MINERVA MEDICA COPYRIGHT

Transcription

1 Hip fracture (HF) is a common event in the geriatric population and is often associated with significant morbidity, mortality and costs for the Healthcare Systems. The growing awareness of HF consequences and the expected rise in the total number of HF worldwide have led to the development and implementation of models of care alternative to the traditional ones for the acute and post-acute management of HF older adults. These services were set to minimize in-hospital complications, streamline hospital care and provide early discharge with the main objectives of improving functional and clinical outcomes, and reducing healthcare costs associated with hip and other fractures. Basically, the main feature that distinguishes these models is the different healthcare professional that retains the responsibility of the care during the acute and postacute phases. This review has been conceived to provide a brief description of the models implemented in the last twenty years, to describe their potential benefits on short- and long-term outcomes, to define the strengths and limitations of these models and the areas of uncertain, and to make some consideration about the future. Actually, on the basis of available studies, it is not possible to define the best model of care for HF older adults. However, the more complex Fundings. No external funding source. Conflicts of interest. Dr A Giusti has received consulting fees from Novartis and Procter & Gamble, speaker fees from, ASMN Hospital, Novartis, Roche/GSK, Abiogen, Chiesi, Eli Lilly, Merck & Co and Stroder, and travel/accommodations expenses reimbursement from Novartis, Servier, Nycomed, Eli Lilly and Roche/GSK. Dr A Barone, Dr M Razzano, Dr M Pizzonia and Dr G Pioli declare no conflict of interests. Epub ahead of print on May 10, Corresponding author: A. Giusti, Department of Gerontology and Musculoskeletal Sciences, Galliera Hospital, Corso Mentana 10, 16128, Genoa, Italy. andrea.giusti@galliera.it EUR J PHYS REHABIL MED 2011;47: Optimal setting and care organization in the management of older adults with hip fracture A. 1, A. BARONE 1, 2, M. RAZZANO 2, M. PIZZONIA 2, G. PIOLI 3 1Department of Gerontology and Musculoskeletal Sciences, Bone Clinic Galliera Hospital, Genoa, Italy 2Orthogeriatic Unit Department of Gerontology and Musculoskeletal Sciences, Galliera Hospital, Genoa, Italy 3Orthogeriatric Unit, Department of Geriatric Medicine Arcispedale Santa Maria Nuova, Reggio Emilia, Italy and sophisticated services, characterized by a multidisciplinary approach demonstrated, in randomizedcontrolled and before-after observational studies, to produce better outcomes compared to the traditional or simplest models. Further research is warranted to confirm long-term functional and clinical benefits of these models and to evaluate their cost-effectiveness. Key words:aged - Hip fracture - Delivery of health care, integrated. Innovative models of care for the management of older orthopaedic patients, particularly hip fracture (HF) older adults, have been developed and implemented since more than 30 years ago, 1-5 with the first randomized-controlled trial (RCT) comparing a traditional model with an orthopaedic-geriatric inpatient service published by Gilchrist et al. in These models of care were set to minimize in-hospital complications, streamline hospital care and provide early discharge with the main objectives of improving functional and clinical outcomes, and reducing the direct and indirect healthcare costs associated with hip and other fractures. In the traditional model (Figure 1A), the fracture patient was admitted to a trauma ward, where the Vol No. 2 EUROPEAN JOURNAL OF PHYSICAL AND REHABILITATION MEDICINE 281

2 Figure 1. Models of integrated care for the management of the older adults presenting with hip fracture. The models are distinguished by the different healthcare professional that retains the responsibility for the management of the patients throughout the care pathway. Early rehabilitation usually refers to the first 3-5 days after surgical repair and includes mobilisation out of bed and early ambulation, while late rehabilitation refers to exercises and training aimed at recovery of prior functional status. In the Traditional model in-hospital stay lasts about two weeks, includes an important part of rehabilitation, and a significant proportion of subjects may be discharged directly home without further rehabilitation (A). The Consultant Team in the Orthopaedic Ward is a variation of the traditional model, with a programmed consultant service throughout in-hospital stay, usually starting after surgery (B). In the more recent models of care (C-E), in-hospital stay is usually shorter (on average a week) and post-acute rehabilitation is performed in different settings, such as other in-hospital wards, out of hospital facilities or home-based services. In the most sophisticated model, the rehabilitation ward is a unit specifically designed for the rehabilitation of geriatric patient (Geriatric Rehabilitation Unit) or even orthopaedic geriatric patients (Geriatric Orthopaedic Rehabilitation Unit). For further details about the specific models of care see the text. orthopaedic surgeon was responsible for the care of the patient, while medical queries and complications were dealt by consultative medical services. 3, 6, 7 The physician s opinion was sought only when the orthopaedic surgeon considered it desirable, and different physicians could see the patient during in-hospital stay. In the post-acute phase, early rehabilitation took usually place within the Orthopaedic Ward. After discharge, the patient could be transferred directly home, to a skilled nursing facility or to a rehabilitation facility, without substantial continuity of care. 282 EUROPEAN JOURNAL OF PHYSICAL AND REHABILITATION MEDICINE June 2011

