Physical Activity of ICU Survivors during Acute Admission: Agreement of the activPAL with Observation

Validity of the activPAL and ActiGraph for measuring sitting time and steps in hospitalised orthopaedic patients with altered weight bearing

PURPOSE

To determine the criterion validity of the activPAL and ActiGraph for measuring steps and sitting/sedentary time, compared to observation, in people hospitalised following orthopaedic lower limb injury who were weight bearing (WB) (i.e., walking) or non-weight bearing (NWB) (i.e., hopping).

MATERIALS AND METHODS

Participants wore an activPAL and ActiGraph on the hip/thigh/unaffected (UA)/affected ankle (AA) while completing bouts of walking and sitting. Lin's concordance correlation coefficient, Bland-Altman methods, and ratio of agreement were used to compare device-measured to observed (videoed) step count, sitting/sedentary time.

RESULTS

In 42 participants, the ActiGraph demonstrated excellent concordance with the observed step count when worn on the ankle (LCC 0.91-0.92) compared to the hip (LCC 0.56) in participants that were WB. The ActiGraph AA achieved the highest concordance (LCC 0.71) with observed steps in participants NWB. The activPAL had poor concordance with observed steps, particularly at slow gait speeds, in participants that were WB (LCC 0.38-0.46), however was less influenced by gait speed and had good concordance in NWB participants (LCC 0.52-0.69). The activPAL (LCC 0.79-0.88) and ActiGraph UA (LCC 0.94) showed excellent concordance with observed sitting and sedentary time, respectively.

CONCLUSIONS

The ActiGraph worn at the ankle provided the most valid measure of steps in people who are WB and NWB following orthopaedic injury, while the activPAL was best for measuring sitting time.Implications for rehabilitationTo accurately measure both steps and sitting time in people with lower limb orthopaedic injuries, a combination of activity monitors should be used (i.e., ActiGraph for steps, activPAL for sitting time).The ActiGraph device when worn on the ankle demonstrated the strongest agreement with observed step count in people who were weight bearing and non-weight bearing.Caution is needed when using thigh- or hip-worn devices in people who walk slowly.

Validation of the Activ8 Activity Monitor for Monitoring Postures, Motions, Transfers, and Steps of Hospitalized Patients

Sedentary behaviors and low physical activity among hospitalized patients have detrimental effects on health and recovery. Wearable activity monitors are a promising tool to promote mobilization and physical activity. However, existing devices have limitations in terms of their outcomes and validity. The Activ8 device was optimized for the hospital setting. This study assessed the concurrent validity of the modified Activ8. Hospital patients performed an activity protocol that included basic (e.g., walking) and functional activities (e.g., room activities), with video recordings serving as the criterion method. The assessed outcomes were time spent walking, standing, upright, sedentary, and newly added elements of steps and transfers. Absolute and relative time differences were calculated, and Wilcoxon and Bland-Altman analyses were conducted. Overall, the observed relative time differences were lower than 2.9% for the basic protocol and 9.6% for the functional protocol. Statistically significant differences were detected in specific categories, including basic standing (p < 0.05), upright time (p < 0.01), and sedentary time (p < 0.01), but they did not exceed the predetermined 10% acceptable threshold. The modified Activ8 device is a valid tool for assessing body postures, motions, steps, and transfer counts in hospitalized patients. This study highlights the potential of wearable activity monitors to accurately monitor and promote PA among hospital patients.

Comparison of postoperative mobilization measurements by activPAL versus Johns Hopkins Highest Level of Mobility scale after major abdominal surgery

BACKGROUND

The Johns Hopkins Highest Level of Mobility scale is a validated tool for assessing patient mobility in the hospital. It has excellent inter-rater and test-retest reliabilities, but it is unknown how accurately Johns Hopkins Highest Level of Mobility documentation reflects the patients' mobility performance in the immediate postoperative period compared to objective measures such as accelerometers.

