Study Design and Population

The Cuenca Adults Study is a longitudinal study aimed at assessing the changes in lifestyle and cardiovascular risk factors that occur during an individual’s time at university [31–33].

The measurements of this study were conducted in 2009–2010 in the Cuenca campus (University of Castilla-La Mancha, Spain). From a total of 963 students in their first year, 770 students were invited to participate. They were all first year students of eight different degrees (Psychology, Law, Nursing, Architecture, Engineering, Business, Fine Arts and Social Work) and only 6 of 12 groups of first year of Education grade which were randomly invited to participate (Fig 1). Only the Nursing degree course was related to health. From the 770 invited students, 683 (88.7%) agreed to participate (504 woman, 73.8%). This sex distribution was similar to the whole campus student population (4,814 students, 63% women, n = 3,022), so that some degrees such as Nursing or Social Work could be mainly preferred by women.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Flowchart showing the progress of students through the study.

https://doi.org/10.1371/journal.pone.0141977.g001

The investigators attended first year courses to explain the aims of the study and invite participants; those interested signed the informed consent.

Students received a T-shirt and one European Credit Transfer System (ECTS.) for having participated in the study.

The university population included students older than 17 years old. Most of them lived in rented flats (40%) or in campus hostels (30%) close to the campus, so they are not far away from the lecturing places. In comparison to the other Castilla-La Mancha campuses, the campus of Cuenca is not very crowded and is located about a mile and a half away from the town centre. A sports service that organizes competitions and other recreational activities is situated in the campus. However, the student participation is quite poor. Only 3% of university women and 51% of men were engaged in competition activities and 19% of women and 23% of men in leisure activities.

The study protocol was approved by the Clinical Research Ethics Committee of Virgen de la Luz Hospital in Cuenca, and all subjects signed informed consent forms prior to their participation in the study.

Anthropometry

The measurements were made in the University by trained investigators to minimize inter-observer variability. Body mass and height were measured following standardized recommendations with an electronic scale (SECA Model 861; Vogel & Halke, Hamburg, Germany, precision = 100 g, range = 0–150 kg), and a stadiometer (Type Seca 222, precision = 0.1 cm, range = 6–230 cm). The average of two measurements was taken. Body mass index (BMI) was calculated as body mass (in kilograms) divided by height² (in metres).

Assessment of physical activity

Participants wore an accelerometer (model MTI/CSA 7164, Actigraph, Shalimar, FL, USA) for seven consecutive days. A minimum of four days with valid data (minimum 600 registered min/day and including at least one weekend day) were set as inclusion criteria [33,34]. The device was worn on the right hip of the participants, attached with an elastic band. Because the accelerometers were not water-resistant, the subjects were asked to remove them before showering or entering a pool. In this study, the interval of recorded time (epoch) was set at 60 seconds, shown as valid for measuring PA in adults [35], and any 10-min intervals of continuous ‘zero’ activity were excluded from the analysis. Data were collected throughout the school year (September to July). Considering the students’ answers in the PA log (mean bedtime and wake up) it was considered ‘Awake time’ from 8.00 to 0.00 h. Accelerometer data were processed (Kinesoft v3.3.55) to provide values for each hour between 8.00 and 0.00 h for total daily and hourly counts per minute (cpm), sedentary time, light PA, moderate PA, vigorous PA, moderate-vigorous PA (MVPA), steps and non-wear time for weekdays and weekend. The intensity of weekly PA was assessed as average cpm. The cut-off values used to define the intensity of PA and, therefore, quantify the mean time in each intensity were the following: sedentary time = <100 cpm, light PA = 100–1951 cpm, moderate PA = 1952–5724, vigorous PA = ≥5725, and MVPA was calculated as the sum of moderate and vigorous PA [36].

Finally, the student percentage that had met different PA recommendations for adults was quantified and assessed. According to different public health guidelines, the sum of the min/day of moderate and vigorous PA and number of steps/day were processed. This was structured from low to high demand criteria of the guidelines [18–23] and by calculating the MVPA and vigorous PA in bouts of at least 10 minutes (allowing a minute or two below level marking):

1. -. Recommendation 1 (R1): ≥30 min of MVPA for most of the week (5 days) [18].
2. -. Recommendation 2 (R2): ≥150 min/week MVPA (R2A) and/or ≥75 min/week vigorous PA in bouts of at least 10 min (R2B) [20,21].
3. -. Recommendation 3 (R3): ≥150 min/week MVPA (R3A) and/or ≥75 min/week vigorous PA (R3B) in bouts of at least 10 min in 5 or more days/week [22].
4. -. Recommendation 4 (R4): ≥30 min/day MVPA × 5 days/week (R4A) and/or ≥20 min/day vigorous PA × 3 days/week (R4B) in bouts of at least 10 min [19].
5. -. Recommendation 5 (R5): 10,000 steps/day [23].

The students who had less than five valid days were excluded from the analysis of recommendations R3A, R3B and R4A.

Calculation of sample size

The sample size (n = 451) was calculated by means of the software StatsDirect estimating a prevalence of overweight/obesity of 25%, a confidence level of 95% and a precision of 4%. Estimating a rate of not response of 10%, the total size of the sample was 496 students.

Statistical Analyses

The variables were checked for normal distribution both through graphical procedures and by the Kolmogorov-Smirnov test. Descriptive data were assessed by the t-test for normally distributed variables and Mann-Whitney’s U for non-normally distributed variables.

