Frail older people have increased vulnerability to poor resolution of homeostasis following a stressor event,1 such as an illness or fall necessitating an admission to hospital. In frail older adults, hospitalization is associated with longer length of stay (LOS),2,3 and sometimes it can lead to physical deconditioning and loss of functional ability.4,5
Pressures on hospital bed occupancy in the English National Health Service are increasing: over the period of 2006/2007 to 2012/2013, hospitals have reported increases in admissions from 12.6 million per year to 14.6 million per year, an increase of 16%.6 These pressures have in part been driven by an increase in the population of frail older people7 and have focused attention on enhanced service delivery models and potential methods by which geriatric physical therapists might contribute to effective cost savings, while retaining a patient-centered approach. Our previous work suggested that higher physical therapy frequency is associated with shorter LOS and greater functional recovery in hospitalized frail older adults.8 Few studies investigated the effect of early mobilization on LOS in similar populations,9,10 as well as the effect of physical therapy within the emergency department in reducing admissions.11 An important distinction exists between early physical therapy assessment and early mobilization. Assisting mobilization is often an intervention carried out by a physical therapist but by no means exclusively. Physical therapy assessment includes assessment of the patient's impairments, activity limitations, and social situation. The information ascertained from the assessment is used to devise a management plan to optimize physical functioning and facilitate discharge from hospital. Few studies have investigated early physical therapy assessment on LOS, functional outcomes, and care needs on discharge. Our aim was to study these associations using a retrospective observational design. We hypothesized that early assessment may reduce LOS through earlier optimization of physical functioning and reduce the need for care on discharge.
Setting and Participants
This was a retrospective observational study in a large tertiary university National Health Service hospital in the United Kingdom. We analyzed all admission episodes of people admitted to the Department of Medicine for the Elderly wards between May 2 and August 26, 2016. Patients who were discharged without being examined by a physical therapist were excluded.
Anonymous routinely collected clinical data were obtained from the hospital electronic medical records. Most data were collected from running an electronic report of variables of interest. Other data were manually searched in the patients' electronic medical records by a member of the physical therapy team and verified independently by a second member. All measures used in this service evaluation audit were routinely collected as part of normal clinical care.
The exposures that we investigated were early assessment by a physical therapist (ie, within 24 hours of admission to hospital) and late assessment (ie, after 24 hours). The definition of assessment did not take into account whether any intervention had been carried out. The definition of early and late was arbitrarily set, although the policy in our Department of Medicine for the Elderly is that every patient should be examined by a physical therapist on the day of admission. A physical therapist is employed at the weekend to assess patients admitted to the Department of Medicine for the Elderly wards; this includes those patients admitted on a Friday, Saturday, or Sunday.
The outcome variables were LOS (days), the Elderly Mobility Scale (EMS),12 walking speed over more than 6 m (m/s), and the need for a new formal care package on discharge (yes or no) or new institutionalization. Information on inpatient mortality was also collected.
The EMS is a 20-point ordinal scale for the assessment of function in frail older patients12,13 (worst: 0 points; best: 20 points). The scale includes the assessment of balance, mobility, and ability to change body positions (eg, from lying to sitting). The interrater reliability of the EMS has been reported as r = 0.88 (P < .001), and it has good convergent validity with the Barthel Index (r = 0.787, P < .001).13 The EMS (which includes walking speed over more than 6 m) is routinely measured by Department of Medicine for the Elderly physical therapists on initial assessment and on day of discharge from hospital.
A new formal care package on discharge is defined as new care provided by an external care agency as opposed to informal arrangements of support with family or friends. Patients are discharged home once they are deemed clinically fit for discharge by the multidisciplinary team and any social support required is in place. New institutionalization is defined as discharge to a care home when patients had been admitted from home.
