Abstract
Keywords
An estimated 10% of Australians ≥65 yr old sustain an injury in any given month (Australian Bureau of Statistics, 2006). This age group constitutes approximately one-quarter of all hospitalized community injury cases in Australia (Cripps & Harrison, 2008). The most common injury in this group is to the hip and thigh (27% of all trauma admissions, with 84.3% of these being hip fractures). Injuries to the head, brain, or face (14%) are the next most frequent (Bradley & Harrison, 2008). Given the long recovery time for hip fractures and brain injuries, a significant proportion of these patients will require further rehabilitation; however, delays are commonly experienced in the transition from an acute care unit to an appropriate rehabilitation facility. The evidence has shown that hospitalized older adults are at significant risk of functional decline because of deconditioning (Clinical Epidemiology and Health Services Evaluation Unit, 2008) and that the risk of negative outcomes increases with the length of stay (LOS) (Chandler & Hadley, 1996; Krasnoff & Painter, 1996). Increasing evidence has shown that rehabilitation interventions to maximize function and facilitate discharge provided directly within trauma units may be effective in minimizing subsequent disability in this population (de Morton, Keating, & Jeffs, 2007).
At the hospital in which this study took place, the demand for trauma services increased substantially from 1996–1997 (737 admissions) to 2004–2005 (2,630 admissions). This growth in admissions, along with an increased awareness among trauma unit staff of the role of occupational therapy in treating mild traumatic brain injury, resulted in an increased demand for occupational therapy services. An analysis of departmental statistics at that time suggested that trauma patients >65 yr old were usually assessed by an occupational therapist only when ready for discharge planning, presumably as a consequence of competing clinical priorities and lack of therapist time. The limited and, often, delayed occupational therapy intervention for these patients was concerning, particularly given the evidence of older adults’ susceptibility to functional decline during hospitalization.
In response to these concerns, a targeted occupational therapy service was established to provide acute rehabilitation input to this group. The service included the allocation of a clinician with specialist expertise in elder care to provide a more tailored and appropriate service for these patients and the increased use of an existing functional conditioning program (FCP; Bynon, Wilding, & Eyres, 2007). The FCP is a daily therapy program delivered by an occupational therapist, physiotherapist, or allied health assistant. The FCP encompasses individual or group sessions of mobility practice, exercise, and activities based on the patient’s premorbid self-care, leisure, or productivity activities and is the subject of a separate publication (Bynon et al., 2007). Referrals to occupational therapy for patients admitted to the trauma service are usually initiated by the unit care coordinator during daily handover meetings. Referrals can also be initiated by the nursing, medical, or other allied health staff.
A range of terms is used to describe functional decline, including loss of function, activities of daily living (ADL) decline, declining function, status decline, ADL status decline, and functional impairment (Hoogerduijn, Schuurmans, Duijnstee, de Rooij, & Grypdonck, 2007). Functional decline is defined as a decrement in physical functioning, cognitive functioning, or both that results in a reduced ability to attend to usual ADLs (Vorhies & Riley, 1993); it is often not strictly caused by the injury or illness that led to hospital admission. It can be compounded by preadmission condition, effects of immobility, comorbidities, and the adverse effects of medical treatment. Functional decline has been associated with a range of negative clinical outcomes, including increased LOS and need for subacute rehabilitation, higher levels of discharge to residential care rather than home, and increased burden on caregivers and community resources (Hoogerduijn et al., 2007).
In hospitalized older people, functional decline may occur rapidly. Evidence suggests it begins within 24 hr of admission (Topp, Ditmyer, King, Doherty, & Hornyak, 2002). It is also a widespread problem, and various studies have reported that it affects between 30% and 60% of older people during their stay in an acute hospital ward (Fortinsky, Covinsky, Palmer, & Landefeld, 1999; Hirsch, Sommers, Olsen, Mullen, & Winograd, 1990; Lamont, Sampson, Matthias, & Kane, 1983; McVey, Becker, Saltz, Feussner, & Cohen, 1989; Murray et al., 1993; Sager et al., 1996). Functional decline can be difficult to reverse. Sager et al. (1996), for example, found that only half of older people who experienced functional decline during a hospital admission had returned to premorbid ADL performance levels at 3 mo after discharge.
