Effects of 8-week endurance and resistance training programmes on cardiovascular stress responses, life stress and coping

ABSTRACT

This study tested the effect of 8-week endurance and resistance training programmes on cardiovascular stress responses, life stress, and coping. Fifty-two untrained but healthy female students were randomised to an 8-week endurance training, an 8-week resistance training, or a wait list control group. Before and after the training intervention, we assessed the groups’ cardiorespiratory fitness (VO2max test), self-reported life stress, coping strategies and cardiovascular reactivity to and recovery from a standardised laboratory stressor. Both endurance and resistance training programmes caused physiological adaptation in terms of increased VO2max after the intervention. For stress and coping parameters, participants in the training groups improved cardiovascular recovery from stress and reported having less stress in their everyday life after the intervention than participants in the control group, while the two training groups did not differ from each other. We did not find any significant differences in heart rate reactivity and coping strategies between the study groups. These results partly support that exercise training has stress-reducing benefits regardless of the type of exercise. Both endurance and resistance exercise activities may be effectively used to improve stress regulation competence while having less impact on changing specific coping strategies.

KEYWORDS:

• Physical activity
• exercise
• sport
• stress
• coping

Regular physical activity and exercise have been found to prevent stress-related diseases (Harber et al., 2017; Li & Siegrist, 2012) and strengthen mental health (Hearing et al., 2016; Lederman et al., 2017; Schulz et al., 2012). These benefits are due to several factors, including a preventive effect whenever exercise contributes to lowered stress perceptions (Gerber et al., 2014) and a coping-promoting effect when exercise helps to endorse adaptive coping behaviours (Cairney et al., 2014). Exercise training has also been identified as a beneficial strategy to buffer stress-related physiological reactions (Gerber & Pühse, 2009; Klaperski et al., 2012), with physically active individuals typically showing lower cortisol increase (Klaperski et al., 2013, 2014; Rimmele et al., 2009, 2007), lower cardiovascular reactivity (Crews & Landers, 1987; Forcier et al., 2006; Klaperski et al., 2013, 2014; Rimmele et al., 2009, 2007), and more rapid cardiovascular recovery to laboratory stressors when compared to their less active counterparts (Jackson & Dishman, 2006).

Notably, the above advantages have been mostly reported on endurance exercise training (e.g., running, biking; Klaperski et al., 2012), with only a few studies looking at effects of resistance exercise training (weight lifting; Gröpel et al., 2018; Spalding et al., 2004). Of the few studies that directly compared the two, Gröpel and colleagues observed similar effects on cardiovascular stress reactivity, whereas Spalding and colleagues found stronger effects in the endurance-trained group. Thus, evidence of whether or not resistance training has the same psychophysiological benefits as endurance training is still unclear. This study tested the effects of endurance and resistance training programmes on cardiovascular stress responses, life stress perceptions and coping.

On the physiological level, researchers have proposed that regular exercise, both endurance and resistance, leads to physiological adaptations which may contribute to reduced physiological reactions to stressors in general (Hamer et al., 2006; Huang et al., 2013). Endurance training enhances aerobic capacity and induces adaptations that increase ventricular filling and decrease myocardial work, thereby improving cardiac performance (i.e., enhanced stroke volume; Spalding et al., 2004). This enables submaximal workloads to be negotiated with greater efficiency (e.g., at a lower heart rate and blood pressure; McArdle et al., 1996), which may generalise from ergogenic to psychogenic challenges (Claytor, 1991; Sinyor et al., 1983). Similar to endurance training, resistance training produces adaptations in the cardiovascular system that lower blood pressure (Spalding et al., 2004) and cause a more rapid return of heart rate to baseline levels following physical exercise (Darr et al., 1988), even though aerobic capacity does not increase as much.

On the psychological level, the role of physical exercise as a factor that reduces self-reported stress experience is well documented (Gerber et al., 2014; Sigfusdottir et al., 2011), but there is a dearth of research on whether exercise training also changes coping. Coping refers to the ‘conscious volitional efforts to regulate emotion, cognition, behaviour, physiology, and the environment in response to stressful events or circumstances’ (Compas et al., 2001, p. 89), generally categorised as engagement and disengagement coping (Carver & Connor-Smith, 2010). Engagement coping is aimed at dealing with the stressor or related emotions and includes strategies such as active coping, support seeking and cognitive restructuring, while disengagement coping is aimed at escaping the stressor or related emotions and comprises strategies such as behavioural disengagement, substance use and denial. Coping strategies are considered adaptive if they improve functioning in a given situation and maladaptive if they result in maintained or increased levels of strain and distress (Zeidner & Saklofske, 1996).

