Journal Logo


Objectively Measured Physical Activity in the 1993 Pelotas (Brazil) Birth Cohort


Author Information
Medicine & Science in Sports & Exercise: December 2012 - Volume 44 - Issue 12 - p 2369-2375
doi: 10.1249/MSS.0b013e3182687d35
  • Free


Physical activity is an important determinant of health outcomes throughout the lifespan. Despite its importance, free-living physical activity is difficult to measure accurately, especially in young people. Theoretically, physical activity is defined as “any bodily movement produced by skeletal muscles resulting in energy expenditure” (3), suggesting that all body movements contribute to the overall levels of activity.

Self-reported instruments (e.g., questionnaires) are frequently used to assess physical activity (27). Although questionnaires have several advantages, they also have important limitations, particularly in studies with children and adolescents, because these age groups may have difficulties in accurately reporting their physical activity levels (17,22). Therefore, alternative methods for assessing physical activity, particularly among youth, are warranted.

In the last decades, the use of accelerometers to determine physical activity patterns has increased markedly (11). These devices allow investigation of the frequency, duration, and intensity of physical activities practiced for a period (e.g., 1 wk). Another advantage of the method is the ability to make comparisons between populations, because the output for a given activity is the same regardless of the setting (20,27).

Some previous studies have used accelerometers to describe the pattern and levels of physical activity in adolescents, but virtually all of them are from high-income countries (2,7,12,15,18). Brazil is a middle-income country undergoing a rapid transition in which adult patterns of physical activity are very different from those observed in wealthier settings (9). It is unknown whether or not objectively measured physical activity of Brazilian adolescents differs from those of adolescents from other countries. The aim of this study was to describe the patterns of objectively measured physical activity in a representative sample of Brazilian adolescents belonging to the 1993 Pelotas Birth Cohort study.


This is a cross-sectional study nested within the 1993 Pelotas (Brazil) Birth Cohort. A detailed description of the cohort and its follow-up is available elsewhere (30,31). In brief, virtually all individuals (99.7%) born in the city of Pelotas (South of Brazil) in the calendar year 1993 whose family lived in the urban area of the city have been followed up for several times. The 5249 cohort members approached at birth were followed up several times during infancy, childhood, and adolescence. At the mean ages of 1 and 6 months, 1 and 4 yr, a subsample of the cohort members was visited. At the mean age of 11 yr, all participants were sought, and 87.5% were located. At the mean age of 13.3 yr, we aimed to locate all individuals (n = 568) with information from all previous follow-up visits for a comprehensive study on physical activity and body composition. A detailed description of the methods used in this follow-up visit is described elsewhere (16). The Federal University of Pelotas Medical School Ethics Committee approved the study protocol, and written informed consent was obtained from parents or guardians and verbal consent was obtained from adolescents before data collection.

Measurement of physical activity.

Physical activity was assessed by the ActiGraph model GT1M (ActiGraph Corporation, Pensacola, FL). Adolescents were asked to wear the accelerometers on the waist, except when showering, bathing, or swimming. Accelerometers were delivered to the individuals on Wednesdays; fieldworkers visited the participants’ home on the following Monday to collect the device. Therefore, for most adolescents (>80%), accelerometer data comprise four consecutive days (Thursday, Friday, Saturday, and Sunday).

Data were downloaded and processed according to manufacturer’s instructions. Epoch was set to 5 s, and data were analyzed using the MAHuffe software ( Days with <600 min of registered data and periods of time above 60 min of consecutive zero counts were excluded (4).

