Sociodemographic determinants of the level of physical activity in residents of Eastern Poland

Abstract

Background

Post-industrial societies that benefit from the development of science, technology and subsequent inventions that relieve people of everyday duties, have more free time, but face a greater temptation of laziness and limited physical activity. Common diseases increasingly resulting from limited physical activity, which is no longer just a way to spend free time, but a necessity in the field of health care. This necessity obliges us to undertake research that allows for recognizing factors influencing the level of physical activity in individual societies.

Objective

The conducted research aimed at identifying sociodemographic factors that would determine the level of physical activity in women and men from the Biała Podlaska district in eastern Poland.

Participants

The group involved 173 adults (71 women and 102 men) from eastern Poland.

Methods

The presented research was conducted in the years 2018–2020 as part of the international EUPASMOS Plus project. The collected sociodemographic data of the respondents and the results of physical activity monitoring with the use of the GPAQ questionnaire and the RM42 accelerometer − 24/7 allowed for an analysis of the factors determining the physical activity undertaken by the respondents, as well as the comparison of the obtained data with the use of the above-mentioned tools. The statystical analysis involved the use of the Shapiro-Wilk test, Mann-Whitney test and Spearman’s rho correlation analysis. The adopted standard significance level of α = 0.05.

Results

The results of the research on sociodemographic factors conditioning the physical activity of the examined persons, obtained from the objective tool (Accelerometer RM42) and a subjective one for measuring physical activity (GPAQ questionnaire) showed some discrepancies. However, the established consistency of the research results using the above-mentioned tools allows for formulating the following conclusions: women from the Biała Podlaska district are more active than men. Older people more often undertake PA of lower intensity, giving up high-intensity efforts. The respondents declaring a higher subjective assessment of their health are more physically active.

Conclusions

The results obtained are varied and depend on the used tool. They indicate an enormous importance of the tool used in the study on physical activity.

Introduction

The World Health Organization (WHO) has been warning for years about insufficient physical activity on a global scale. Estimates from 2010 indicated that 27.5% of adults [1] and 81% of youth [2] do not meet the recommendations for health-promoting physical activity [3]. This trend has also continued over the last decade. As shown by the latest WHO data, 31% of adults and 80% of adolescents do not meet the recommended levels of physical activity [4]. The studies from recent years also show how much the pandemic has influenced physical activity [5, 6].

Physical activity is of great importance at every stage of a person’s life. All the more alarming are the results of the latest research carried out in 17 primary schools in Poland, which shows that as many as 94% of children and young people in Poland do not have sufficient physical activity skills. This is alarming information, which should prompt reflection both on the physical fitness of the youngest, as well as on the quality of physical education lessons and the substantive preparation of the teachers conducting these lessons [7].

Kantanista et al. also highlighted in their 2022 study the insufficient level of physical activity in children and adolescents from the Czech Republic, Hungary, Poland, and Slovakia (mainly girls), and the need for public health interventions aimed at increasing physical activity [8].

Confirming the importance of physical activity for quality of life is the study by Lin et al. As the meta-analysis shows, physical activity promotes successful aging among middle-aged and older adults [9].

Meanwhile, regular physical activity is considered as a preventive and therapeutic behavior in the case of non-communicable diseases, including type 2 diabetes, cancer, cardiovascular diseases and mental illnesses. This is confirmed by the results of recent studies [10–13]. One of their conclusions is that the research results might serve as a guide for healthcare providers in enhancing physical activity adherence among patients with non-communicable diseases through an illness perception approach.

Among the many factors determining physical activity, the WHO distinguished, among others, gender - showing that in most countries women are less active than men [14]. Another important factor is economic conditions, the analysis of which showed differences in physical activity among people with a higher and lower economic status [14]. Results from European studies indicate that physical activity decreases with age [15–17]. Education and acquired knowledge have also been identified as important determinants of physical activity uptake. According to the results of a study, people with higher education undertake physical activity more frequently [17, 18]. Self-assessment of one’s health and quality of life has been identified as another important determinant of physical activity uptake, particularly in older people [19, 20]. The importance of physical activity in the prevention of non-communicable diseases, as well as the relatively small percentage of people undertaking health-promoting physical activity encourages research in this area, i.e. in the areas of research groups from individual countries, as well as a search for such research methods and tools that will enable the collection of reliable results comparable with other populations. The results of the studies conducted around the world confirm the beneficial effect of physical activity on the broadly understood health of respondents [21–27].

Being aware of the risks associated with the lack of physical activity, the European Union implemented an evidence-based policy in subsequent years aimed at promoting physical activity and reducing sedentary behaviours on a population scale [28–30].

The following paper presents the results of the research data on monitoring physical activity using the RM42 accelerometer, the results of the subjective assessment of physical activity using the GPAQ questionnaire, and sociodemographic factors that influence the participation of women and men in physical activity in the Biała Podlaska district in eastern Poland.

Aim of the study and research questions

Physical activity, if undertaken regularly, can be an excellent tool for health prevention. This is evidenced by the recommendations regarding health-promoting physical activity published by the World Health Organization. However, the problem of today’s societies is very often a sedentary lifestyle and, as a consequence, insufficient physical activity. Lack of exercise (hypokinesia) together with other negative factors cause many non-communicable diseases. The aim of the analyses was to indicate the sociodemographic factors that determine the level of physical activity of people living in the Biała Podlaska district, in the area of ​​eastern Poland.

Based on the formulated objective, the following research question was posed:

What sociodemographic factors determine the level of physical activity of the residents in the Biała Podlaska district in eastern Poland?

Materials and methods

Before carrying out the study, the authors familiarised themselves with the available research methods in order to select the most appropriate ones from among them.

First of all, it should be noted that subjective and objective methods are used to assess physical activity [31]. In the case of subjective (estimation) methods, respondents are asked to subjectively estimate the physical activity undertaken per day, per week, per month. In the case of objective (measurement) methods, physical activity is measured objectively by monitoring the physiological responses of the body to physical exertion [32] or by recording kinematic parameters of movement (mainly accelerations of the whole body or its parts).

