Ideal timing to transfer from an acute care hospital to an interdisciplinary inpatient rehabilitation program following a stroke: an exploratory study

The demand for rehabilitation services following stroke is rapidly growing since the population at risk for vasculocerebral accident is continuously increasing and the progress in acute clinical management have resulted in higher survival rates over the past few years [1]. Due to limited financial and specialized human resources, the publicly-funded and universal-access health care system is challenged by this increasing demand for inpatient stroke rehabilitation services [2]. Eventually, rapid accessibility to organized inpatient stroke rehabilitation services may become compromised, especially during the initial recovery period, and may lead to a prolonged length of stays in acute care facilities for individuals surviving a stroke [3, 4]. It is believed that this potential unnecessary delay spent in acute care facilities, commonly refer to as nonmedical bed-days (not resulting from medically unstable acute or chronic comorbidity), may inhibit the crucial motor recovery process taking place shortly after a stroke [5, 6]. Over the past few years, the development of comprehensive and interdisciplinary stroke units in most acute care facilities deserve additional attention since it may influence rehabilitation outcomes [7]. Physical therapy, occupational therapy, and speech and language therapy services are now rapidly initiated after the onset of the stroke and offered throughout the acute care hospital stay [7]. On these specialized interdisciplinary units, therapeutic interventions target specific goals to be achieved according to specific treatment plan developed for each patient [7]. As the stroke survivors become medically stable, therapy intensity usually increases in preparation for the transfer to an inpatient stroke rehabilitation program [7].

The association between the timing of rehabilitation efforts and rehabilitation outcomes has been repeatedly documented [8–12]. Numerous studies have concluded that individuals with stroke who are admitted earlier to inpatient stroke rehabilitation programs usually present the most favourable functional recovery [8–10]. However, this association has been challenged given evidences showing that the time elapsed between the onset of the stroke and the admission to inpatient rehabilitation program is rather a strong indicator (confounder) of the health status of individuals who suffered a stroke [11, 12]. Clinically, stroke survivors, who are medically ready for discharge earlier, usually experienced only minor decline in the integrity of their motor, sensory or cognitive functions [11, 12]. As a consequence, these patients may be admitted sooner to inpatient rehabilitation and may reach the most favourable rehabilitation outcomes, in terms of degree of disability and discharge destination. Age- and gender-specific variability have also been identified as strong prognostic factors influencing stroke rehabilitation outcomes [13–15].

It is important to document the ideal timing to initiate intensive inpatient stroke rehabilitation after the neurological event. Therefore, the objective of this study was to examine the specific influence of short, moderate and long onset-admission intervals (OAI) on rehabilitation outcomes across homogeneous subgroups of patients who were admitted to a standardized interdisciplinary inpatient stroke rehabilitation program.

Finally, total, motor, and cognitive-efficiency scores generated from total, motor and cognitive absolute FIM gain scores and inpatient rehabilitation LOS measured for all patients and for the short, medium and long OAI subgroups are presented in Table 4. Similar total (P = 0.120), motor (P = 0.103), and cognitive-efficiency scores (P = 0.674) were reached between the OAI subgroups.

Inpatient rehabilitation outcomes (FIM scores, FIM change scores, LOS, FIM efficiency score) were not influenced by the OAI in this study. Surprisingly, the impact of the OAI on stroke rehabilitation outcomes was found to be suppressed when controlling for the degree of severity, gender, and age. These results challenge the consensus that stroke survivors who are rapidly admitted to inpatient rehabilitation potentially reach more favourable outcomes than those admitted later, and the presumption that delayed inpatient stroke rehabilitation admission may prompt adverse effect on rehabilitation outcomes. Overall, these results confirm that if interdisciplinary rehabilitation services are rapidly initiated after the stroke, OAI loses importance independently of the setting (acute hospital vs rehabilitation centre) although a small effect of timing (OAI) might still remain (Efficiency scores: FIM total P = 0.120; FIM motor P = 0.103). This small effect definitively deserves attention as this can be of clinical importance even if no statistically significant difference was found. This trend corroborates that individuals who are medically stable following a first CVA should rapidly be transferred to a rehabilitation facility offering an intensive stroke rehabilitation program. Under no circumstance should the admission to a rehabilitation facility from an acute care facility be delayed.

