To our knowledge, this is one of the first studies measuring physical activity during and after hospitalization with an accelerometer in a heterogeneous group of older patients hospitalized for acute illness. Our results show that during hospitalization, the patients spent most of their time engaged in sedentary behavior and took fewer steps and were less physically active than after discharge. The included patients moved less than 900 steps per day and were therefore at risk of hospitalization-associated functional decline [17]. During hospitalization, the diurnal profiles for steps and uptime showed no notable variation in activity after 9 AM whereas we found a characteristic diurnal profile for steps and uptime after discharge and four weeks after discharge with most physical activity occurring between 9-11 AM, 14-16 PM and 17-18 PM. Hence, differences were seen in the number of steps taken, the time spent in uptime, and peaks in activity during the day. However, no remarkable difference was seen between discharge and four weeks after discharge. During hospitalization, the patients spent 93% of the daytime engaged in sedentary behavior and spent 1.7 hours upright This is consistent with a review of studies in acute-care settings in which the daily time spent lying or sitting accounted for 89–99%, and the duration of uptime was 1–2 hours per day [23]. A second important finding was that during hospitalization only one minor peak in uptime occurred at 9 AM. This suggests that the patients only get out of the bed in the morning, maybe for the morning toileting or breakfast, and spent most time engaged in sedentary behavior for the rest of the day. However, right after discharge and four weeks later, several peaks in activity occur at similar timepoints throughout the day. The doubling in the duration of uptime, and the consequent decrease in sedentary time after discharge, in conjunction with the lack of a second peak in activity during hospitalization suggests that sedentary behavior is a result of a culture of bed rest at the departments and that the patients could potentially be more physical active.
A study by Mai et al. [50] analyzed the diurnal physical activity profile in non-hospitalized, community-dwelling individuals older than 70 years. Consistent with our study, they identified two peaks, one at 10-11 AM and a second at 3-4 PM; sex, age, morbidity, and season had no moderating effects. Limited mobility was the only factor that significantly moderated the profile, reducing the number of peaks to one, which could supports the lack of a second peak in our group during hospitalization, since the median DEMMI score in the included patients was 57 on admission, which reflect limited mobility. In contrast to our study, a recent study by Tasheva et al. [51], which assessed the distribution of physical activity levels in older patients hospitalized for acute medical illness, found three peaks of physical activity during the day: between 8-10 AM, at 12 PM, and at 6 PM. Tasheva et al. proposed that older inpatients are primarily active during meals, as reflected in the three peak times. Although the patients in our study did receive three meals per day, this was not reflected in their peak times and may indicate that meals were consumed close to or in bed. This is well in line with recent studies by our research group, Pedersen et al. 2020 [52] and Stefánsdóttir et al. [53], that investigated facilitators and barriers for mobility during hospitalization. They found that the barriers for mobility in this patient group were the provision of excessive service and care by the department, a culture of bed rest, and lack of encouragement by health care professionals to motivate the patients to increase activity. Also, Stefansdottir et al. reported that the staff brought food, beverages, and clothes to patients, including those who were able to get out of bed and walk.
We consider all of these barriers explanations for the lack of a second peak during hospitalization in our study. After discharge and four weeks after discharge, the patients started physical activity one hour earlier and had higher levels of physical activity in the morning and afternoon hours. Moreover, the peaks persisted longer and declined more gradually. These findings are consistent with those of Zisberg et al. [28], who showed that the timing of getting dressed in the morning moved an hour and a half during hospitalization, and most basic activities were reduced in frequency and duration. Our findings indicate that older patients could potentially be more physically active during hospitalization and emphasize the need for interventions that encourage more physical activity during hospitalization. It could be in the form of simple routine activities, such as patients eating their breakfast out of bed in a common room and changing clothes by themselves. Another starting point for the effort to increase patients’ physical activity during hospitalization could be to encourage physical activity around 10-11 AM, 3 PM, and 5-6 PM, which would be more consistent with their habitual. Our findings are also consistent with a recent study by Kolk et al. [54], which showed a doubling in steps only one day after discharge compared with one day prior to discharge. Presumably, the relatively high level of activity after discharge in our study, with a median of 2207 steps per day after discharge and 2622 four weeks after discharge, indicates that patients’ return to their basic routines occurs earlier than expected, our results indicate a recovery of activity the first week after discharge.
It is known that older patients are vulnerable to disturbances in their routines and that this can lead to delirium during hospitalization [55]. Thus, suggested efforts should be made to re-establish routines among those at risk of loss of functional decline and delirium. Since our results showed a wide range in steps, uptime and sedentary behavior, a personalized intervention strategy would be a reasonable means of optimizing physical activity during hospitalization. Recently, guidelines on physical activity for admitted older patients have also highlighted the importance of integrating physical activity throughout daily care, with a focus on functionality and activities of daily living; and bearing in mind that it is important for patients and staff to share the responsibility of promoting physical activity and minimizing sedentary behavior [16]. An alternative method of providing clinical care to a segment of this group of older patients has recently been suggested in a systematic review [56], which found that hospital-at-home (HaH) treatment may be a clinically effective approach and suggested that this treatment method may result in less functional decline in patients than the traditional ward-based treatment method. However, further research is needed, and the implementation of this alternative method of treatment (HaH) would necessitate significant changes to the current practice as well as time, it can takes several years making structural changes in the healthcare systems. In the short term, a relevant indicator is needed to identify patients at a high risk of inactivity during their time in the hospital. In the long term, systematic changes in the hospital environment and care setting are needed where the responsibility to encourage physical activity should be a shared responsibility and delegated to all health professionals.
Strengths and limitations:
This study’s major strength was the longitudinal measurement of physical activity during hospitalization, at discharge, and at four weeks after discharge in a heterogeneous cohort of older adults hospitalized for acute illness. This study has some limitations. First, we assessed number of steps using the activPAL3™ activity monitor, which has an uncertainty in detecting walking at speeds less than 0.67 m/s [40,41]. At baseline, the patients in the current study had a median walking speed of 0.67 (IQR: 0.48-0.87), and therefore likely that walking was underestimated in some patients. An underestimation would affect the absolute level of steps, but not affect the distribution of activity throughout the day as a result of we are looking at uptime (walking and standing), therefore the diurnal physical activity profiles may be regarded as a true reflection of older adults' patterns of daily physical activity.