If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Patterns of Objectively Measured Sedentary Behavior and Physical Activity and Their Association with Changes in Physical and Functional Performance in Geriatric Rehabilitation Inpatients
Department of Human Movement Sciences, @AgeAmsterdam, Faculty of Behavioural and Movement Sciences, Vrije Universiteit Amsterdam, Amsterdam Movement Sciences, the Netherlands
Department of Human Movement Sciences, @AgeAmsterdam, Faculty of Behavioural and Movement Sciences, Vrije Universiteit Amsterdam, Amsterdam Movement Sciences, the Netherlands
Department of Internal Medicine, Amstelland Hospital, Amstelveen, the NetherlandsDepartment of Internal Medicine, Amsterdam UMC, VU University Medical Center, Amsterdam, the Netherlands
Amsterdam UMC, Vrije Universiteit Amsterdam, Department of Rehabilitation Medicine, Amsterdam Neuroscience and Amsterdam Movement Sciences, Amsterdam, the Netherlands
Department of Human Movement Sciences, @AgeAmsterdam, Faculty of Behavioural and Movement Sciences, Vrije Universiteit Amsterdam, Amsterdam Movement Sciences, the Netherlands
Address correspondence to Andrea B. Maier, MD, PhD, @Age, Faculty of Behavioral and Movement Sciences, Vrije Universiteit Amsterdam, Amsterdam Movement Sciences, van der Boechorststraat 7, 1081 BT, Amsterdam, The Netherlands.
Department of Human Movement Sciences, @AgeAmsterdam, Faculty of Behavioural and Movement Sciences, Vrije Universiteit Amsterdam, Amsterdam Movement Sciences, the NetherlandsDepartment of Medicine and Aged Care, @AgeMelbourne, The Royal Melbourne Hospital, The University of Melbourne, Parkville, Victoria, AustraliaHealthy Longevity Program, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, SingaporeCenter for Healthy Longevity, @AgeSingapore, National University Health System, Singapore, Singapore
To examine whether The Ending PyJama (PJ) Paralysis campaign, focused on increasing in-hospital physical activity, affects objectively measured sedentary behavior and physical activity patterns and if these are associated with changes in physical and functional performance in geriatric rehabilitation inpatients.
Design
Quasi-experimental study.
Setting and Participants
Within the REStORing health of acutely unwell adulTs (RESORT) observational, longitudinal cohort of geriatric rehabilitation inpatients, the Ending PJ Paralysis campaign was implemented on 2 out of 4 wards.
Methods
Objectively measured sedentary behavior and physical activity were measured by an inertial sensor (ActivPAL4) for 1 week, comparing control (non-PJ) and intervention (PJ) groups using linear mixed models. Mean sedentary behavior and physical activity measures and their association with physical and functional performance changes were investigated by linear regression analyses, stratified by low vs high performance at admission using the median as a cut-off.
Results
A total of 145 (n = 68 non-PJ and n = 77 PJ) inpatients with a mean age of 83.0 (7.7) years (55.9% female inpatients) were included. The median nonupright time was 23.1 [22.1-23.6] and 23.0 [21.8-23.6] hours/day for non-PJ and PJ groups, respectively. Objectively measured sedentary behavior and physical activity measures did not significantly change over measurement days and were independent of the Ending PJ Paralysis campaign. For inpatients with low performance at admission, lower sedentary behavior [B(SE) −0.013 (0.005) to −0.157 (0.045), P < .01] and higher physical activity [B(SE) 0.033 (0.007) to 0.814 (0.200), P < .01] measures were associated with improved physical performance. In addition, lower sedentary behaviour [B(SE) = -0.058 (0.024), P < .05 and higher physical activity [B (SE) 0.060 (0.024) to 0.683 (0.182), P < .05] were associated with improved instrumental functional performance.
Conclusions and Implications
In geriatric rehabilitation inpatients, the Ending PJ Paralysis campaign did not affect objectively measured sedentary behavior and physical activity patterns. Lower mean sedentary behaviour and higher physical activity measures were associated with improved physical and functional performance in inpatients with low performance.
It is hypothesized that these associations also hold for geriatric rehabilitation inpatients. In contrast, a systematic review in this population showed that interventions explicitly aiming to increase daily objectively measured physical activity behavior were ineffective, although only three studies were included.
How effective are interventions to increase physical activity levels among older inpatients receiving rehabilitation, without increasing the amount of therapy? A systematic review.
In this campaign, nursing staff encourage inpatients to be more physically active by getting dressed in day-clothes, having meals out of bed, and partaking in additional walks during the day. The Ending PJ Paralysis campaign was implemented in several countries.
The first studies showed conflicting results: positive effects with a 37% reduction in falls, 86% reduction in pressure injuries, 80% reduction in inpatient complaints, and a reduction of 1.5 days in length of stay,
Whether or not the Ending PJ paralysis campaign affects objectively measured sedentary behavior and physical activity patterns in geriatric rehabilitation inpatients and whether these are associated with changes in physical and functional performance is unknown. Patterns of sedentary behavior and physical activity are best studied using inertial sensors because self-reported measures under- and overestimate actual physical activity.
Reliability and validity of the international physical activity questionnaire compared to calibrated accelerometer cut-off points in the quantification of sedentary behaviour and physical activity in older adults.
The primary aim of this study in geriatric rehabilitation inpatients was to describe whether objectively measured sedentary behavior and physical activity patterns over 1 week were affected by the Ending PJ Paralysis campaign, taking the nursing staff availability at the bed-side and the time of day into account. Secondarily, associations between objectively measured sedentary behavior and physical activity measures and changes in physical and functional performance were investigated.
Methods
Study Design and Setting
The REStORing health of acutely unwell adulTs (RESORT) is an observational, longitudinal cohort of geriatric rehabilitation inpatients admitted at the Royal Park Campus of the Royal Melbourne Hospital (Melbourne, Victoria, Australia), which provides inpatient hospital-based care in 4 different wards. Geriatric rehabilitation inpatients were transferred from acute care wards toward these post-acute rehabilitation wards. Those unable to provide informed consent, without a legal proxy to consent or undergoing palliative care were excluded. Inpatients were assessed by a Comprehensive Geriatric Assessment within 48 hours of admission, which involves a multidisciplinary diagnostic process that assesses health domains, including medical, cognitive, physical, functional, and social parameters.
Inpatients within the RESORT cohort were considered for inclusion in the Ending PJ Paralysis campaign, using a quasi-experimental design comparing the control (non-PJ) group receiving usual care and intervention (PJ) group. The Ending PJ Paralysis campaign was adopted on one-half of the geriatric rehabilitation wards from June 3, 2019 to March 29, 2020. The campaign aimed to have at least (1) 80% of inpatients dressed in day-clothes by 11 o'clock, (2) 80% of inpatients wear appropriate footwear when out of bed, (3) 80% of inpatients eat lunch and dinner sitting out of bed, and (4) a 50% increase in participation of daily physical activity.