A review of the evidence for functional conditioning and exercise programs for older adult inpatients in acute hospital settings found that a multidisciplinary intervention that included exercise resulted in a small but significant reduction in acute hospital LOS and a small but significant increase in the proportion of patients discharged directly to home (de Morton et al., 2007). Similarly, a systematic review and meta-analysis of the evidence of the effectiveness of specialized geriatric units for acute hospitalizations found that these services produced a functional benefit compared with conventional hospital care and increased the likelihood of the person returning home after discharge (Baztan, Suarez-Garcia, Lopez-Arrieta, Rodriguez-Manas, & Rodriguez-Artalejo, 2009).
Much of the published evidence for exercise or functional conditioning programs, however, is specific to patients admitted for acute medical illness or elective surgery, and the studies of people admitted after traumatic injury are limited. One cohort study that followed 532 patients >50 yr old after hip fracture (Siu et al., 2006) found that of the subgroup of participants who were previously ambulating independently, those who commenced mobilization later had lower scores on the FIM™ at 2 mo after injury. Results must be interpreted with caution, however, because this study was observational in design, groups were not randomly assigned, and the study included people of a broad age range. Although Siu et al. reported that 252 participants were previously living independently in the community without paid help, they did not conduct a separate analysis of results for this group, nor did they include LOS or discharge destination as outcome measures.
Our aim in this study is to describe the impact of a targeted occupational therapy intervention and increased use of the FCP on patient LOS, discharge destination, departmental resource use, and functional status in older people admitted as the result of trauma. Specifically, we hypothesized that the provision of targeted occupational therapy and the FCP service to older patients in the trauma unit would be associated with (1) a reduced hospital LOS, (2) a higher proportion of patients discharged to home, and (3) an increase in number and timeliness of referrals to occupational therapy for this population.
Method
Study Design
We compared all eligible consecutive cases admitted in the first 6 mo of 2007 who consented to be enrolled in the study with a cohort of historical controls. The historical control group consisted of consecutive patients referred for occupational therapy services (i.e., medically stable) during the first 6 mo of 2006; the intervention group consisted of a sample of patients admitted during the corresponding period in 2007 (after implementation of the targeted occupational therapy trauma service for older adults). We chose the same months to minimize the effect of seasonal variations in admission trends. During the study period, access to rehabilitation services did not change. We investigated differences in LOS, discharge destination, and resource utilization patterns for occupational therapy and the FCP in the two groups.
Setting
This study was conducted at The Alfred, a 500-bed urban teaching hospital in Melbourne, Victoria, Australia. It is one of only two adult trauma hospitals in the state of Victoria, and it provides specialist services for >50% of the state’s adult trauma cases. The trauma unit admits >5,000 injured patients per year, of whom approximately 20% are classed as major trauma (Injury Severity Score >15; Baker & O’Neill, 1976). Patients >65 yr old account for 20% of all admissions. The study setting provides acute services only and has no trauma rehabilitation facilities on campus.
Participants
Participants in this study were all ≥65 yr old, admitted to the acute trauma service, living independently in the community before admission, and identified by the multidisciplinary team as requiring occupational therapy assessment. Approximately 10% were transferred from other city or regional acute hospitals because of the need for specialist trauma services but had been living independently before the trauma. The intervention group participants (n = 50) were recruited consecutively over a 6-mo period between January and June 2007. Patients who were unable to communicate or provide informed verbal consent were not considered for inclusion. The historical control group data (n = 105) were drawn from the hospital database and were restricted to patients meeting the preceding criteria and admitted between January and June 2006. We undertook a file review to identify and exclude everyone in this group who was cognitively impaired (as identified by standardized testing using the Neurobehavioral Cognitive Status Examination; Kiernan, Mueller, Langston, & Van Dyke, 1987) or noted as unable to communicate. This review was done to maximize comparability of groups.
Consent was not required from the control group because this study involved secondary use of previously collected information. The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki (Shephard, 1976) and the National Statement on Ethical Conduct in Human Research of the Australian National Health and Medical Research Council (2007) and was approved by the hospital’s Human Research and Ethics Committee.
Details of the demographic characteristics of both groups are presented in Table 1. In accordance with Bradley and Harrison (2008), we grouped injuries by the region of the body injured (head, thorax and chest, abdomen, low back, lumbar spine, pelvis, upper limb, or lower limb), according to the principal diagnosis recorded for that patient on the hospital’s electronic database. If a person was admitted because of multiple injuries, the most serious injury was coded as the principal diagnosis.