Researchers have proposed that physical activity and exercise can promote specific coping strategies. For example, Kim and McKenzie (2014) found that physical activity promoted problem-focused coping and Al Sudani and Budzynska (2015) reported that physical activity correlated with task-oriented coping and social diversion. Other researchers found that physical activity itself has been described as a coping strategy (Garber, 2017; Long, 1993), yet individuals who endorsed physical activity as a means of coping also tended to endorse other coping behaviours (Cairney et al., 2014). The implication is that the use of some coping strategies may change in response to enhanced physical activity and exercise. However, the above studies were limited by self-reported levels of physical activity, with no objective measurement or manipulation of actual physical activity behaviour. Also, the studies asked participants about the level of physical activity in general, without differentiating between endurance and resistance exercise activities.

The aim of this study was twofold. First, we investigated the effect of 8-week endurance and resistance training programmes on cardiovascular stress reactivity and recovery by including three female groups differing in the type of training (endurance-trained participants, resistance-trained participants and untrained controls). Second, we examined the effect of the two training programmes on life stress perceptions and coping strategies. The endurance training consisted of low- and high-impact aerobic dance, which is an effective method to increase physical endurance (Wilmore & Knuttgen, 2003), and the resistance training comprised a progressive weight-lifting programme, which is an effective intervention to improve physical strength (Liu & Latham, 2009). We chose the 8-week period because it had been found that 6 to 8 weeks of either training were enough to both increase aerobic capacity and strength and improve well-being and cardiovascular adaptation to psychological stressor (Delextrat et al., 2016; Spalding et al., 2004). We hypothesised that both endurance- and resistance-trained participants would show lower cardiovascular stress reactivity to and more rapid recovery from psychological stressor, and less life stress, as a consequence of the training. Regarding coping strategies, we used an exploratory rather than confirmatory approach, as the evidence on physical activity and coping is unclear. On the one hand, physical activity itself has been described as a coping strategy. On the other hand, physical activity has also promoted other coping behaviours and thus at least some coping strategies could be used more (or less) frequently in response to the exercise training programmes.

An additional aim of this study was to test the above hypotheses with a female sample. The majority of studies on physical activity and stress were conducted with men, while female samples are currently underrepresented (Klaperski et al., 2013). However, findings from male samples cannot be automatically generalised to female population, as gender strongly affects physiological and psychological stress response (Jackson & Dishman, 2006). For example, Kelly et al. (2008) found that women reported greater psychological stress than men on a standardised stress test, although both groups showed similar physiological stress responses (cortisol and heart rate reactivity). Women are on average less physically active than men (Colley et al., 2011), but more likely than men to use exercise as a way of coping with stress (Cairney et al., 2014). More research on physical activity, stress and coping is thus needed with female samples. In sum, the current study contributes to the literature by comparing the effects of endurance and resistance training programmes on cardiovascular stress responses, life stress perceptions and coping in a sample of female participants.

Methods

Participants

Participants were 65 female students (M_age = 21.02 years, SD = 1.65) recruited in cooperation with the university leisure sports centre for women. Inclusion criteria were: essentially healthy (not suffering from any known somatic or psychiatric disease), not having performed any regular exercise training during the last year and studying full time. General exclusion criteria were medication intake, substance abuse and smoking. The study was approved by the first author’s university review board. Before entering the study, all participants gave written informed consent and were informed about their right to discontinue participation at any time.

Procedure

Using a computer-generated list of random numbers (Microsoft Excel; Microsoft, Redmond, WA), participants were randomised to one of the three study groups: an 8-week endurance training group (EG), an 8-week resistance training group (RG) and a wait list control group (WCG; receiving no training intervention). One week prior to (pre-test) and after the training intervention (post-test), all participants completed individual testing sessions that comprised, in this order, psychological questionnaires on stress and coping strategies, a stress test with assessment of cardiovascular stress responses, assessment of body mass index (BMI) and a 2-km UKK walk test for assessing cardiorespiratory fitness. The pre-test and post-test assessments were identical.