The intervention group was exposed to a multidisciplinary intervention, including extensive staff and inpatient education, a promotional campaign, and the introduction of communal dining and walking trails. Further details on the Ending PJ Paralysis study are mentioned elsewhere.
Hypothesized was that the intervention was dependent on (1) the availability of nursing staff, as nurses encourage physical activity in inpatients throughout the day; and (2) the time of day. Nursing shifts were divided into groups representing low, intermediate, and high nursing staff availability, based on a combination of hand-over times, patient care load and breaks (Supplementary Table 1). To explore further distributions over the day, the morning (6 am‒12 pm), afternoon (12 pm‒6 pm), and evening (6 pm‒12 am) were separated. Waking time was set from 7 pm to 9 pm . As part of the Ending PJ Paralysis campaign, objectively measured sedentary behavior and physical activity were assessed from October 22, 2019 to March 29, 2020. All inpatients without a bilateral lower extremity paralysis were considered eligible and no baseline level of ambulation status was required.
Inpatient Characteristics at Admission
Inpatient medical records were used to extract age, sex, number of medications, and the length of stay (in days) in geriatric rehabilitation. Use of a walking aid and a history of at least 1 fall in the past year were self-reported or extracted from medical records. Standing height was assessed to the nearest 0.1 cm using a stadiometer if the inpatient could stand. Otherwise, knee height was assessed using a measuring rod and height was calculated using the Chumlea equation for Caucasians.
Weight was assessed to the nearest 0.1 kg either by using a standing scale, seated scale, or a weighted hoist, depending on the inpatient's ambulation status. Body mass index was calculated by body mass (kg) divided by height squared (m) and expressed in kg/m2. The primary reason for hospital admission was categorized into musculoskeletal, cardiovascular or respiratory, neurologic, infectious, and other reasons. Comorbidity was assessed by the Charlson Comorbidity Index (range 0‒37)
and/or the Rowland Universal Dementia Assessment Scale if further cognitive testing was indicated. Cognitive impairment was defined as either a dementia diagnosis reported in medical records, a MMSE score <24/30, a MoCA score <26/30 or a Rowland Universal Dementia Assessment Scale score <23/30. The risk of delirium was assessed by the Short Confusion Assessment Method.
Objectively Measured Sedentary Behavior and Physical Activity
The ActivPAL4 (PAL Technologies Ltd) was used as an inertial sensor to assess objectively measured daily sedentary beahavior and physical activity patterns. The ActivPAL consists of a tri-axial capacitive accelerometer with a range of ±4 g, which collected data at a sample frequency of 20 Hz. On day 5 of admission (range: 3‒7), the ActivPAL sensor was attached to the right thigh for 1 week, or until hospital discharge. A valid day of measurements was defined as 20/24 hours of wear. Inpatients were included if they reported at least 1 valid day. The ActivPAL software (Generation 8, PAL Technologies Ltd) was used and a custom code obtained objectively measured sedentary behavior and physical activity measures for every 30 minutes. Daily objectively measured sedentary behavior patterns were described by time spent nonupright (sum of sitting and lying), sitting and lying in hours/day, and physical activity patterns by time spent upright (sum of standing and stepping), standing and stepping time in minutes/day, and the number of steps and sit-to-stand transitions per day.
Physical and Functional Performance
Physical performance was assessed by the Short Physical Performance Battery (SPPB, score range 0‒12) combining balance, a timed 4-meter walk [gait speed (m/s)] and the timed chair stand test.
A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission.
The maximum value in kilograms was used for analyses. Inpatients who were unable were allocated 0.00 m/s or 0.0 kg. Functional performance was measured using the Katz index of activities of daily living (ADL, range 0‒6)
Change (Δ) in physical and functional performance during geriatric rehabilitation was defined as the discharge performance score minus the admission performance score.
Statistical Analyses
Descriptive statistics for continuous variables with a Gaussian (normal) distribution were presented as means with standard deviations (SDs) and a non-Gaussian (skewed) distribution as medians with interquartile ranges (IQR). Categorical variables were presented as numbers with percentages, n (%). Baseline characteristics between the non-PJ and PJ groups were compared using independent-samples t-tests (normal distribution), Mann-Whitney U tests (skewed distribution), χ2- tests, or Fisher exact tests (categorical variables). Deciles based on the mean number of steps per day were made to visualize patterns.
The change of objectively measured sedentary behavior and physical activity measures over days was analyzed using generalized negative binomial mixed models for count variables, ie, number of steps and sit-to-stand transitions. Linear mixed models were conducted to assess the change in noncount, ie, sedentary behavior and physical activity measures. Independent variables included measurement days, the Ending PJ Paralysis campaign and the interaction between measurement days and the Ending PJ Paralysis campaign. A random intercept on patient level was included. The following variables were added to the model to adjust for possible confounding: age, sex, comorbidity (Cumulative Illness Rating Scale score), ambulation status (Functional Ambulation Classification score), and weekend day (binary).
For all valid days, the duration per hour in objectively measured sedentary behavior and physical activity measures were calculated and compared between low, intermediate, and high availability of nursing staff during waking hours and the time of day (morning, afternoon, evening) using Friedman tests, including pairwise comparisons with a Bonferroni correction to adjust for multiple testing. Differences in objectively measured sedentary behavior and physical activity measures between non-PJ and PJ groups per level of nursing staff availability and the time of day were tested by Mann-Whitney U tests.
To investigate the association between mean objectively measured sedentary behavior and physical activity measures over days and changes in physical and functional performance during geriatric inpatient rehabilitation, multivariable linear regression analyses were performed. Moderator analyses were conducted to investigate the effect of low vs high physical and functional performance at admission. The median of the specific performance measure was used as a cut-off, to conduct approximately equal groups of low and high performers. All sedentary behavior and physical activity measures were divided by 10, except for step count which was divided by 100, for interpretation purposes. All analyses were performed using an age and sex-adjusted model (model 1) and additionally adjusted for comorbidity (CIRS-score) (model 2). If the moderator analysis did not show a significant effect, analyses were not stratified and additionally adjusted for physical or functional performance at admission (model 3). Sensitivity analyses investigating differences in patterns of objectively measured sedentary behavior and physical activity measures between inpatients with low vs high physical performance were conducted.
The statistical significance level was set at α = 0.05. Analyses were performed using the IBM SPSS Statistics for Windows, v 27.0 (IBM Corp.).
