Multiple sclerosis is a chronic inflammatory autoimmune disease of the central nervous system (CNS) (1, 2). The immune system is activated against the self and damages myelin layers of axons of the CNS. This damage and its location may result in symptoms such as impaired vision and motor functions, sensory loss, and fatigue, impaired sleep, and cognitive decline. Often, the onset of disease is between 18 and 49 years, though early (before 18 years) and late onset (after 49 years) are also observed (3).

Several reviews (4–7) compared sleep patterns between persons with multiple sclerosis (PwMS) and healthy controls. These reviews reported that compared to healthy controls, PwMS had impaired sleep such as symptoms of insomnia, restless legs syndrome, and sleep-disordered breathing, along with increased daytime sleepiness and fatigue.

Restoring sleep is important for daytime functioning such as cognitive processing (8–11) or emotion regulation (12–18). This holds also true for PwMS. However, surprisingly, longitudinal data on sleep in PwMS are scarce, and to our knowledge, longitudinal data on objective sleep measurements in PmMS are mainly missing. These observations hold also true as regards the associations between sleep patterns and physical activity behaviors. For non-clinical samples (19–28), clinical samples (29, 30), and samples with PwMS (31–36), favorable sleep indices were associated with higher physical activity indices. Despite this, longitudinal studies on the development of sleep and physical activity indices over time are scarce. In a first study (37), we showed that about 2 years after disease onset, patterns of subjective sleep remained basically unchanged; that is to say, PwMS reporting restoring sleep at disease onset also reported unchanged restoring sleep 2 years later. As regards physical activity patterns, over time, both sedentary lifestyle and vigorous physical activity indices decreased, while moderate physical activity indices significantly increased. In a further study, objective sleep was assessed under naturalistic conditions in PwMS's familiar home setting (32): Both objective and subjective sleep dimensions were associated with higher moderate physical activity indices.

Data from interventional studies showed that physical activity improved depression, fatigue, and paresthesia (33, 38). Likewise, higher physical activity levels were associated with favorable objective and subjective sleep dimensions (32), higher emotion regulation and empathy (35), and improved sexual behavior (at least among female PwMS) (39).

With the present study, we aimed at expanding previous research on the following dimensions: First, we investigated, if and if so, to what extent objectively assessed sleep at baseline was associated with objectively assessed sleep on average 13.5 months later at follow-up and assessed under naturalistic conditions in PwMS's home setting. Second, we investigated to what extent physical activity indices remained stable or altered from baseline to follow-up. Third, we explored if and to what extent both objective sleep indices (at baseline and follow-up) and physical activity patterns (at baseline and follow-up) were related to each other. To do so, we combined findings from two previous studies and publications (32, 33). While objective sleep and subjective physical activity indices were main outcome dimensions, symptoms of depression and fatigue and EDSS scores were secondary outcome variables.

The following two hypotheses were formulated. First, based on previous findings (37), we assumed that also objective sleep patterns remained stable over a time lapse of on average 13.5 months. Second, following previous results (37, 40–42), we assumed that physical activity indices remained stable over a time lapse of on average 13.5 months.

The key findings of this small longitudinal study among persons with multiple sclerosis (PwMS) and performed under naturalistic conditions were that about 13.5 months after discharge from a rehabilitation center, descriptively, symptoms of depression remained fairly stable; descriptively, fatigue and subjective sleep scores worsened; and objectively, measured wake time after sleep onset statistically significantly improved. Further, descriptively, moderate and vigorous physical activity levels decreased. The present data add to the current literature in that we did not only replicate previous works but also expanded on them: Stability and changes as regards psychological, illness-related, physical activity-related and subjective and objective sleep-related dimensions did evolve in a differentiated fashion.

Two hypotheses were formulated and these are considered in turn.

