Iranian Journal of Ergonomics

Increased cardio-respiratory fitness (CRF) is one of the key factors related to physical fitness that is associated with reduced mortality and metabolic diseases [1]. CRF is an important indicator of the ability of the heart and arteries to pump oxygen-rich blood to the tissues during exercise and the consumption of as much oxygen in these tissues, therefore, CRF is essential in most sports [3, 2]. This index has been reported as an indicator for measuring health status and predicting cardiovascular diseases [5, 4]. Researchers consider the gold standard index of CRF evaluation to be maximum oxygen consumption or VO_2peak [7, 6]. The standard CRF assessment method is direct VO_2peak measurement using a respirator analyzer in maximal graded exercise test using a treadmill [8].
However, the possibility of using these exercise tests for patients, sports injuries and people who have been away from exercise for a long time is dangerous and sometimes impossible. The use of field or laboratory sports tests in large populations and epidemiological studies or when it requires rapid VO_2peak estimation is limited despite its relatively high accuracy and sensitivity [20].
In order to design and validate different multivariate regression predictor equations, in different populations, researchers have found a close relationship between anthropometric and physiological variables such as age, sex, BSA, infertility, resting heart rate with VO_2peak [26-24, 22]. This method of non-exercise predictive equations that estimate people at different ages without VO_2peak physical activity can be useful in the elderly, especially in patients with cardiovascular disease, hypertension, and metabolic patients [27, 10]. According to research, in Iran, less design and validation of non-exercise equations predicting VO_2peak has been done and most of the existing equations are valid for foreign countries [28-30]. Therefore, in the present study, the design of indigenous estimation equations of non-exercise estimation with high determination coefficient and low predictive error of VO_2peak predictor and validation by standard method of performing sports tests during disabling activity along with respiratory gas analysis in adolescent boys population is investigated.

 

In this applied study, we voluntarily selected 156 healthy male students aged 13 to 17 years from among the selected secondary schools in a targeted cluster method from districts one and two of Hamadan city, Iran.
Students were first asked to complete a PARQ health status questionnaire in consultation with their parents and complete the PARQ.
In this study, 104 people were included in the model design group to design the VO_2peak non-exercise equation. We selected about 35 statistical samples for each independent variable (age, BMI, resting heart rate). Also, 52 people were included in the VO_2peak indigenous equations to verify the non-exercise indigenous equations. The present research process was approved by the Research Ethics Committee of Hamadan University of Medical Sciences (Code of Ethics Committee IR.BASU.REC.1398.006 :).
Kolmogorov-Smirnov test (K-S) was used to examine the normal distribution of data. Parametric statistics and appropriate descriptive statistics were used to test the research hypotheses [32]. Pearson correlation was used to evaluate the relationship between VO_2peak and independent variables (age, weight, BMI, body surface area and resting heart rate). To design the VO_2peak non-exercise estimation equation, step-wise multivariate regression was used to design the model. The selection criteria for regression equation were considering the components of coefficient of determination (R2), standard estimation error (SEE) and standard relative estimation error (100% (mean VO_2peak measured / SEE) =% SEE) [33]. In order to evaluate the accuracy and efficiency of the non-exercise native equation designed in the present study, the estimated VO_2peak obtained from the non-exercise native equations was compared with the actual VO_2peak measured in the subjects of the validation group (n = 52). This comparison was made using Pearson correlation test and paired t-test. Statistical analysis was performed using SPSS software version 24 (SPSS Inc., Chicago, Ill., USA) at the level of P-value <0.05.

Significant correlations were observed between the measured VO_2peak and the anthropometric and physiological variables (R=0.122 - 0.799, P<0.001). Also, a valid non-exercise linear equation for boys' VO_2peak prediction was designed with variables such as age, BMI and resting heart rate (SEE = 3.59 mL/kg/min, R² = 0.712, P<0.001). The estimated VO_2peak from equations had a significant correlation with the obtained criterion value. (R = 0.707 – 0.730, P<0.01) (Tables 1-3).
 