Results
Table 1 summarizes the characteristics of the 145 included inpatients (55.9% female) with a mean age 83.0 (SD 7.7) years. The median SPPB score was 1 [0‒4] and 1 [0‒5] for non-PJ (n = 68) and PJ (n = 77) groups and did not differ. Median length of stay was 17 days [IQR 12-30]. Primary reason for admission varied from musculoskeletal (49%) to cardiovascular and respiratory (14.5%) and neurologic (13.8%) diseases.
Table 1Inpatient Characteristics of the Total Population and Stratified by the Ending PyJama Paralysis Campaign
The mean of the objectively measured sedentary behavior and physical activity measure over measurement days. Independent samples t-test for normally distributed variables, Mann-Whitney U-test for non-normally distributed variables, χ2 or Fisher exact tests for categorical variables were used to compare non-PJ to PJ-group. P <.05 presented in bold.
The mean of the objectively measured sedentary behavior and physical activity measure over measurement days. Independent samples t-test for normally distributed variables, Mann-Whitney U-test for non-normally distributed variables, χ2 or Fisher exact tests for categorical variables were used to compare non-PJ to PJ-group. P <.05 presented in bold.
The mean of the objectively measured sedentary behavior and physical activity measure over measurement days. Independent samples t-test for normally distributed variables, Mann-Whitney U-test for non-normally distributed variables, χ2 or Fisher exact tests for categorical variables were used to compare non-PJ to PJ-group. P <.05 presented in bold.
The mean of the objectively measured sedentary behavior and physical activity measure over measurement days. Independent samples t-test for normally distributed variables, Mann-Whitney U-test for non-normally distributed variables, χ2 or Fisher exact tests for categorical variables were used to compare non-PJ to PJ-group. P <.05 presented in bold.
The mean of the objectively measured sedentary behavior and physical activity measure over measurement days. Independent samples t-test for normally distributed variables, Mann-Whitney U-test for non-normally distributed variables, χ2 or Fisher exact tests for categorical variables were used to compare non-PJ to PJ-group. P <.05 presented in bold.
The mean of the objectively measured sedentary behavior and physical activity measure over measurement days. Independent samples t-test for normally distributed variables, Mann-Whitney U-test for non-normally distributed variables, χ2 or Fisher exact tests for categorical variables were used to compare non-PJ to PJ-group. P <.05 presented in bold.
The mean of the objectively measured sedentary behavior and physical activity measure over measurement days. Independent samples t-test for normally distributed variables, Mann-Whitney U-test for non-normally distributed variables, χ2 or Fisher exact tests for categorical variables were used to compare non-PJ to PJ-group. P <.05 presented in bold.
The mean of the objectively measured sedentary behavior and physical activity measure over measurement days. Independent samples t-test for normally distributed variables, Mann-Whitney U-test for non-normally distributed variables, χ2 or Fisher exact tests for categorical variables were used to compare non-PJ to PJ-group. P <.05 presented in bold.
145
20 [10-30]
18 [9-30]
21 [10-30]
.345
Physical and functional performance at admission
Hand grip strength (kg)
131
Female
72
13.1 (6.9)
10.0 (6.2)
15.7 (6.4)
.001
Male
59
20.5 (8.9)
21.5 (6.4)
19.6 (10.9)
.366
SPPB score [0-12]
136
1 [0-4]
1 [0-4]
1 [0-5]
.924
Gait speed (m/s)
140
0.18 [0.00-0.49]
0.00 [0.00-0.47]
0.27 [0.00-0.52]
.247
ADL score [0-6]
145
2 [1-3]
2 [1-2]
2 [1-3]
.474
IADL score [0-8]
145
1 [0-2]
1 [0-1]
1 [1-2]
.003
Change in physical and functional performance during geriatric rehabilitation
∗ The mean of the objectively measured sedentary behavior and physical activity measure over measurement days. Independent samples t-test for normally distributed variables, Mann-Whitney U-test for non-normally distributed variables, χ2 or Fisher exact tests for categorical variables were used to compare non-PJ to PJ-group. P <.05 presented in bold.
Objectively Measured Sedentary Behavior and Physical Activity Measures
The median wearing duration of the ActivPAL4 was 6 [5-6] days. Median nonupright time was 23.1 [22.1-23.6] and 23.0 [21.8-23.6] hours/day and the median number of steps was 291 [42-871] and 471 [79-951] per day, for the non-PJ and PJ groups, respectively. Figure 1 shows objectively measured sedentary behavior and physical activity patterns of one day for 10 representative inpatients. Long periods of nonupright time were found, even in inpatients representing higher deciles, representing more physically active inpatients.
Fig. 1Sedentary behavior and physical activity patterns of one representative day for 10 inpatients representing the deciles made based on the number of steps.
The changes in objectively measured sedentary behavior and physical activity measures over days in the fully adjusted models are presented in Tables 2 and 3, and showed no significant effect of measurement days on these measures, except for stepping time. Table 2 shows the changes of objectively measured sedentary behavior and physical activity measures over days comparing the respective day with day 1. Table 3 shows comparisons between the non-PJ and PJ groups. In the total population, inpatients spent 2.9‒6.7 minutes more in stepping time on day 6-8 (P ≤ .048) when compared with day 1 (Table 4). Neither changes over days for sedentary behavior or physical activity measures (Table 4) nor differences between non-PJ and PJ groups were found (Supplementary Table 2).
Table 2Results of the Fully Adjusted Models for Patterns of Objectively Measured Sedentary Behavior and Physical Activity Measures Over Measurement Days (n = 140)
Table 3Results of the Fully Adjusted Models for Patterns of Objectively Measured Sedentary Behavior and Physical Activity Measures Over Measurement Days, Including the Ending PJ Paralysis Campaign (n = 140)
Supplementary Tables 3 and 4 shows an overview of objectively measured sedentary behavior and physical activity measures in minutes per hour of ActivPAL wear during waking hours, stratified by nursing staff availability at the bed-side and the time of day. There was no dose-response relationship between nursing staff's availability at the bed-side and sedentary behavior and physical activity measures. Over the day, all physical activity measures were highest in the morning and nonupright time was highest in the evening. No differences were found between non-PJ and PJ groups (Supplementary Tables 5 and 6). Figure 2 shows the mean number of steps per 30-minute period over the day, showing a higher number of steps in the morning and before lunch for both groups.
Fig. 2Visualization of the mean number of steps per 30-minute period over the day for the non-PJ and PJ groups, including visualization of the periods of high nursing staff availability at the bed-side. For each 30-minute period, group medians (50th percentiles) are presented by points connected by solid lines and upper (75th percentile) and lower (25th percentile) limits of interquartile ranges are represented by dotted lines.
