Older age and mechanical ventilation are associated with discharge to a care facility after hospitalization in an intensive care unit (ICU) (Feng et al., 2009; Gehlbach et al., 2011). Low mobility and inactivity among older adults throughout hospitalization have been previously described as an epidemic, and are also associated with adverse outcomes, including functional decline and discharge to a care facility (Brown et al., 2009). This profound inactivity has prompted expert recommendations on activity for older adults during hospitalization for acute medical illness (Baldwin et al., 2020); however, currently there has been little research or guidance on inactivity, specifically among hospitalized older adults recovering from acute critical illness, to facilitate substantial changes in care practices.
Older adults discharged from the ICU may undergo significant changes in their lives secondary to critical illness, including deficits in one or more domains of physical, psychological, or cognitive function, known as post-ICU syndrome (Elliott et al., 2014). The existing literature on activity among older adults during hospitalization has not examined the association between post-ICU activity levels and discharge outcomes at time of hospital discharge among older adults discharged from the ICU. One prior study has examined the association between physical activity levels and discharge outcomes among hospitalized older adults (Tasheva et al., 2020); however, the study did not exclusively focus on older adults who required ICU-level care and/or mechanical ventilation. Thus, the aim of the current study is to explore the relationships between early post-ICU activity, discharge disposition, and hospital length of stay (LOS), specifically among a cohort of hospitalized older adults discharged from the ICU. We hypothesized that lower levels of activity would negatively affect discharge disposition and hospital LOS.
Sample and Setting
In the current study, we enrolled 30 English-speaking hospitalized older adults (aged ≥65 years) who were functionally independent prior to hospitalization, mechanically ventilated while in the ICU, and recently transferred out of the ICU (within 24 to 48 hours post-ICU discharge). We recruited from all 12 medical-surgical units, regardless of ICU admission diagnoses, at a Level 1 trauma hospital. The university's Institutional Review Board and affiliated Level 1 trauma hospital approved our protocol prior to study initiation.
Inclusion and Exclusion Criteria
All participants must have been community-dwelling older adults (i.e., admitted from home) and functionally independent prior to hospital admission. The Katz Index of Activities of Daily Living (Katz Index) assessed participants' baseline functional ability (i.e., prior to their hospital admission) to independently perform the six activities of daily living: bathing, dressing, toileting, transferring, continence, and feeding (Katz et al., 1963). During recruitment, potential participants were asked to retrospectively comment on their baseline ability to perform each of these activities of daily living independently 2 weeks prior to hospital admission, based on a validated method (Covinsky et al., 2000). The Katz Index has demonstrated high reliability, with Cronbach's alpha coefficients ranging from 0.87 to 0.94 (Wallace & Shelkey, 2008). Potential participants who scored <6 on the Katz Index were deemed ineligible; a score of 6 of 6 indicates functional independence without any supervision or assistance.
Additional exclusion criteria were pre-existing diagnosis of dementia, imminent death, active palliative care or hospice orders, and/or spinal cord injury. Individuals admitted from a long-term care facility, assisted living facility, or skilled nursing/rehabilitation facility, or those who received private/home health care at home prior to admission, were excluded from this study.
Post-ICU Activity. Wrist actigraphy (Actiwatch Spectrum, Philips Respironics) collected data on early post-ICU daytime activity and nighttime activity consecutively throughout a 2-night/1-day observation period, beginning at the time of study enrollment. An actigraph is a lightweight monitoring device, designed for long-term monitoring of gross motor activity and can be worn on the individual's dominant limb. Actigraphy has been used to measure activity patterns in hospitalized older inpatients (Lim et al., 2018) and adult ICU patients (Schwab et al., 2019) based on accelerometry algorithms using activity counts. Activity counts were generated for each 15-second epoch: if an activity count fell below the threshold designated as “wake,” then the epoch was automatically scored as rest or sleep. We defined “post-ICU daytime activity” as mean activity counts/min between 6:00 a.m. and 9:59 p.m. over one daytime observation period. We defined “post-ICU nighttime activity” as the mean activity counts/min over two consecutive nighttime periods between 10:00 p.m. and 5:59 a.m. We chose a continuous actigraphy observation period of two consecutive nighttime periods and one daytime period, because most higher-functioning older adults discharged from the ICU were discharged from the hospital within 2 to 3 days of ICU discharge. Daytime and nighttime hours were chosen based on established unit routines at the hospital site: for example, on most units, vital signs were typically scheduled for 6:00 a.m., 10:00 a.m., 2:00 p.m., 6:00 p.m., 10:00 p.m., and 2:00 a.m.
