To our knowledge, this study is the first to investigate the psychological responses to acute exercise in patients with stress-induced exhaustion disorder (ED). Our primary aim was to compare these responses between ED patients and healthy controls, facilitating an understanding of typical exercise reactions in the ED group in contrast to non-clinical individuals. Our findings show that a short bout of aerobic exercise can alleviate symptoms experienced by ED patients, consistent with previous research on the effects of exercise on symptoms in patients with depression and anxiety disorders [13,14,15]. As expected, ED patients, compared to healthy controls, reported higher general levels of fatigue, anxiety, and stress, and lower levels of energy, reflecting the common symptoms of ED. The exercise effects in ED patients differed from those in healthy controls in several ways. ED patients perceived the exercise as more strenuous but experienced greater reductions in fatigue and stress immediately after exercise, with effects also sustained after a 30-min rest. Additionally, ED patients showed a more pronounced energy increase post-exercise. Furthermore, we aimed to explore whether psychological responses varied between low and moderate exercise intensities. We found that ED patients exhibited similar psychological responses to both intensities.

During the exercise sessions, the ED patients experienced the activity as more strenuous than the healthy controls. This was expected given that mentally fatigued individuals rate exercise as more strenuous [38] and that marked physical weakness is a symptom of ED (ICD-10-SE). Hence, the higher self-reported exertion in ED patients may be attributed to central fatigue [39], an issue reported in studies on similar disorders [5, 40], rather than solely the physiological processes in exercise seen in overall healthy individuals. Central nervous system-related fatigue affects exercise performance together with peripheral fatigue also in healthy individuals but can be more pronounced in disorders such as ME/CFS and depression [5]. Previous studies in ME/CFS patients have shown higher perceived exertion ratings in relation to heart rate during exercise despite no defect in neuromuscular function, which has been attributed in part to central fatigue [40].

Interestingly, the elevation of exertion did not correspond with that of psychological discomfort. Although ED patients reported greater overall psychological discomfort, they did not experience a higher increase in discomfort compared to healthy controls. This is consistent with previous research in women with ED, showing higher overall SUDS ratings but no difference in psychological distress response compared to a control group when exposed to a mental stressor [41]. Hence, ED patients seem to respond similarly to mental and physical stressors in terms of psychological discomfort. As our study measured both perceived exertion and psychological discomfort during the exercise, the complexity of how exercise is experienced was highlighted. The notion that more strenuous physical activity does not necessarily correspond to increased psychological discomfort has support from previous research, which shows that affective displeasure does not correlate linearly with exercise intensity but instead plateaus at higher intensities above the ventilatory threshold [12], typically correlating to RPE ratings around 13 [42].

Despite the exercise being perceived as more strenuous by the ED group compared to non-clinical individuals, it elicited more beneficial psychological effects post-exercise. Previous research performed primarily on healthy individuals has found that acute exercise consistently increases energy [18]. However, results on fatigue are more heterogeneous, showing that fatigue reduction occurs when baseline fatigue is normal or high and when energy increase post-exercise is substantial [18]. The reductions in fatigue and increased energy in ED patients are hence consistent with these results. A study on college students with elevated fatigue [43] found similar reductions in fatigue after exercise at 50% of V̇O2 max, resembling our study’s moderate-intensity exercise, but not at 75% of V̇O2 max. The fatigue reductions in the ED group contrast with the elevated fatigue in ME/CFS patients post-exercise [16], a condition similar to ED characterized by persistent fatigue. This highlights differences between the two conditions. While ED patients experienced lowered fatigue post-exercise, controls did not, likely due to a floor effect, as their pre-exercise fatigue levels were already low (see Table S8, Additional file 1). In previous acute exercise studies showing fatigue-reducing effects, pre-exercise POMS fatigue scores averaged 6.42 [18], allowing for noticeable reductions.

Conversely, the control group showed increased fatigue 30 min after moderate-intensity exercise. Typically, post-exercise fatigue increases occur when the baseline fatigue level is low [18], as in this study’s control group, but are otherwise also associated with high-intensity and longer-duration exercise [18]. Sustained reductions in fatigue have been observed in both healthy individuals [44] and those with substance abuse [45]. However, a difference from these studies is that participants generally experienced an immediate fatigue-reducing effect, which was not observed in the control group of this study.

Both groups experienced elevated energy post-exercise, in line with previous research [18]. For ED patients, this effect was significant at both intensities, while for healthy controls, it was significant only at low intensity, though approaching significance at moderate intensity (p = 0.089). The lack of a pronounced energy increase from moderate-intensity exercise in the control group might contribute to the explanation of the delayed fatigue observed 30 min post-exercise, since energy has been shown to moderate changes in fatigue [18].