As none of the PA-related variables had a normal distribution, the Wilcoxon test was used to compare PA intensities on weekdays and weekends. Since an interaction between sex and the studied variables was not observed (all p>0.05), all analyses were performed for men and women together.

All statistical analyses were performed with the statistical software IBM^® SPSS^® Statistics 19, and the level of significance was set at p<0.05.

Physical activity levels and patterns

Compliance with different physical activity recommendations

Our results showed, in keeping with other similar studies [3,4,13,46], that young adults do not meet the official PA recommendations for health. In addition, when PA guidelines include more flexible criteria [18], the compliance increases markedly [15]. Previous studies have also shown that low levels of PA are especially evident when bouts of 10 min are considered [28]. If we compare R1 with R4A, we observe these have the same criteria, but without the bouts requirement, and compliance decreases from 30.3% to 0.5%. This indicates that university students accumulate PA during periods of very short duration and/or with frequent interruptions. A study of the Portuguese population [5] also showed that when blocks of 10 min of continuous activity of moderate or vigorous intensity were considered, the prevalence of the active population dramatically decreased, varying from ~70% of people aged 18–29 to ~5% accumulating at least 30 min of PA a day. However, this decrease was much lower in persons between 40 and 64 years as they showed a greater predominance of more continuous activities in leisure time as opposed to intermittent activities, probably more dominant in younger ages. Although there are limited epidemiologic studies that have examined the association with health outcomes between accumulation of PA in 10-minute bouts and without bouts, this encouraging evidence indicates that short bouts of exercise can increase cardiorespiratory fitness, promote weight loss, and improve various biologic health outcomes [47,48]. However, a study of 3,250 adults aged 18 years or older [49]affirmed that bouts of PA lasting > or = 10 min may be a more time-efficient strategy to decrease BMI and waist circumference.

The same thing occurred when we considered activity on less than five days/week, where university students performed 150 minutes of MVPA in bouts of 10 min. The compliance decreased from 5.4% (R2A) to 0.5% (R3A). Scheers et al. [15]showed that 94.2% of men and 86.5% 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. Although prevalence in compliance with the recommendations is much higher than in our study, this reflects that removing the frequency and duration requirements resulted in a considerable increase in the prevalence of sufficient MVPA. This fact deserves reflection, for that reason, we know that for a habit to occur, the behaviour needs to be repeated regularly, so strategies must be created for university students not only to increase their PA duration, but also the number of days a week when it is performed.

The percentage of adults who achieved at least 60 or 75 min of vigorous PA per week was zero, with or without the bouts criteria, showing that university students do not perform vigorous PA. Studies in healthy people and in individuals with coronary artery disease suggest that vigorous PA, measured subjectively, provides greater benefits than moderate PA [50,51], so it would be necessary to establish strategies to increase the practice of vigorous PA during this stage. A relevant strategy would be to alert university students that the time taken to implement the recommendations of vigorous-intensity PA is lower than with moderate intensity, so they would not need to spend much time weekly. These results should be taken with caution as PA recommendations are based on epidemiological data that is collected using self-report measures [18] and cannot be directly translated to data using objective methods.

The compliance to get the daily average 10,000 steps was 28.1% (R5). Our results were much lower than a study in South Africa [53], which showed that 77.9% men and 59.4% women met the guidelines. Moreover, the degree of compliance with PA guidelines in university students was also very low, as was that in other studies conducted in Spanish adults [24,26], although these studies used questionnaires to gather their data.

Limitations

Some limitations should be recognized. First of all, it is possible that accelerometers may produce some reactivity by the participants (Hawthorne effect) in wearing the device; however, unlike self-reports, accelerometer estimates do not suffer from bias due to social desirability and recall problems. In addition, we reduced this effect since the data recorded on the first day of data collection was not considered. Secondly, PA levels and the degree of compliance with the recommendations may be influenced by uncontrolled factors in our study, such as socioeconomic level or weight status. Thirdly, we do not have any reliability data for the people conducting the measures. Fourthly, we are aware that the definition of accelerometer non-wear time could change average wearing time, average accelerometer counts and also might underestimate sedentary time [52]. Thus, our results might have been different if we had used a different definition of accelerometer non-wear time, because when we increase the time period of consecutive zero for exclusion (from 20 to 40 to 60 minutes) total wearing time increases. Note that probably excluding 10 minutes of consecutive zero values as non-wear time might have reduced the estimates of valid wear time for each measurement day, and as consequence would have reduced the number of days that meet the criterion of five or more valid days, underestimating the overall sample size. However, there were no age, sex or anthropometric variable differences between students who had five or more valid days and those who did not, so we think that the representativeness of the sample was guaranteed. Fifth, the lack of agreement between statements for minimum data of accelerometer recordings and recommendations for PA in most guidelines could affect our results, since while it is usually accepted that 4 days of accelerometer data recording is enough to measure PA behaviour [34], accumulating at least 150 minutes of moderate-to-vigorous PA in five or more days a week is recommended in most international guidelines. In order to overcome these limitations, a recent publication stressed the need for technological advances in devices. These advances will allow the accelerometer to be worn on the wrist to increase adherence, to be programmed in the 5 hertz (Hz) to 100 Hz range to better record all kinds of movements and with an interval of recorded time (epochs) of 1 second, so that the activity time will grow exponentially [53]. Finally, accelerometers that had been combined with other sensors (measuring physiological responses) would have guaranteed a more accurate PA measurement.