Discharge Outcomes. Discharge to home (i.e., home with minimal assistance or home with home health care) versus discharge to facility (i.e., inpatient rehabilitation facility, skilled nursing facility, or long-term acute care hospital) was collected via prospective chart review. Total hospital LOS (days) was also calculated, via prospective chart review, at time of discharge.
Covariates. Length of mechanical ventilation in days while in the ICU was collected via retrospective chart review. Severity of illness on ICU admission was measured by the Acute Physiology, Age, and Chronic Health Evaluation (APACHE III) (Knaus et al., 1991) via retrospective chart review. We collected three additional variables at the time of enrollment: cognitive flexibility, grip strength, and dexterity. We assessed cognitive flexibility using the National Institutes of Health (NIH) Toolbox Cognition Battery Dimensional Change Card Sort Test (DCCST) (Weintraub et al., 2013), grip strength using the NIH Toolbox Motor Battery Grip Strength Test (GST) (Reuben et al., 2013), and dexterity using the NIH Toolbox Motor Battery 9-Hole Pegboard Dexterity Test (PDT) (Reuben et al., 2013). The NIH Toolbox assessments are easily administered on an iPad®, via the NIH Toolbox iPad application. The DCCST, a measure of executive function, asks the individual to match or sort sets of two pictures that differ in color to target pictures. The GST is a measure of grip strength using handgrip dynamometry. The PDT, a measure of dexterity and fine motor coordination, instructs individuals to place and remove all nine pegs in the given pegboard, one at a time, as quickly as possible. We chose these three assessments of cognition and motor function from the NIH Toolbox because cognitive and physical impairment are described in the literature as feature components of post-ICU syndrome (Elliott et al., 2014), and are strongly associated with discharge to a care facility following critical illness (Gehlbach et al., 2011). Fully corrected T scores (adjusted for age, sex, race/ethnicity, and level of education) on these NIH Toolbox measures were used for analyses.
Data were analyzed using IBM SPSS Statistics version 26. Descriptive statistics were reported as mean and standard deviation. Median values and interquartile ranges (IQR) were calculated for post-ICU activity and hospital LOS. Independent samples t tests examined differences in post-ICU daytime activity between individuals discharged home (home, or home with home health care) versus those discharged to a facility (in-patient rehabilitation facility, skilled nursing facility, or long-term acute care hospital). Pearson correlation coefficients examined bivariate correlations; covariates with correlations <0.2 were considered for inclusion in exploratory regression analyses. Exploratory regression models examined relationships between predictor variables (post-ICU daytime activity and nighttime activity) and the outcome variable (hospital LOS), adjusting for five covariates (length of mechanical ventilation, APACHE III, DCCST, GST, and PDT).
Table 1 summarizes the demographic and clinical characteristics of the study sample. During study recruitment, 32 possible participants were approached for consent after ICU discharge: one was not interested in study participation, and the other did not agree to wrist actigraphy. Among the 30 enrolled participants, post-ICU daytime activity averaged 41.21 activity counts/min (SD = 28.24, range = 5.8 to 91.2 [median = 31.25; IQR = 48.37]), and nighttime activity averaged 29.27 activity counts/min (SD = 28.98, range = 4 to 137.52 [median = 17.76; IQR = 36.35]). Three participants were transferred back to the ICU after the study period; all three were eventually discharged to skilled nursing facilities. Ten (33.3%) participants were discharged home, whereas five (16.7%) were discharged to inpatient rehabilitation facilities, 12 (40%) to skilled nursing facilities, and one (3.3%) to a long-term acute care hospital. Two (6.7%) participants were discharged to out-of-network hospitals for continuity of care. The mean hospital LOS was 24.32 days (SD = 17.1, range = 5 to 71 [median = 22.5; IQR = 21.5]).