3 The first models introduced were variations of the traditional one and were, therefore, quite simple (Figure 1B). They were based on the designation of a specific and sometimes trained team for the management of the patients during the acute and post-acute rehabilitative phases, on the regular input by a consultant team, and on the implementation of early discharge programs. 3, 6, 7 Over the years these models evolved, being replaced by multidisciplinary and co-ordinated approaches that demonstrated to be more effective to meet patients complex care needs (Figure 1C-E). 3, 6-10 Particularly in the last ten years a number of innovative joint orthopaedic-geriatric experiences have been embraced in some countries and evaluated in small studies. 3 These experiences have been designated with different names, such as ortho-geriatric units (OGU), Co-managed geriatric fracture centres, interdisciplinary/multidisciplinary clinical/ care HF pathway or geriatric HF clinical pathway, which in most of the cases, but not always, distinguished dissimilar models in terms of setting and organization However, the basic idea underlying the development of these projects was quite always the same: the definition of a multidisciplinary team dedicated to the care of the fracture (mainly hip fracture) older adult, to promote continuity of care, rapid management of the undercurrent medical problems, early mobilization and rehabilitation, and coordinated discharge planning. As it will be clear from the description of the several experiences presented in this review, differences between these models mainly rely on the typology of the healthcare professionals involved, the level of responsibility and contribution of these figures, the continuity of care throughout the overall care pathway and the different settings in which the acute and postacute phases take place. 3, 4, 7-9 Although a variety of experiences have now been described, the alternative services reported in the available literature can be broken down into the few models described in the following chapters. 3, 6-9 These models are distinguished by the different healthcare professional that retains the responsibility of the management of the patients throughout the care pathway (Figure 1). The main objectives of the present review will be: to provide a brief description of the models implemented in the last twenty years, with particular consideration to the most recent and outstanding; to Vol No. 2 EUROPEAN JOURNAL OF PHYSICAL AND REHABILITATION MEDICINE 283 describe the potential benefits of these experiences, with the attention focused on primary and relevant outcomes such as short- and long-term mortality, functional recovery, quality of life and, when possible, cost-effectiveness; to define the strengths and limitations of these models and the areas of uncertain; and, finally, to make some consideration about the future, by setting the agenda for further research on this topic. General considerations Models of integrated care Most of the studies undertaken to evaluate and define the benefits of innovative models for the care for fracture older adults, have been focused primarily on patients with HF over years of age. Thus, our review will concentrate mainly on the optimal settings and care organization for the management of HF older adults. The studies are really heterogeneous in nature, particularly with respect to the design (randomised controlled trials versus before-after observational trials) and to the outcomes assessed (e.g., functional recovery assessed at different time points with dissimilar tools). 9 Therefore, except for the few cases in which RCTs were undertaken, it is really challenging to define which model is optimal or superior to another one following a strictly evidence-based approach. Moreover, in some experiences both the setting and the organization were not defined on the basis of evidence-based guidelines, but were chosen and delineated considering the healthcare resources available, the willingness to pay, and, most important, the organization of the Healthcare System in a given Country All these limitations hamper our possibility to define the optimal setting and care organization. However, the results of the different studies allow us to emphasize some of the most outstanding and innovative characteristics of these models, and those principles of organization and implementation that seem to add value to the traditional management of the older adults with hip fracture. In this contest, studies have found a significant reduction in postoperative complications, length of hospital stay (LOS), functional disability and mortality up to one year by combining surgical and medical care. Importantly, all these trials clearly

4 demonstrated that a comanaged model has not any detrimental consequences, and that can be implemented in every developed country and within existing hospital organization and frameworks. Consultant team skilled in the care of older people in the orthopaedic ward Setting and care organization This is a variation of the traditional model and one of the first services implemented in different developed countries (Figure 1B). Basically, it is characterized by regular inputs from a consultant team including several professionals, with the overall responsibility of the care under the orthopaedic surgical staff. 2, The team held weekly or more frequent rounds to develop and monitor treatment plans. The prevention and management of common problems and complications of elderly patients with fracture were based on the choices of the individual surgeon or physician. The orthopaedic surgeon settled early mobilisation and discharge timing and location. Main results In RCTs, no significant benefits were achieved when this model was compared to usual care. 3 Gilchrist et al. observed no difference in length of inhospital stay, mortality and placement of patients six months after discharge, between HF subjects who received standard care and those who were followed by a consultant team skilled in the care of older people. 