The physical activity variables analyzed were mean counts per minute (cpm), as an indicator of daily average physical activity intensity, and time spent in different physical activity intensities. For the latter, the Evenson et al. (6) thresholds of cpm were used: 0 to 100 cpm, sedentary activities; 101 to 2295 cpm, light activities; 2296 to 4012 cpm, moderate-intensity activities; and >4012 cpm, vigorous-intensity activities. These thresholds are recommended to adolescents because they have higher ability to accurately classify physical activity intensities than other cut points (28). Intensity thresholds were scaled down (division by 12) to accommodate the 5-s epoch setting. Finally, a continuous score of moderate-to-vigorous activities was built by summing the time spent (in minutes per day) in these two intensities. This score was further dichotomized into two categories: achieving or not achieving current physical activity guidelines for adolescents (i.e., 60 min·d−1 of moderate-to-vigorous intensity activities) (24). Further details of the physical activity measurement in this study are available elsewhere (16).

Independent variables.

Adolescents answered a pretested questionnaire through face-to-face interviews in their homes administered by trained interviewers. Sociodemographic variables were sex, skin color (based on interviewer’s perception), schooling, and socioeconomic (SES) level of the family. Schooling was defined as inadequate if the adolescent was at grade 4 or below, because it represents at least 2 yr late for the normal age-to-grade relationship. SES status of the family was expressed using the criterion of the Brazilian Association of Research Institutes, which takes into account schooling of the head of the family, household assets, and presence of servants at home. This classification divides families into five groups, from A (highest level) to E (lowest level).

In addition, two physical activity–related variables were collected during the interview: mode of transportation to school (categorized into inactive or active) and perception of adolescent’s own level of physical activity compared with his/her friends of the same age (below average, average, and above average).

Statistical analyses.

Descriptive analyses of categorical variables are shown as absolute frequencies and percentages, whereas continuous variables are presented as means and SD.

All analyses were adjusted for the registered time and expressed as a percentage (i.e., time spent in each intensity was divided by the registered time and multiplied by 100). Differences between time spent in different physical activity intensities and average intensity (mean cpm) with independent variables were investigated by t-tests or ANOVA. The association between the percentage of individuals achieving 60 min·d−1 of moderate-to-vigorous intensity activities with the independent variables was verified by the Fisher exact test. All analyses were performed by Stata 10 (StataCorp, College Station, TX, ), and a P level <0.05 was denoted as statistical significance.


Of the 568 eligible adolescents, 511 were located and interviewed and 486 (95.1%) provided valid accelerometer data. The mean age was 12.9 yr (SD = 0.36 yr), and 51% were boys. Table 1 shows descriptive characteristics of the sample. We observed no major differences between the original cohort and those of the subsample that provided valid accelerometer data in terms of gender, SES status, and skin color distribution. Few adolescents belonged to the tail of SES distribution (2.7% to the highest and 6.8% to the lowest), and 67.8% were categorized as white. In terms of self-reported physical activity, 25.9% of the adolescents reported an inactive mode of transportation to school, and 66.7% perceived their physical activity level to be comparable with that of friends of the same age.

Description of the sample.

The mean registered time of accelerometer data was similar for boys (1198 min·d−1) and girls (1183 min·d−1, P value = 0.13). The mean times spent in sedentary, light, moderate, and vigorous activities were 962 (SD = 114), 176 (SD = 46), 36 (SD = 16), and 17 (SD = 10) min·d−1, respectively. Table 2 shows the proportion of time spent in each of the physical activity intensities. Skin color and perception of physical activity level were not associated with any of the physical activity variables. The time spent in sedentary pursuits was higher in girls, those with higher SES status, and those who reported inactive mode of transportation to school (Table 2).

Proportion of time spent in sedentary, light, moderate, and vigorous activities.

Differences were also observed between sexes, categories for SES, and mode of transport to school for the time spent in moderate and vigorous activities (Table 2). For example, boys and those who had an active mode of transportation to school had higher scores of minutes per day of moderate-to-vigorous activities than their counterparts.

Table 3 shows the mean cpm and prevalence of ≥60 min·d−1 of moderate-to-vigorous activities stratified according to sex, ethnicity, mode of transportation to school, SES status, and perception of physical activity. Analyses of these data are consistent with each other, because those groups with higher levels of average physical activity intensity (cpm) also showed higher proportions of achieving ≥60 min·d−1 of moderate-to-vigorous activities. For example, a strong association was observed between physical activity and SES; the mean cpm and the proportion of individuals accumulating more than 60 min of moderate-to-vigorous physical activity (MVPA) per day increased in a graded manner with lower SES. Similarly, physical activity was higher in those who reported active mode of transportation to school.