Objectively Measured Sedentary Behavior and physical Activity Measures and Changes in Physical and Functional Performance
The associations between objectively measured sedentary behavior and physical activity measures and changes in physical and functional performance were dependent on performance levels at admission (Supplementary Table 7). A higher mean nonupright time of 10 min/d was associated with declined SPPB scores of −0.157 [standard error (SE) 0.045) points, P = .001], gait speed −0.013 (0.005) m/s, P = .008 and IADL scores (−0.058 (0.024) points, P = .015). Higher mean objectively measured physical activity measures were associated with improved SPPB scores, gait speed and IADL scores. Higher mean number of sit-to-stand transitions were associated with improved ADL scores.
Sensitivity Analyses
Inpatients with low physical performance were more sedentary and less active than patients with high physical performance and showed a greater increase in physical and functional performance (Supplementary Table 8). Patterns of objectively measured sedentary behavior and physical activity measures over days did not differ, except for stepping time showing a higher stepping time on day 5 and 6 with respect to day 1 in inpatients with high physical performance (Supplementary Tables 9 and 10).
The effect of the Ending PJ Paralysis campaign did not differ between inpatients with low vs high physical performance at admission (Supplementary Tables 11 and 12).
Discussion
In geriatric rehabilitation inpatients, patterns of objectively measured sedentary behavior and physical activity measures were not affected by the Ending PJ Paralysis campaign and did not change over measurement days. Geriatric rehabilitation inpatients were very physically inactive. Lower objectively measured sedentary behavior and higher physical activity measures were present in the morning when compared with the afternoon and evening. For inpatients with low physical or functional performance at admission, both lower sedentary behavior and higher physical activity measures were associated with improved physical and functional performance during geriatric rehabilitation.
Ending PJ Paralysis Campaign Effect
Although the Ending PJ paralysis campaign embraced recent recommendations to increase in-hospital physical activity,
Recommendations for older adults' physical activity and sedentary behaviour during hospitalisation for an acute medical illness: an international Delphi study.
the intervention could not influence objectively measured sedentary behavior and physical activity patterns. Between different levels of nursing staff availability at the bed-side, objectively measured sedentary behavior and physical activity measures significantly differ. However, differences were minimal and no dose-response relationship across levels of nursing staff availability was found. Therefore, these differences were considered not clinically relevant. Allied health professionals' availability may have influenced objectively measured sedentary behavior and physical activity patterns, but was not considered in this study. Another explanation could be that the Ending PJ Paralysis campaign was not intensive enough to change sedentary behavior and physical activity. In a recent systematic review, interventions aimed at increasing nontherapy physical activity in geriatric rehabilitation were found to be ineffective.
How effective are interventions to increase physical activity levels among older inpatients receiving rehabilitation, without increasing the amount of therapy? A systematic review.
In acutely hospitalized older adults, intensive exercise interventions, characterized by supervised sessions that lasted between 15 and 30 minutes and were performed 5 to 7 days a week, have shown to be effective as physical and functional performance improved.
However, the effects on objectively measured sedentary behavior and physical activity were not assessed. Innovative health care models such as rehabilitation in the home
Geriatric Rehabilitation Inpatients Roam at Home! A Matched Cohort Study of Objectively Measured Physical Activity and Sedentary Behavior in Home-Based and Hospital-Based Settings.
Future studies should assess the impact on objectively measured sedentary behavior and physical activity and may address the possible added beneficial effect of using these measures as a feedback tool.
Objectively Measured Sedentary Behavior and Physical Activity Measures and Changes in Physical and Functional Performance
Even in these frail and highly inactive geriatric rehabilitation inpatients, shown by moderately to severe frailty scores and a median of 402 steps per day, both lower objectively measured sedentary behavior and higher physical activity were associated with improved physical and functional performance for those with low performance at admission only. Improvements in physical performance are important, as these are associated with lower institutionalization and mortality rates 3 months after discharge from geriatric rehabilitation.
The association of changes in physical performance during geriatric rehabilitation and short-term hospital readmission, institutionalisatin and mortality: RESORT.
The disparity in the association between physical activity and outcome for older adults with low vs high physical performance at admission was also found in sedentary older community-dwelling participants of the Lifestyle Interventions and Independence for Elders study.
These results may indicate that older adults with low physical and/or functional performance have a greater benefit from physical activity.
Strength and Limitations
This is the first study evaluating the effect of the Ending PJ Paralysis campaign on objectively measured sedentary behavior and physical activity measures. A strength of this study is the use of the ActivPAL4, as accelerometers are able to assess sedentary behavior and physical activity reliably in hospitalized older adults
Reliability and validity of the international physical activity questionnaire compared to calibrated accelerometer cut-off points in the quantification of sedentary behaviour and physical activity in older adults.
A limitation is the current measurement period which might have been too short to capture any changes in objectively measured sedentary behavior and physical activity patterns. This study's quasi-experimental design could also be a limitation, introducing possible bias as inpatients were not randomized over intervention groups. Another limitation is that we were not able to consider allied health care use in our analyses. Finally, inpatients were included regardless of their admission diagnosis, highlighting the generalizability of our results.
Conclusions and Implications
In geriatric rehabilitation inpatients, the Ending PJ Paralysis campaign did not affect patterns of objectively measured sedentary behavior and physical activity. Inpatients were very physically inactive and showed less sedentary behavior and more physical activity in the morning when compared with the afternoon and evening. Even in this highly inactive population, lower mean objectively measured sedentary behavior and higher mean physical activity measures were associated with improved physical and functional performance for inpatients with low performance at hospital admission.
Acknowledgments
The authors thank the multidisciplinary team members of the Royal Melbourne Hospital, Royal Park Campus involved in the RESORT cohort for their clinical work and the @Age team for their role in the data collection and processing.