2 Similarly, Naglie et al. found that the presence of a consultant team in the orthopaedic ward did not result in a significantly higher proportion of patients alive without decline in ambulation or with no change in the place of residence at 6 months. 19 It should be noted, however, that in these studies the consultant team contribution started postoperatively, 2, 19 and that slightly better results were reported when the consultant service participated early from hospital admission with daily visits. 21, 22 Swanson et al. observed a reduction in the length of in-hospital stay in HF patients who underwent an early intervention program with a multidisciplinary team, compared to those managed traditionally, with no significant difference in mortality, functional recovery or social support 6 months after hospital discharge. 21 In another study (before-after observational trial), undertaken to evaluate the effectiveness of a newly implemented HF service, the authors observed a shorter in-hospital stay, fewer medical complications and a higher proportion of HF patients treated surgically within 24 hours, after the introduction of the model. 22 Comments Although the implementation of a consultant team skilled in the care of older people in the orthopedic ward for the management of HF elderly seems to add some short-term benefit to the traditional model of care, RCTs failed to demonstrate a real beneficial effect of such service on long-term and primary outcomes such as mortality, functional recovery and consume of community services after the fracture. Finally, studies had a limited time of follow-up (six months) and did not consent to evaluate the costeffectiveness of this model. Comanaged care Setting and care organization This is the most sophisticated and probably complex model implemented to date (Figure 1E). It has been developed in different countries, mainly in North America and Europe, 11-13, 16, 17, and, as it appears by the different experiences reported in the available literature, it has evolved over 10- to 15-year periods with gradual improvements being added with time. 11, 12, 16, 23, 24, 26 The key and characterizing concept is the co-management of the fractured patients by a physician skilled in the management of older adults and the orthopaedic surgeon that share the responsibility and leadership from admission in the OGU to discharge. An interdisciplinary team including several healthcare professionals supports the codirection. Standardized treatment paths are implemented. The two healthcare figures directly involved (physician skilled in the care of older adults and surgeon) see the patient daily, write their own orders and communicate frequently, thus reducing the risk of delays, inappropriate care and iatrogenic errors, and promoting an optimal clinical coordination. Even orders and choices that traditionally are of surgical competence, such as timing of surgery or 284 EUROPEAN JOURNAL OF PHYSICAL AND REHABILITATION MEDICINE June 2011

5 surgical agenda, are usually shared and discussed, to optimize the management of the patients. A detailed description of the standardized care, orders and protocols reported in the literature is out of the aims of the present article. Basically, the care pathway may be best described by the term patientcentred, protocol-driven, standardised care as recently suggested by Kates et al. 23 Main results The beneficial effects (short- and long-term functional and clinical outcomes) of this innovative model have been illustrated in few well-designed RCTs and before-after prospective observational studies. Vidan M et al. randomized 319 patients (mean age±standard deviation [SD]=81.9±7.7; 82% women; 24% demented) presenting with a HF to a single University Hospital to receive a co-managed intervention (N.=155) or usual care (N.=164). 25 In-hospital mortality and major medical complications were significantly (P=0.03 and P=0.003, respectively) lower in the intervention group (0.6% and 45.2%, respectively) compared to the usual care group (5.5% and 61.7%, respectively). Even after adjusting for several confounding factors, the comanaged intervention was independently associated with a reduction of the risk of death or major complications during hospitalization (OR: 0.55, 95% CI: ). Although both LOS (intervention, 16 days; usual care, 18 days) and 12-month mortality (intervention, 19.9%; usual care, 25.3%) were reduced in the intervention group, these differences were not significant. Finally, more patients from the co-managed care group (57%) achieved partial recovery at 3-month compared to those in the usual care group (44%, P=0.03), while no differences in functional recovery were found at 6- and 12-month. In another quasi-experimental, prospective, randomized trial, undertaken in 224 HF patients (mean age of 85 to 87 years; 83% female; 54% with cognitive impairment) admitted to a tertiary University Hospital, a newly implemented geriatric-orthopaedic comanagement program was compared to usual care with geriatric consultant. 27 Significant differences between the co-managed intervention and the usual care were found in the median number of days from admission to geriatric assessment (1 versus 4 days, P<0.001), from admission to surgery (5 versus 6 days, P<0.001), and in length of acute (12 versus Vol No. 2 EUROPEAN JOURNAL OF PHYSICAL AND REHABILITATION MEDICINE days, P<0.001) and total hospital stay (14 versus 20 days, P<0.01). A non-significant higher proportion of subjects in the intervention group were able to walk alone or with aid at discharge (intervention, 78%; usual care, 69%). Khan R et al., compared short-term outcomes in 745 patients admitted within a 5-year period in the OGU of a single University Hospital, before and after the implementation of a combined geriatricorthopaedic model. 