Exercise training intervention

Both exercise training groups had three 60-min training sessions per week (24 sessions in total), all of which were supervised in a group format and led by an experienced fitness coach. We used two fitness coaches, one led all endurance training sessions and the other one led all resistance training sessions. Each training session started with a 7-min warm-up, followed with a standardised endurance (EG) or resistance (RG) exercise protocol (45 min) and ended with an 8-min stretching phase to cool down. The exercise protocols for both training groups are in the Appendix. Participants in the endurance training group performed a combination of low- and high-impact aerobic dance at a tempo of 130–148 beats per minute (BPM) in the first 4 weeks and progressed to high-impact aerobic dance at a tempo of 148–160 BPM in the second 4 weeks. Participants in the resistance training group performed a full‐body programme with eight exercises per session (bench press, wide-grip pulldown behind the neck, lying leg curl, dumbbell biceps curl, leg extension, dumbbell side lateral raise, triceps bench dips and sit-ups). The training programme comprised three to five sets of eight to 12 repetitions, with a 2‐min recovery phase between the sets, and was progressive, as the load started by 40% of the one-repetition maximum (1RM) and was continually increased by 10% in the third, fifth and seventh training week. The 1RM test was conducted before the training intervention and recalculated after 4 weeks of the intervention.

Psychological questionnaires

Two well-validated, widely used questionnaires were administered. First, we used the one-dimensional Perceived Stress Scale (PSS; Cohen & Williamson, 1988) to measure the frequency with which participants found their lives unpredictable, uncontrollable and overloading during the last month, and themselves not able to cope with it (e.g., “How often have you found that you could not cope with all the things that you had to do?”). The PSS consists of 10 items that are answered on a 5-point Likert scale from 1 (never) to 5 (very often). Prior research has supported the reliability and validity of the PSS for student and employee samples across different cultures (cf. Lee, 2012). In the present study, Cronbach’s alphas were good, with .82 and .86 in the pre-test and the post-test, respectively. Second, we administered the Coping Orientation for Problem Experiences inventory (COPE; Carver et al., 1989) to assess participants’ coping strategies that they had been using during the last month. The COPE consists of 60 items that assess 15 different coping strategies: positive reinterpretation and growth, mental disengagement, focus on and venting of emotions, use of instrumental social support, active coping, denial, religious coping, humour, behavioural disengagement, restraint, use of emotional social support, substance use, acceptance, suppression of competing activities and planning. Participants responded to the COPE using a 4-point Likert scale from 1 (I did not do this at all) to 4 (I did this a lot). Carver et al. (1989) reported adequate validity, test–retest reliability and internal consistency of the COPE. In the present study, internal consistencies for the majority of the COPE subscales were in line with psychometric standards (αs > .70), except for mental disengagement (α = .60), denial (α = .68) and suppression of competing activities (α = .62).

Stress test and autonomic measures

Participants were exposed to a standardised stress protocol from the iSense relaxation training software (Comesa, Austria), which consisted of four phases: baseline, preparatory, stress and relaxation. In the baseline phase (135 s), participants were asked to calm down (“Please sit back and relax”). Next, in the preparatory phase (50 s), they were informed about an upcoming stress stimulus (“The stress stimulus is coming soon”). The stress stimulus was then presented for 15 seconds and included both visual and acoustic cues (flickering pictures of spiders and snakes with an unpleasant noise). Finally, in the post-stress, relaxation phase (70 s), participants were again asked to calm down. All instructions were presented by the software on the computer screen and via headphones. Participants’ heart rate was monitored continuously during the stress protocol using a wireless chest heart rate transmitter and a wrist monitor recorder (Suunto T6 c, Suunto, Finland). For analyses, heart rates within each phase were averaged. We calculated the “area under the individual response curve with respect to the increase” (AUC_I) using the trapezoid formula (Pruessner et al., 2003) as an index of heart rate reactivity. The AUC_I allows a sensitive measurement of physiological changes over time and has been often used as a reliable indicator of reactivity to stress (Hellhammer et al., 2007). For heart rate recovery, we calculated the time (in seconds) at which the post-stress values reached the baseline values; shorter time indicated a more rapid recovery from stress.