Due to their high availability, low cost and relative ease of survey implementation, subjective methods are among the most commonly used for large groups of respondents. It should be noted that in this case, a particularly heavy responsibility rests with the person conducting the survey. This person should be adequately prepared in terms of content and have an appropriate approach to respondents.

The work undertaken within the EUPASMOS Plus project aimed at collecting the most comprehensive data on the physical activity of the respondents. The analysis of the latest literature helped to create a unified questionnaire used for global studies - GPAQ (Global Physical Activity Questionnaire) [33], which was selected from among the subjective tools for measuring physical activity. This is a questionnaire whose basic validation criterion was the International Physical Activity Questionnaire (IPAQ), considered to be the best validated and most frequently used questionnaire for measuring physical activity [34].

In connection with the researchers’ indications of cultural differences between respondents from different countries and cultures, which may affect different understanding of the terminology used in the questionnaire, and consequently the obtained results [35–37], we paid special attention to adapting the questionnaire in terms of language to respondents from individual countries.

For the objective measurement of physical activity, RM42 accelerometers were used, which were constructed and validated at the UKK Research Institute in Tempere, Finland.

Acceleration data were acquired in a range of ± 16 G at a sampling rate of 100 Hz. In accordance with the guidelines [38] physical activity analysis was based on the mean amplitude deviation (MAD) in six-second epochs [38]. Subsequent MAD values were converted to METs (3.5 mL/kg/min oxygen consumption). Thereafter the epoch-wise MET values were further smoothed by calculating an exponential moving average for each epoch time point [39, 40] and were analysed in 6-s epochs. The following cut points wereseted: 3.0 METs ≤ MPA < 6.0 METs and VPA ≥ 6.0 METs.

In order to avoid any errors in the computer analysis of raw data collected using the RM42 accelerometer, the probability of which was suggested by other researchers [40–42] a checklist was created to verify whether a given measurement day can be considered valid and the collected data can be analyzed. The operating procedure paid attention to such parameters as the minimum measurement period, the cut-off point in terms of effort intensity, the criterion indicating that the accelerometer was not worn, the minimum time of wearing the device during the day and the number of days on which it was worn.

Research procedure

The interview based on the GPAQ questionnaire was conducted using the “Face to Face” method, using digital support - a specially designed computer platform CAPI (Computer Assisted Personal Interviews), which included a digital version of the GPAQ questionnaire and questions regarding the socio-demographic characteristics of the study participants. Objective measurements of physical activity were performed using RM42 accelerometers in accordance with the approved research protocol. Each respondent was instructed on how to use the accelerometer, received the device and wore it for the next seven days (24/7), during which an objective measure of physical activity was recorded. The devices were worn by the study participants at hips height, on the right side. For the comfort of the subjects and in order to position the accelerometer appropriately, the subjects were given an adjustable strap with a pocket for the RM42 accelerometer along with the device. Each subject was given instructions printed on a sheet of paper, which, among other things, emphasised that the accelerometer must be worn on the right hip, in a special pocket, and that the closure of the pocket (Velcro) should face upwards and inwards (so that the measuring device does not fall out). The strap itself should fit tightly on the hip, preventing the accelerometer from moving around on the hip. While sleeping, the device was worn on an adjustable wrist strap. The accelerometer was positioned on the non-dominant hand at the time, placed in a pocket located on the outside of the wrist.

The study material involved a group of 173 people of both sexes (71 women and 102 men) aged 18+, participants of the EUPASMOS Plus study. The selection of groups was made in a purposeful manner. The sample size was determined in accordance with the COSMIN recommendations [43]. The inclusion criteria were dictated by ethical principles (written consent to participate in the study), in accordance with the assumption of the study, the respondents were healthy adults and were divided into four age groups: 18–34, 35–49, 50–64 and 65 +. The health condition of the respondents had no effect on the undertaken physical activity. People who did not meet the above inclusion criteria could not be included in the study.

• the inclusion criteria included:

  • submitting a declaration of consent to participate in the study;
  • age 18 years or more;
  • ano health contraindications to undertaking health-promoting physical activity.
• exclusion criteria included:

  • refusal to participate in the study;
  • age below 18;
  • health condition precluding participation in any physical activity.

The entire research procedure was carried out by individuals who were members of the EUPASMOS Plus research team.

In accordance with ethical principles and due to the sensitivity of the data and the potential for traceability, all data collected was coded and then collated into two Excel sheets:

1. “EUPASMOS_Database_QuestionnairesValidation_Poland”

2. „EUPASMOS_Database_DataCollection_Poland”.

The above presented study was approved of by resolution no. 6/2018, the bioethics committee of the PSW in Biała Podlaska.

The research conducted in the years 2018–2020 was part of the international EUPASMOS Plus program, financed by the European Commission under the Erasmus + Program grant no. 603,328-EPP-1–2018−1-PT-SPO-SCP. The aim of the research was to develop a harmonized system for monitoring physical activity, sedentary behavior and participation in sports. The research process was carried out based on a strictly defined methodology including: anthropometric and sociodemographic studies and monitoring of physical activity using the RM42 accelerometer. The subjects were trained in how to put on and wear the accelerometer and document cases of not wearing the device. Each person also kept a physical activity diary, in which recreational and sports activity was reported. The subjects were required to wear the accelerometer for 7 consecutive days (24/7). On the eighth day, the study participants were asked to report to the research laboratory to submit accelerometers and physical activity diaries. Interviews using physical activity questionnaires were also conducted that day. Participation in the study was free of charge and voluntary.

Statistical methods

The data collected during the research process were coded in a database that included both quantitative and categorical variables, and then subjected to in-depth statistical analysis using TIBCO Software Inc. (2017). The selected program analyzing the data was Statistica data analysis software system, version 13. and IBM SPSS Statistics 26 package. In order to answer the research question, basic descriptive statistics analysis was performed with the Shapiro-Wilk test, Mann-Whitney test and Spearman’s rho correlation analysis. The level of significance was considered to be α = 0.05.

An assessment of the correspondence of the empirical distribution with the normal distribution was carried out using the Shapiro-Wilk test. This test has a high statistical power especially for small and medium-sized samples, as in the case of the research conducted.