13 Subjects in the pre-implementation period (mean age 81 years; 80% female; 56% living at home at the time of fracture) were quite similar to those hospitalised and managed with the new model of care (mean age 82 years; 85% female; 53% living at home at the time of fracture). No significant differences were in in-hospital mortality (before 10.4% versus after 11.1%), LOS (mean days±sd, before 26±26 versus after 27±23) and discharge to pre-admission place of residence (before 79.5% versus after 79.5%) between the two groups. In 2006, our group have published the results of a similar study undertaken in 819 HF patients admitted over a 3-year period to a Public Non-Teaching Hospital. 11 Patients were managed with standard care (2 groups of patients: control-1, N.=272; control-2, N.=295) or with a co-managed approach (N.=252) on an alternate annual basis. The mean age (years) ±SD (control-1, 84±7; intervention 84±7; control-2 85±7; P=0.21) and the proportion of women (control-1, 77%; intervention, 81%; control-2 78%; P=0.21) were comparable between the three groups. No significant difference in the LOS was observed between the groups. However, the rate of in-hospital mortality for the intervention group was lower than for control-1 (4.8% versus 9.9%, P=0.03) and similar to that of control-2 (6.8%, P=0.34). Finally, the comparison of survival curves (1-year follow-up) showed a significant difference over time (P=0.04) in favour of the intervention group, and, one year after admission, the survival rate was about 10% higher in the geriatric orthopaedic co-managed group than in the control groups (75% versus 65% [control-1], P=0.01, and versus 67% [control-2], P=0.04). Finally, Friedman SM et al. compared short-term outcomes between patients hospitalized with a proximal femur fracture in the University of Rochester (New York), where a Co-managed Geriatric Fracture Centre (GFC) was set up, and those admitted to a local institution without such service. 16 GFC

6 subjects (N.=193) were significantly older (mean age ±SD: intervention 85±8 years versus usual care 82±9 years, P=0.002), were less likely to reside in the community (intervention 39% versus usual care 82%, P<0.01), and had more comorbid conditions (Charlson comorbidity score, mean±sd: intervention 3.4±2.2 versus usual care 2.6±1.6, P<0.001) and dementia (intervention 54% versus usual care 22%, P<0.001) than usual care controls (N.=121). Despite these differences and after adjusting for baseline confounders, the older adults treated in the GFC had shorter times to surgery (mean hours ±SD: intervention 24±17 versus usual care 37±64, P=0.02 after adjustment), shorter LOS (mean days±sd: intervention 4.6±3.3 versus usual care 8.3±6.3, P<0.001 after adjustment), fewer cardiac complications (intervention 1.0% versus usual care 7.4%, P=0.03) and fewer cases of thromboembolism (intervention 0.5% versus usual care 5.0%, P=0.03), delirium (intervention 24.4% versus usual care 32.2%, P<0.001) and postoperative infection (intervention 2.3% versus usual care 19.8%; P<0.001). There was no difference in in-hospital mortality (intervention 1.6% versus usual care 2.5%) and 30-day readmission rate (intervention 9.8% versus usual care 13.2%). Comments The implementation of a co-managed model of care for the elderly patients with HF appears to offer many benefits to the patients, reducing short- and long-term adverse events. Although the results of the different studies are not always consistent and sometimes skewed, in general this service has demonstrated to reduce short-term mortality and complications, to increase survival up to one-year, and in part to improve functional outcomes, compared to the traditional model where only the orthopaedic surgeon is responsible of the management of the patient. Indeed, to implement such program are needed considerable efforts, consistent administrative support and strong physician leadership. 12, 23, 24 Persistence and determination are essential for its success; and, once implemented, the program requires a continuous evaluation in order to add improvements with time. Given the relevant resources needed to implement a comanaged model, additional studies are needed for a better understanding of its impact on long-term functional outcomes (return to prefracture residence, consume of community resources, rate of institutionalization after one year) and to evaluate its cost-effectiveness, considering long-term direct and indirect cost of hip fracture. Finally, as recently suggested, further studies are regarded to evaluate whether this service is translatable and applicable to any hospital organization and frameworks, since it is commonly believed that a reasonable hospital volume (e.g., 100 cases of hip fractures) is needed. 23 Interdisciplinary care/clinical pathway Setting and care organization A huge number of services and experiences described in the literature and identified with different designations (e.g., geriatric hip fracture clinical pathway, multidisciplinary integrated care pathway, perioperative evidence-based clinical pathway, comprehensive hip fracture program) should be comprised in this group. 