Demographic and Clinical Characteristics (N = 30)
Post-ICU Activity and Discharge Disposition
Among 28 participants with available discharge data, there were significant differences in post-ICU daytime activity by discharge disposition (t = 2.054, p = 0.050). Post-ICU daytime activity was greater among those discharged home (home with minimal assistance or with home health care: 54.42 activity counts/min [SD = 29.3, range = 5.8 to 89.5]) than those discharged to a facility (inpatient rehabilitation facility, skilled nursing facility, or long-term acute care hospital: 33.26 activity counts/min [SD = 24.26, range = 7.7 to 91.2]). There were no significant differences in nighttime activity by discharge disposition. Figure 1 illustrates trends between post-ICU daytime activity and LOS by discharge disposition.
Trends between post-intensive care unit (ICU) daytime activity, discharge disposition, and hospital length of stay.
Note. Post-ICU daytime activity: mean activity counts/min between 6:00 a.m. and 9:59 p.m.; hospital length of stay: total length of inpatient hospitalization in days.
Post-ICU Activity and Length of Stay
A total of 25 participants had complete data on all variables in the regression analyses: one participant was excluded due to incomplete actigraphy observation, two participants were excluded due to inability to complete the NIH Toolbox measures, and two participants were excluded due to loss of follow up after transfer to out-of-network hospitals. The regression model exploring the relationship between post-ICU daytime activity and LOS was significant (R2 = 0.708, p < 0.001). Post-ICU daytime activity was negatively associated with hospital LOS (β = −0.322, p = 0.041) after adjusting for covariates. The unique variance for post-ICU daytime activity was 7.84%: lower post-ICU daytime activity was associated with longer LOS. Post-ICU nighttime activity was not significantly associated with hospital LOS (β = −0.279, p = 0.066). Table 2 summarizes the results of these exploratory regression analyses.
Associations Between Post-Intensive Care Unit (ICU) Activity and Length of Hospital Stay
Greater post-ICU daytime activity was associated with better hospital discharge outcomes, specifically discharge to home instead of to a care facility, among our study cohort of hospitalized older adults discharged from the ICU. Our study found that post-ICU daytime activity was much lower among those who were eventually discharged to a facility. Moreover, daytime activity among older adults discharged from the ICU during the early post-ICU period may be a prognostic indicator of hospital LOS. Lower post-ICU daytime activity was prospectively associated with longer LOS. Of note, we included only hospitalized older adults discharged from the ICU who, at baseline, were functionally independent prior to hospital admission, which strengthens the implications of our findings.
In our study, daytime activity (mean = 41 activity counts/min) was remarkably lower than that reported by other studies of hospitalized older adults. Beveridge et al. (2015) reported 77 activity counts/min, and Kessler et al. (2019) reported 129 activity counts/min. An important differentiation is that our study focused exclusively on older adults after ICU discharge, whereas these studies focused on hospitalized older adults who did not require intensive care. More than one half (53.3%) of older adults discharged from the ICU in our study were admitted to a surgical or trauma ICU and underwent a major procedure during ICU hospitalization. During our actigraphy observation period, eight (26.7%) participants received three or more doses of opioids and/or benzodiazepines for acute pain—administration of these medications may affect activity levels. In addition, in our study, approximately 57% of older adults discharged from the ICU were discharged to a facility, compared to only 8% of the sample in the study by Beveridge et al. (2015). This difference suggests that hospitalized older adults who survived an ICU stay and mechanical ventilation continue to display sedentary behavior and inactivity beyond the ICU setting and supports our finding that daytime activity may affect discharge disposition.
Results from our study support those found in similar studies of hospitalized older adults. Activity objectively measured by actigraphy or accelerometry has been linked to hospital-acquired disability (Pavon et al., 2020; Tasheva et al., 2020). Further, older adult patients who are ambulatory may tend to have a shorter hospital LOS than those who are non-ambulatory (Pedersen et al., 2012). These studies, however, did not exclusively focus on a cohort of hospitalized older adults discharged from the ICU and their unique discharge or LOS outcomes. When compared to general hospitalized older inpatients, older adults discharged from the ICU tend to have inpatient stays complicated by higher severity of illness, and therefore are more likely to require longer courses of treatment and higher acuity of discharge disposition. Thus, it may be prudent to aggressively promote daytime activity and minimize daytime sedentary behavior and/or daytime sleepiness among older adults discharged from the ICU throughout post-ICU hospitalization.