The result of the Shapiro-Wilk test for most of the entered variables turned out to be statistically significant, which means that their distributions significantly deviate from the normal distribution. It should also be noted that the skewness of the distribution of these variables often exceeded the absolute value of 2, which means that their distributions are asymmetric to a significant extent. Therefore, it was considered justified to conduct an analysis based on nonparametric tests.

For comparisons by gender and place of residence, the Mann-Whitney U test was used. This test is a non-parametric alternative to the T-student test and is used to compare two different independent test groups. This test analyses the difference in medians between the test groups and is robust to violations of the normality of distribution and non-equality of variance. When examining the relationship between quantitative variables, Spearman’s rho correlation coefficient (also known as Spearman’s rank coefficient) was used. It is a measure of the relationship between the two quantitative variables under study. It is used similarly to the Mann-Whitney U-test when the normality of the distribution of the variables under study is violated.

The effect size was calculated using r_g - Glass’s bivariate correlation coefficient. A post hoc Power (1-β) statistical power analysis was then performed. Compute achieved power - given α, sample size, and effect size.

Sociodemographic characteristics

The group of the respondents involved people aged 19 to 78 (M = 46.07; SD = 18.19), the majority of whom were men (59.0%). The surveyed persons lived together with 1–8 people (Me = 2.00) (Table 1).

Table 1.

Basic descriptive statistics of quantitative sociodemographic variables

	M	Me	SD	Min.	Maks.
Age	46,07	47,00	18,19	19,00	78,00
Number of persons in the household	2,80	2,00	1,36	1,00	8,00

The majority of the respondents lived in houses (57.8%) located in small towns or suburbs (78.0%). Almost one third of them declared higher education degree, master’s degree (35.3%), or secondary education (33.5%). The declared health status of most people (54.9%) was “good” (Table 2).

Table 2.

Percentage distributions of nominal and ordinal sociodemographic variables

		N	%
Gender	Men	102	59,0
	Women	71	41,0
Documented education	Secondary	58	33,5
	Post-Secondary	30	17,3
	Bachelor’s or equivalent	11	6,4
	Master’s or equivalent	61	35,3
	PhD or equivalent	13	7,5
Type of residence	Flat	73	42,2
	House	100	57,8
Place of residence	Big city	1	0,6
	Small town	135	78,0
	Countryside	37	21,4
Health condition	Very good	29	16,8
	Good	95	54,9
	Moderate	48	27,7
	Bad	1	0,6
Age group	< 30 years	43	24,9
	30–39 years	23	13,3
	40–49 years	28	16,2
	≥ 50 years	79	45,7

Analysis of the research results

Sociodemographic factors influencing the level of physical activity in the respondents assessed using the RM42 accelerometer

In order to search for a relationship between the accelerometer data and sociodemographic variables, the Mann-Whitney test was performed to check the relationship with gender and place of residence, and Spearman’s rho correlation analysis was performed to check the relationship with other variables.

First, the differences between the genders in terms of accelerometer variables were verified. The results of the analysis are presented in Table 3.

Table 3.

Comparison of women and men’s physical activity with regard to the data from accelerometers

	Men (n = 102)			Women (n = 71)
	average rank	Mdn	IQR	average rank	Mdn	IQR	Z	p	r g	Power(1-β)
LPA	84,92	4:20:51	01:35:57	89,99	4:18:27	02:23:08	−0,66	0,512	0,06	0,10
MVPA	77,04	0:45:05	00:27:55	101,31	1:01:56	00:59:00	−3,14	0,002	0,28	0,55
VPA	82,19	0:00:08	00:01:02	93,92	0:00:20	00:04:27	−1,56	0,120	0,14	0,22
PA0-5	90,85	2:52:39	00:57:05	81,46	2:42:51	01:12:05	−1,21	0,225	0,11	0,17
PA5-10	83,83	0:56:46	00:30:34	91,55	0:58:59	00:36:59	−1,00	0,319	0,09	0,14
PA10+	78,29	1:05:23	01:02:23	99,51	1:28:33	01:28:51	−2,74	0,006	0,25	0,47
Steps	81,01	7170,57	3250,47	95,61	8358,86	5091,43	−1,89	0,059	0,17	0,28

LPA low intensity physical activity, MVPA moderate intensity physical activity, VPA high intensity physical activity, PA0-5 physical activity from 0 to 5 minutes, PA5-10 physical activity up to 10 minutes, PA10+ physical activity over 10 minutes, Steps number of steps

The analysis showed statistically significant differences between women and men in terms of moderate-intensity exercise (MVPA) and physical activity lasting more than 10 min (PA10+). It turned out that women were characterized by significantly longer daily time of moderately intensive physical activity, and longer time of physical activity above 10 min compared to men.

Statistical analysis showed that the power of the effect for MVPA and PA 10 + is close to average (respectivep rg = 0.28 and rg = 0.25).

Based on the results of the post hoc Power (1-β) statistical power assessment, we can conclude that, in the case of MVPA, the data obtained detect existing relationships with 55% probability. By contrast, in the case of PA10+, this probability turned out to be much lower, at 47%. We can therefore conclude that the chance of the existence of real gender-based correlations is moderate and, for confirmation, re-testing with a larger sample would be advisable.

In terms of the remaining variables, no statistically significant differences between the sexes were noted.

The next correlation that was checked was whether there was a difference between the physical activity undertaken by rural and urban residents with regard to the accelerometer data. The results of the analysis are presented in Table 4.

Table 4.