14, 15, Although they might appear a little heterogeneous, there are some relevant features ruling these programs that render them analogous and that consent to consider them as variations of an essential model (Figure 1C). The main characteristic that distinguishes this model from the others described is the absence of a single true leadership in the management of the older adult presenting with a fracture, and the presence of different healthcare professionals involved with specific responsibilities. From hospital admission to discharge, the patient follows a clinical pathway ruled primarily by standard orders and protocols based on most updated evidence-based guidelines. Once in the emergency department, the patient can undergo an early assessment from the anaesthesiologist, can be directly transferred to the orthopaedic ward, or can even be first admitted to an acute care unit for further clinical stabilization, when clinically not fit for surgery. 28, 29, 36 In general, the care of the fracture older adults may take place in a single ward or can be fragmented in different and subsequent settings. The postoperative phase may vary significantly from one experience to the other, mainly due to differences in the organization of Healthcare Systems and in available resources. For example, after a short period of early in-hospital rehabilitation the subject can be rapidly transferred to his/her usual environment for further 286 EUROPEAN JOURNAL OF PHYSICAL AND REHABILITATION MEDICINE June 2011

7 rehabilitation under the responsibility of a rehabilitation physician. 33 In other cases, rehabilitation is undertaken in the Orthopaedic Ward for a longer period of time by a multidisciplinary team responsible for assessment and treatment adjustments. 36 An interdisciplinary team, with leaders from different departments is involved in the clinical, surgical and rehabilitative management of the fractured adults. The team follows the rules of the multidisciplinary approach, with meetings and rounds, continuous communication and coordination between members, education of the clinical staff and discussion/sharing of clinical choices, but there is not an actual leader or director of the care pathway, although in some experiences the orthopedic surgeon maintains an higher level of responsibility. Vol No. 2 EUROPEAN JOURNAL OF PHYSICAL AND REHABILITATION MEDICINE 287 Main results Most of the trials available are prospective or retrospective before-after studies and only very few RCTs have been undertaken. In a RCT carried out in a single University Hospital in Taiwan, Shyu et al. compared a newly implemented interdisciplinary intervention program (patients N.=68) for HF patients with routine care (patients N.=69). 33 Mean age (SD) of the HF subjects was 77.6 (7.7) years, about 69% were female and 31% were cognitively impaired. Outcomes were initially compared at 1 and 3 months. 33 No significant differences were found between the two groups in the number of hospital readmissions, emergency department visits and institutionalizations at the times of followup. Significantly more subjects in the experimental group than in the usual care group recovered their previous walking ability by the first (55% versus 37%, P = 0.004) and third month (78% versus 62%, P=0.001) after discharge, while no differences were found in pain intensity, mortality (at 3-month, 0% versus 1.6%, respectively) or recurrence of falls. The intervention group demonstrated higher scores in the ability to perform basic activities of daily living (indicating a better recovery) compared to the control group at 1 and 3 months. Health-related quality of life was similar between patients who were randomized to the new program and those managed with usual care 1 month after discharge, while at 3 months subjects in the intervention group had significantly fewer depression symptoms (Geriatric Depression Scale score ±SD: intervention, 3.6±3.0, usual care 5.4±4.2, P=0.008), higher vitality and physical functioning. The beneficial effects of this interdisciplinary program for HF patients have been further evaluated on the long-term (1 and 2 years). 31, 32 One year after discharge, the experimental group demonstrated significantly greater functional improvements and fewer depressive symptoms than the control group. Most interesting, these positive effects demonstrated to last up to two years after hospital discharge. 31, 32 Subjects in the intervention group had significant better performance in the basic activities of daily living, better recovery of walking ability (OR 2.72, 95% CI , for the intervention group recovering previous walking ability), fewer falls (OR 0.56, 95% CI , for the intervention group to have a subsequent fall), fewer depressive symptoms and better ratios of hip flexion than the control group 24 months after discharge. 31 No differences were found in 2-year mortality (intervention 11.3% versus control 15.9%), rates of hospital readmission (intervention 11.1% versus control 7.8%) and institutionalization (intervention 2.4% versus control 2.9%) between the two groups. Pedersen et al. evaluated the rates of post-operative complications, LOS and 1-year mortality, before and after introduction of a comprehensive multidisciplinary fast-track treatment and care program for HF patients in the Orthopaedic Ward of a single University Hospital (Copenhagen). 36 Overall, 357 subjects admitted before the implementation of the service were compared to 178 patients who were treated with the optimized HF program (mean age 84 years for the women and 77 years for the men, 24% males and 29% with a diagnosis of dementia). Data from the two groups were collected using retrospective chart review. After the implementation of the HF program the rate of in-hospital post-operative complications felt from 33% to 20% (OR 0.61, 95% CI , P=0.002). This decrease was related to the lower rate of urinary tract infection, pneumonia and confusion in the intervention group. Also LOS decreased significantly with the introduction of the new service (before 16 ±18 versus after 10±9 days, P<0.001). Twelve months mortality was similar between the two groups (before 29% versus after 23%, P=0.2). However, when only community-dwellers were considered one-year mortality was significantly decreased after the implementation of the new program (before 23% versus after 12%, P=0.02). In a larger study, undertaken in the University of