Admission variables collected for descriptive purposes were age, gender, the existence of a formal care package on admission (yes or no), number of falls in past 12 months, whether the patient had daily contact with a family member or friend (yes/no, as reported by the patient or a next of kin), preadmission abode, the Clinical Frailty Scale (CFS),14 the Charlson Comorbidity Index (CCI) (non–age adjusted),15 specific comorbidities, the Emergency Department Modified Early Warning Score (ED-MEWS, highest recorded in the ED),16 C-reactive protein level on admission, and the 4-item version of the Abbreviated Mental Test (AMT4).17
The CFS has been routinely collected in our center since 2013, thanks to a local Commissioning for Quality and Innovation scheme (https://www.england.nhs.uk/nhs-standard-contract/cquin/) that mandated that all patients 75 years of age or older admitted to the hospital via the emergency pathway be screened for frailty using the CFS within 72 hours of admission. Our center uses the 9-point CFS (http://geriatricresearch.medicine.dal.ca/clinical_frailty_scale.htm). The reported interrater reliability of the CFS is high with an intraclass correlation coefficient 0.97 (P < .001), and it has high convergent validity with the Frailty Index (Pearson coefficient = 0.80, P < .01).14
The CCI is based on patients' diagnoses as coded by the World Health Organization's International Classification of Diseases, Tenth Revision. The CCI has been validated for use in acutely hospitalized older adults, with areas under the receiver operating characteristic curve to predict mortality of 0.66 at 3 months after admission, 0.70 at 1 year, and 0.73 at 5 years.18
The ED-MEWS scores are routinely collected by nursing staff in ED and are considered as a measure of acute illness severity.16 Our ED-MEWS and its scoring protocol are shown in Table 1. An ED-MEWS score of 4 or more has been shown to be an independent predictor of survival time (hazard ratio = 2.87, 95% confidence interval: 2.27-3.62, P < .001).19 C-reactive protein is a measure of acute inflammation and is a recognized clinical measure of illness severity.20,21
The 4-item version of the Abbreviated Mental Test (AMT4)17 is routinely collected in our center as part of a Dementia/Delirium Commissioning for Quality and Innovation, which aims at detecting cognitive impairment on admission to hospital. The AMT4 consists of 4 questions regarding the patient's age, date of birth, the place that the person is currently located, and the current year. The AMT4 score showed a statistically significant correlation with AMT score (Somers' d statistic 0.90, P < .001).17
Anonymized data were analyzed with IBM SPSS Statistics (version 22) software. Descriptive statistics were given as count (with percentage) or mean (with standard deviation). For continuous variables with a nonnormal distribution, we reported median values with interquartile ranges. Differences in the characteristics and outcomes of patients who received early versus late physical therapy were evaluated using unpaired Student t, χ2, or Mann-Whitney tests as appropriate. Missing values for each variable were reported and treated as missing in each analysis. The level of statistical significance was set at P < .05, and P < .1 was considered as statistical trend.
The association between the delay to physical therapy assessment and LOS was evaluated using a Cox proportional hazards regression model. Patients admitted from a residential or nursing home were excluded, and we included only those admitted from their own home. Cox proportional hazards regression can account for the censoring of some participants who do not experience the outcome within the study time frame. This type of regression is most commonly used to analyze survival data, where time to an event such as death or recurrence of disease is modeled. In this study the “event” was set as “Discharge to Usual Residence” within 21 days of hospital admission. Those who were not discharged to their usual residence within 21 days were censored. To differentiate those who died and to prevent informative censoring (ie, at death) those who died were given an imputed LOS value of 21.01 days (ie, just over the maximum follow-up time allowed) and were, therefore, censored only at the end of the study. The decision to choose 21 days as the cutoff was made because LOS had a very skewed distribution with a long tail at the right end. A preliminary analysis of our data showed that by 21 days, more than 80% of patients had been discharged from the hospital. Clinical experience tells us that the majority of patients not discharged by this point are considered “stranded,” that is, factors not related to the patient's physiological status keep them in the hospital. They can be delayed from going home for a number of non–patient-related reasons such as requiring care but none being available, or requiring institutionalization but there not being any places in their locality. By choosing the 21-day cutoff point, we aimed to focus on the impact of physical therapy on the early optimization of patients' physiological status and functional abilities. In this study, the hazard ratio represents the likelihood of being discharged back to usual residence within 21 days of admission.
This study was registered as a service evaluation audit with our center's Safety and Quality Support Department (Project Register Number 5205). Formal confirmation was received that approval from the ethics committee was not required.
There were 1022 hospital episodes over the study period. Of those, we excluded 19 who were discharged without being examined by a physical therapist. Of the remaining 1003, 584 (58.2%) were examined within 24 hours of admission (early assessment) and 419 (41.8%) after 24 hours of admission (late assessment).