Comparison of the physical activity of rural and urban residents with regard to the accelerometer data

	City (n = 135)			Countryside (n = 37)
	Average rank	Mdn	IQR	Average rank	Mdn	IQR	Z	p	r g	Power(1-β)
LPA	89,09	4:20:47	01:54:47	77,04	3:59:44	01:48:10	−1,30	0,192	0,14	0,18
MVPA	86,58	0:50:01	00:40:23	86,20	0:47:16	00:38:33	−0,04	0,967	< 0,01	0,05
VPA	85,77	0:00:09	00:02:49	89,16	0:00:15	00:01:44	−0,38	0,706	0,04	0,08
PA0-5	88,81	2:52:32	00:58:42	78,05	2:42:18	01:11:19	−1,16	0,244	0,13	0,17
PA5-10	89,75	0:58:59	00:30:13	74,65	0:52:25	00:25:51	−1,63	0,102	0,18	0,24
PA10+	88,48	1:18:21	01:06:42	79,27	1:07:19	01:12:58	−1,00	0,319	0,11	0,14
Steps	88,13	7661,14	3936,00	80,54	7357,29	3073,58	−0,82	0,411	0,09	0,12

LPA low intensity physical activity, MVPA moderate intensity physical activity, VPA high intensity physical activity, PA0-5 physical activity from 0 to 5 minutes, PA5-10 physical activity up to 10 minutes, PA10+ physical activity over 10 minutes, Steps number of steps

The analysis did not show any statistically significant differences between city dwellers and rural dwellers with regard to the accelerometer measurement results, which indicates that, regardless of the place of residence, the subjects showed similar physical activity parameters.

The next examined correlation was the connection between physical activity and age, education and health status of the respondents. The results of the analysis are presented in Table 5.

Table 5.

Correlation of the accelerometer data with regard to age, education and health status

	Age		Documented education		Health condition
	r	Power (1-β)	r	Power (1-β)	r	Power (1-β)
LPA	0,16*	0,68	0,10	0,37	−0,05	0,16
MVPA	−0,36***	0,99	0,04	0,13	−0,28***	0,98
VPA	−0,51***	1,00	0,09	0,32	−0,38***	0,99
PA0-5	0,17*	0,73	0,17*	0,73	−0,02	0,38
PA5-10	< 0,01	0,06	0,08	0,28	−0,11	0,42
PA10+	−0,09	0,32	0,03	0,11	−0,19*	0,81
Steps	−0,26***	0,97	0,07	0,23	−0,23**	0,93

LPA low intensity physical activity, MVPA moderate intensity physical activity, VPA high intensity physical activity, PA0-5 physical activity from 0 to 5 minutes, PA5-10 physical activity up to 10 minutes, PA10+ physical activity over 10 minutes, Steps number of steps

* p < 0,050; ** p < 0,010; *** p < 0,001

The analysis showed statistically significant negative correspondence between age and moderate and vigorous exercise (MVPA, VPA), as well as the number of steps (Steps). The results also indicated statistically significant positive correlation between age and light physical exercise (LPA).

An assessment of post hoc power statistical power (1-β) confirmed the high probability of age correlation for MVPA, VPA and STEPS (0,97 − 1).

Weak positive correlations were observed between the education degree and physical activity up to 5 min (PA0-5). It also turned out that with the increase in the subjective assessment of health, moderate and vigorous physical activity (MVPA, VPA), physical activity above 10 min (PA10+) and the number of steps decreased. Again, the high probability of the above correlation was confirmed by a post hoc Power (1-β) statistical power assessment (0.93–0.99).

Sociodemographic factors influencing the level of physical activity of respondents in the studies conducted using the GPAQ questionnaire

In order to compare the results of the objective (accelerometer) and subjective (GPAQ) studies, it was checked whether there was any correspondence between the data from the GPAQ questionnaire and the sociodemographic variables of the examined persons.

For this purpose, the Mann-Whitney test was used to check the relationship with gender and place of residence, and Spearman’s rho correlation analysis was used to check the relationship with other variables.

First, the differences between the genders in the questionnaire variables were verified. The results of the analysis are presented in Table 6.

Table 6.

Comparison of physical activity of women and men with regard to the variables from the GPAQ questionnaire

	Men (n = 95)			Women (n = 65)
	average rank	Mdn	IQR	average rank	Mdn	IQR	Z	p	r g	Power (1-β)
GPAQ
WorkVigDays	67,12	0,00	0,00	81,33	0,00	1,00	−2,62	0,009	0,18	0,29
WorkVigMinDay	67,91	0,00	0,00	80,21	0,00	30,00	−2,56	0,011	0,15	0,23
WorkModDays	65,45	0,00	4,00	83,70	3,00	5,00	−2,72	0,007	0,23	0,40
WorkModMinDay	63,95	0,00	60,00	85,82	60,00	180,00	−3,24	0,001	0,27	0,49
TravelDays	72,30	3,00	5,00	73,99	3,00	5,00	−0,24	0,808	0,02	0,06
TravelMinDay	69,03	30,00	60,00	78,63	40,00	86,25	−1,38	0,169	0,12	0,18
SportsVigDays	65,79	0,50	3,00	81,90	2,00	4,00	−2,39	0,017	0,20	0,33
SportsVigMinDay	67,05	0,00	55,00	81,43	30,00	60,00	−2,17	0,030	0,18	0,29
SportsModDays	70,33	1,00	3,00	76,78	2,00	3,00	−0,94	0,346	0,08	0,12
SportsModMinDay	69,18	20,00	60,00	78,41	30,00	60,00	−1,36	0,175	0,12	0,18
MET (min/week)	64,12	1800,00	3440,00	85,58	4680,00	7900,00	−3,03	0,002	0,27	0,49

WorkVigDays frequency of intense physical effort at work, WorkVigMinDay length of intense physical effort at work, WorkModDays frequency of moderate physical effort at work, WorkModMinDay length of moderate physical effort at work, TravelDays frequency of getting around/travel, TravelMinDay time spent on getting around/travel, SportsVigDays frequency of intense physical effort while doing sports, SportsVigMinDay length of intense physical effort while doing sports, SportsModDays frequency of moderate physical effort while doing sports, SportsModMinDay length of moderate physical effort while doing sports, MET (min/week) metabolic equivalent value

The analysis of the results obtained from the GPAQ questionnaires showed statistically significant differences between women and men in terms of such variables as the length and frequency of intensive and moderate physical effort at work (WorkVigDays, WorkVigMinDay, WorkModDays, WorkModMinDay), the length and frequency of intensive physical effort during sports (SportsVigDays, SportsVigMinDay), as well as MET-min/week. It turned out that in each of the recorded cases, women were characterized by significantly higher values of the indicators compared to men.

However, statistical analysis showed that the strength of the effect for the above-mentioned variables is weak (rg = 0.02–0.27).

Based on the results of the post hoc Power (1-β) statistical power assessment, we can conclude that, in the case of effort at work, the data obtained detect relationships with low probability.