METHODS

In this single-center observational study, consented adults undergoing open abdominal surgery wore a research-grade accelerometer, activPAL, starting immediately postoperatively until hospital discharge or up to 7 days. We collected the Johns Hopkins Highest Level of Mobility scores documented by hospital staff via retrospective chart review and evaluated their accuracy in describing the type, frequency, and volume of postoperative out-of-bed mobilization using the activPAL as the criterion.

RESULTS

We analyzed data from 56 participants. The activPAL showed that participants spent 97.7% of their time lying in bed or sitting in a chair. Meanwhile, the Johns Hopkins Highest Level of Mobility documentation of preambulatory activities (scores 1-5) was rare. The activPAL detected 4 times more out-of-bed mobilization than routine Johns Hopkins Highest Level of Mobility documentation. Whereas the frequency of activPAL-measured out-of-bed mobilization increased steadily to a median of 9 sessions by postoperative day 6, the number of Johns Hopkins Highest Level of Mobility documentation remained around twice daily. ActivPAL measurements demonstrated that Johns Hopkins Highest Level of Mobility documentation of ambulatory sessions (scores 6-8) was accurate.

CONCLUSIONS

We found that routine Johns Hopkins Highest Level of Mobility documentation did not accurately detect preambulatory activities or the overall frequency of out-of-bed mobility sessions, poorly reflecting the highly sedentary behaviors of the acute postoperative inpatients and highlighting the need to improve clinical documentation or use alternative methods to track postoperative mobilization.

Daily variation in physical activity during mechanical ventilation and stay in the intensive care unit

BACKGROUND

Early mobilisation of mechanically ventilated patients during their stay at an intensive care unit (ICU) can improve physical recovery. Yet, an objective and specified description of physical activities while in the ICU is lacking. Therefore, our aim was to describe the objectively assessed type, quantity, and daily variation of physical activity among mechanically ventilated patients while in the ICU.

METHOD

In an observational study in two mixed medical/surgical ICUs, we measured body posture in 39 patients on mechanical ventilation using a thigh- and chest-worn accelerometer while in the ICU. The accelerometer describes time spent lying, sitting, moving, in-bed cycling, standing and walking. Descriptive analysis of physical activity and daily variation was done using STATA.

RESULTS

We found that mechanically ventilated patients spend 20/24 h lying in bed, 3 h sitting and only 1 h standing, moving, walking or bicycling while in the ICU. Intervals of non-lying time appeared from 9.00 to 12.00 and again from 18.00 to 21.30, with peaks at the hours of 9.00 and 18.00.

CONCLUSION

ICU patients on mechanical ventilation were primarily sedentary. Physical activity of mechanically ventilated patients seems to be related to nurse- and/or physiotherapy-initiated activities. There is a need to create an awareness of improving clinical routines, towards active mobilisation throughout the day, for this vulnerable patient population during their stay in the ICU.

Validity of the activPAL monitor to measure stepping activity and activity intensity: A systematic review

BACKGROUND

Accumulating step counts and engaging in moderate-to-vigorous intensity physical activity is positively associated with numerous health benefits. The activPAL is a thigh-worn monitor that is frequently used to measure physical activity.

RESEARCH QUESTION

Can the activPAL accurately measure stepping activity and identify physical activity intensity?

METHODS

We systematically reviewed validation studies examining the accuracy of activPAL physical activity outcomes relative to a criterion measure in adults (>18 years). Citations were not restricted to language or date of publication. Sources were searched up to May 16, 2021 and included Scopus, EMBASE, MEDLINE, CINAHL, and Academic Search Premier. The study was pre-registered in Prospero (ID# CRD42021248240). Study quality was determined using a modified Hagströmer Bowles checklist.

RESULTS

Thirty-nine studies (20 laboratory arms, 17 semi-structured arms, 11 uncontrolled protocol arms; 1272 total participants) met the inclusion criteria. Most studies demonstrated a high validity of the activPAL to measure steps across laboratory (12/15 arms), semi-structured (10/13 arms) and uncontrolled conditions (5/7 arms). Studies that demonstrated low validity were generally conducted in unhealthy populations, included slower walking speeds, and/or short walking distances. Few studies indicated that the activPAL accurately measured physical activity intensity across laboratory (0/6 arms), semi-structured (0/5 arms) and uncontrolled conditions (2/5 arms). Using the default settings, the activPAL overestimates light-intensity activity but underestimates moderate-to-vigorous intensity activity. The overall study quality was 11.5 ± 2.0 out of 19.