We propose that perhaps promotion of daytime activity could mitigate worse discharge outcomes (i.e., cost and patient/caregiver burden associated with higher acuity of discharge disposition) associated with critical illness hospitalization. It is possible that promotion of daytime activity impacts outcomes in part by improving nighttime sleep. One prior study found that greater sleep duration, measured by actigraphy-observed rest/wake patterns, was associated with longer hospital LOS among older adults discharged from the ICU—suggesting that prolonged low-quality/fragmented sleep and inactivity may lead to worse outcomes (Elías et al., 2020). Clinical interventions to avoid daytime naps, discourage sedentary behavior, and promote nighttime sleep consolidation may be paired with interventions to increase daytime activity and mobility as often and as safely as possible.
Up to 90% of older adults discharged from the ICU are discharged with at least one geriatric syndrome, which increases risk of new institutionalization in a care facility (Tang et al., 2017). Furthermore, older adults are at particularly high risk for developing the long-term motor, cognitive, and/or psychological sequelae associated with post-ICU syndrome (Elliott et al., 2014). Physical activity and mobility throughout recovery from critical illness beyond the ICU is strongly recommended.
Limitations of the current study include the short actigraphy observation period and small sample size; these limitations are inherent in an exploratory study design. The small sample size required comparative analyses of post-ICU activity by discharge disposition via independent samples t tests, instead of regression analyses with adjustment for covariates. Interpretations should be made with caution. The current study design does not imply causality in the relationship between activity and discharge outcomes. Future research should include greater sample sizes and longer periods of actigraphy observation or longitudinal follow up to observe activity characteristics and trends over the course of recovery from critical illness.
Hospitalized older adults with critical illness discharged from the ICU are profoundly inactive despite transition of care out of the ICU. Higher post-ICU daytime activity could identify older adults who are likely to be discharged home instead of to a care facility. Similarly, lower post-ICU daytime activity may be a potential predictor of longer hospital LOS; hence, we suggest that daytime inactivity may be a modifiable risk factor for prolonged LOS. Increasing daytime activity may decrease health care costs associated with poor outcomes for older adults discharged from the ICU. Future research could explore the effects of multimodal interventions that promote daytime activity, decrease daytime naps and prolonged rest/immobility periods, and optimize nighttime sleep consolidation. Ultimately, these findings may reflect that daytime activity could be a target for future intervention studies to promote discharge outcomes for hospitalized older adults discharged from the ICU.
- Baldwin, C. E., Phillips, A. C., Edney, S. M. & Lewis, L. K. (2020). Recommendations for older adults' physical activity and sedentary behaviour during hospitalisation for an acute medical illness: An international Delphi study. The International Journal of Behavioral Nutrition and Physical Activity, 17(1), 69–69 doi:10.1186/s12966-020-00970-3 [CrossRef] PMID:32450879