Besides a significantly more pronounced energy drop from post-exercise to 30 min after exercise at moderate intensity for both groups, and elevated fatigue in the control group 30 min after moderate-intensity exercise, the short exercise bouts did not produce significantly different psychological responses based on intensity in either group. Previous research on intensity effects on affective states shows mixed results [12]. Many studies report no intensity effects, while others have found varied or negative effects on higher exercise intensities, particularly for energy and fatigue. However, a meta-analysis suggested greater anxiety-reducing effects at higher intensities due to the exposure, habituation, and reappraisal of bodily sensations similar to anxiety reactions (e.g., elevated heart rate, elevated breathing rate) [34]. Consequently, low-intensity exercise usually does not produce beneficial changes in fatigue, energy, and anxiety, as measured in this study. A more mechanistic study design could shed light on the mechanisms underlying these patterns in ED patients.

Previous research generally shows small reductions in state anxiety [34], and an anxiety-reducing effect of the exercise was also observed in this study. The time x group interaction only approached significance (p = 0.077), making it uncertain how the two groups differed in their responses, but importantly, the ED group did not show increased anxiety after exercise (see Table S8, Additional file 1). Interestingly, anxiety levels decreased further during the 30-min rest period following exercise, an effect not observed in previous studies, such as a meta-analysis including the same anxiety measure used here [34], or a recent study on depression patients that included a post-exercise rest period [46]. A partial explanation could be that the rest period was particularly valued by the ED patients due to their symptoms of exhaustion. Although not a part of the data collection protocol, anecdotal data indicated that many ED patients viewed the mandatory rest as a welcome treat. Additionally, the self-selected component of the quiet rest, where participants could choose to rest, sleep, read provided magazines, or bring their own reading material, may have contributed to the rest being viewed as an overall positive experience. The stress response curve was similar to that of anxiety, indicating an overlap between the two constructs. A recent meta-analysis on stress reactivity to a mental stressor found no reliable changes in self-reported stress after acute exercise [47], hence again highlighting the need to study mechanisms underlying responses.

Significant group differences were observed in exercise-related measures (see Table 1), with the ED group reporting lower physical activity levels over the past year and exhibiting lower fitness levels compared to the control group. Baseline physical activity and fitness have been shown to mediate exercise responses, although the data is primarily based on non-clinical populations. While research on the impact of physical activity levels on acute exercise effects regarding fatigue and energy is limited [18], one study found that regular exercisers showed significant changes in energy and fatigue, while non-exercisers did not [48]. A meta-analysis found that sedentary individuals experienced the largest reductions in anxiety following exercise [34]. Additionally, both fit and unfit participants reported positive affective responses to low-intensity exercise compared to high-intensity exercise [49]. Overall, the available data is heterogeneous, suggesting that the relationship between physical activity, fitness, and exercise responses is complex and potentially confounded by unreported factors. While we cannot confirm if the ED patients in this study fully represent the broader ED population in these aspects, a comparison of the cardiorespiratory fitness level in our ED group (34.1; see Table 1) with that of ED patients in a randomized trial comparing exercise training, cognitive training, and a control condition shows comparable baseline fitness levels (34.66–35.96) [50].

We believe that the measures 6 and 24 h after exercise were important, as clinical observations suggest that concerns about delayed post-exercise fatigue and reduced energy often discourage physical activity and exercise in individuals with ED. Recent qualitative research [4] has supported this. Given that the psychological effects of acute exercise are transient, with positive activated affect like energy typically lasting up to 30 min [51], we did not expect improvements later in the day or the next morning. However, we considered potential delayed negative psychological effects, similar to those reported by ME/CFS patients [16], especially since ED patients have reported such expectations [4]. Contrary to these assumptions, neither group showed increased fatigue, anxiety, stress, or decreased energy later in the day or the following morning compared to baseline, which is encouraging findings. It is important to note that this only concerns the short-term effects of single exercise sessions, and no conclusions can be drawn about the effects of repeated exercise. One factor potentially influencing the delayed ratings is the quiet rest post-exercise. It is possible that a rest period after exercise affected the ED group more since lack of recovery, i.e., “psychophysiological unwinding after effort expenditure” [52], is central in the link between exposure to stressors and negative effects of stress. Hence, the delayed measurements show how exercise followed by 30 min of quiet rest affected the participants, not solely exercise.

Fatigue, being a central clinical characteristic of ED [3, 4], can be defined as a “persistent sense of physical, emotional, and/or cognitive tiredness or exhaustion” [31] and implies a lack of ability to initialize and maintain mental and physical tasks that require effort and self-motivation. This has implications for the real-world transferability of the high compliance and low dropout rates in the exercise sessions, as participants had support in initiating the exercise, and researchers were present throughout the sessions. Social support is an important factor for ED patients with less exercise experience in establishing an exercise routine [53]. Previous research shows that mental fatigue affects the choice of sedentary activities over physical activities, with the only exception being low intensities [54]. So, although our results have positive immediate effects on symptoms, the effort required to initiate and maintain short exercise sessions might be a considerable obstacle. This must be addressed in intervention design.

There is limited data on how generalizable acute exercise studies are to long-term exercise. Thus, our study does not answer how ED patients would react psychologically to regular exercise, warranting further studies. Acute exercise effects can however increase the chance of continuous exercise adherence, with previous research showing that lower levels of fatigue post-exercise correlate with more frequent exercise months later [55]. The beneficial effects of single exercise sessions observed here might thus potentially increase compliance with exercise in ED patients, by enhancing external motivation and limiting negative expectations of exercise consequences.