CONCLUSION

Despite heterogeneous methodological and statistical approaches, the included studies generally provide supporting evidence that the activPAL can accurately detect stepping activity but not physical activity intensity. Strategies that use alternative data processing methods have been developed to better characterize physical activity intensity, but all methods still underestimate vigorous-intensity activity.

Validity of the ActivPAL monitor to distinguish postures: A systematic review

BACKGROUND

Posture has been recently integrated into activity guidelines, advising people to limit their sedentary time and break up sedentary postures with standing/stepping as much as possible. The thigh-worn activPAL is a frequently used objective measure of posture, but its validity has only been investigated by individual studies and has not been systematically reviewed.

RESEARCH QUESTION

Can the activPAL accurately characterize different postures?

METHODS

A rigorous systematic review protocol was conducted, including multiple study screeners and determiners of study quality. To be included, validation studies had to examine the accuracy of an activPAL posture outcome relative to a criterion measure (e.g., direct observation) in adults (>18 years). Citations were not restricted to language or date of publication. Sources were searched on May 16, 2021 and included Scopus, EMBASE, MEDLINE, CINAHL, and Academic Search Premier. The study was pre-registered in Prospero (ID# CRD42021248240). Study quality was determined using a modified Hagströmer Bowles checklist. The results are presented narratively.

RESULTS

Twenty-four studies (18 semi-structured laboratory arms, 8 uncontrolled protocol arms; 476 participants) met the inclusion criteria. Some studies (5/24) incorporated dual-monitor (trunk: 4/5; shin: 1/5) configurations. While heterogenous statistical procedures were implemented, most studies (n = 22/24) demonstrated a high validity (e.g., percent agreement >90%, no fixed bias, etc.) of the activPAL to measure sedentary and/or upright postures across semi-structured (17/18 arms) and uncontrolled study designs (7/8 arms). Specific experimental protocol factors (i.e., seat height, fidgeting, non-direct observation criterion comparator) likely explain the divergent reports that observed valid versus invalid findings. The study quality was 11.3 (standard deviation: 2.3) out of 19.

CONCLUSION

Despite heterogeneous methodological and statistical approaches, the included studies generally provide supporting evidence that the activPAL can accurately distinguish between sedentary and standing postures. Multiple activPAL monitor configurations (e.g., thigh and torso) are needed to better characterize sitting versus lying postures.

Fine Detection of Human Motion During Activities of Daily Living as a Clinical Indicator for the Detection and Early Treatment of Chronic Diseases: The E-Mob Project

Methods to measure physical activity and sedentary behaviors typically quantify the amount of time devoted to these activities. Among patients with chronic diseases, these methods can provide interesting behavioral information, but generally do not capture detailed body motion and fine movement behaviors. Fine detection of motion may provide additional information about functional decline that is of clinical interest in chronic diseases. This perspective paper highlights the need for more developed and sophisticated tools to better identify and track the decomposition, structuration, and sequencing of the daily movements of humans. The primary goal is to provide a reliable and useful clinical diagnostic and predictive indicator of the stage and evolution of chronic diseases, in order to prevent related comorbidities and complications among patients.

Optimization and Validation of a Classification Algorithm for Assessment of Physical Activity in Hospitalized Patients