Supplementary Data
Supplementary Table 1Overview of Expected Availability of Nursing Staff at the Bed-Side During Waking Hours
Supplementary Table 2Comparison of Patterns of Objectively Measured Sedentary Behavior and Physical Activity Measures Over Measurement Days Between Non-PJ and PJ Groups
Day
Nonupright Time (min/d), Unadjusted N = 145
Nonupright Time (min/d), Adjusted N = 140
PJ Group vs Non-PJ Group Intervention vs control
PJ Group vs Non-PJ Group Intervention vs control
Coefficient
95% CI
P
Coefficient
95% CI
P
2 vs 1
12.324
−4.077 to 28.726
.141
6.813
−8.080 to 21.706
.369
3 vs 1
13.845
−2.741 to 0.432
.102
8.650
−6.417 to 23.717
.260
4 vs 1
9.385
−7.666 to 26.437
.280
10.634
−4.874 to 26.141
.179
5 vs 1
4.931
−12.596 to 22.458
.581
0.162
−15.755 to 16.078
.984
6 vs 1
3.913
−14.082 to 21.907
.670
2.453
−13.906 to 18.812
.769
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
Sitting time (min/d), unadjusted N = 145
Sitting time (min/d), adjusted N = 140
2 vs 1
−42.596
−156.029 to 70.836
.461
−53.561
−167.945 to 60.822
.358
3 vs 1
−44.783
−159.406 to 69.841
.443
−71.072
−186.711 to 44.567
.228
4 vs 1
40.082
−77.673 to 157.836
.504
29.471
−89.470 to 148.412
.627
5 vs 1
5.250
−115.750 to 126.249
.932
−20.292
−142.343 to 101.760
.744
6 vs 1
−3.627
−127.836 to 120.581
.954
−29.639
−155.062 to 95.784
.643
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
Lying time (min/d), unadjusted N = 145
Lying time (min/d), adjusted N = 140
2 vs 1
54.951
−58.722 to 168.624
.343
59.937
−54.533 to 174.407
.304
3 vs 1
59.590
−55.295 to 174.475
.309
80.256
−35.485 to 195.998
.174
4 vs 1
−30.107
−148.149 to 87.935
.617
−18.346
−137.408 to 100.715
.762
5 vs 1
0.654
−120.651 to 121.958
.992
21.340
−100.842 to 143.522
.732
6 vs 1
8.217
−116.309 to 132.743
.897
32.912
−92.649 to 158.474
.607
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
Upright time (min/d), unadjusted N = 145
Upright time (min/d), adjusted N = 140
2 vs 1
−12.315
−28.571 to 3.941
.137
−6.772
−21.504 to 7.960
.563
3 vs 1
−11.196
−27.636 to 5.244
.182
−5.893
−20.797 to 9.011
.587
4 vs 1
−9.346
−26.247 to 7.554
.278
−10.643
−25.983 to 4.696
.174
5 vs 1
−8.862
−26.234 to 8.510
.317
−4.355
−20.099 to 11.389
.438
6 vs 1
−6.081
−23.916 to 11.755
.504
−4.775
−20.957 to 11.407
.367
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
Day
Standing Time (min/d), Unadjusted N = 145
Standing Time (min/d), Adjusted N = 140
PJ Group vs Non-PJ Group Intervention vs Control
PJ Group vs Non-PJ Group Intervention vs Control
Coefficient
95% CI
P
Coefficient
95% CI
P
2 vs 1
−13.572
−29.097 to 1.952
.087
−8.202
−22.035 to 5.631
.245
3 vs 1
−11.985
−27.684 to 3.715
.134
−6.306
−20.300 to 7.689
.377
4 vs 1
−9.590
−25.729 to 6.549
.244
−11.344
−25.747 to 3.059
.122
5 vs 1
−10.565
−27.154 to 6.025
.212
−6.197
−20.980 to 8.586
.411
6 vs 1
−7.436
−24.468 to 9.596
.392
−6.187
−21.380 to 9.007
.424
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
Stepping time (min/d), unadjusted N = 145
Stepping time (min/d), adjusted N = 140
2 vs 1
1.223
−1.367 to 3.813
.354
1.426
−1.166 to 4.017
.281
3 vs 1
0.707
−1.912 to 3.326
.596
0.376
−2.246 to 2.997
.779
4 vs 1
0.150
−2.542 to 2.843
.913
0.649
−2.049 to 3.348
.637
5 vs 1
1.603
−1.164 to 4.371
.256
1.783
−0.986 to 4.553
.207
6 vs 1
1.251
−1.590 to 4.092
.388
1.350
−1.497 to 4.196
.352
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
Day
Steps(number/d), unadjusted N = 145
Steps (number/d), adjusted N = 140
Rate Ratio
95% CI
P
Rate Ratio
95% CI
P
2 vs 1
1.062
0.544 to 2.075
.860
1.051
0.532 to 2.075
.886
3 vs 1
1.009
0.513 to 1.986
.979
1.019
0.512 to 2.027
.958
4 vs 1
1.045
0.521 to 2.098
.901
1.134
0.558 to 2.303
.728
5 vs 1
0.845
0.413 to 1.729
.643
0.875
0.423 to 1.811
.718
6 vs 1
0.747
0.358 to 1.558
.437
0.750
0.355 to 1.582
.449
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
STS (number/d), unadjusted N = 145
STS (number/d), adjusted N = 140
2 vs 1
0.875
0.563 to 1.359
.551
0.889
0.600 to 1.318
.558
3 vs 1
0.924
0.593 to 1.442
.728
0.923
0.621 to 1.373
.962
4 vs 1
1.014
0.642 to 1.604
.951
1.044
0.693 to 1.572
.838
5 vs 1
0.926
0.578 to 1.484
.750
0.957
0.628 to 1.459
.838
6 vs 1
0.892
0.550 to 1.445
.641
0.918
0.597 to 1.414
.698
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
STS, sit-to-stand transitions.
Adjusted for age, sex, comorbidity, and ambulation status and weekend day. Physical performance assessed by Short Physical Performance Battery. n/a: Comparison between non-PJ and PJ group is not applicable as all patients wearing the ActivPAL for 7 days or more were part of the PJ group.
Supplementary Table 3Overview of Objectively Measured Sedentary Behavior and Physical Activity Measures in Minutes per Hour of ActivPAL Wear Between 7 am and 9 pm, Stratified By Nursing Staff Availability At the Bed-Side
Nursing Staff Availability
Across Groups
Low vs Intermediate
Low vs High
Intermediate vs High
Low
Intermediate
High
P
P
P
P
Nonupright time (min/h)
57.2 [53.0-58.9]
56.4 [51.4-58.4]
56.4 [52.7-58.5]
<.001
<.001
.052
.019
Sitting time (min/h)
34.6 [9.43-45.3]
34.4 [9.38-47.1]
33.5 [9.83-42.9]
<.001
.011
.283
.001
Lying time (min/h)
18.8 [9.17-48.9]
13.8 [1.39-48.2]
19.1 [9.73-47.1]
<.001
<.001
1.000
.001
Upright time (min/h)
2.78 [1.11-7.03]
3.58 [1.61-8.62]
3.56 [1.48-7.31]
<.001
<.001
.052
.019
Standing time (min/h)
2.37 [0.91-5.84]
2.93 [1.24-7.31]
2.95 [1.39-6.01]
<.001
<.001
.038
.096
Stepping time (min/h)
0.33 [0.06-0.82]
0.49 [0.07-1.31]
0.35 [0.07-0.86]
<.001
<.001
.998
.001
Steps (number/h)
15.6 [2.67-51.8]
28.4 [2.80-76.0]
20.5 [3.37-49.8]
<.001
<.001
1.000
.001
Sit-to-Stand transitions (number/h)
1.06 [0.50-1.67]
1.33 [0.60-2.23]
1.20 [0.62-1.78]
<.001
<.001
.077
.120
Differences in objectively measured sedentary behavior and physical activity measures between low, intermediate and high nursing staff availability at the bed-side were tested by a Friedman test. Pairwise comparisons have been adjusted by the Bonferroni correction.
