day do you consume?” The main meals represented meals that were conventionally served on a plate.
2.3. Statistical Analysis
Statistical analyses were performed using the Statistical Package SPSS 17.0.
Data are presented as mean ± standard deviation unless stated otherwise.
Analysis of the variance with Bonferroni post-hoc tests, was used to assess BMI differences between sitting/PA groups.
Linear Regression analysis was performed to assess unstandardized regression coefficients and standard errors predicting BMI. Variables entered in the models were significantly correlated with BMI (in the bivariated analysis—data not shown). Model 1 was adjusted for age, education level; smoking, alcohol consumption, sleep duration, meal frequency, island of residence; model 2 was further adjusted for total PA time.
Statistical significance was set at p < 0.05.
Of the 4,091 men included in this study, 96.3% were employed full-time. Of those, 73.4% of men had a low level of education (≤9 school years) and 71% were manual workers.
Participants had an average BMI of 26.3 ± 3.7 kg/m2
and spent on an average 25.5 ± 17.5 % of their waking hours sitting (about 250 min/day). Descriptive characteristics are presented in Table 1
As depicted in Table 2
, men with high sitting time and insufficient PA presented a highest mean of BMI −26.8 ± 3.8 kg/m2
(p < 0.05 for all).
Linear regression analysis showed that after adjustments for potential confounders total sitting time was positively associated with BMI (B = 0.078; p < 0.001—model 2).