Intervention and Control Group Patient Characteristics
Note. ISS = injury severity score (>15 indicates major trauma); M = mean; SD = standard deviation.
Intervention
After receipt of an electronic referral by the care coordinator, nurse, or medical staff (generated by means of the computerized patient records system), all participants received an initial assessment by the occupational therapist, and issues affecting discharge were documented and addressed. In the case of the intervention group, this assessment was conducted by a specialist elder-care occupational therapist. In all cases, the initial occupational therapy assessment consisted of an interview with the patient, caregivers, or both to determine the preadmission and current levels of occupational performance. Subsequent standardized and nonstandardized assessments (e.g., the Neurobehavioral Cognitive Status Examination or Westmead Post Traumatic Amnesia scale; Shores, Marosszeky, Sandanam, & Batchelor, 1986) were conducted as determined by the occupational therapist on the basis of impairment and the patient’s individual circumstances. This assessment information, in collaboration with the patient and interdisciplinary team, was used to determine relevant treatment goals, an intervention program (which included the FCP), and a discharge plan.
Outcome Measurements
Primary outcome was LOS, which was recorded in total hours and then calculated into days. Discharge destination was coded as home, rehabilitation facility, other acute hospital, or elder care facility. These data were obtained from the hospital’s trauma database. Total number of referrals to occupational therapy were recorded for the intervention period (January–June 2007) and compared with referral rates from January–June 2006.
Data Analysis
All data were analyzed using SPSS Version 18.0 (SPSS, Inc., Chicago). We compared characteristics of the treatment and historical control groups using t tests for normally distributed variables (with log transformation as required) and χ2 tests for categorical variables. Total number of referrals to occupational therapy was calculated as a proportion of the total number of people >65 yr old admitted to the trauma unit during the two study periods. Total referrals to FCP were calculated as a proportion of the included samples. For those receiving FCP, we calculated mean total hours of FCP. A two-sided p of .05 was considered to be statistically significant.
Results
Groups were similar to each other at baseline on all measures (see Table 1). Results for key outcome measures are reported in Table 2.
Comparison of LOS and Occupational Therapy/FCP Usage Between Groups
Note. FCP = functional conditioning program; LOS = length of stay; M = mean; SD = standard deviation.
Geometric means calculated as data-required log-transformation. bEffect size = 0.29 units.
p < .05.
Length of Stay
LOS was found to be well approximated by a log-normal distribution, and so it was log transformed before analysis. Results were compared using a Student’s paired t test and reported as geometric means (95% confidence interval). The difference in means (M) was 1.7 (intervention M = 5.2, range = 4.0–6.7; control M = 6.9, range = 6.0–8.1), p = .04, meaning that the intervention group’s mean LOS was 25% shorter than that of the control group (ratio = 0.75, range = 0.56–0.98). We observed an effect size of .29 units, equivalent to 35% of a standard deviation (0.82 units), indicating a small effect size (Cohen, 1988).
Discharge Destination
We assessed data for discharge destination using a χ2 test for equal proportion as well as a Cochran-Mantel-Haenszel test (Mantel & Haenszel, 1959) for trend because discharge destination showed some natural ordering. Although a larger proportion of the intervention group was discharged home (62%, compared with 48.6% of the control group), the differences were not statistically significant (p = .16). Full results are presented in Table 3.
Discharge Destination by Group
N = 50. b N = 105.
Referrals to Occupational Therapy and Use of the Functional Conditioning Program
Referral rates of older adults’ trauma admissions to occupational therapy increased significantly over the study period, from a baseline figure of 44% (January–June 2006) to 55% (January–June 2007; p = .05). We found no difference in the intervention and control groups between occasions of service and total hours of occupational therapy, but time from admission to first occupational therapy service did differ significantly: The intervention group was seen, on average, 1.5 days sooner than the control group (p < .003). The proportion of patients referred to the FCP increased significantly during the study period (20%, compared with 2.85% of the control group), although mean sessions per referred patient remained the same.