(23%−49%). Similarly, for sporting activities (29%−33%) as well as for MET-min/week (49%). In order to demonstrate real correlations, there is a need for further studies conducted on a larger sample.

Next, it was checked whether there were any differences between rural and urban residents in terms of physical activity declared in the GPAQ questionnaire. The results of the analysis are presented in Table 7.

Table 7.

Comparison of physical activity of urban and rural residents with regard to the variables from the GPAQ questionnaire

	City (n = 124)			Countryside (n = 36)
	average rank	Mdn	IQR	average rank	Mdn	IQR	Z	p	r g	Power (1-β)
GPAQ
WorkVigDays	71,90	0,00	0,00	76,74	0,00	1,00	−0,76	0,446	0,06	0,09
WorkVigMinDay	72,29	0,00	0,00	75,41	0,00	0,00	−0,55	0,581	0,04	0,08
WorkModDays	74,96	1,50	5,00	66,35	0,00	5,00	−1,09	0,275	0,11	0,14
WorkModMinDay	75,03	30,00	90,00	66,11	0,00	65,00	−1,13	0,260	0,11	0,14
TravelDays	76,67	4,00	5,00	60,55	3,00	4,50	−1,97	0,049*	0,20	0,27
TravelMinDay	76,87	30,00	57,50	59,86	30,00	50,00	−2,08	0,038*	0,21	0,29
SportsVigDays	73,29	1,00	3,00	69,85	1,00	3,00	−0,43	0,664	0,04	0,08
SportsVigMinDay	75,28	10,00	60,00	65,26	0,00	55,00	−1,29	0,198	0,13	0,16
SportsModDays	73,05	2,00	3,00	72,82	1,00	3,50	−0,03	0,977	< 0,01	0,05
SportsModMinDay	74,70	30,00	60,00	67,24	20,00	60,00	−0,93	0,351	0,09	0,12
MET (min/week)	71,90	0,00	0,00	76,74	0,00	1,00	−0,76	0,446	0,06	0,09

WorkVigDays frequency of intense physical effort at work, WorkVigMinDay length of intense physical effort at work, WorkModDays frequency of moderate physical effort at work, WorkModMinDay length of moderate physical effort at work, TravelDays frequency of getting around/travel, TravelMinDay time spent on getting around/travel, SportsVigDays frequency of intense physical effort while doing sports, SportsVigMinDay length of intense physical effort while doing sports, SportsModDays frequency of moderate physical effort while doing sports, SportsModMinDay length of moderate physical effort while doing sports, (min/week) metabolic equivalent value

* p < 0,050

The analysis showed statistically significant differences between the physical activity of urban and rural residents. It turned out that urban residents were characterized by significantly higher values ​​in terms of the frequency and length (duration) of active getting around/movement, compared to rural residents.

Statistical analysis, however, showed that the strength of the effect for the aforementioned variables was weak (rg = 0.20–0.21).

The results of the post hoc Power (1-β) statistical power assessment, however, showed a low probability (27%−29%) of demonstrating a true relationship.

In the case of the remaining analyzed variables, no statistically significant differences were noted, which means that regardless of the place of residence, the respondents showed similar declared parameters of physical activity.

The next verified correlation was the existence of relationship between the physical activity results from the GPAQ questionnaire and age, education and health status. The results of the analysis are presented in Table 8.

Table 8.

Correlation of variables from the GPAQ questionnaire with age, education and health status

	Age		Documented education		Health condition
	r	Power (1-β)	r	Power (1-β)	r	Power (1-β)
GPAQ
WorkVigDays	−0,15	0,63	−0,06	0,19	−0,11	0,42
WorkVigMinDay	−0,11	0,42	−0,03	0,11	−0,08	0,28
WorkModDays	0,05	0,16	−0,27***	0,98	0,12	0,47
WorkModMinDay	0,10	0,37	−0,29***	0,99	0,09	0,32
TravelDays	−0,04	0,13	−0,12	0,47	−0,02	0,08
TravelMinDay	0,03	0,11	−0,09	0,32	−0,01	0,06
SportsVigDays	−0,21**	0,88	0,27**	0,98	−0,23**	0,93
SportsVigMinDay	−0,23**	0,93	0,18*	0,77	−0,24**	0,94
SportsModDays	0,01	0,06	0,13	0,53	−0,15	0,63
SportsModMinDay	−0,04	0,13	0,15	0,63	−0,13	0,53
MET (min/week)	0,03	0,11	−0,19*	0,81	−0,11	0,42

WorkVigDays frequency of intense physical effort at work, WorkVigMinDay length of intense physical effort at work, WorkModDays frequency of moderate physical effort at work, WorkModMinDay length of moderate physical effort at work, TravelDays frequency of getting around/travel, TravelMinDay time spent on getting around/travel, SportsVigDays frequency of intense physical effort while doing sports, SportsVigMinDay length of intense physical effort while doing sports, SportsModDays frequency of moderate physical effort while doing sports, SportsModMinDay length of moderate physical effort while doing sports, MET (min/week) metabolic equivalent value

* p < 0,050; ** p < 0,010; *** p < 0,001

The results revealed statistically significant negative relationships between age and the duration and frequency of intense physical activity (SportsVigDays, SportsVigMinDay). A similar trend was observed between documented education and the frequency and duration of moderate physical activity at work (WorkModDays, WorkModMinDay) and the metabolic equivalent value (MET).

The analysis also showed statistically significant positive relationships between documented education and the frequency and duration of intense physical activity during sports (SportsVigDays, SportsVigMinDay).

It also turned out that with an increase in the subjective assessment of health, the frequency and duration of intense physical activity increased (SportsVigDays, SportsVigMinDay).

The post hoc Power (1-β) statistical power assessment confirmed the high probability of the above-mentioned correlations (1-β = 0.77–0.99).

In the case of the remaining pairs of variables, no statistically significant relationships were noted. It is also worth noting that the observed correlations mostly turned out to be weak and moderately strong.

Discussion

The aim of the study was to indicate the sociodemographic factors that determine the level of physical activity of people living in the Biała Podlaska district, in the eastern part of Poland. Physical activity was analyzed in relation to factors such as: gender, education, place of residence, subjective assessment of health, age.