Patient admission characteristics are reported in Tables 2 and 3. No significant differences were seen between the 2 groups, except that the following comorbidities were more frequent in the late assessment group: myocardial infarction, congestive heart failure, metastatic cancer, and depression.
Patient outcomes are reported in Table 4. The median (interquartile range) LOS of the early assessment group was 6.7 (3.1-13.7) versus 10.0 (4.2-20.1) days in the late assessment group (P < .001). The other significant difference between the 2 groups was the number of patients requiring a new package of care on discharge: 110 (20.3%) in the early assessment group versus 105 (27.0%) in the late assessment group (P = .016). There were no other statistically significant differences, although there was a trend observed with those in the early assessment group being apparently less likely to require new institutionalization on discharge: 4.1% versus 6.7%, P = .073.
The result of the Cox proportional hazards regression model studying the association between delay to physical therapy and discharge to usual residence (excluding those already living in a residential or nursing home prior to admission) is presented in Figure 1. The hazard ratio for early assessment compared with late assessment was 1.29 (95% confidence interval: 1.12-1.48, P < .001) and can be interpreted as a 29% increase in the probability of discharge to usual residence for those in the early assessment group compared with those in the late assessment group. Table 5 presents the results of the Cox regression, with different covariates added to the model. After controlling for age, sex, ED-MEWS, CCI, and the EMS score on admission, results were still significant: 1.34 (95% confidence interval: 1.16-1.55), P < .001.
This retrospective observational study examined the association between early physical therapy assessment and hospital and functional outcomes in acutely hospitalized older adults. In our busy National Health Service geriatric wards, the majority of eligible patients (58.2%) were examined by the physical therapist within 24 hours of admission. Early physical therapy assessment was associated with a shorter LOS, reduced need for care on discharge, a trend toward reduced new institutionalization, and equal amount of functional recovery by discharge. Causality cannot be inferred from this observational study, but results would suggest that it is worth investigating in prospective studies whether physical therapy intervention within the first 24 hours of admission is beneficial. Our findings are in keeping with previous work reporting an association between early mobilization and reduced LOS,9,10 and with previous evidence that early physical rehabilitation for acutely hospitalized older adults may lead to functional benefits and can be safely executed.22,23 Indeed, in other specialty areas such as stroke, it appears that mobilization within 24 hours of admission has become the “norm” in recent years.24 The reason why there was no difference in functional outcomes may be due to discharge criteria; for a patient to be deemed clinically fit for discharge, his or her physical function and amount of recovery is a factor taken into account by the geriatric multidisciplinary team. In the majority of cases, the multidisciplinary team looks for the patients to be close to their preadmission level of function. Given the similarities in baseline characteristics, it is probable that both groups had the same amount of recovery to be made, and the longer LOS seen in the late assessment group may be in part due to the slower functional recovery.
The reasons as to why some patients were not seen within 24 hours of admission are not clear from our design. Although Table 2 reports no significant differences between groups in key patient characteristics, an interesting finding is that specific comorbidities such as acute myocardial infarction, congestive heart failure, metastatic cancer, and depression seemed to be more prevalent in the late assessment group (Table 3). It is possible that in some cases, the lateness of the physical therapy assessment may have been due to a medical contraindication arising from acute cardiovascular instability. In other cases, the delay in seeing the therapist may have been due to patients being too unwell or psychologically averse to therapy (eg, depressed or withdrawn). A limitation is that our database did not contain the principal diagnosis for the admission, and this may have shed light into these subtle patient differences. Otherwise, patients in the 2 categories were treated by similar multidisciplinary teams and we have no reasons to believe that the care received by the 2 groups differed. However, we cannot exclude the effects of day-to-day variations in staffing and number of admissions.
The main limitation of our study is the lack of randomization or blinding. As a result, we cannot make any assertions regarding the causality of our findings, definitive statements of association, or the generalizability beyond our hospital. In addition, we only recorded measures of function on admission and discharge. Further measures at other time-points may have given us an indication of the rate of functional recovery. All we can infer regarding functional change is that both groups had a similar overall amount of recovery, and the earlier discharge in the early assessment group did not appear to represent risk-taking behavior by clinicians (ie, patients were not discharged earlier without having made a similar amount of functional recovery as those in the late assessment group). We have isolated one aspect of the “dose” of physical therapy input; the AVERT studies have highlighted the potential importance of studying other aspects of the dose of physical therapy input, such as frequency and duration.25 A limitation of routinely collected clinical data obtained from the hospital electronic medical records is the risk of bias characteristic of retrospective studies.