8 Southampton, the effectiveness of implementing a care pathway for HF in older people was evaluated in a before-after trial. 15 Seven hundred sixty-eight patients (before N.=395; after N.=373) were considered (mean age ±SD, 83±7 years; 23% males; 21% with a history of dementia). Subjects in the intervention group showed a significant longer LOS and improved clinical outcomes compared to those in the control group. Mean LOS (SD) in the orthopedic ward was 16 (19) days before and 23 (24) days after the implementation of the care pathway (P<0.001). A higher proportion of subjects in the experimental service were able to walk alone at discharge (before 63% versus after 73%, P=0.03). Although the overall complication rate was similar, the incidence of wound infections, pressure sores and urinary tract infection decreased after the new service was set up. These beneficial effects were associated with a higher number of occupational therapist contacts during the intervention phase. The authors found no differences in 30-day mortality, institutionalization at discharge (before 19% versus after 13%, P=0.058) and hospital readmissions within 30 days of discharge. In agreement with these results, 15, 36 Beaupre et al. 34 found a significant reduction of post-operative complications after the implementation of a perioperative evidence-based clinical pathway for HF management. This is actually one of the largest prospective before-after study undertook to date, with more than HF patients enrolled (mean age 82 year-old, 24% men). Although the risk adjusted in-hospital mortality was similar before and after the introduction of the care pathway, patients who were managed with the new service presented a significant lower incidence of post-operative heart failure (OR 0.24, 95% CI ), pulmonary edema (OR 0.51, 95% CI ), cardiac arrhythmias (OR 0.22, 95% CI ) and pressure sores (OR 0.20, 95% CI ) compared to those treated with a traditional approach. Overall LOS and costs did not differ between the groups. Length of stay in the orthopedic ward was higher in the intervention group (10 days) compared to the control group (8.5 days, P=0.001). On the contrary, inpatient rehabilitation was longer in the control group (before 28 days versus after 22 days, P=0.04), and a higher proportion of subjects who were enrolled in the new program were discharged directly home or to long-term care without the need of further rehabilitation. The beneficial effects of an integrated care pathway on in-hospital morbidity have been further confirmed in a similar but smaller study carried out in Sweden, in which the authors were also able to demonstrate a significant 40% reduction in average total cost for HF management after the implementation of the integrated pathway (mean cost of treatment, before versus after 7 209, P<0.001). 37, 39 Gholve et al. (United Kingdom) compared shortterm outcomes of a multidisciplinary integrated care pathway (N.=151 patients) for HF with those of a fast tracking procedure (N.=143 patients). 35 Many hospital (75%) in the United Kingdom adopted a fast track scheme to admit patients with HF, based on an Audit Commission Report in There is no globally accepted definition for fast track but essentially it is characterised by optimum patients assessment, baseline investigations, appropriate referrals and admission time within 60 minutes of arrival in the Accident and Emergency Department. The care pathway adopted was implemented including the principles of the fast tracking and the features of the interdisciplinary care pathway above described. The authors found a significant decrease of in hospital stay (before 19 days versus after 15 days, P<0.001) and a non-significant reduction in 30-day mortality and in the number of patients operated within 24 hours from admission. In the University Hospital of Copenhagen was developed an orthopaedic-anaesthesiologist liaison for the peri-operative and early post-operative rehabilitative management of HF older adults. 41 The benefits of such program were evaluated in an observational before-after study undertaken in 224 subjects. The authors failed to demonstrate a significant positive effect of the orthopaedic-anaesthesiologic service on post-operative morbidity, LOS, in-hospital (before 12% versus after 7%, P=0.24) and 30-day mortality. However, as it could be expected, the anaesthesiologist presence reduced the total number of in-hospital visits by external non-surgical specialists (before 128 versus after 50, P=0.02). Comments Interdisciplinary clinical pathways have been successfully implemented in conditions such as elective joint replacement surgery, coronary and cancer treatment programs. 42 Correctly, they are considered a mean of providing high quality care with the re- 288 EUROPEAN JOURNAL OF PHYSICAL AND REHABILITATION MEDICINE June 2011

9 sources available, although cost-benefit analyses are still needed. The data presented demonstrate that an interdisciplinary care pathway has beneficial effects on shortterm outcomes, such as post-operative in-hospital complications, early recovery of functional ability and, in some, albeit not all, experiences, LOS. 14, 15, 33-36, 39 These short-term benefits have been reached without an increase in direct hospital costs, 34 and, at least in one case, a significant reduction in total costs for in-hospital HF management was demonstrated. 37 Most of the studies failed to demonstrate a significant effect on short- and long-term mortality. This last observation cannot be easily explained, and should be interpreted taking into account that only two studies lasted one-year or more, 31, 32, 36 and that few trials were probably powered to detect a significant difference in cumulative mortality. 