- Beveridge, C., Knutson, K., Spampinato, L., Flores, A., Meltzer, D. O., Van Cauter, E. & Arora, V. M. (2015). Daytime physical activity and sleep in hospitalized older adults: Association with demographic characteristics and disease severity. Journal of the American Geriatrics Society, 63(7), 1391–1400 doi:10.1111/jgs.13520 [CrossRef] PMID:26131982
- Brown, C. J., Redden, D. T., Flood, K. L. & Allman, R. M. (2009). The underrecognized epidemic of low mobility during hospitalization of older adults. Journal of the American Geriatrics Society, 57(9), 1660–1665 doi:10.1111/j.1532-5415.2009.02393.x [CrossRef] PMID:19682121
- Covinsky, K. E., Palmer, R. M., Counsell, S. R., Pine, Z. M., Walter, L. C. & Chren, M. M. (2000). Functional status before hospitalization in acutely ill older adults: Validity and clinical importance of retrospective reports. Journal of the American Geriatrics Society, 48(2), 164–169 doi:10.1111/j.1532-5415.2000.tb03907.x [CrossRef] PMID:10682945
- Elías, M. N., Munro, C. L., Liang, Z., Padilla Fortunatti, C. F., Calero, K. & Ji, M. (2020). Nighttime sleep duration is associated with length of stay outcomes among older adult survivors of critical illness. Dimensions of Critical Care Nursing, 39(3), 145–154 doi:10.1097/DCC.0000000000000411 [CrossRef] PMID:32251163
- Elliott, D., Davidson, J. E., Harvey, M. A., Bemis-Dougherty, A., Hopkins, R. O., Iwashyna, T. J., Wagner, J., Weinert, C., Wunsch, H., Bienvenu, O. J., Black, G., Brady, S., Brodsky, M. B., Deutschman, C., Doepp, D., Flatley, C., Fosnight, S., Gittler, M., Gomez, B. T. & Needham, D. M. (2014). Exploring the scope of post-intensive care syndrome therapy and care: Engagement of non-critical care providers and survivors in a second stakeholders meeting. Critical Care Medicine, 42(12), 2518–2526 doi:10.1097/CCM.0000000000000525 [CrossRef] PMID:25083984
- Feng, Y., Amoateng-Adjepong, Y., Kaufman, D., Gheorghe, C. & Manthous, C. A. (2009). Age, duration of mechanical ventilation, and outcomes of patients who are critically ill. Chest, 136(3), 759–764 doi:10.1378/chest.09-0515 [CrossRef] PMID:19736189
- Gehlbach, B. K., Salamanca, V. R., Levitt, J. E., Sachs, G. A., Sweeney, M. K., Pohlman, A. S., Charbeneau, J., Krishnan, J. A. & Hall, J. B. (2011). Patient-related factors associated with hospital discharge to a care facility after critical illness. American Journal of Critical Care, 20(5), 378–386 doi:10.4037/ajcc2011827 [CrossRef] PMID:21885459
- Katz, S., Ford, A. B., Moskowitz, R. W., Jackson, B. A. & Jaffe, M. W. (1963). Studies of illness in the aged. The index of ADL: A standardized measure of biological and psychosocial function. Journal of the American Medical Association, 185, 914–919 doi:10.1001/jama.1963.03060120024016 [CrossRef] PMID:14044222
- Kessler, R., Knutson, K. L., Mokhlesi, B., Anderson, S. L., Shah, M., Meltzer, D. O. & Arora, V. M. (2019). Sleep and activity patterns in older patients discharged from the hospital. Sleep (Basel), 42(11), zsz153. doi:10.1093/sleep/zsz153 [CrossRef] PMID:31310317
- Knaus, W. A., Wagner, D. P., Draper, E. A., Zimmerman, J. E., Bergner, M., Bastos, P. G., Sirio, C. A., Murphy, D. J., Lotring, T., Damiano, A. & Harrell, F. E. Jr. . (1991). The APACHE III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults. Chest, 100(6), 1619–1636 doi:10.1378/chest.100.6.1619 [CrossRef] PMID:1959406
- Lim, S. E. R., Ibrahim, K., Sayer, A. A. & Roberts, H. C. (2018). Assessment of physical activity of hospitalised older adults: A systematic review. The Journal of Nutrition, Health & Aging, 22(3), 377–386 doi:10.1007/s12603-017-0931-2 [CrossRef] PMID:29484351