Cardiorespiratory fitness test

Participants performed the 2-km UKK walk test to indirectly measure aerobic capacity (Oja & Tuxworth, 1995). This test provides an index of aerobic capacity and an estimate of maximal oxygen uptake (VO2max) and has been validated for adults who are free from illnesses that disable walking and from cardiovascular illnesses (Laukkanen et al., 2000; Oja et al., 1991). The walks took place on a 400-m outdoor-track. The instruction for the walks was “Walk the distance as fast as you can, but do not risk your health”. Heart rate was monitored throughout the walks by a wireless chest heart rate transmitter and a wrist monitor recorder (Suunto T6 c, Suunto, Finland). The mean rate during the last 60 s of the walk was considered the walking heart rate. Participants were asked to refrain from fat eating, alcohol drinking and intensive physical activity for at least 1 day prior to the test day.

Data analysis

We tested for systematic differences in age, BMI, VO2max, heart rate reactivity, heart rate recovery, perceived life stress and coping strategies among the three study groups at the pre-test and the post-test by means of separate analyses of variance (ANOVAs). In case of significant results, we performed post hoc tests with Bonferroni correction to explore differences between any two pairs of groups. Furthermore, we analysed differences among the groups over time by means of separate repeated measure ANOVAs with Group as the between-subject factor (3 groups: endurance training vs. resistance training vs. control group) and Time as the within-subject factor (repeated measures: 4 for heart rate during the stress protocol, and 2 for VO2max, heart rate reactivity, heart rate recovery, perceived life stress and coping strategies). Repeated-measures results were verified with Greenhouse-Geisser corrections where the Mauchly test of sphericity determined heterogeneity of covariance. In case of significant results, we used paired t-tests in each group to assess changes from the pre-test to the post-test. All analyses were performed using SPSS 24.0 (IBM Corp.; Armonk, NY). Data are presented as mean ± SEM. The level of significance was set at p ≤ .05 (two-tailed). Partial eta squared (η_p²) was used as an indicator of effect size for ANOVA calculations, and Cohen’s d_z was used for paired t-tests. Partial eta squared of 0.01 indicated a small, 0.059 a medium and 0.138 a large effect size, respectively. For Cohen’s d_z, values of 0.20 indicated a small, 0.50 a medium and 0.80 a large effect size, respectively. An a priori sample-size calculation with G*Power (Faul et al., 2009) for three groups and two repeated measurements, based on middle effect size (f = 0.25), power = .80 and α = .05, resulted in a minimal sample size of 42 participants.

Results

Six persons from the training groups did not complete all training session (EG: n = 4; RG: n = 2), and seven persons from the control group did not show up for the post-test and were therefore excluded from the analyses. Thus, the final sample consisted of 52 participants. The characteristics of the sample are presented in Table 1. The groups did not significantly differ in age and BMI (see Table 1 for exact p values), but there was a significant Group × Time interaction on BMI, F(2, 49) = 12.18, p < .001, η_p² = 0.33. BMI decreased significantly from the pre-test to the post-test in both the endurance, t(17) = 4.10, p = .001, d_z = 0.96, and the resistance training groups, t(20) = 5.01, p < .001, d_z = 1.09, whereas no significant change occurred in the control group. Furthermore, the groups did not differ in pre-test VO2max, but significant differences emerged in post-test VO2max, F(2, 49) = 9.48, p < .001, η_p² = 0.28. Post hoc tests revealed that both the endurance and the resistance training groups had significantly higher VO2max than the control group in the post-test (p < .001 and p = .008, respectively), whereas the difference between the endurance and the resistance groups was not significant. This was further qualified by a significant Group × Time interaction, F(2, 49) = 24.85, p < .001, η_p² = 0.50. Analyses of VO2max within the groups revealed that VO2max improved significantly from the pre-test to the post-test in both the endurance, t(17) = 11.76, p < .001, d_z = 2.95, and the resistance training groups, t(20) = 4.34, p < .001, d_z = 0.96, whereas no significant change occurred in the control group.

Table 1. Characteristics of the study groups.