15, 34, 36 For the same reasons, it seems tricky to draw considerations about functional outcomes (recovery of basic activities of daily living and pre-fracture ambulation, institutionalization, and return to usual place of residence). Notably, the only RCT with a 2-year follow up demonstrated a higher short- and longterm functional recovery in patients who were treated with the experimental model compared to those managed with the traditional model In addition, two before-after trial showed significantly improved functional outcomes after the implementation of an interdisciplinary care pathway. 15, In conclusion, although interdisciplinary care/ clinical pathway seems to be a valuable and promising model to improve clinical outcomes of older adults with HF, further studies are warranted to evaluate its long terms benefit on survival and functional recovery. Geriatrician-led fracture service with orthopaedic consultant Setting and care organization The overall management and healthcare pathway take place in a Geriatric Ward and/or under the complete responsibility of the geriatrician (Figure 1D). 18, The central role of the geriatrician as the primary attending physician for all patients from hospital admission to discharge distinguishes the Vol No. 2 EUROPEAN JOURNAL OF PHYSICAL AND REHABILITATION MEDICINE 289 service: he evaluates the fractured patient on admission and during in-hospital stay, coordinates the timing of surgery, procedures, diagnostics, treatments and transition/discharge planning. The orthopaedic surgeon is a consulting physician that follows the patients until complete wound healing. An interdisciplinary team, including different healthcare professionals, is integrated in the service and participates in the care of the patients. Standardized orders and protocols are implemented. Usually, the HF patient is admitted directly from the emergency department, evaluated and prepared for surgery in the geriatric ward, transferred to the operating room and then returned to the geriatric ward. On the basis of the clinical/rehabilitative pathway following these pre- and perioperative phases two different experiences have been described: 1) in the Sheba model implemented by Adunsky and colleagues in 1999 (Israel), the patient is cared throughout the acute and post-acute rehabilitative phases in the same setting, with an overall LOS quite high (around 30 days); ) in the geriatrician-led hip fracture program developed by Miura et al. (Portland, Oregon, United States), the HF subject is rapidly transferred to a community skilled-nursing facility early after surgery (3 days) for further rehabilitation, with a mean LOS lower than 5 days, but with fragmentation of care. 18 Actually, this really relevant difference in the organization of the postacute phase seems to be attributable to the organization of the Healthcare System in the two countries where these programs were undertaken, as well as to the resources available. Main results In a partially concurrent prospective study, Adusnky et al., compared functional outcomes between the newly implemented Sheba model and a traditional service where the HF patient was first admitted to the Orthopaedic Ward for surgical management and than shortly transferred to a rehabilitative ward for ongoing rehabilitation. 44 Patients who were treated with the innovative model of care demonstrated a shorter LOS and, after controlling for several covariates, better functional outcomes (OR: 1.97, 95%CI: , for successful rehabilitation in Sheba group). In a subsequent, non-controlled analysis of the Sheba model, undertaken in 592 patients after 5-year of experience, the authors sup-

10 ported the applicability and feasibility of the service in elderly subjects after HF, as a model being able to cover the various and complex needs of these patients. 43 In a before-after observational study, recently published, Miura et al. demonstrated the short-term beneficial effect of a Geriatric-Led HF Program in terms of LOS, time to surgery (about 50% subjects operated in less than 24 hours) and costs (reduction in direct and indirect costs of about $2,000). 18 Comments The Sheba model is one of the unique experiences described of a fully geriatric-led HF service. On the basis of the papers published by Adunsky et al., this approach seems feasible, applicable and efficacious in term of functional outcomes In this model, all the care takes place in the same setting with the same intensity of care. Although this is for sure a strongpoint, by reducing the detrimental effects of fragmentation of care, it seems that this setting organization is not acceptable in terms of costeffectiveness for the Healthcare Systems of European countries, where indeed the trend in the last ten years was to separate the settings of the acute and intermediate care. In this contest, a separation of the intensity of care during the in-hospital phase might be equally effective and economically more suitable. This last consideration is at least in part supported by the data from Miura and collaborators, who demonstrated a significant reduction in the direct and indirect cost when the geriatric leadership was implemented only for the acute phase. 