Supplementary Table 4Overview of Objectively Measured Sedentary Behavior And Physical Activity Measures In Minutes Per Hour Of ActivPAL Wear Between 6 am and 12 pm, Stratified by the Time of Day
Morning
Afternoon
Evening
Across Groups
Morning vs Afternoon
Morning vs Evening
Afternoon vs Evening
P
P
P
P
Nonupright time (min/h)
56.0 [52.7-58.4]
57.0 [53.1-58.9]
58.4 [55.9-59.4]
<.001
.016
<.001
<.001
Sitting time (min/h)
25.6 [8.83-37.1]
34.2 [9.70-49.5]
17.4 [3.61-27.1]
<.001
<.001
<.001
<.001
Lying time (min/h)
24.1 [14.2-49.4]
18.9 [1.83-47.8]
40.0 [28.9-54.8]
<.001
<.001
<.001
<.001
Upright time (min/h)
4.00 [1.59-7.30]
2.96 [1.15-6.92]
1.63 [0.64-4.08]
<.001
.016
<.001
<.001
Standing time (min/h)
3.24 [1.42-6.33]
2.33 [0.99-5.64]
1.43 [0.60-3.40]
<.001
.005
<.001
<.001
Stepping time (min/h)
0.43 [0.08-0.90]
0.35 [0.08-0.83]
0.17 [0.03-0.49]
<.001
.066
<.001
<.001
Steps (number/h)
23.7 [3.50-55.5]
18.7 [3.08-53.9]
9.00 [1.08-27.8]
<.001
.193
<.001
<.001
Sit-to-Stand transitions (number/h)
1.33 [0.63-1.96]
1.03 [0.49-1.67]
0.56 [0.29-1.04]
<.001
.014
<.001
<.001
Morning: 6 am‒12 pm; Afternoon: 12 pm‒6 pm; Evening: 6 pm‒12 Am. Differences in objectively measured sedentary behavior and physical activity measures between the morning, afternoon and evening were tested by a Friedman test. Pairwise comparisons have been adjusted by the Bonferroni correction.
Supplementary Table 5Differences in Objectively Measured Sedentary Behavior and Physical Activity Measures Between Non-PJ and PJ Groups Per Level Of Nursing Staff Availability (Low, Intermediate, or High) At The Bed-Side
Non-PJ Groups (Control) n = 68
PJ Groups (Intervention) n = 77
P
Nonupright time (min/h)
Low
57.4 [54.3‒58.9]
56.8 [52.5‒58.9]
.550
Intermediate
55.8 [52.0‒58.5]
56.6 [50.6‒58.3]
.566
High
56.8 [53.8‒58.7]
56.1 [51.4‒58.3]
.322
Sitting time (min/h)
Low
31.5 [9.60‒45.1]
36.5 [8.64‒45.6]
.419
Intermediate
35.3 [11.7‒46.3]
33.9 [5.84‒49.0]
.529
High
30.5 [10.2‒43.2]
34.3 [6.71‒42.4]
.926
Lying time (min/h)
Low
24.8 [9.54‒49.8]
15.5 [8.33‒47.6]
.279
Intermediate
23.0 [5.02‒47.3]
11.9 [0.00‒49.8]
.282
High
25.7 [10.7‒47.1]
16.7 [9.08‒47.5]
.421
Upright time (min/h)
Low
2.65 [1.09‒5.72]
3.18 [1.12‒7.50]
.550
Intermediate
4.16 [1.51‒8.04]
3.43 [1.66‒9.43]
.566
High
3.21 [1.32‒6.19]
3.87 [1.70‒8.59]
.322
Standing time (min/h)
Low
2.33 [0.90‒4.74]
2.45 [0.91‒6.58]
.721
Intermediate
3.46 [1.10‒6.65]
2.80 [1.32‒7.92]
.698
High
2.90 [1.26‒5.32]
2.95 [1.54‒7.60]
.381
Stepping time (min/h)
Low
0.24 [0.07‒0.66]
0.47 [0.05‒0.94]
.233
Intermediate
0.40 [0.05‒1.37]
0.55 [0.11‒1.31]
.461
High
0.29 [0.06‒0.75]
0.37 [0.12‒0.97]
.183
Steps (number/h)
Low
13.2 [2.69‒39.8]
24.4 [2.10‒56.6]
.239
Intermediate
28.3 [2.30‒72.5]
28.4 [4.96‒78.5]
.426
High
18.9 [2.01‒41.7]
21.5 [4.44‒61.7]
.188
Sit‒to‒stand transitions (number/h)
Low
0.89 [0.50‒1.54]
1.13 [0.50‒1.76]
.416
Intermediate
1.20 [0.42‒2.32]
1.53 [0.70‒2.17]
.502
High
1.15 [0.54‒1.69]
1.21 [0.70‒1.85]
.255
Values are presented as median [IQR]. Differences between non-PJ and PJ groups were tested by a Mann-Whitney U test.