	Endurance (n = 18)	Resistance (n = 21)	Control (n = 13)	p^a	ηp^2a	p^b	ηp^2b
Age (years)	21.22 ± 0.39	21.24 ± 0.42	20.38 ± 0.27	.283	0.05
BMI (kg/m^2)
Pre-Test	21.59 ± 0.55	21.64 ± 0.49	20.49 ± 0.52	.285	0.05	<.001	0.33
Post-Test	21.19 ± 0.55***	21.24 ± 0.46***	20.68 ± 0.55	.738	0.01
VO2max (ml/kg/min)
Pre-Test	35.30 ± 0.77	35.91 ± 0.72	34.74 ± 1.28	.659	0.02	<.001	0.50
Post-Test	39.25 ± 0.73***	37.82 ± 0.60***	34.16 ± 1.15	<.001	0.28

N = 52. Data are presented as mean ± SEM; significant differences within groups (paired t-tests) are highlighted by bold type and partly indicated as follows: ***p <.001.

Abbreviations: BMI = Body Mass Index; VO2max = maximal oxygen uptake.

^aExact significances (p values) and effect sizes (η_p² values) for the ANOVA at pre-test and post-test.

^bExact significances (p values) and effect sizes (η_p² values) for the Group × Time repeated measures ANOVA.

Heart rate responses to stress test

Mean heart rate levels (bpm) are presented in Figure 1. The stress protocol induced a significant increase in heart rate in all three groups, both in the pre-test, F(2.27, 111.03) = 130.51, p < .001, η_p² = 0.73, and the post-test F(2.09, 102.41) = 149.74, p < .001, η_p² = 0.75. The groups did not differ in heart rate levels during the protocol in either test phase. Heart rate reactivity, as indicated by AUC_I, did not change significantly in the three groups from the pre-test to the post-test (Table 2). However, a significant interaction effect emerged for heart rate recovery, F(2, 49) = 3.47, p = .04, η_p² = 0.12. While there was no difference among the three groups in the pre-test, the groups showed different recovery times in the post-test, F(2, 49) = 6.23, p = .004, η_p² = 0.20. Post hoc tests revealed that both the endurance and the resistance training groups had significantly shorter recovery time than the control group in the post-test (p = .05 and p = .003, respectively), whereas the difference between the endurance and the resistance groups was not significant. Further analyses of heart rate recovery within the groups revealed that recovery times decreased significantly from the pre-test to the post-test in both the endurance, t(17) = 2.80, p = .01, d_z = 0.66, and the resistance training groups, t(20) = 2.43, p = .03, d_z = 0.53, whereas no change occurred in the control group (Table 2).

Table 2. Mean levels of heart rate reactivity, heart rate recovery and perceived life stress across the study groups.

	Endurance (n = 18)	Resistance (n = 21)	Untrained (n = 13)	p^a	ηp^2a	p^b	ηp^2b
HR Reactivity (AUCI)
Pre-Test	957.85 ± 123.83	803.45 ± 143.91	577.31 ± 89.11	.171	0.07	.222	0.06
Post-Test	777.36 ± 113.54	776.73 ± 92.07	819.71 ± 156.95	.962	0.00
HR Recovery (seconds)
Pre-Test	23.44 ± 5.23	14.71 ± 3.68	16.62 ± 5.09	.354	0.04	.039	0.12
Post-Test	10.00 ± 2.97*	5.24 ± 1.59*	22.92 ± 6.35	.004	0.20
Perceived Stress (PSS)
Pre-Test	2.69 ± 0.53	2.76 ± 0.10	2.82 ± 0.13	.778	0.01	.005	0.20
Post-Test	2.35 ± 0.08*	2.28 ± 0.09***	2.87 ± 0.11	<.001	0.29

N = 52. Data are presented as mean ± SEM; significant differences within groups (paired t-tests) are highlighted by bold type and partly indicated as follows: *p <.05; ***p <.001.

Abbreviations: HR = heart rate; AUC_I = area under the curve with respect to the increase; PSS = perceived stress scale (range: 1 = never to 5 = very often).

^aExact significances (p values) and effect sizes (η_p² values) for the ANOVA at pre-test and post-test.

^bExact significances (p values) and effect sizes (η_p² values) for the Group × Time repeated measures ANOVA.

Figure 1. Mean heart rates during the stress protocol across the study groups. Error bars are standard errors of the mean (SEM).