Total physical activity (cpm) and prevalence of ≥60 min·d−1 of moderate-to-vigorous activities according to independent variables.

Figure 1 shows the distribution of time spent in moderate-to-vigorous activities stratified by sex. Both mean and median for moderate-to-vigorous activities were higher in boys. Figure 2 presents the distribution of time spent in MVPA stratified by SES status. It is clear that both mean and median scores increase toward lower SES status. The box of the highest SES status is slightly flatter than the others, indicating a lower interquartile range.

Time spent in MVPA in boys and girls. Vertical dashed line represents 60 min·d−1 of MVPA.
Time spent in MVPA according to SES level of the family. Horizontal dashed line represents 60 min·d−1 of MVPA.

Figure 3 presents the distribution of time spent in light-, moderate-, and vigorous-intensity physical activities stratified by MVPA categories. Those who accumulated ≥60 min·d−1 of moderate-to-vigorous activities have roughly twice the amount of both moderate and vigorous activities than those who perform 30 to 59 min·d−1 of MVPA. Similarly, compared with the least active group (0 to 29 min·d−1 of moderate-to-vigorous activities), those with ≥60 min·d−1 of moderate-to-vigorous activities spent four times more minutes in both moderate- and vigorous-intensity activities. The time spent in light-intensity activities also increased with higher levels of MVPA (approximately 30 min by group). The time spent sedentary was 1018 min for the category of 0 to 29 min·d−1 of MVPA, 974 for the category of 30 to 59 min·d−1 of MVPA, and 918 min for the category of ≥60 min·d−1 of MVPA (P < 0.001).

Distribution of time spent in different physical activity intensities stratified by MVPA categories.


To the best of our knowledge, this study is the first to measure the physical activity of Brazilian adolescents by accelerometry. The high response rate achieved indicates that it is feasible to use these instruments in low-to-middle income countries despite their high cost compared with questionnaires. More than 80% of time was spent in sedentary activities, and 34.2% of our sample met the current recommendation of 60 min·d−1 of MVPA, mainly through moderate-intensity physical activities. Higher physical activity levels were observed among those with an active mode of transportation to school and from lower SES groups.

Our results differ from previous observations in this cohort when physical activity was assessed by self-report. For example, Hallal et al. (8) observed 58.2% of 10- to 12-yr olds did not achieve 300 min·wk−1 of MVPA. Although defined differently, in the present study, a slightly higher proportion (65.8%) did not accumulate 60 min of MVPA per day. The observed differences are likely due to differences in the methods used to estimate physical activity. Furthermore, the intensity thresholds used to define MVPA influence the prevalence estimates of sufficiently active youth. For example, analyzing our data set using criteria threshold of 2000 cpm for MVPA (15), the prevalence of inactive youth (<60 min·d−1 of MVPA) decreased from 65.8% to 49.2%.

Other studies with Brazilian youths using different types of questionnaires also report markedly different prevalence of physical activity (13,25). Self-report instruments do not pick up all short bouts of MVPA, which are picked up by accelerometry, particularly when using a short epoch such as 5 s. Furthermore, self-report instruments do not usually assess total physical activity but rather focus on specific domains of physical activity (e.g., leisure-time physical activities). Thus, comparing the prevalence of physical activity from studies using objective and subjective measures should be done with caution.