The analyses of the research results indicated discrepancies between the results obtained from the objective research tool, which was the RM42 accelerometer, and the subjective tool, i.e. the GPAQ questionnaire.

Gender

The analyses of the data collected with the use of the RM42 accelerometer showed that women living in the Biała Podlaska district spend more time on moderate-intensity physical activity during the day and engage in physical activity lasting more than 10 min for a longer period of time compared to men.

The analyses of the data obtained from the GPAQ questionnaire also indicated greater physical activity of the examined women compared to men. However, statistically significant differences in this case concerned other variables. Female respondents predominated in the duration and frequency of intensive and moderate physical effort at work (WorkVigDays, WorkVigMinDay, WorkModDays, WorkModMinDay). They more often undertook intensive physical effort while practicing sports which they did for a longer time (SpostrVigDays, SportsVigMinDay). Women also achieved a more favorable result in the analyses of the MET-min/week indicator compared to men.

However, it should be noted that for both the data obtained from the RM42 accelerometer measurements and those collected with the GPAQ questionnaire, statistical analyses showed that the strength of the effect was weak or close to average. On the other hand, based on the results of the post hoc Power (1-β) statistical power assessment, it was found that the data obtained in the case of the accelerometer had a 47–55% probability of detecting an existing relationship, while in the case of the GPAQ, the results indicated a range of only 6–49% probability of detecting an existing relationship.

The results obtained, indicating that the female subjects were more physically active compared to the male subjects, contradict previous reports suggesting that adult men are more physically active than women [14, 44–46].

Similar results were shown by the studies of the physical activity of respondents from Portugal [47]. Also, the results presented above are worth comparing with Jędrzejczyk’s study published in 2021 [48]. In his work, the author presented a review of the published research results on the assessment of the level of physical activity of adults. The results of the cited studies show that men declared a higher level of physical activity compared to women in all studies included in the sample. However, the author clearly pointed out the need for detailed verification of the data he referred to, which shows the need for further research in this area.

Place of residence

Further differences in the results of analyses of data obtained from the RM42 accelerometer and the GPAQ questionnaire were shown by the analysis of the impact of place of residence on physical the undertaken activity.

The analysis of the data obtained from the RM42 accelerometer did not show statistically significant differences between city and village residents. Regardless of the place of residence, the subjects were characterized by similar parameters of physical activity. The analysis of the results obtained from the data collected with the use of the GPAQ questionnaire led to other conclusions. According to these subjective studies, people from the Biała Podlaska district, from the area of ​​eastern Poland, living in the city and in the countryside significantly differ in the level of the undertaken physical activity. Significantly higher values ​​were shown in city residents, compared to people living in the countryside. This was confirmed by the analysis of the frequency and duration of physical activity related to moving around. However, an assessment of the statistical power of the post hoc Power (1-β) indicated a low probability of demonstrating true relationships (27%−29% probability).

Similar conclusions were obtained from the studies by Biernat and Buchholtz [49]. The results of their studies conducted using the IPAQ questionnaire in 2014–2015 on a representative sample of Polish women and men aged 15–69 years clearly showed that the smaller the town in which the respondents lived, the lower their physical activity was. The conclusions from the studies by Biernat and Bucholtz were also confirmed by the CBOS study from 2018 [50], which showed that among the factors characterizing physically passive people, living in smaller towns was indicated. These conclusions are also confirmed by the study from the Multi-Sport program from 2019, which showed that physically active people are primarily those living in large cities [51].

Age

Another analyzed factor was age. The analysis of the data collected using the RM42 accelerometer showed statistically significant negative relationships between age, moderately intensive and vigorous exercise (MVPA, VPA) and the number of steps (Steps). It was observed that older people are, the less physically active they become (the older the person, the lower the AF).

Positive, statistically significant relationships were found between age and light physical activity (LPA) undertaken by the respondents. Thus, older people more often chose low-intensity activity as opposed to moderate and high-intensity activity. Similar trends were shown in the analysis of the data from the GPAQ. Again, in this case, it was shown that the older the examined person, the less often they undertook intensive physical activity, and more often they undertook lower-intensity effort. This relationship also concerned age in comparison with the length and frequency of intensive effort (SportsVigDays, SportsVigMinDay).

It should be noted that for both the analyses of the RM42 accelerometer data and those of the GPAQ, a high probability of the aforementioned correlations was confirmed based on the post hoc Power (1-β) statistical power assessment.

Similar conclusions were presented in the Report of the Public Health Committee of the Polish Academy of Sciences from 2021 [52], which showed that the results of many studies indicate that the lowest level of physical activity was visible in people over 50 years of age. Much the same conclusions were obtained in the previously cited studies by Biernat and Buchholtz from 2014 to 2015, the results of which showed that Polish women and men aged 15–69 become more physically passive with the increase in the age of their children. The least physically active were parents of children aged 0–6 years, while the highest PA was recorded in parents of children of high school or university age [49].

Education

The analysis of the relationship between education and undertaken physical activity examined using the RM42 accelerometer did not show any statistically significant relationships between these variables. Only weak, positive relationships were noted between education and physical activity lasting up to 5 min (PA0-5).

However, the analyses of the data collected using the GPAQ questionnaire led to completely different conclusions. In this case, statistically significant relationships, both negative and positive, between physical activity and education were shown. Thus, the higher the educational degree of the respondents, the less frequent and shorter their declared moderate-intensity physical activity at work (WorkModDays, WorkModMinDay). This also results in a lower value of the metabolic equivalent MET.

In the case of intensive efforts undertaken during sports (SportsVigDays, SportsVigMinDay), their frequency and duration increase with education.

It should be noted that the results of the above analyses were confirmed by a post hoc Power (1-β) statistical power assessment, indicating a high probability of correlation (77–99%).

The above results may indicate that people with higher educational degrees do not have the opportunity to participate in physical activity during their work (e.g. office work). However, they make up for this deficit in a conscious way in their free time, undertaking high-intensity efforts, thus meeting the recommendations for health-promoting physical activity.