Supplementary Table 6Differences in Objectively Measured Sedentary Behavior and Physical Activity Measures Between Non-PJ and PJ Groups for the Morning, Afternoon, and Evening
Non-PJ Groups (Control) n = 68
PJ Groups (Intervention) n = 77
P
Non-upright time (min/h)
Morning
55.4 [52.8‒58.6]
56.2 [52.6‒58.3]
.521
Afternoon
57.0 [54.5‒59.0]
57.3 [52.0‒58.6]
.358
Evening
58.3 [56.6‒59.4]
58.6 [55.2‒59.3]
.513
Sitting time (min/h)
Morning
27.9 [9.28‒37.7]
24.8 [7.68‒36.3]
.814
Afternoon
32.6 [9.82‒49.2]
37.7 [8.50‒49.6]
.426
Evening
16.6 [3.79‒27.2]
17.4 [2.95‒27.6]
.981
Lying time (min/h)
Morning
25.1 [14.4‒49.8]
23.3 [13.7‒47.8]
.714
Afternoon
22.1 [4.35‒47.9]
12.5 [0.00‒48.4]
.194
Evening
41.3 [29.6‒55.1]
39.7 [27.5‒53.0]
.705
Upright time (min/h)
Morning
4.63 [1.40‒7.25]
3.81 [1.66‒7.44]
.521
Afternoon
3.02 [0.97‒5.47]
2.70 [1.42‒8.01]
.358
Evening
1.67 [0.63‒3.38]
1.44 [0.74‒4.84]
.513
Standing time (min/h)
Morning
4.08 [1.30‒6.35]
3.04 [1.49‒6.36]
.620
Afternoon
2.52 [0.89‒4.24]
2.22 [1.09‒6.64]
.503
Evening
1.52 [0.61‒2.96]
1.37 [0.51‒3.95]
.571
Stepping time (min/h)
Morning
0.33 [0.06‒0.90]
0.48 [0.11‒0.91]
.243
Afternoon
0.29 [0.05‒0.70]
0.43 [0.09‒1.12]
.180
Evening
0.14 [0.02‒0.44]
0.18 [0.03‒0.51]
.590
Steps (number/h)
Morning
21.2 [3.06‒49.6]
25.9 [4.37‒58.7]
.276
Afternoon
15.9 [2.51‒40.0]
24.2 [3.31‒66.7]
.184
Evening
9.22 [0.82‒23.2]
8.89 [1.58‒32.0]
.704
Sit-to-stand transitions (number/h)
Morning
1.28 [0.53‒2.04]
1.33 [0.79‒1.96]
.426
Afternoon
0.89 [0.37‒1.47]
1.08 [0.53‒1.67]
.314
Evening
0.50 [0.23‒0.83]
0.67 [0.32‒1.11]
.250
Values are presented as median [IQR]. Differences between non-PJ and PJ groups were tested by a Mann-Whitney U test. Morning: 6 am‒12 pm; Afternoon: 12 pm‒6 pm; Evening: 6 pm‒12 am
Supplementary Table 7The Association Between Mean Objectively Measured Sedentary Behavior and Physical Activity Measures and the Change in Physical or Functional Performance Measures During Geriatric Rehabilitation, in the Whole Population or Stratified by Baseline Performance
Additionally adjusted for baseline physical performance as baseline physical performance was not an effect-modifier and therefore analyses were not stratified. P < .05 presented in bold.
Additionally adjusted for baseline physical performance as baseline physical performance was not an effect-modifier and therefore analyses were not stratified. P < .05 presented in bold.
Additionally adjusted for baseline physical performance as baseline physical performance was not an effect-modifier and therefore analyses were not stratified. P < .05 presented in bold.
Additionally adjusted for baseline physical performance as baseline physical performance was not an effect-modifier and therefore analyses were not stratified. P < .05 presented in bold.
Additionally adjusted for baseline physical performance as baseline physical performance was not an effect-modifier and therefore analyses were not stratified. P < .05 presented in bold.
Additionally adjusted for baseline physical performance as baseline physical performance was not an effect-modifier and therefore analyses were not stratified. P < .05 presented in bold.
Additionally adjusted for baseline physical performance as baseline physical performance was not an effect-modifier and therefore analyses were not stratified. P < .05 presented in bold.
Additionally adjusted for baseline physical performance as baseline physical performance was not an effect-modifier and therefore analyses were not stratified. P < .05 presented in bold.
.001
Model 1: age + sex
−0.122 (0.179)
.497
0.005 (0.019)
.775
−0.065 (0.124)
.600
Model 2: Model 1 + comorbidity
−0.195 (0.173)
.264
0.003 (0.019)
.861
−0.065 (0.126)
.605
B, unstandardized beta regression coefficient; N, number of inpatients; Δ, delta.
∗ Additionally adjusted for baseline physical performance as baseline physical performance was not an effect-modifier and therefore analyses were not stratified. P < .05 presented in bold.
Supplementary Table 8Descriptives of Objectively Measured Sedentary Behavior and Physical Activity Measures and the Change in Physical and Functional Performance Stratified by Baseline Physical Performance (SPPB Score)
Supplementary Table 9Results of the Adjusted (n = 132) Models for Patterns of Objectively Measured Sedentary Behavior and Physical Activity Measures Over Days, Including the Effect of Baseline Physical Performance (SPPB score).
Supplementary Table 11Results of the Adjusted Models for Patterns of Objectively Measured Sedentary Behavior and Physical Activity Measures Over Days Including the Ending PJ Paralysis Effect, Stratified for Inpatients With Low vs High Baseline Physical Performance (SPPB Score) (n = 132)
Nonupright Time (min/d)
Sitting Time (min/d)
Lying Time (min/d)
Upright Time (min/d)
Standing Time (min/d)
Stepping Time (min/d)
Steps (number/d)
STS (number/d)
P
P
P
P
P
P
P
P
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
n = 71
n = 61
n = 71
n = 61
n = 71
n = 61
n = 71
n = 61
n = 71
n = 61
n = 71
n = 61
n = 71
n = 61
n = 71
n = 61
Day
.487
.272
.060
.370
.049
.467
.058
.310
.045
.583
.124
.002
.580
.952
.947
.996
Ending PJ Paralysis
.618
.391
.780
.952
.845
.739
.623
.419
.694
.432
.412
.600
.981
.623
.861
.683
Day∗ Ending PJ Paralysis
.835
.425
.349
.623
.350
.732
.468
.444
.502
.467
.837
.322
.776
.997
.998
.950
Age
.875
.173
.161
.215
.186
.537
.924
.199
.852
.186
.573
.673
.708
.661
.943
.479
Sex
.618
.396
.303
.485
.293
.387
.716
.369
.682
.369
.854
.657
.694
.995
.668
.003
Comorbidity
.072
.114
.622
.685
.822
.847
.057
.111
.086
.203
.057
.008
.067
.002
.162
.382
Ambulation status
<.001
.171
.096
.870
.037
.565
<.001
.144
<.001
.186
<.001
.122
<.001
.122
<.001
.348
Weekend d
.003
.315
.920
.887
.674
.979
.002
.407
.007
.432
<.001
.578
.005
.212
.003
.065
CIRS, Cumulative Illness Rating Scale; FAC, Functional Ambulation Classification; STS, sit-to-stand transitions. Physical performance, binary low vs high based on SPPB score; Comorbidity = CIRS score; Ambulation status = FAC score.