- Pavon, J. M., Sloane, R. J., Pieper, C. F., Colón-Emeric, C. S., Cohen, H. J., Gallagher, D., Hall, K. S., Morey, M. C., McCarty, M. & Hastings, S. N. (2020). Accelerometer-measured hospital physical activity and hospital-acquired disability in older adults. Journal of the American Geriatrics Society, 68(2), 261–265 doi:10.1111/jgs.16231 [CrossRef] PMID:31747050
- Pedersen, M. M., Bodilsen, A. C., Petersen, J., Beyer, N., Andersen, O., Lawson-Smith, L., Kehlet, H. & Bandholm, T. (2012). Twenty-four-hour mobility during acute hospitalization in older medical patients. The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences, 68(3), 331–337 doi:10.1093/gerona/gls165 [CrossRef]
- Reuben, D. B., Magasi, S., McCreath, H. E., Bohannon, R. W., Wang, Y.-C., Bubela, D. J., Rymer, W. Z., Beaumont, J., Rine, R. M., Lai, J. S. & Gershon, R. C. (2013). Motor assessment using the NIH Toolbox. Neurology, 80(11, Suppl. 3), S65–S75 doi:10.1212/WNL.0b013e3182872e01 [CrossRef] PMID:23479547
- Schwab, K. E., To, A. Q., Chang, J., Ronish, B., Needham, D. M., Martin, J. L. & Kamdar, B. B. (2019). Actigraphy to measure physical activity in the intensive care unit: A systematic review. Journal of Intensive Care Medicine, 35(1), 1323–1331 doi:10.1177/0885066619863654 [CrossRef] PMID:31331220
- Tang, H. J., Tang, H. J., Hu, F. W. & Chen, C. H. (2017). Changes of geriatric syndromes in older adults survived from intensive care unit. Geriatric Nursing, 38(3), 219–224 doi:10.1016/j.gerinurse.2016.10.011 [CrossRef]
- Tasheva, P., Vollenweider, P., Kraege, V., Roulet, G., Lamy, O., Marques-Vidal, P. & Méan, M. (2020). Association between physical activity levels in the hospital setting and hospital-acquired functional decline in elderly patients. JAMA Network Open, 3(1), e1920185 doi:10.1001/jamanetworkopen.2019.20185 [CrossRef] PMID:32003817
- Wallace, M. & Shelkey, M. (2008). Monitoring functional status in hospitalized older adults. The American Journal of Nursing, 108(4), 64–71 doi:10.1097/01.NAJ.0000314811.46029.3d [CrossRef] PMID:18367933
- Weintraub, S., Dikmen, S. S., Heaton, R. K., Tulsky, D. S., Zelazo, P. D., Bauer, P. J., Carlozzi, N. E., Slotkin, J., Blitz, D., Wallner-Allen, K., Fox, N. A., Beaumont, J. L., Mungas, D., Nowinski, C. J., Richler, J., Deocampo, J. A., Anderson, J. E., Manly, J. J., Borosh, B. & Gershon, R. C. (2013). Cognition assessment using the NIH Toolbox. Neurology, 80(11, Suppl. 3), S54–S64 doi:10.1212/WNL.0b013e3182872ded [CrossRef] PMID:23479546
Demographic and Clinical Characteristics (N = 30)
|Characteristic||Mean (SD) (Range)|
|Age (years)||71.37 (5.35) (65 to 86)|
|Length of ICU stay (days)||11.8 (11.49) (2 to 51)|
|Length of mechanical ventilation (days)||5.18 (7.4) (0.5 to 36)|
|Total length of hospital stay (days)||24.32 (17.1) (5 to 71)|
|APACHE III scorea||95.93 (32.01) (46 to 157)|
|NIH Toolbox Assessmentb (T-score)|
| DCCST||38.81 (9.2) (24 to 64)|
| PDT||30.04 (8.94) (19 to 51)|
| GST||32.86 (14.19) (4 to 59)|
| Male||19 (63.3)|
| Female||11 (36.7)|
| White||26 (86.7)|
| Black or African American||4 (13.3)|
| Hispanic or Latino||3 (10)|
| Skilled nursing facility||12 (40)|
| Home with home health care||9 (30)|
| Inpatient rehabilitation facility||5 (16.7)|
| Transferred to out-of-network hospital/lost to follow up||2 (6.7)|
| Home with minimal assistance||1 (3.3)|
| Long-term acute care hospital||1 (3.3)|
Associations Between Post-Intensive Care Unit (ICU) Activity and Length of Hospital Stay
|Predictor Variable||Outcome Variable||Modela|
|R2||F||p Value||β||95% CI||p Value|
|Post-ICU daytime activityb||Length of hospital stayd||0.708||F(6, 18) = 7.286||0.001||−0.322||[−0.391, −0.009]||0.041|
|Post-ICU nighttime activityc||Length of hospital stayd||0.689||F(6, 17) = 6.267||0.001||−0.279||[−0.355, 0.013]||0.066|