Low amounts of physical activity (PA) and prolonged periods of sedentary activity are common in hospitalized patients. Objective PA monitoring is needed to prevent the negative effects of inactivity, but a suitable algorithm is lacking. The aim of this study is to optimize and validate a classification algorithm that discriminates between sedentary, standing, and dynamic activities, and records postural transitions in hospitalized patients under free-living conditions. Optimization and validation in comparison to video analysis were performed in orthopedic and acutely hospitalized elderly patients with an accelerometer worn on the upper leg. Data segmentation window size (WS), amount of PA threshold (PA Th) and sensor orientation threshold (SO Th) were optimized in 25 patients, validation was performed in another 25. Sensitivity, specificity, accuracy, and (absolute) percentage error were used to assess the algorithm’s performance. Optimization resulted in the best performance with parameter settings: WS 4 s, PA Th 4.3 counts per second, SO Th 0.8 g. Validation showed that all activities were classified within acceptable limits (>80% sensitivity, specificity and accuracy, ±10% error), except for the classification of standing activity. As patients need to increase their PA and interrupt sedentary behavior, the algorithm is suitable for classifying PA in hospitalized patients.

Using Consumer-Grade Physical Activity Trackers to Measure Frailty Transitions in Older Critical Care Survivors: Exploratory Observational Study

Background

Critical illness has been suggested as a sentinel event for frailty development in at-risk older adults. Frail critical illness survivors are affected by increased adverse health outcomes, but monitoring the recovery after intensive care unit (ICU) discharge is challenging. Clinicians and funders of health care systems envision an increased role of wearable devices in monitoring clinically relevant measures, as sensor technology is advancing rapidly. The use of wearable devices has also generated great interest among older patients, and they are the fastest growing group of consumer-grade wearable device users. Recent research studies indicate that consumer-grade wearable devices offer the possibility of measuring frailty.

Objective

This study aims to examine the data collected from wearable devices for the progression of frailty among critical illness survivors.

Methods

An observational study was conducted with 12 older survivors of critical illness from Kingston General Hospital in Canada. Frailty was measured using the Clinical Frailty Scale (CFS) at ICU admission, hospital discharge, and 4-week follow-up. A wearable device was worn between hospital discharge and 4-week follow-up. The wearable device collected data on step count, physical activity, sleep, and heart rate (HR). Patient assessments were reviewed, including the severity of illness, cognition level, delirium, activities of daily living, and comorbidity.

Results

The CFS scores increased significantly following critical illness compared with the pre-ICU frailty level (P=.02; d=−0.53). Survivors who were frail over the 4-week follow-up period had significantly lower daily step counts than survivors who were not frail (P=.02; d=1.81). There was no difference in sleep and HR measures. Daily step count was strongly correlated with the CFS at 4-week follow-up (r=−0.72; P=.04). The average HR was strongly correlated with the CFS at hospital discharge (r=−0.72; P=.046). The HR SD was strongly correlated (r=0.78; P=.02) with the change in CFS from ICU admission to 4-week follow-up. No association was found between the CFS and sleep measures. The pattern of increasing step count over the 4-week follow-up period was correlated with worsening of frailty (r=.62; P=.03).

Conclusions

This study demonstrated an association between frailty and data generated from a consumer-grade wearable device. Daily step count and HR showed a strong association with the frailty progression of the survivors of critical illness over time. Understanding this association could unlock a new avenue for clinicians to monitor and identify a vulnerable subset of the older adult population that might benefit from an early intervention.

Quantifying Mobility in the ICU: Comparison of Electronic Health Record Documentation and Accelerometer-Based Sensors to Clinician-Annotated Video

Supplemental Digital Content is available in the text.

To compare the accuracy of electronic health record clinician documentation and accelerometer-based sensors with a gold standard dataset derived from clinician-annotated video to quantify early mobility activities in adult ICU patients.

Actigraphy to Measure Physical Activity in the Intensive Care Unit: A Systematic Review

OBJECTIVE

In the intensive care unit (ICU), prolonged inactivity is common, increasing patients' risk for adverse outcomes, including ICU-acquired weakness. Hence, interventions to minimize inactivity are gaining popularity, highlighting actigraphy, a measure of activity involving a wristwatch-like accelerometer, as a method to inform these efforts. Therefore, we performed a systematic review of studies that used actigraphy to measure patient activity in the ICU setting.

DATA SOURCES

We searched PubMed, EMBASE, CINAHL, Cochrane Library, and ProQuest from inception until December 2016.

STUDY SELECTION

Two reviewers independently screened studies for inclusion. A study was eligible for inclusion if it was published in a peer-reviewed journal and used actigraphy to measure activity in ≥5 ICU patients.