Patients admitted to Department of Medicine for the Elderly wards undergo inpatient comprehensive geriatric assessment. There is evidence that frail patients undergoing comprehensive geriatric assessment in the hospital are more likely to be alive and at home after hospital discharge.26 Our study suggests that physical therapy is likely a key part of comprehensive geriatric assessment, and earlier input may be associated with better hospital outcomes. The reasons for reduced hospital LOS are not clear and causality cannot be inferred from our findings. Frail older patients are particularly susceptible to functional loss during acute illness via direct inflammatory damage to the musculoskeletal and central nervous systems.27–29 Furthermore, lack of physical activity and bed rest seen in this population30,31 has been shown to result in rapid muscle atrophy.32 It may be that early physical therapy assessment encourages increased physical activity by direct intervention, education, and improving patient confidence with self-administered exercise and as a result reduces hospital deconditioning, leading to faster functional recovery. Interestingly, our data in Table 4 demonstrate reduced need for care on discharge and a trend for reduced numbers of patients in the early assessment group who required a new package of care on discharge. This may potentially represent reduced deconditioning in the early assessment group.
Our findings may not be generalizable beyond our hospital, but they make a worthwhile contribution to what the UK Medical Research Council defines as the “development phase” of the development and evaluation of a complex intervention,33 in this case the acute care of frail older patients. Prospective interventional studies are necessary to clarify the importance of early physical therapy input in the outcomes of hospitalized frail older people, including the prevention of hospital-related deconditioning.
We set out to investigate the association of early physical therapy input with LOS, functional outcomes, and care needs on discharge. We found that there was an association with reduced LOS, need for formal care on discharge from hospital, and a trend toward reduced new institutionalization on discharge from hospital. This may be due to preventing hospital deconditioning; however, further prospective research is needed to establish causality and, if appropriate, investigate causal mechanisms.
The authors thank all the members of the Department of Medicine for the Elderly teams in their hospital, without whom this initiative would have not been possible. Licensed access to the Trust's information systems is also gratefully acknowledged.
1. Clegg A, Young J, Iliffe S, Rikkert MO, Rockwood K. Frailty in elderly people. Lancet. 2013;381(9868):752–762.
2. Basic D, Shanley C. Frailty in an older inpatient population: using the Clinical Frailty Scale to predict patient outcomes. J Aging Health. 2015;27(4):670–685.
3. Wallis SJ, Wall J, Biram RW, Romero-Ortuno R. Association of the Clinical Frailty Scale with hospital outcomes. QJM. 2015;108(12):943–949.
4. Hoogerduijn JG, Buurman BM, Korevaar JC, Grobbee DE, de Rooij SE, Schuurmans MJ. The prediction of functional decline in older hospitalised patients. Age Ageing. 2012;41(3):381–387.
5. Hartley P, Adamson J, Cunningham C, Embleton G, Romero-Ortuno R. Clinical frailty and functional trajectories in hospitalized older adults
: a retrospective observational study. Geriatr Gerontol Int. 2016. [Epub ahead of print].
6. Smith P, McKeon A, Blunt I, Edwards N. NHS Hospitals Under Pressure: Trends in Acute Activity up to 2022. London: Nuffield Trust; 2014.
7. Soong J, Poots AJ, Scott S, et al Quantifying the prevalence of frailty in English hospitals. BMJ Open. 2015;5(10):e008456.
8. Hartley P, Adamson J, Cunningham C, Embleton G, Romero-Ortuno R. Higher physiotherapy frequency is associated with shorter length of stay and greater functional recovery in hospitalized frail older adults
: a retrospective observational study. J Frailty Aging. 2016;5(2):121–125.
9. Fisher SR, Kuo YF, Graham JE, Ottenbacher KJ, Ostir GV. Early ambulation and length of stay in older adults
hospitalized for acute illness. Arch Intern Med. 2010;170(21):1942–1943.
10. Mundy LM, Leet TL, Darst K, Schnitzler MA, Dunagan WC. Early mobilization of patients hospitalized with community-acquired pneumonia. Chest. 2003;124(3):883–889.