Life stress perception and coping strategies

The groups did not significantly differ in the pre-test levels of perceived life stress, but significant differences emerged in the post-test, F(2, 49) = 10.23, p < .001, η_p² = 0.29 (Table 2). Post hoc tests revealed that both the endurance and the resistance training groups reported significantly less life stress than the control group in the post-test (p = .002 and p < .001, respectively), whereas the difference between the endurance and the resistance groups was not significant. This was further qualified by a significant Group × Time interaction, F(2, 49) = 5.93, p = .01, η_p² = 0.20. Analyses of perceived life stress within the groups revealed that life stress declined significantly from the pre-test to the post-test in both the endurance, t(17) = 2.81, p = .01, d_z = 0.65, and the resistance training groups, t(20) = 5.64, p < .001, d_z = 1.26, whereas no significant change occurred in the control group. Regarding coping strategies, the groups did not differ in any of the tested coping strategies in either test phase (Table 3). Moreover, there was no significant Group × Time interaction, indicating that coping strategies did not change among the groups from before to after the training. Analyses of coping strategies within the groups revealed two significant changes, both in the resistance training group; participants in this group reported using positive reinterpretation, t(20) = 2.14, p = .045, d_z = 0.47, and active coping, t(20) = 2.21, p = .04, d_z = 0.49, more frequently in the post-test than in the pre-test.

Table 3. Mean scores of coping strategies across the study groups.

	Endurance (n = 18)	Resistance (n = 21)	Untrained (n = 13)	p^a	ηp^2a	p^b	ηp^2b
Positive Reinterpretation
Pre-Test	2.95 ± 0.12	2.81 ± 0.15	2.98 ± 0.09	.587	0.02	.119	0.08
Post-Test	3.01 ± 0.15	2.98 ± 0.15*	2.90 ± 0.13	.887	0.01
Mental Disengagement
Pre-Test	2.44 ± 0.10	2.36 ± 0.12	2.52 ± 0.13	.639	0.02	.417	0.04
Post-Test	2.39 ± 0.13	2.35 ± 0.13	2.33 ± 0.15	.949	0.00
Venting of Emotion
Pre-Test	2.46 ± 0.13	2.42 ± 0.16	2.52 ± 0.20	.912	0.00	.850	0.01
Post-Test	2.39 ± 0.14	2.25 ± 0.16	2.42 ± 0.18	.707	0.01
Instrumental Social Support
Pre-Test	2.53 ± 0.11	2.64 ± 0.15	2.56 ± 0.24	.859	0.01	.933	0.00
Post-Test	2.47 ± 0.12	2.58 ± 0.15	2.44 ± 0.19	.781	0.01
Active Coping
Pre-Test	2.81 ± 0.14	3.00 ± 0.11	2.88 ± 0.20	.599	0.02	.446	0.03
Post-Test	3.08 ± 0.13	3.19 ± 0.08*	2.92 ± 0.20	.381	0.04
Denial
Pre-Test	1.75 ± 0.13	1.88 ± 0.12	1.58 ± 0.13	.290	0.05	.271	0.05
Post-Test	1.72 ± 0.12	1.75 ± 0.10	1.75 ± 0.14	.981	0.00
Religious Coping
Pre-Test	1.72 ± 0.19	2.26 ± 0.21	2.25 ± 0.30	.159	0.07	.429	0.03
Post-Test	1.81 ± 0.20	2.26 ± 0.22	2.15 ± 0.29	.329	0.04
Humour
Pre-Test	2.07 ± 0.13	2.21 ± 0.14	1.88 ± 0.15	.286	0.05	.540	0.03
Post-Test	2.03 ± 0.15	2.31 ± 0.17	1.75 ± 0.18	.083	0.10
Behavioural Disengagement
Pre-Test	1.69 ± 0.13	1.65 ± 0.11	1.63 ± 0.21	.959	0.00	.188	0.07
Post-Test	1.49 ± 0.12	1.61 ± 0.10	1.69 ± 0.20	.584	0.02
Restraint
Pre-Test	2.43 ± 0.12	2.68 ± 0.12	2.54 ± 0.17	.392	0.04	.351	0.04
Post-Test	2.47 ± 0.11	2.56 ± 0.09	2.63 ± 0.16	.660	0.02
Emotional Social Support
Pre-Test	2.68 ± 0.12	2.57 ± 0.18	2.83 ± 0.24	.620	0.02	.987	0.00
Post-Test	2.68 ± 0.14	2.58 ± 0.19	2.81 ± 0.25	.720	0.01
Substance Use
Pre-Test	1.21 ± 0.09	1.21 ± 0.11	1.11 ± 0.08	.775	0.01	.117	0.08
Post-Test	1.36 ± 0.19	1.14 ± 0.10	1.10 ± 0.08	.381	0.04
Acceptance
Pre-Test	2.42 ± 0.16	2.48 ± 0.10	2.27 ± 0.18	.609	0.02	.246	0.06
Post-Test	2.49 ± 0.15	2.35 ± 0.11	2.37 ± 0.15	.720	0.01
Suppression of Activities
Pre-Test	2.47 ± 0.13	2.61 ± 0.07	2.37 ± 0.15	.348	0.04	.929	0.00
Post-Test	2.57 ± 0.09	2.65 ± 0.11	2.40 ± 0.14	.339	0.04
Planning
Pre-Test	2.97 ± 0.12	3.07 ± 0.13	3.12 ± 0.22	.801	0.01	.263	0.05
Post-Test	3.11 ± 0.12	3.17 ± 0.10	2.96 ± 0.20	.561	0.02