Table 1. Assessing the accuracy of foreign VO_2peak estimation equations in Iranian boys

^**P-value<0.01; ^*P-value<0.05

Figure 1.  Comparison of standard and estimated VO_2peak in the validation group (Issue 1: VO_2peak measured standard, Numbers 2-4: VO_2peak estimated with native non-exercise equations, Numbers 5-7: VO_2peak estimated with foreign non-exercise equations). Significant difference between measured and estimated VO_2peak P-value <0.01
 

Table 2. Correlation matrix of the studied variables in the linear model

^**P-value<0.01; ^*P-value<0.05

Table 3. VO_2peak non-exercise estimation equations

The results of this study showed that native non-exercise regression equations have a high validity in estimating the aerobic capacity of 13 to 17 year old boys (P<0.001 and R2 = 0.63-71). In this study, 3 native non- exercise equations were designed and validated to estimate VO_2peak for healthy boys. Based on our results, the independent variables of age, infertility and HR(rest) in the range of 63 to 71% explain the variability of boys' VO_2peak (P<0.001 mL / kg / min, SEE = 4.5 - 3.59). In the present study, a high correlation was found between BMI and VO_2peak, which was consistent with the results of George et al. [16]. But Kind et al. [34] reported that the effect of waist circumference on VO_2peak was stronger than BMI in men. The reason for the difference with the findings of the present study is probably due to the high effect of abdominal obesity at older ages on VO_2peak [35]. Also, a high correlation was reported between HR(rest) and VO_2peak, which was consistent with the results of George et al. [16] and Schantz et al. [36] but inconsistent with the results of Hess [37]. This difference may be in the level of readiness, gender and nutrition of the subjects. However, no significant correlation was observed between age and VO_2peak, which was consistent with the results of Wei et al. [38] but inconsistent with the results of Fleg et al. [39].
 Although the variables used in the non-exercise equations of previous studies are different, anthropometric variables, age, heart rate while resting and doing activities, previous physical activity, fat percentage, genetics and smoking have usually been the most important factors influencing Vo_2peak [14-14, 11, 10]. In the present study, a significant correlation was observed between Vo_2peak obtained from the standard method and measured variables (Table 3). Scientific studies have identified body composition factors such as weight, body fat percentage and restlessness as significant variables in linear and nonlinear regression equations of Vo_2peak estimation that can have a positive or negative correlation [47-47, 17].
Our findings indicated that the use of more than one variable in predictive equations tends to create higher correlation values between predicted and measured VO_2peak (Table 4). However, the study of alignment components in the multivariate regression pattern of VO_2peak non-exercise estimation equations showed the occurrence of an effective alignment between the independent variables of illness, BMI and weight in the equations, which removed these variables from the native equations of this study (VIF = 298.3, tolerance=0.003). The validity and high accuracy of the three indigenous non-exercise predictor equations measured by standard VO_2peak were confirmed by a suitable statistical sample in this study (SEE = 3.59 - 4.59 and R = 0.79 - 0.84). In Equation No. 3, by selecting the physiologically and physiologically dependent variables of age, BMI and HR(rest), the optimal correlation between aerobic capacity by non-exercise method (SEE = 3.59 and R = 0.84) and the standard method was obtained, which is compared with two native linear equations. The other was even smaller in predictive error.

 According to the results of the present study, it is possible to use aerobic capacity estimation equations, which is a simple, accurate and safe tool in assessing the baseline cardiorespiratory fitness (VO_2peak). It can be used as a daily clinical practice in large populations of adolescent boys and even heart patients and the elderly with the aim of promoting health, cardiovascular health, preventive actions and planning aerobic exercise program.
 

This study was conducted from the faculty validation of the Vice Chancellor for Research and Technology of Bu Ali Sina University. The authors would like to thank the staff of the Education Department of Hamadan, the Association of Parents and Teachers of the schools and the students who cooperated well in the implementation of this project.

 

The authors declared no conflict of interest.