Discussion
Our finding of reduced LOS (from 9.29 to 7.28 days) is similar to findings of other studies measuring the impact of multidisciplinary intervention (which included occupational therapy and physiotherapy service and early exercise) in similar older adult populations (Asplund et al., 2000; Collard, Bachman, & Beatrice, 1985; Landefeld, Palmer, Kresevic, Fortinsky, & Kowal, 1995), although all of these studies recruited people with a medical ailment rather than our study’s recruitment of people with traumatic injury. These findings contrast with those of another study (Curtis, Zou, Morris, & Black, 2006), which examined the impact of a trauma case management service (consisting of experienced trauma nursing staff who were key coordinators for individual cases from admission to discharge). In that study, LOS actually increased from 9 to 10 days in patients >64 yr old.
Although patients were more likely to be discharged to their own home after the implementation of the targeted service, this finding was not statistically significant. This result contrasts with the findings of a systematic review of interventions, including exercise, for acutely hospitalized older medical patients (de Morton et al., 2007); the review concluded that “multidisciplinary intervention significantly increased the proportion of patients discharged directly home compared to geriatric rehabilitation/transfer to another acute hospital/sheltered living/nursing home” (p. 12). Individual studies included in de Morton et al.’s (2007) analysis found higher proportions of discharge to home than in our study: 62% compared with 83% (Asplund et al., 2000), 73% (Collard et al., 1985), and 85% (Landefeld et al., 1995). Again, the population included in these studies differed from our group in that they were hospitalized for a medical condition rather than trauma, and this difference could explain the variation in results.
As hypothesized, referral rates to occupational therapy were significantly higher after the targeted occupational therapy program was implemented; however, the number of occupational therapy sessions and average time spent with each patient remained the same. One can argue that the timing of the occupational therapy service and higher utilization rate of the FCP were the key factors in improving the primary outcome, because patients were seen, on average, 1.82 days after admission (1.5 days sooner than they had been previously) and were 7 times more likely to be referred to the FCP. As stated previously, deconditioning can start in the second day postadmission, further highlighting the need for early intervention to minimize the risks.
Several factors may account for the improvements in LOS found in this study; however, the provision of an occupational therapy service and FCP specifically targeted at early intervention to prevent functional decline is possibly responsible for at least some of these improvements. This possibility is consistent with findings by Hagsten, Svensson, and Gardulf (2004), who found that patients with hip fractures who received more intensive occupational therapy interventions during their acute hospital stay had improved performance in self-care tasks at time of discharge compared with patients who received conventional care. Ultimately, an interactive effect among multiple factors (including input from other professional disciplines in the interdisciplinary team) is most likely to have produced the changes identified in this study. The role of patient-related factors, such as changes in outcome expectancy associated with progress through the system toward discharge, in mediating these changes must also be considered.
What This Study Adds to the Evidence
In this study, we specifically examined the impact of targeted occupational therapy and FCP services on the LOS and discharge destination of people >65 yr old admitted after traumatic injury. Much of the research to date has focused on patients admitted for acute medical reasons, so this study addressed an identified gap in the evidence.
Implications for Occupational Therapy Practice
We found that earlier occupational therapy assessment and intervention (and initiation of the FCP) resulted in reduced LOS and a trend toward discharge home rather than to another care facility. These findings provide the following helpful clinical guidelines for this population:
Occupational therapy has a pivotal role in identifying and addressing issues of occupational performance, and this should commence at the earliest opportunity.
Early engagement in meaningful occupations (including self-care and leisure) is likely to have minimized functional decline and contributed to the positive patient outcomes reported in this study.
Limitations of This Study
Several methodological limitations in the study’s design need to be considered. First, the intervention group consisted of people who agreed to take part in the study, and they may differ from our historical control group, despite all efforts to exclude those with cognitive impairment or an inability to communicate. The analysis of baseline variables, however, showed that they were matched on several variables, including age, gender, admission reason, injury severity score, and body part injured.
Conclusion
Results suggest that the provision of targeted occupational therapy and functional conditioning service to older patients in the trauma unit is associated with a reduced hospital LOS. It also resulted in an increase in number and timeliness of referrals to occupational therapy from this population, despite no significant change in individual treatment time for the treating therapist. Further research, such as a randomized controlled trial, is required to determine the true effect.
Footnotes
Acknowledgments
We thank Michael Bailey (biostatistician) for advice and assistance with statistical analysis and Louise Niggemeyer (and team) from the Alfred Trauma Service for the historical data.