DATA EXTRACTION

Two reviewers independently performed data abstraction and risk of bias assessment. Abstracted actigraphy-based activity data included total activity time and activity counts.

RESULTS

Of 16 studies (607 ICU patients) identified, 14 (88%) were observational, 2 (12%) were randomized control trials, and 5 (31%) were published after 2009. Mean patient activity levels per 15 to 60 second epoch ranged from 25 to 37 daytime and 2 to 19 nighttime movements. Actigraphy was evaluated in the context of ICU and post-ICU outcomes in 11 (69%) and 5 (31%) studies, respectively, and demonstrated potential associations between actigraphy-based activity levels and delirium, sedation, pain, anxiety, time to extubation, and length of stay.

CONCLUSION

Actigraphy has demonstrated that patients are profoundly inactive in the ICU with actigraphy-based activity levels potentially associated with important measures, such as delirium, sedation, and length of stay. Larger and more rigorous studies are needed to further evaluate these associations and the overall utility of actigraphy in the ICU setting.

A study investigating the validity of an accelerometer in quantification of step count in adult hospital inpatients recovering from critical illness

OBJECTIVE

To determine the validity of the ActiGraph GT3X accelerometer in step count quantification when compared to observed step count in hospitalised adults recovering from critical illness.

SETTING

Large National Health Service (NHS) Hospitals Trust.

SUBJECTS

In total, 20 hospital ward-based adults (age: mean 62.3, SD 11.5) who had required greater than 48 hours of mechanical ventilation in the intensive care unit.

MAIN MEASURES

Participants walked self-selected distances and speeds as part of a semi-structured movement protocol not exceeding 3 hours. Two ActiGraph GT3X accelerometers were worn, one on the thigh and one on the ankle of the non-dominant leg. Accelerometer-recorded step counts were compared against observed step counts.

RESULTS

In total, 31 separate walking episodes were analysed. A mean (SD) of 45.87 (±19.72) steps was calculated for observed step count (range 15-90). Mean differences (95% limits of agreement) of -0.84 steps (-3.88 to 2.2) for the ankle placement and -17.7 steps (-40.63 to 5.25) for the thigh were calculated. Intraclass correlation coefficients (95% confidence intervals) of 0.99 (0.99 to 1.0) and 0.46 (-0.1 to 0.78) were determined for the ankle and thigh, respectively. Placement sites were well tolerated by 95% of participants.

CONCLUSION

An ankle-mounted ActiGraph GT3X accelerometer demonstrates validity in quantification of step count in hospitalised adults recovering from critical illness. A thigh placement was not considered valid.

Validation of a wireless patch sensor to monitor mobility tested in both an experimental and a hospital setup: A cross-sectional study

Purpose

To assess the concurrent validity of a wireless patch sensor to monitor time lying, sitting/standing, and walking in an experimental and a hospital setup.

Methods

Healthy adults participated in two testing sessions: an experimental and real-world hospital setup. Data on time lying, sitting/standing, and walking was collected with the HealthPatch and concurrent video recordings. Validity was assessed in three ways: 1. test for mean differences between HealthPatch data and reference values; 2. Intraclass Correlation Coefficient analysis (ICC 3.1 agreement); and 3. test for mean differences between posture detection accuracies.

Results

Thirty-one males were included. Significant mean differences were found between HealthPatch data and reference values for sitting/standing (mean 14.4 minutes, reference: 12.0 minutes, p<0.01) and walking (mean 6.4 minutes, reference: 9.0 minutes, p<0.01) in the experimental setup. Good correlations were found between the HealthPatch data and video data for lying (ICC: 0.824) and sitting/standing (ICC: 0.715) in the hospital setup. Posture detection accuracies of the HealthPatch were significantly higher for lying and sitting/standing in the experimental setup.

Conclusions

Overall, the results show a good validity of the HealthPatch to monitor lying and poor validity to monitor sitting/standing or walking. In addition, the validity outcomes were less favourable in the hospital setup.