Interestingly, the factors associated with physical activity are very similar, regardless of the method used to measure physical activity (i.e., questionnaire or accelerometer). For example, in agreement with the results from the present study, Hallal et al. (8) also found lower level of physical activity among girls, those of higher SES status, and those who had an inactive mode of transportation to school. The difference in physical activity across socioeconomic strata is interesting. We have previously shown that in Brazilian adults, this association depends on the domain of activity examined (i.e., activity performed in the leisure time, domestics, commuting, and work). Considering all domains, self-reported overall physical activity is higher among lower SES groups (9). In contrast, if only leisure-time physical activity is considered, physical activity is higher among higher SES groups (1). This pattern appears similar in adolescents. Although adolescents from higher SES status have higher level of leisure-time physical activity, adolescents from lower SES have higher levels of commuting physical activity (8,21).

Data on the pattern of physical activity measured by accelerometry have increased remarkably during the last decades, particularly from high-income countries. Data from the National Health and Nutrition Examination Survey (USA) reported a lower prevalence of meeting adolescent physical activity guidelines than the current study (26). In contrast, other studies have reported higher levels of physical activity. Nader et al. (14) reported the shape of distribution and time spent in MVPA in a cohort of US adolescents followed between the age of 9 and 15 yr. Although data are not directly comparable because of the use of different intensity thresholds when defining time spent in MVPA, at the age of 12 yr, the distribution of MVPA in the American adolescents had a shape similar to that observed in the current study. However, US adolescents spent a slightly higher time in moderate-to-vigorous activities, and a higher proportion of individuals achieved 60 min·d−1 of MVPA than the youth in the present study (14). Furthermore, the magnitude of difference in physical activity level between boys and girls was higher in the current sample compared with the American adolescents. The contrasting results of these two American studies highlight the limited external validity of data on physical activity level of the populations, even when assessed by the same type of accelerometry. Using a similar methodology as in the current study (same device, two weekdays and weekend days monitored) but different intensity thresholds (2000 cpm to moderate intensity and 5000 cpm to vigorous intensity), Nilsson et al. (15) reported data from a sample of 1327 individuals (9 and 15 yr old) from Norway, Estonia, and Portugal. It was observed that the proportion of daytime spent in MVPA in boys and girls was 12% and 10%, respectively, whereas the proportion of time spent in sedentary activities was 42% and 58%, respectively. When reanalyzing the data from the current study using those intensity thresholds applied by Nilsson et al. (15), the proportion of time spent in MVPA and sedentary activities was 5.3% and 80.7%, respectively. Other studies also indicate higher levels of PA expressed as average intensity (mean cpm) for European adolescents than those observed in our study (2,10,18). Keeping in mind differences between studies in their definition of intensity thresholds and indicators of nonwear time, our results suggest that Brazilian adolescents may have lower levels of physical activity than European adolescents (15). Future studies aimed at pooling and reanalyzing data from different countries are needed to examine true differences between populations.

Some strengths and methodological aspects of this study should be highlighted. This is one of the first studies characterizing objectively the measure of physical activity in a large representative sample of adolescents from a developing country. We used a valid and reliable accelerometer to assess physical activity for 4 d. A short epoch was used, as recommended by the literature (5). The intensity thresholds used to estimate sedentary time and MVPA have been previously shown to be valid and accurate to predict physical activity intensities (28). Despite some difficulties (16), physical activity was monitored for weekdays and the whole weekends. Although there is some evidence that more days of monitoring might be better (29), such a decision could result in higher refusal rates or lower compliance in wearing the monitors. In fact, compared with other studies (14,23,26), a higher adherence to wearing the monitors was achieved in our study. Adherence to wearing the device is important to maintain the representativeness of activities performed throughout the measurement period. However, our monitoring period (Thursday to Sunday) may not be fully representative of the entire week because physical activity level in the weekend is generally lower than during weekdays (14). Thus, slightly higher level of physical activity might be expected in our sample if all weekdays were monitored. Furthermore, our sample was derived from a single city and is likely that adolescents from other regions of Brazil have different levels of physical activity, which has been previously been observed by self-reported instruments (8,19,21). Finally, the sample is part of a large, well-characterized birth cohort study, and objective measures of physical activity will be important for future studies in the cohort.