11. Jesudason C, Stiller K, McInnes M, Sullivan T. A physiotherapy service to an emergency extended care unit does not decrease admission rates to hospital: a randomised trial. Emerg Med J. 2012;29(8):664–669.
12. Smith R. Validation and reliability of the Elderly Mobility Scale. Physiotherapy. 1994;80(11):744–747.
13. Prosser L, Canby A. Further validation of the Elderly Mobility Scale for measurement of mobility of hospitalized elderly people. Clin Rehabil. 1997;11(4):338–343.
14. Rockwood K, Song X, MacKnight C, et al A global clinical measure of fitness and frailty in elderly people. CMAJ. 2005;173(5):489–495.
15. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–383.
16. Subbe CP, Kruger M, Rutherford P, Gemmel L. Validation of a modified Early Warning Score in medical admissions. QJM. 2001;94(10):521–526.
17. Swain DG, Nightingale PG. Evaluation of a shortened version of the Abbreviated Mental Test in a series of elderly patients. Clin Rehabil. 1997;11(3):243–248.
18. Frenkel WJ, Jongerius EJ, Mandjes-van Uitert MJ, van Munster BC, de Rooij SE. Validation of the Charlson Comorbidity Index in acutely hospitalized elderly adults: a prospective cohort study. J Am Geriatr Soc. 2014;62(2):342–346.
19. Romero-Ortuno R, Wallis S, Biram R, Keevil V. Clinical frailty adds to acute illness severity in predicting mortality in hospitalized older adults
: an observational study. Eur J Intern Med. 2016;35:24–34.
20. Clyne B, Olshaker JS. The C-reactive protein. J Emerg Med. 1999;17(6):1019–1025.
21. Povoa P, Almeida E, Moreira P, et al C-reactive protein as an indicator of sepsis. Intensive Care Med. 1998;24(10):1052–1056.
22. Kosse NM, Dutmer AL, Dasenbrock L, Bauer JM, Lamoth CJ. Effectiveness and feasibility of early physical rehabilitation programs for geriatric hospitalized patients: a systematic review. BMC Geriatr. 2013;13:107.
23. Nolan J, Thomas S. Targeted individual exercise programmes for older medical patients are feasible, and may change hospital and patient outcomes: a service improvement project. BMC Health Serv Res. 2008;8(1):250.
24. Bernhardt J, Langhorne P, Lindley RI, et al Avert Trial Collaboration group. Efficacy and safety of very early mobilisation within 24 h of stroke onset (AVERT): a randomised controlled trial. Lancet. 2015;386(9988):46–55.
25. Bernhardt J, Churilov L, Ellery F, et al Prespecified dose-response analysis for A Very Early Rehabilitation Trial (AVERT). Neurology. 2016;86(23):2138–2145.
26. Ellis G, Whitehead MA, O'Neill D, Langhorne P, Robinson D. Comprehensive geriatric assessment for older adults
admitted to hospital. Cochrane Database Syst Rev. 2011(7):CD006211.
27. Bartley JM, Pan SJ, Keilich SR, et al Aging augments the impact of influenza respiratory tract infection on mobility impairments, muscle-localized inflammation, and muscle atrophy. Aging (Albany NY). 2016;8(4):620–635.
28. Baylis D, Bartlett DB, Patel HP, Roberts HC. Understanding how we age: insights into inflammaging. Longev Healthspan. 2013;2(1):8.
29. Braun TP, Zhu X, Szumowski M, et al Central nervous system inflammation induces muscle atrophy via activation of the hypothalamic-pituitary-adrenal axis. J Exp Med. 2011;208(12):2449–2463.
30. Brown CJ, Friedkin RJ, Inouye SK. Prevalence and outcomes of low mobility in hospitalized older patients. J Am Geriatr Soc. 2004;52(8):1263–1270.
31. Theou O, Maclean M, Rockwood K. How much time hospitalized older patients spend upright? Gerontologist. 2014;54(suppl 2):S280.
32. Kortebein P, Symons TB, Ferrando A, et al Functional impact of 10 days of bed rest in healthy older adults
. J Gerontol A Biol Sci Med Sci. 2008;63(10):1076–1081.
33. Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M. Developing and evaluating complex interventions: the new Medical Research Council guidance. Int J Nurs Stud. 2013;50(5):587–592.
Keywords:© 2019 Academy of Geriatric Physical Therapy, APTA
acute care; England; older adults; outcome assessments; physical therapy