Both publicly-funded and universal-access acute care and rehabilitation hospitals may not be as proactive at initiating discharge procedures as privately-funded healthcare institutions since less constraints (e.g.: financial) influence this process. In fact, the average OAI of 30.97 (SD 18.12) days calculated in this study was almost three times more elevated than the average of 11 days recently reported in the United-States of America (USA) following a stroke [17]. This reduced OAI reported in the USA may result, in part, from the presence of numerous constrained for admission to rehabilitation (e.g.: financial) and from the limited number of severe cases (FIM-Total score lower than 40) admitted to rehabilitation facilities despite evidence of slow recovery among these individuals. Highly disabled stroke clients are likely to undergo unfavourable outcome, but unexpected improvement cannot be ruled out [27]. Nonetheless, adherence to specific validated criteria to determine readiness for medical discharge following a stroke may lead to a reduction of the OAI, which may in turn contribute to cost reductions in acute care hospital [5, 28, 29]. Concerning the inpatient stroke rehabilitation LOS, it is obvious that the publicly-funded and universal-access health care system, as seen in Canada, also permits a more prolonged LOS during inpatient rehabilitation than what is reported in the USA. For example, the average LOS of 50.81 (SD 24.51) days calculated in this study was more than two times the average of 20 (SD 13) days recently reported in the USA following a stroke [17]. Reasons similar to those supporting reduced OAI in the USA may also partially explain the short LOS observed in this country. A reduced LOS may definitively have an effect on the efficiency scores reported in this study. Since limited financial and human resources available currently prevent the expansion of inpatient stroke rehabilitation programs, refinement of their discharge criteria, which would influence the LOS, may simultaneously optimise inpatient stroke rehabilitation efficiency and improve accessibility to these programs. Rapid access to rehabilitation programs for all individuals who sustain a stroke should never be jeopardized from an ethical, biological, clinical or administrative point of view.

Strength of rehabilitation interventions offered in acute care facilities also deserves additional attention to better determine its impact on OAI of clients referred to inpatient rehabilitation programs [30]. This study demonstrates that OAI did not have a significant impact on rehabilitation LOS (P = 0.096) among individuals with stroke. Yet, the fact that individuals with a long OAI remained almost 10 days longer than those with short or moderate OAI in rehabilitation may represent a clinically and administrative relevant difference. Moreover, there is no evidence suggesting that rehabilitation is ineffective if begun at a later time. In fact, recent publication even proposes that rehabilitation in chronic stroke clients should not be neglected and represents a promising option [31].

It is clear that inpatient rehabilitation after a stroke is effective for all individuals although variability is observed among OAI groups. All patients equally gained from inpatient stroke rehabilitation indifferently of OAIs although individuals with short OAI had an extra 8 total-FIM units than those with moderate or long OAI (P = 0.083). More precisely, the comparable FIM efficiency scores found between the OAI subgroups suggest that similar rates of functional improvement may be expected despite differences in OAI. Alternative services, such as ambulatory hospital-based rehabilitation services or early supported discharge with home rehabilitation, may represent interesting clinical approaches to limit the OAI (nonmedical bed days), especially for individuals with mild disability [32, 33]. This definitively needs to be further studied since superior therapeutic efficacy and effectiveness may be reached during these programs when compared to inpatient stroke rehabilitation.

Since it is still extremely difficult to discriminate functional gains emerging from spontaneous recovery from those resulting from therapeutic interventions, caution is advised when analyzing the various levels of therapeutic efficiency measured. Spontaneous recovery might occur at different times for individuals with various degree of disability following a stroke and might overlap with recuperation attributable to therapeutic interventions (rehabilitation and adaptation processes). It is believed that spontaneous recovery usually results from three separate, but interactive, processes which include resolution of diaschisis (loss of function in remote areas anatomically connected to region of lesion), behavioural compensation and neuroplasticity [34, 35]. More precisely, the mechanisms underlying the neuroplasticity has been variously attributed to redundancy (parallel distributed pathways), changes in synaptic strength, axonal sprouting with formation of new synapses, assumption of function by contralateral homologous cortex, and substitution of uncrossed pathways [36].

Some evidences suggest a possible association between the severity of the vasculocerebral event and the duration of the spontaneous recovery process [37]. Courses of spontaneous recovery process associated to rapid initiation of rehabilitation efforts might fluctuate among individuals presenting various levels of disability following a stroke, refuting the general belief that early rehabilitation is associated to greater gains for all of them. As highlighted by this study, this general belief may be modulated by the severity of impairment and disability levels following a stroke. Consequently, the timing of intensive functional rehabilitation deserves more attention to ensure best possible cumulative gains resulting from spontaneous recovery and therapeutic interventions. Other elements such as purity, specificity, dose, intensity, duration, and timing of rehabilitation interventions also require added consideration to better understand treatment efficiency in stroke rehabilitation [30].

After controlling for the degree of disability, gender, and age, this matched block design study showed that OAI may not be a relevant prognostic factor of inpatient stroke rehabilitation outcomes for patients referred from acute care facilities where rehabilitation services are rapidly initiated after a stroke. The impacts of early or delayed initiation of inpatient rehabilitation following a stroke may not favourably or adversely affect rehabilitation outcomes, respectively. However, it certainly remains critical to rapidly initiate rehabilitation services after the onset of a stroke at the acute care hospital, according to specific goals to be achieved for each patient, because any delay may adversely influence rehabilitation outcomes. Perhaps most importantly, these results may have important implications in the continuum of care of individuals with stroke between acute care and rehabilitation hospitals. Finally, the result of the present study not only contrasts from previous literature but also supports the need of future prospective studies focusing on the influence of OAI on inpatient stroke rehabilitation outcomes and their retention, even improvement, over time after discharge.