One factor not manipulated in the study was the duration of exercise, which is another aspect influencing the psychological responses to exercise. A meta-analysis on acute exercise effects on fatigue and energy [31] showed that while exercise consistently increases feelings of energy, fatigue increases with longer duration, specifically beyond 20 min. However, a study comparing various combinations of durations of acute exercise (5–60 min) and subsequent rest periods showed no difference in fatigue ratings [44]. The exercise bouts in that study produced mean RPE ratings between 9.7 and 11.9, hence comparable to the low-intensity exercise in our study (see Table S7, Additional file 1). But considering that fatigue is central to ED, it might be that longer durations of exercise beyond 20 min affect this group more than many other populations. This highlights that the combination of intensity and duration affects the psychological responses and must be considered jointly.

Estimation of aerobic capacity was performed using an indirect method. As such, it contains errors larger than those related to direct or maximal test methods. However, maximal tests were deemed unsuitable for the ED group and any comparison to the controls could have been attributed to differences in ability to reach maximal effort. Nonetheless, the errors related to the submaximal, indirect test may have resulted in over- or underestimations, potentially causing participants in both groups to work at a higher or lower work rate than the intended 40% and 55%. We regard this as a random error that increases the variance and potentially results in an underestimation of real changes or differences, but not creating systematic group or time differences. Concerning blood pressure, we have no record on which arm it was measured, so there may have been variations. Since these measures were used solely for exclusion purposes and no participant was approaching the exclusion limit, we consider this a minor limitation. Another limitation regarding internal validity is that some participants may have slept during the resting period after the exercise bouts, which was not systematically registered.

We did not include a measure of positive affect during exercise, which could have given a more comprehensive understanding of the psychological responses [12]. Future studies should address this. However, one strength of our study is the inclusion of some psychological measures during exercise, which is often overlooked in acute exercise research. Since psychological responses are dynamic and not linear, this, combined with delayed measurements, provides a more complete picture of the psychological responses to exercise in ED patients.

The sample consisted primarily of women, with two men in each group. This limits the generalizability of the results to men. Gender differences in psychological responses to acute exercise are understudied. Still, one study [56] on young adults found such differences, with women showing greater improvements than men in fatigue and energy after 30 min of vigorous aerobic exercise. Generally, those with worse baseline values for variables such as anxiety [34] and energy level [51] experience significant beneficial effects of the exercise. Since women reported higher baseline problems with these variables, the observed differences might be due to the larger potential for benefits rather than to gender differences.

Experimental studies in a laboratory environment always have the issue of generalizability or external validity. Factors that could have contributed to negative experiences during the exercise (e.g. level of exertion and psychological discomfort) include that the exercise took place in a windowless room without distractions such as music or scenery. Exercising outdoors vs. indoors has been proposed to elicit lower RPE, and listening to music has also been shown to have beneficial effects on RPE [57]. The presence of a researcher during exercise might have affected the participants. Non-systematic observations indicated that some ED patients felt nervous about exercising, especially at moderate intensity, but felt safer in the controlled situation with a researcher present.

Another limitation of this study is the issue of multiple testing. While Bonferroni corrections were applied to control for Type I errors, the large number of tests still introduces some risk of false findings. However, given that the results generally pointed in the same direction, the likelihood of Type I errors is likely reduced. On the other hand, using a conservative correction may have limited the ability to detect smaller effects, particularly in the control group, which could be a source of Type II error. Despite this, the clear direction of the results supports the overall validity of the findings.

However, a strength of the study is the comparison of the ED patients to a healthy control group, which allowed reliable conclusions about the psychological responses to exercise specific to ED patients. This makes adjusting the exercise recommendations easier. Additionally, counterbalancing exercise session order based on intensity enhances the reliability of conclusions about intensity effects, avoiding potential novelty effects on psychological discomfort, stress, and anxiety.

Lastly, we had concerns that only ED patients with low symptom levels and limited impairment in important areas of functioning would participate in the study. This was, however, not substantiated, as demonstrated in the baseline exhaustion ratings and sickness absence (see Table 1). The large interest shown by individuals with an ED diagnosis in participating in the study indicates a large interest among them in exercise as an intervention.

This study aimed to shed some initial light on how acute exercise affects patients with ED, but several questions remain unanswered. Future studies could include exercise with different durations and higher exercise intensities, as well as explore the influence of different durations of rest post-exercise. The focus was on psychological responses to acute exercise, without exploring underlying processes that govern these responses in ED patients. Studies with a mechanistic focus can contribute to understanding the processes that underlie the psychological changes during and after exercise in ED patients. Research with a focus on susceptibility profiles is also needed, enabling more tailored exercise programs for the different needs of different ED patients. Potential moderators could include symptom severity, exercise level, fitness level, and BMI. While this study focused on aerobic exercise, other types of training, such as strength training, could also have beneficial psychological effects, and warrant further research.