N = 52. Data are presented as mean ± SEM; significant differences within groups (paired t-tests) are highlighted by bold type and partly indicated as follows: *p <.05.

^aExact significances (p values) and effect sizes (η_p² values) for the ANOVA at pre-test and post-test.

^bExact significances (p values) and effect sizes (η_p² values) for the Group × Time repeated measures ANOVA.

Discussion

This study tested the effect of 8-week endurance and resistance training programmes on cardiovascular stress responses, life stress perceptions and coping. Both training programmes included standardized exercise protocols that led to a significant increase of VO2max in both training groups (with no significant difference between the groups). The results showed that both endurance and resistance training programmes improved heart rate recovery. Furthermore, participants who completed the training reported experiencing less stress in their life thereafter. We did not find any significant differences in heart rate reactivity and coping strategies from before to after the training programmes.

Recent researchers have demonstrated that regular endurance training improves cardiovascular responsiveness to acute stress (Jackson & Dishman, 2006; Klaperski et al., 2013, 2014; Rimmele et al., 2009, 2007). Our results support these findings with regard to heart rate recovery, but not with regard to reactivity. We did not observe a difference between trained and untrained participants in post-test heart rate AUC_I. However, in the endurance training group, heart rate responses were imposed on a significantly lower baseline level post-test (71.8 bpm) as compared to pre-test (83.4 bpm). This would imply that absolute heart levels during stress were also lowered as a result of the endurance training (Sinyor et al., 1986). The reason why this did not translate in lower reactivity may lay in the stress test used. Even though our stress protocol induced a significant increase in heart rate in all study groups, the absolute increase (+15 bpm) was lower as compared to the more demanding Trier Social Stress Test applied by recent researchers (over +20 bpm; Klaperski et al., 2013, 2014; Rimmele et al., 2007). Hence, with our less “stressful” protocol, we might be on the lower side of detecting a reactivity effect.

In this study, we directly compared the effects of endurance and resistance trainings. This was an important contribution given that only few researchers to date have simultaneously tested the effects of endurance and resistance trainings on cardiovascular stress responses (Gröpel et al., 2018; Spalding et al., 2004). We found that both endurance and resistance training groups showed more rapid recovery times after stressor cessation than untrained controls. Notably, the two training groups did not differ from each other, indicating that both training types can result in the same health-related benefits. Similar results emerged when using subjective measures of stress. The training groups reported having less stress in their everyday life than the control group after (but not before) the training period. This was in line with prior evidence showing that active individuals were less likely to report stress compared with those who had low exercise levels (Aldana et al., 1996; Gerber et al., 2014; Sigfusdottir et al., 2011). Potential explanations for why physical exercise lowers stress perceptions are that being enrolled in physical activity training can prevent feelings of loneliness (by broadening social networks) and reduce susceptibility when confronted with external demands (i.e., higher stress tolerance; Gerber & Pühse, 2009). As we implemented a training group format and exercise protocols with progressively increasing physical demands in the current study, both explanatory mechanisms are likely to account for our results.