Our study also has some limitations that should be considered when interpreting the results. There is no gold standard to measure physical activity, and although it is generally accepted that objective measures are more accurate than self-reports, they also have limitations. For example, accelerometers are known to imprecisely estimate the intensity of certain types of activities such as cycling and strength conditioning exercise (27). However, these activities are only practiced by a small percentage of Brazilian adolescents (8).

In conclusion, our data show that objectively measured physical activity levels of Brazilian adolescents from the 1993 Pelotas Birth Cohort are higher than those that have been reported using self-report instruments, yet a large proportion of adolescents do not achieve current guidelines of health-related physical activity. Active commuting to school may be a target for interventions aimed at increasing physical activity. Physical activity levels are higher in lower SES groups than that in higher groups.

This project was partially funded by a study partially financed by the National Research Council (CNPq, Brazil) and State Research Council (FAPERGS, Brazil) and was supported by the Wellcome Trust’s initiative entitled “Major Awards for Latin America on Health Consequences of Population Change.” Earlier phases of the 1993 cohort study were funded by the European Union, the National Program for Centers of Excellence (Brazil), and the Ministry of Health (Brazil).

The authors are particularly thankful to all individuals of the 1993 Pelotas (Brazil) Birth Cohort and their families.

The authors declare they have no conflict of interest.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.