With the first hypothesis, we assumed that also objective sleep patterns remained unchanged over a time lapse of on average 13.5 months, and the general pattern confirmed this assumption and thus mirrored previous findings (37). However, we expand upon previous data in that the follow-up objective sleep assessment occurred under naturalistic and familiar conditions. Further, wake time after sleep onset statistically significantly decreased. Also, this result demands particular attention: While the quality of the data does not allow a deeper understanding of the underlying psychophysiological mechanisms, it appears that at baseline, some participants had a particularly long wake time after sleep onset (see Table 2), while others did not. This huge range of wake time duration might be astonishing, given that baseline values were assessed under laboratory-like conditions at the rehabilitation center. In contrast, we also note that at follow-up, light sleep duration significantly increased. Combining both the decreased wake-time duration after sleep onset and the longer light sleep duration, one might speculate that the longer light sleep duration increased at the expense of a shorter wake time after sleep onset. Further, this kind of “shift” had an influence neither on sleep efficiency nor on the subjectively perceived sleep quality.

With the second hypothesis, we assumed that physical activity indices remained unchanged over a time lapse of on average 13.5 months, though the answer is not that evident and we could not fully confirm previous findings (37). While from a pure statistical point of view (significant p-values; effect sizes), no change was observed; descriptively, both moderate and vigorous physical activity levels decreased. The explanation of this pattern of results demands more attention. First, as shown in Table 2, standard deviations were large at both time points, suggesting thus that some participants were particularly physically active, while others were not. Given this, the zero result might be the sum of particularly high and particularly low physical activity levels. Second, in the present study, baseline values equaled to the end assessment of a 3- to 4-week lasting stay at the rehabilitation center. As described elsewhere (33), the stay at the rehabilitation center consisted of regular physiotherapy and physical activity interventions. Given this, it appears plausible that both moderate and vigorous physical activity levels under guided and monitored conditions were higher, compared to an in-home and self-paced context.

The novelty of the results should be balanced against the following limitations: First, the sample size was small, though we also focused on effect size calculations, which by definition are not sensitive to sample sizes. Second, given the voluntary character of the study, only participants able and willing to comply with the study conditions took part in the study; it follows that the entire pattern of results might be biased. In this view, third, it is also conceivable that further factors such as medication intake, social structures, employment, or stable daytime structures could have biased the present pattern of results. Likewise, fourth, current change of MS status could have unfavorably interfered with dimensions of the follow-up assessment; while a current attack was not explicitly an exclusion criterium, change in MS status and diseased progress occur rather slowly (3), and participation at the study under the condition of an attack appears rather unlikely. Fifth, the female:male ratio was 11:5; given the small sample size, statistical calculations to take gender as further independent factor were not taken into consideration. It follows that a systematic gender bias could not be ruled out.

The pattern of results of the present small study conducted under naturalistic conditions suggests that during a time lapse of about 13.5 months, objective sleep patterns, physical activity levels, and psychological well-being might change at an individual and fine-grained level. We hold that the present results are of clinical and practical importance for treatment counseling: PwMS can be reassured that their sleep quality does not deteriorate, once they have left a rehabilitation center. Further, they should be encouraged to keeping their physical activity levels as stable and as high as possible.

The raw data supporting the conclusions of this article will be made available by the authors to experts in the field upon request.

The studies involving human participants were reviewed and approved by Ethikkommission Nordwest und Zentralschweiz (EKNZ, Basel, Switzerland): 2016-1347. The patients/participants provided their written informed consent to participate in this study.

DSB, RG, JK, JB, DC, OR, DN, MG, and SB: conceptualization. DSB, RM, JK, OR, DN, and MG: data curation. DSB, RG, JK, JB, RM, DC, OR, DN, MG, and SB: investigation and methodology. DSB, RG, JB, OR, DC, MG, and SB: project administration. DSB, RG, OR, MG, and SB: supervision. DSB, RG, JK, RM, JB, OR, DN, DC, MG, and SB: writing the draft and final version. All authors read and agreed to the publication in the present form. All authors have read and agreed to the published version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.