Time spent lying, sitting, and upright during hospitalization after stroke: a prospective observation study

One of the core elements of care for dedicated comprehensive stroke units is acute medical treatment combined with early rehabilitation [1–3]. Early rehabilitation with mobilization out of bed during hospitalization has shown to be associated with better functional outcomes for patients after stroke [4–8]. This may reduce the loss of muscle mass, increase muscle strength [6, 9], avoid complications [3, 4, 10], exploit the plasticity of the brain [11], improve neurological functioning [12], and improve gait function [13]. Although time spent out of bed early after stroke varies significantly between hospitals [14–16], the amount of physical activity is generally low during hospital stays [17, 18].

Optimal timing and intensity of early mobilization is, however, not clear, as previous studies have shown an association between mobilization within the first 24 h and a negative trend towards increased dependency [12, 19] and increased mortality [20]. On the other hand, short and frequent mobilizations have been associated with improved outcome [21].

Most studies of physical activity during hospitalization have been single-day observational studies during daytime hours. Recently, objective measurements using body-worn single sensor activity monitors have been applied, enabling continuous monitoring of physical activity over time [22, 23]. So far, only a few studies have used activity monitors to distinguish between different activities in the acute phase for stroke patients [24]. In one of these studies, Strømmen et al. (2014) found significantly increased activity in the lower extremities for hospitalized stroke patients from the first day until discharge [25]. The increased activity was significantly associated with lower severity of stroke, whereas decreased activity was found with increasing age [25]. However, the study did not discriminate between times spent in different positions, which is crucial when measuring physical activity in the acute phase after stroke. Applying two single sensors from the ActivPAL sensor system, one on the chest and one on the thigh, makes it possible to quantify the amount of time spent in different positions. This method has previously been proved valid for stroke patients [26].

The main aim of this study was to describe the amount of time spent in lying, sitting and upright (standing or walking) positions early after stroke and how these activity levels changed during the hospital stay, both regarding total time per day and duration for each mobilization. Another aim was to examine which factors were associated with total time and bouts of sitting and upright activity. Our primary hypothesis was that patients would gradually increase time spent in an upright position, and thereby increase the duration of each upright bout.

The flow of patients is shown in Fig. 1. Of the 105 patients who met the inclusion criteria, 47 were excluded, mainly because of early discharge.

Figure 2 illustrates the combinations of activities for each day, showing an increase in time spent out of bed during the hospital stay. Time spent upright remained low throughout the observation period.

Although not significant, LMM analysis indicated a daily increased duration of 4.32 min for each sitting bout (p = 0.051). Adjusted for time, this increase in time was associated with both a higher NIHSS score (p = 0.018) and increasing age (p = 0.028).

In this study, we found that stroke patients increased their time spent sitting and upright during the initial hospital stay, with a corresponding decrease in time spent in a lying position. The duration of sitting bouts also increased over time, while the duration of upright bouts remained constant. Furthermore, overall increased time spent upright was associated with increasing independency (mRS) and improved physical function (SPPB), while prolonged sitting bouts were associated with higher age and more severe stroke.

Despite a statistically significant increase in time spent upright during the hospital stay, we may ask whether an estimated 3.75 min’s more upright activity per day was a clinically significant change. This means that time spent upright will increase by almost 40 min during a 10-day hospital stay. A previous study has shown that every 5-min increase in time spent in bed was associated with a 4% deterioration on the mRS score three months later [6]. Hence, the daily increase in time spent upright could potentially have a great impact on functional recovery over time.

On the other hand, we found no association between stroke severity and time spent out of bed. This finding might be explained by the fact that activity measures during hospitalization also mirror the mobilization routines in the wards [33], showing that patients are mobilized out of bed as soon as they are medically approved, independent of the patients’ ability to mobilize themselves.

The duration of each sitting bout also increased by more than 4 min per day during the hospital stay. More surprisingly, the increased duration of sitting bouts was associated with higher age and more severe stroke, while no associations were found for bouts in upright activity. These findings might be explained by the ability of less affected and younger patients to change position more often by themselves. This is interesting because it adds further knowledge to the ongoing debate regarding the intensity of early mobilization. Also, clinicians should be aware that patients with the most severe strokes who are not able to move independently are at increased risk of sedentary behavior, which might be harmful, thereby increasing the risk of poor outcome [6].

Our results show that stroke patients on average spend a significant proportion of each 24 h in a lying position (Fig. 2). Considering the 24-h monitoring, patients would normally spend at least 30% of the time sleeping. Still, patients spent almost 30% of the remaining time in bed. This corresponds with the high amount of time spent lying previously reported early after stroke [14, 17, 25]. One factor that could influence the low overall activity amongst our participants is the exclusion of those expected to be discharged within three days after inclusion, as these patients are likely to be more active. However, as we aimed to study change in physical activity over time, patients with short monitoring time had to be excluded.

The major strength of the present study was the prospective study design with continuous monitoring of activity, from two activity monitors, over 24 h for several days during the hospital stay. This allowed us to discriminate between lying, sitting and upright positions, which is of major importance in the acute phase after stroke. The use of such a protocol has been validated in the stroke population earlier, showing a high accuracy compared to video observation [26]. In contrast to observation of activity by, for example, behavioral mapping during daytime hours, usually for 9–10 h [7, 14, 17], continuous monitoring will account for out-of-bed time during 24 h. It is also a strength that all included patients received evidence-based treatment in a comprehensive stroke unit combining acute medical treatment and early rehabilitation. Therefore, we know that our patients were mobilized according to recommendations in the national and international guidelines [27, 34], increasing the validity of the study.

A limitation of this study was the wide window for inclusion, ranging from one to seven days post stroke. This criterion was chosen for practical reasons, as recruitment was delayed during weekends and because the diagnosis was delayed in some patients (if the MR images were inconclusive or if patients were not admitted to hospital immediately after the first stroke symptoms). Nevertheless, apart from a significant difference in NIHSS between early and late admission to hospital, indicating that patients with the most severe symptoms are admitted to hospital earlier, we found no significant differences in the demographic data at baseline. However, the variation in time from onset of stroke to inclusion is accounted for in the LMM analysis. Our prolonged recruitment period (2013–2016), with three different time periods may also have influenced our results. The same routines were, however, applied for all stroke patients in the acute stroke treatment in the time periods. Therefore, all included patients, regardless of when they were recruited, should have received treatment focused on early rehabilitation. Another limitation is the lack of possibility to draw a causal relationship between the clinical parameters and the activity categories. Our secondary analysis indicate that the clinical parameters were influencing the activity levels. However, it is also possible to reverse this relationship, and to argue that the activity levels might influence the clinical parameters. Hence, it is important to have this possible bilateral relationship in mind when interpreting the results. Finally, the accelerometers used in this project do not discriminate between active and passive mobilization, and based on our inclusion criteria, mobilization may have been conducted with or without support from another person. This could influence both length and frequency of mobilization, as an active mobilization is naturally more tiring for the patient. However, in a comprehensive stroke unit, hospital staff are trained to activate stroke patients as soon as medically accepted, to encourage active rehabilitation. This means that even though some patients receive more support than others, all patients should be challenged according to their functional level.

This study showed that time spent in sitting and upright positions increased throughout the hospital stay in patients admitted to hospital following a stroke. We found that these changes were associated with improved physical function and a higher degree of independence. There was also an increased duration of bouts spent sitting during hospitalization, with longer bouts associated with increasing age and more severe strokes. Future research should focus on defining the optimal dose of activity in the acute phase in order to improve function in the long term after stroke.