1. Azevedo MR, Horta BL, Gigante DP, Victora CG, Barros FC. Factors associated to leisure-time sedentary lifestyle in adults of 1982 birth cohort, Pelotas, Southern Brazil. Rev Saude Publica. 2008; 42 (2 suppl): 70–7.
2. Aznar S, Naylor PJ, Silva P, et al.. Patterns of physical activity in Spanish children: a descriptive pilot study. Child Care Health Dev. 2011; 37 (3): 322–8.
3. Caspersen CJ, Powell KE, Christenson GM. Physical activity, exercise, and physical fitness: definitions and distinctions for health-related research. Public Health Rep. 1985; 100 (2): 126–31.
4. Choi L, Liu Z, Matthews CE, Buchowski MS. Validation of accelerometer wear and nonwear time classification algorithm. Med Sci Sports Exerc. 2011; 43 (2): 357–64.
5. Edwardson CL, Gorely T. Epoch length and its effect on physical activity intensity. Med Sci Sports Exerc. 2010; 42 (5): 928–34.
6. Evenson KR, Catellier DJ, Gill K, Ondrak KS, McMurray RG. Calibration of two objective measures of physical activity for children. J Sports Sci. 2008; 26 (14): 1557–65.
7. Hagstromer M, Troiano RP, Sjostrom M, Berrigan D. Levels and patterns of objectively assessed physical activity—a comparison between Sweden and the United States. Am J Epidemiol. 2010; 171 (10): 1055–64.
8. Hallal PC, Bertoldi AD, Goncalves H, Victora CG. Prevalence of sedentary lifestyle and associated factors in adolescents 10 to 12 years of age. Cad Saude Publica. 2006; 22 (6): 1277–87.
9. Hallal PC, Victora CG, Wells JC, Lima RC. Physical inactivity: prevalence and associated variables in Brazilian adults. Med Sci Sports Exerc. 2003; 35 (11): 1894–900.
10. Martinez-Gomez D, Welk GJ, Calle ME, Marcos A, Veiga OL. Preliminary evidence of physical activity levels measured by accelerometer in Spanish adolescents: the AFINOS Study. Nutr Hosp. 2009; 24 (2): 226–32.
11. McClain JJ, Tudor-Locke C. Objective monitoring of physical activity in children: considerations for instrument selection. J Sci Med Sport. 2009; 12 (5): 526–33.
12. Metzger JS, Catellier DJ, Evenson KR, Treuth MS, Rosamond WD, Siega-Riz AM. Patterns of objectively measured physical activity in the United States. Med Sci Sports Exerc. 2008; 40 (4): 630–8.
13. Moraes AC, Fernandes CA, Elias RG, Nakashima AT, Reichert FF, Falcao MC. Prevalence of physical inactivity and associated factors in adolescents. Rev Assoc Med Bras. 2009; 55 (5): 523–8.
14. Nader PR, Bradley RH, Houts RM, McRitchie SL, O’Brien M. Moderate-to-vigorous physical activity from ages 9 to 15 years. JAMA. 2008; 300 (3): 295–305.
15. Nilsson A, Andersen LB, Ommundsen Y, et al.. Correlates of objectively assessed physical activity and sedentary time in children: a cross-sectional study (The European Youth Heart Study). BMC Public Health. 2009; 9: 322.
16. Reichert FF, Menezes AM, Kingdom Wells JC, Ekelund E, Rodrigues FM, Hallal PC. A methodological model for collecting high-quality data on physical activity in developing settings-the experience of the 1993 Pelotas (Brazil) Birth Cohort study. J Phys Act Health. 2009; 6 (3): 360–6.
17. Reilly JJ, Penpraze V, Hislop J, Davies G, Grant S, Paton JY. Objective measurement of physical activity and sedentary behaviour: review with new data. Arch Dis Child. 2008; 93 (7): 614–9.
18. Riddoch CJ, Bo Andersen L, Wedderkopp N, et al.. Physical activity levels and patterns of 9- and 15-yr-old European children. Med Sci Sports Exerc. 2004; 36 (1): 86–92.
19. Romanzini M, Reichert FF, Lopes Ada S, Petroski EL, de Farias JC Jr. Prevalence of cardiovascular risk factors in adolescents. Cad Saude Publica. 2008; 24 (11): 2573–81.
20. Rowlands AV. Accelerometer assessment of physical activity in children: an update. Pediatr Exerc Sci. 2007; 19 (3): 252–66.
21. Santos CM, de Souza Wanderley R Jr, Barros SS, de Farias JC Jr, de Barros MV. Prevalence of physical inactivity and associated factors among adolescents commuting to school. Cad Saude Publica. 2010; 26 (7): 1419–30.
22. Sirard JR, Pate RR. Physical activity assessment in children and adolescents. Sports Med. 2001; 31 (6): 439–54.
23. Sirard JR, Slater ME. Compliance with wearing physical activity accelerometers in high school students. J Phys Act Health. 2009; 6 (1 suppl): S148–55.
24. Strong WB, Malina RM, Blimkie CJ, et al.. Evidence based physical activity for school-age youth. J Pediatr. 2005; 146 (6): 732–7.
25. Tenorio MC, Barros MV, Tassitano RM, Bezerra J, Tenorio JM, Hallal PC. Physical activity and sedentary behavior among adolescent high school students. Rev Bras Epidemiol. 2010; 13 (1): 105–17.
26. Troiano RP, Berrigan D, Dodd KW, Masse LC, Tilert T, McDowell M. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc. 2008; 40 (1): 181–8.
27. Trost SG. Measurement of physical activity in children and adolescents. Am J Lifestyle Med. 2007; 1 (4): 299–314.
28. Trost SG, Loprinzi PD, Moore R, Pfeiffer KA. Comparison of accelerometer cut points for predicting activity intensity in youth. Med Sci Sports Exerc. 2011; 43 (7): 1360–8.
29. Trost SG, Pate RR, Freedson PS, Sallis JF, Taylor WC. Using objective physical activity measures with youth: how many days of monitoring are needed? Med Sci Sports Exerc. 2000; 32 (2): 426–31.
30. Victora CG, Araujo CL, Menezes AM, et al.. Methodological aspects of the 1993 Pelotas (Brazil) Birth Cohort Study. Rev Saude Publica. 2006; 40 (1): 39–46.
31. Victora CG, Hallal PC, Araujo CL, Menezes AM, Wells JC, Barros FC. Cohort profile: the 1993 Pelotas (Brazil) birth cohort study. Int J Epidemiol. 2008; 37 (4): 704–9.


©2012The American College of Sports Medicine