Similar conclusions are suggested by the results of the CBOS study from 2018, according to which people with lower educational degrees are characterized by a lower level of physical activity [50]. The above trends are also confirmed by the analyses of studies conducted among residents of Elbląg, according to which people with lower educational degrees engage in physical activity less often than those with higher degrees. The authors of the study explained this relationship assuming that the respondents with higher degrees are more aware of the positive impact of physical activity on their health [53]. Also in the case of studies conducted among respondents from the Mielec district, the analyses showed that the level of physical activity of the respondents increased with the increase in the declared degree of education [54].

However, the opposite conclusions were presented in the work of Puciato et al., who studied physical activity of the residents of Katowice. The data analysis showed that the lower the degree of education, the greater the probability of undertaking health-promoting physical activity [55]. This example may be related to the specificity of employment related to this region, which is specific on a national scale.

Subjective health assessment

Another aspect that was analyzed was the relationship between the subjective assessment of health status and physical activity. According to the results obtained from RM42 accelerometers, it was shown that with the increase in the subjective assessment of health status, moderate and intense physical activity (MVPA, VPA), physical activity over 10 min (PA10+) and the number of steps also increased.

The analysis of the data from GPAQ showed similar relationships in the case of intensive efforts undertaken by the examined persons - with the increase in the subjective assessment of health status, the frequency and duration of intensive physical effort increased (SportsVigDays, SportsVigMinDay).

Confirmation of the above results is provided by post hoc Power (1-β) statistical power assessments, which confirmed the high probability of the above-mentioned correlations showing a very high probability (93–99%).

The presented results of the analyzed data constitute further evidence that the respondents declaring themselves as physically active people assess their subjective health status better.

Over the years, more and more researchers have proven that physical activity is one of the most important factors influencing not only health but also the quality and length of life - our life (56–58). Subsequent studies have shown the beneficial effect of physical activity, among others, in the prevention of such groups of diseases as cardiovascular diseases (59–63), cancer diseases (64–69), diabetes (70–73), diseases of the skeletal system (72, 74, 75), as well as depression (76–78).

Limitations

RM42 accelerometer

As mentioned in the methodological section, a minimum checklist of operating procedures for preparing the accelerometer for operation was used for the study, thus standardizing the parameters. Implementing such standardization was necessary due to the numerous scientific evidence indicating that small differences in the way accelerometer data are processed, such as different pulse recording periods (e.g. 15 s vs. 60 s) or adopting different cut-off points to determine the intensity, can radically change the estimates of physical activity levels [41, 42]. Harmonizing these parameters provides a basis for analyzing accelerometer data obtained using the RM42 accelerometer in other countries within the EUPASMOS project - fulfilling the requirement of data comparability on an EU scale without the objection that the observed differences between populations result from differences in measurement methodology.

GPAQ

The differences in the results obtained from the GPAQ questionnaire and the accelerometer indicate the imperfection of the first of the mentioned tools, which is pointed out by researchers [79]. The limitations concern primarily subjectivity - the fact that the source of the collected data is subjective information from respondents.

The imperfection of human memory, incorrect interpretation of the questions contained in the questionnaire, or the desire to declare undertaking physical activity more often and more intensely than in reality, may be sources that distort the research results.

This fact also illustrates how important it is to conduct this type of face-to-face interviews, thanks to which the person conducting the study can clarify issues that are doubtful from the point of view of the examined person and make sure that individual questions have been correctly understood. Another factor that may affect the differences in data collected using the subjective and objective method is the fact that in the case of the GPAQ, efforts lasting at least 10 min are taken into account, while the accelerometer monitors physical activity 24 h a day, 7 days a week.

Limited sample size

The sample size was determined in accordance with COSMIN recommendations (80), in accordance with recruitment potential and the sample size recommended for validation of psychometric tools (81).

The results presented in the manuscript are the result of research undertaken by the Polish team belonging to the EUPASMOS plus project. The research group consisted of 173 adults (71 women and 102 men) from four age groups (18–34, 35–49, 50–64 and 65+), from the area of ​​eastern Poland. In combination with the results of research from other European countries associated in this project, a database was created, collected according to a standardized methodology. The aim of the project was to design a system allowing for the collection of reliable data, on the basis of which it would be possible to implement adequate actions promoting physical activity. According to the team, this project should be developed and implemented in all EU Member States. This will enable the application of a standardised framework for monitoring physical activity, including the use of appropriate tools, comparison of data between countries and common conclusions on the promotion of an active lifestyle.

It should be emphasised that, with a view to preserving the methodological transparency and prospective value of the study, the authors took steps to mitigate the limitations mentioned. In order to minimise the subjectivity of the survey questionnaire used (GPAQ), it was validated in the context of the population studied and the research objectives. In addition, the survey was implemented according to strictly defined rules, under the control of trained researchers. Respondents were given detailed instructions on how to participate in the study (including how to complete the questionnaire). The interpretation of the processed accelerometer data could in future be supplemented with contextual data, such as the forms of physical activity undertaken by the respondent during the recording of movement, and the size of the study group in subsequent surveys could be increased (which increases statistical power and authorises generalisations of the study results).

With methodological transparency in mind, the authors, guided by the principle of transparency, have previously published a study report containing a detailed description of the methodology, including potential limitations and their impact on the results (82).

Future research directions

The study indicates the need for further research (larger study sample) on the relationship between physical activity and gender and place of residence. For these variables (in both the RM42 accelerometer and GPAQ surveys), no satisfactory probability of real relationships was shown on the basis of the results of the post hoc Power (1-β) statistical power assessment.

The research process conducted highlights the importance of methodological reliability, especially in the context of using subjective tools such as the GPAQ questionnaire. The discrepancies identified between the results obtained with the GPAQ and the RM42 accelerometer data indicate the limited reliability of the results of the former tool. This is also confirmed by the study by Jakicic et al. (83).

This conclusion may be, on the one hand, worrying and, at the same time, an important indicator for the analysis of both the research carried out so far and the selection of appropriate research tools in the future. This is because there is a high probability that inadequately carried out research, using subjective research tools, may be falsified and, consequently, the citation of such results would be meaningless.