Supplementary Table 12Comparison of Patterns of Objectively Measured Sedentary Behavior and Physical Activity Measures Over Measurement Days Between Non-PJ and PJ Groups, Stratified by Baseline Physical Performance
Day
Low Baseline Performance
High Baseline Performance
Nonupright Time (min/d), Adjusted n = 71
Nonupright Time (min/d), Adjusted n = 61
PJ Group vs Non-PJ Group Intervention vs Control
PJ Group vs Non-PJ Group Intervention vs Control
Coefficient
95% CI
P
Coefficient
95% CI
P
2 vs 1
6.270
−12.574 to 25.114
.513
8.089
−18.018 to 34.197
.542
3 vs 1
10.035
−8.810 to 28.879
.296
5.911
−20.945 to 32.767
.665
4 vs 1
12.177
−6.949 to 31.302
.211
8.595
−19.570 to 36.760
.549
5 vs 1
10.116
−9.631 to 29.864
.314
−17.226
−45.944 to 11.491
.239
6 vs 1
10.706
−9.577 to 30.988
.300
−10.454
−40.100 to 19.191
.488
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
Sitting time (min/d), adjusted n = 71
Sitting time (min/d), adjusted n = 61
2 vs 1
−104.288
−275.855 to 67.280
.233
−12.006
−168.750 to 144.737
.880
3 vs 1
−119.948
−291.522 to 51.625
.170
−1.661
−162.485 to 159.163
.984
4 vs 1
22.857
−151.252 to 196.967
.796
76.788
−91.573 to 245.150
.370
5 vs 1
0.303
−179.442 to 180.047
.997
−17.076
−188.732 to 154.581
.845
6 vs 1
40.382
−144.214 to 224.978
.667
−96.700
−273.921 to 80.522
.284
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
Lying time (min/d), adjusted n = 71
Lying time (min/d), adjusted n = 61
2 vs 1
110.042
−61.374 to 281.458
.208
20.066
−137.495 to 177.627
.802
3 vs 1
129.475
−41.946 to 300.897
.138
9.977
−151.809 to 171.763
.903
4 vs 1
−11.344
−185.304 to 162.616
.898
−66.954
−236.416 to 102.509
.437
5 vs 1
9.154
−170.443 to 188.751
.920
2.011
−170.773 to 174.794
.982
6 vs 1
−30.313
−214.761 to 154.136
.747
87.627
−90.754 to 266.008
.334
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
Upright time (min/d), adjusted n = 71
Upright time (min/d), adjusted n = 61
2 vs 1
−6.273
−22.988 to 10.442
.461
−7.970
−35.444 to 19.504
.569
3 vs 1
−4.687
−21.402 to 12.028
.582
−5.931
−34.190 to 22.329
.680
4 vs 1
−8.547
−25.512 to 8.418
.322
−13.484
−43.118 to 16.151
.371
5 vs 1
−16.747
−34.265 to 0.771
.061
15.905
−14.311 to 46.121
.301
6 vs 1
−13.962
−31.955 to 4.031
.128
9.021
−22.172 to 40.214
.570
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
Standing time (min/d), unadjusted n = 71
Standing time (min/d), adjusted n = 61
2 vs 1
−5.078
−21.456 to 11.299
.542
−12.289
−37.579 to 13.000
.340
3 vs 1
−4.208
−20.586 to 12.169
.614
−7.294
−33.307 to 18.719
.581
4 vs 1
−8.292
−24.915 to 8.330
.327
−15.144
−42.422 to 12.135
.275
5 vs 1
−15.563
−32.727 to 1.601
.075
9.837
−17.977 to 37.651
.487
6 vs 1
−13.337
−30.966 to 4.292
.138
4.727
−23.985 to 33.440
.746
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
Day
Low Baseline Performance
High Baseline Performance
Stepping Time (min/d), Adjusted n = 71
Stepping Time (min/d), Adjusted n = 61
PJ Group vs Non-PJ Group Intervention vs control
PJ Group vs Non-PJ Group Intervention vs control
Coefficient
95% CI
P
Coefficient
95% CI
P
2 vs 1
−1.192
−3.227 to 0.843
.250
4.301
−1.068 to 9.669
.116
3 vs 1
−0.476
−2.512 to 1.559
.646
1.282
−4.238 to 6.802
.648
4 vs 1
−0.248
−2.313 to 1.818
.814
1.549
−4.238 to 7.337
.599
5 vs 1
−1.169
−3.302 to 0.965
.282
5.946
0.046 to 11.847
.048
6 vs 1
−0.630
−2.821 to 1.562
.572
4.175
−1.916 to 10.267
.178
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
9 vs 1
n/a
n/a
Day
Steps (number/d), adjusted n = 71
Steps (number/d), adjusted n = 61
Rate Ratio
95% CI
P
Rate Ratio
95% CI
P
2 vs 1
0.950
0.280 to 3.218
.934
1.099
0.567 to 2.129
.780
3 vs 1
0.958
0.282 to 3.252
.946
1.029
0.522 to 2.027
.935
4 vs 1
1.252
0.362 to 4.334
.722
1.009
0.496 to 2.052
.980
5 vs 1
0.702
0.195 to 2.534
.588
1.176
0.570 to 2.426
.659
6 vs 1
0.472
0.127 to 1.752
.261
1.141
0.541 to 2.408
.728
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
STS (number/d), adjusted n = 71
STS (number/d), adjusted n = 61
2 vs 1
0.881
0.443 to 1.753
.719
0.916
0.603 to 1.391
.679
3 vs 1
0.978
0.493 to 1.941
.950
0.877
0.572 to 1.345
.547
4 vs 1
1.004
0.499 to 2.021
.991
1.085
0.694 to 1.697
.719
5 vs 1
1.031
0.499 to 2.129
.934
1.025
0.649 to 1.619
.916
6 vs 1
0.925
0.441 to 1.942
.837
0.996
0.622 to 1.595
.987
7 vs 1
n/a
n/a
8 vs 1
n/a
n/a
9 vs 1
n/a
n/a
STS, sit-to-stand transitions.
Adjusted for age, sex, comorbidity, ambulation status, and weekend day. Physical performance assessed by Short Physical Performance Battery. n/a: Comparison between non-PJ and PJ group is not applicable as all patients wearing the ActivPAL for 7 days or more were part of the PJ group.
How effective are interventions to increase physical activity levels among older inpatients receiving rehabilitation, without increasing the amount of therapy? A systematic review.
Reliability and validity of the international physical activity questionnaire compared to calibrated accelerometer cut-off points in the quantification of sedentary behaviour and physical activity in older adults.
A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission.
Recommendations for older adults' physical activity and sedentary behaviour during hospitalisation for an acute medical illness: an international Delphi study.
Geriatric Rehabilitation Inpatients Roam at Home! A Matched Cohort Study of Objectively Measured Physical Activity and Sedentary Behavior in Home-Based and Hospital-Based Settings.
The association of changes in physical performance during geriatric rehabilitation and short-term hospital readmission, institutionalisatin and mortality: RESORT.
This research was funded by an unrestricted grant of the University of Melbourne received by Prof. Andrea B. Maier and the Medical Research Future Fund (MRFF) provided by the Melbourne Academic Center for Health (MACH). This work was supported by the European Union's Horizon 2020 research and innovation program PreventIT (grant agreement number 689238). The funders had no role in the design, execution, analysis and interpretation of data, or writing of the study.
This study was approved by the Melbourne Health Human Research Ethics Committee (HREC/17/MH/103) with all ethical guidelines followed in full accordance with the Declaration of Helsinki.
00050-6&id=gr1.jpg){kind=link}
00050-6&id=gr2.jpg){kind=link}