The present study extended prior research by including a number of specific coping strategies. We were interested in whether the increase of physical activity behaviour also changes specific coping behaviour. Overall, coping strategies did not change as a consequence of exercise training; the training and control groups reported using the same coping behaviour over time. Although subsequent analyses revealed that participants in the resistance training group used positive reinterpretation and active coping more frequently after the training than before, these differences were only found when testing within-group effects with no between-group comparison. Hence, we cannot conclude that exercise training changes coping strategies. A potential explanation is that physical exercise is a coping mechanism per se, without having much impact on other coping strategies (Garber, 2017). Thus, rather than expecting mediation effects through engagement or disengagement coping behaviour, future researchers should address the use of exercise as a coping strategy itself.

Another explanation is that even though physical exercise might promote coping strategies (Cairney et al., 2014; Kim & McKenzie, 2014; Al Sudani & Budzynska, 2015), the 8-week period was too short to see the difference. Coping strategies were previously found to have trait-like qualities (Powers et al., 2002; Zimmer-Gembeck & Skinner, 2010), but were less stable over time than, for example, personality traits (Carver & Connor-Smith, 2010; Murberg et al., 2002). Nielsen and Knardahl (2014) found a high overall stability of coping behaviour over a 2-year time period, but also fluctuations within some specific strategies. Consequently, it is likely that some coping strategies are malleable and that physical activity may promote them, but it presumably takes much longer than 8 weeks of regular exercise to see the effect. In addition, the physical activity intervention could have a stronger effect in younger samples, as coping repertoires increase with age, with the greatest development emerging from early childhood to adolescence (Zimmer-Gembeck & Skinner, 2010). Future research should therefore test the effect of physical activity on coping strategies in younger samples and over a longer time period.

Limitations

As mentioned earlier, the stress protocol used in this study had not as strong impact on heart rate increase as the protocols used in prior research on cardiovascular stress responses (+15 bpm vs. +20 bpm and more). The findings thus allow only limited conclusions regarding heart rate reactivity, as it cannot be ruled out that reactivity effects were masked by stress induction limitations. Furthermore, the generalisability of our findings is limited to young healthy women having a rather high educational level. Replications with other samples would provide more insight into the effect of gender, age, education and clinical factors on exercise-related stress adaptations.

Conclusion

We tested whether people would improve cardiovascular stress responsiveness, report less stress in life and change coping strategies after completing an 8-week endurance or resistance training programme. We found that participants in both endurance and resistance training groups experienced less stress in their everyday life after the training and also exhibited more rapid heart rate recovery from laboratory-induced stressor than participants in the control group. We did not find any significant differences in heart rate reactivity and coping strategies. These findings partly support that exercise training may have stress-reducing effects regardless of the type of exercise. Both endurance exercise activities such as aerobic or jogging and resistance exercise activities such as lifting weights can be effectively used to improve stress regulation competence and, in turn, obtain health-related benefits.

Acknowledgments

We would like to thank Gabriela Štefániková and Veronika Adamská for their help with organization of the trainings. External grant funding was not used to fund this work. Internal funds by the Comenius University in Bratislava were used to cover the costs of the endurance and resistance exercise trainings.

Disclosure statement

No potential conflict of interest was reported by the authors.

Appendix

Endurance and resistance training protocols used in the study

Endurance training

Table

Week	Tempo (BPM)	Aerobic Dance Intensity
1	130–140	High-impact/Low-impact (basic steps)
2	130–140	High-impact/Low-impact (basic steps)
3	max 148	High-impact/Low-impact (basic and turn steps)
4	max 148	High-impact/Low-impact (basic and turn steps)
5	148–160	High-impact (basic steps)
6	148–160	High-impact (basic and turn steps)
7	148–160	High-impact (advanced steps)
8	148–160	High-impact (advanced and turn steps)

Participants performed a combination of low- and high-impact aerobic dance in the first 4 weeks and progressed to high-impact aerobic dance with more advanced steps in the second 4 weeks.

Resistance training

Table

Week	Intensity (% of 1RM)	Number of Sets	Repetitions
1	40	3	8–10
2	40	3	10–12
3	50	4	8–10
4	50	4	10–12
5	60	4	8–10
6	60	4	10–12
7	70	5	8–10
8	70	5	10–12

Participants performed a full‐body programme with eight exercises per session: bench press, wide-grip pulldown behind the neck, lying leg curl, dumbbell biceps curl, leg extension, dumbbell side lateral raise, triceps bench dips and sit-ups.