Even if the research group carries out the research process in a professional and responsible manner, the problem may be the researcher who, consciously or less consciously, may falsify the data - deliberately entering overestimated values of physical activity (not wanting to appear worse in comparison to the research group), or not remembering exactly what type of activity they undertook, with what intensity it should be associated, and how long the particular effort was. It is also likely that the interpretation of the different intensities of physical activity may differ (despite the explanations in the questionnaire) - for a physically inactive person, MPV may be associated with VPA, whereas for a person who regularly undertakes physical activity (more fit, resistant to fatigue), VPA interpreted by a physically inactive person, may be equivalent to MPA.

Another problem may be physical activity only being reported when sustained for a minimum of 10 min. According to the recommendations, only such efforts are taken into account by the subject, and here again there may be a dilemma as to whether the previous week’s effort lasted for 9–10 min, and thus whether it should be mentioned in the questionnaire or not. This problem does not exist with objective methods, which report physical activity regardless of its duration.

Therefore, future research should rely more on objective measurement methods, such as accelerometers, which eliminate the risk of falsification of results due to unreliable memory or respondents’ tendency to portray themselves in a more favourable light.

The technological sophistication of accelerometers, which take into account parameters such as acceleration over a certain range, the appropriate sampling rate, the use of the MAD index or MET conversions, minimises the possibility of error. Therefore, their use in future studies seems crucial to obtain reliable and precise physical activity data for the populations studied.

The challenge for future objective research may be the choice of methodology, which requires more money to purchase equipment and technical expertise. However, it is worth remembering that the overarching goal of this research is to improve the health and quality of life of entire populations. This perspective may be a compelling argument for investing in more advanced measurement technologies.

Conclusions

The results obtained from the RM42 accelerometer and the GPAQ questionnaire are varied and dependent on the used tool. Still, they indicate the great importance of the tool used in the study of physical activity.

The analysis of the data collected with the use of the RM42 accelerometer allowed for the formulating the following conclusions regarding the sociodemographic factors influencing the level of physical activity of residents of the Biała Podlaska district in eastern Poland:

1. Women living in the Biała Podlaska district are more active than men - they spend more time during the day on moderately intensive physical activity, and they engage in physical activity lasting more than 10 min for a longer period of time compared to men.

2. Regardless of the place of residence, the respondents show similar parameters of physical activity.

3. Older people showed lower physical activity, but the frequency of low-intensity activity increased in their case, as opposed to moderate and high-intensity activity.

4. No statistically significant differences were found between the educational degree and the undertaken physical activity by the respondents.

5. The higher the subjective assessment of health, the higher the recorded physical activity.

The data collected through the GPAQ questionnaire allowed for the formulation of the following conclusions regarding the above-mentioned factors:

1. Women living in the Biała Podlaska district are more active than men (more favorable results in analyses of the length and frequency of intensive and moderate physical effort at work (WorkVigDays, WorkVigMinDays, WorkModDays, WorkModMinDays), the length and frequency of intensive physical effort during sports (SpostrVigDays, SportsVigMinDays) and in the analyses of the MET-min/week indicator compared to men).

2. People from the Biała Podlaska district living in cities engage in physical activity more often than the respondents living in the countryside.

3. Older people showed lower physical activity, while the frequency of low-intensity activity increased in their case as opposed to moderate and high-intensity activity. Older people also had lower values ​​in the length and frequency of intensive effort while doing sports. sports (SportsVigDays, SportsVigMinDay).

4. The respondents with higher educational degree declared less frequent and shorter moderate-intensity physical activity at work (WorkModDays, WorkModMinDay). With higher educational experience. Furhtermore, with the higher the educational degree, the more frequent and longer the undertaken intensive efforts when doing sports (SportsVigDays, SportsVigMinDay) increases.

5. With the increase in the subjective assessment of health, the frequency and duration of intensive physical effort (SportsVigDays, SportsVigMinDay) undertaken by people from the Biała Podlaska district increases.

The presented results of the analyses of the data collected with the use of the RM42 accelerometer and the GPAQ questionnaire allow for the formulation of generalized conclusions based on the results that were consistent in both objective and subjective studies:

• Women from the Biała Podlaska district are more active than men;

• As age increases, the surveyed individuals more often undertake physical activity of lower intensity, while giving up high-intensity efforts;

• The respondents declaring a higher subjective assessment of their health are more physically active.

Abbreviations

EUPASMOS

European Union Physical Activity and Sport Monitoring System

GPAQ

Global Physical Activity Questionnaire

PA

Physical Activity

LPA

Light physical activity

MVPA

Moderate and vigorous physical activity

VPA

Vigorous physical activity

Steps

Number of steps

WHO

World Health Organization

MET

Metabolic Equivalent of Task

WorkVigDays

Number of days with vigorous physical activity at work

WorkVigMinDay

Number of days with vigorous physical activity at work

WorkModDays

Number of days with Moderate physical activity at work

WorkModMinDay

range of time with Moderate physical activity at work

TravelDays

Number of days with travel to and from places

TravelMinDay

Time spend for travel to and from places

SportsVigDays

Number of days with vigorous–intensity sports, fitness or recreational (leisure) activities

SportsVigMinDay

Duration of time spent on vigorous–intensity sports, fitness or recreational (leisure) activities

SportsModDays

Number of days with moderate–intensity sports, fitness or recreational (leisure) activities

SportsModMinDay

Duration of time spent on moderate–intensity sports, fitness or recreational (leisure) activities

CBOS

Centrum Badania Opinii Społecznej (Public Opinion Research Center)

Authors’ contributions

MB and BB conceptualized the study and was involved in design, analysis, interpretation, and manuscript writing. All authors made a substantial contribution with a specific task MB, BB, AS and PR the extraction of data, EN, DT, JB-K and MS analysis and interpretation, MB and BB drafting of the manuscript, and JB-K, MS, and DT critical revision of important intellectual content. All the authors read and approved the final manuscript.

Funding

The research was part of the international EUPASMOS Plus program, financed by the European Commission under the Erasmus + Program grant no. 603328-EPP-1-2018-1-PT-SPO-SCP.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The above presented study was approved of by resolution no. 6/2018, the bioethics committee of the PSW in Biała Podlaska.

Each participant gave written consent to participate in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.