18 In would be interesting to evaluate the long-term functional and clinical outcomes of this Geriatrician-Led HF program, since as it appears from the description of the service after the acute phase the HF patients are transferred to another setting where a different team manages the continuation of the rehabilitation. Background Early supported discharge (homebased rehabilitation and geriatric orthopedic rehabilitation unit) The concept of using forms of intermediate care (skilled nursing or community facilities, early home- based care) to allow earlier discharge from the acute ward, in order to alleviate pressure on hospital bed capacity, LOS and concerns about bed blockers has been introduced in the United States since a long time. 46 In 2001, the average length of stay in English NHS hospitals for diseases such as stroke (27 days) or hip fracture (26 days) was more than double the average for US (7 days), and this was due to an higher proportion of patients discharged to intermediate care in the US. 46 The UK figures were similar to other European countries, where usually patients with a disabling condition (e.g., hip fracture) underwent a period of in-hospital rehabilitation in the acute care ward of admission, before being discharged for late ongoing rehabilitation. 25, 26, The need of cut Healthcare costs and rationalize the consume of resources induced European countries to develop new strategies to reduce acute inhospital stay and promote early discharge in hip fracture patients, while maintaining an acceptable quality of care and without worsening short- and long-term outcomes. For hip fracture, home-based supported discharge and geriatric orthopaedic rehabilitation units represented the more consistently implemented approaches to face these needs, imposed by the healthcare reorganization in Europe. 49, Therefore, these innovative schemes should be considered as possible post-acute transitions that could be used in tandem with any of the aforementioned models, rather than a real and comprehensive model of care for the management of HF older adults. 7 Home-based supported rehabilitation Early discharge and home-based rehabilitation (HBR) after hip fracture have been implemented since , 57, 58 These services have been developed in Europe, Australia and North America, and are considered safe and cost-effective alternatives to in-patient rehabilitation. 17, 53 Indeed, the implementation of HBR programs after hip fracture is highly dependent on the local organization of the Healthcare System and on the resources available. 17, 49, 53, 57, 58 Thus, these services may be introduced exclusively in areas where already exists an hospital at home community nursing service. Patients potentially suitable for early discharge at home are usually those living at home with relatives or with other types of informal support before hos- 290 EUROPEAN JOURNAL OF PHYSICAL AND REHABILITATION MEDICINE June 2011

11 pital admission, and medically fit to be discharged in a non-inpatient setting (clinically stable without relevant acute illness). 33, 53, 54, 57, 58 In general, they are assessed on admission for suitability, to be informed about the service (patient and relatives) and to set up a discharge planning. In some experiences, a trained geriatric nurse, a physiotherapist or an occupational therapist visited the patient s home before discharge to fit the necessary equipment and the rehabilitation setting. 33, 57, 58 Than, early after surgery, the patient is transferred directly home for rehabilitation. A multiprofessional team is usually involved in the care of the older adult in collaboration with the general practitioner. This team may be the same that cared about the patient during hospital stay, 33 or a completely new one. 53 The healthcare professionals involved in the HBR held weekly meeting to evaluate the progresses and problems arising during the rehabilitation. The specific content, frequency and duration of the training/rehabilitative program varies from one experience to another. Basically, physical therapists provide the same treatments that are usually supplied in hospital or in skilled nursing facilities, with strengthening and range-of-motion exercises and gait/transfer training. A number of RCTs and prospective observational studies have evaluated the potential benefits of HBR , Published RCTs demonstrated that HBR services in elderly patients after hip fracture produce comparable results in terms of functional outcomes and reduce length of in-hospital stay compared to traditional programs Moreover, some of these trials showed slightly better results, in terms of ambulation ability, balance confidence, independence in basic activities of daily living, health-related quality of life and caregiver burden, in HF patients discharged directly home for rehabilitation , 64 Limitations of these RCTs were the small samples and the restrictive inclusion criteria for enrolment, with the exclusion of patients with pre-fracture cognitive decline or disability, that indeed represent a significant proportion of HF elderly. To evaluate the feasibility and efficacy of HBR after hip fracture in the real clinical practice, our group undertook two prospective studies with 12-month follow up in an unselected sample of HF older adults. 53, 54 Summarizing, we found that a HBR program is a feasible and efficacious alternative to traditional rehabilitation in an unbiased population of HF elderly living with Vol No. 2 EUROPEAN JOURNAL OF PHYSICAL AND REHABILITATION MEDICINE 291 relative, irrespective of pre-fracture functional and cognitive status, 53 and that HBR services should be implemented and proposed even to HF older adults with moderate to severe dementia. 54 Geriatric orthopedic rehabilitation units The Geriatric Orthopedic Rehabilitation Unit is a variation of the traditional Geriatric Rehabilitation Unit, fully dedicated to the care of older adults presenting with an orthopaedic condition. As previously noted, these services may follow the traditional pre- and peri-operative care or one of the aforementioned models. Usually, when the orthopaedic surgeon or the interdisciplinary team judges the patient fit to be moved to a rehabilitation ward, he/ she is rapidly transferred to a Geriatric Orthopaedic Rehabilitation Unit. 55, 56, Basically, the presence of a physician skilled in the care of older adults as supervisor and leader distinguishes this service from other rehabilitation programs. The basic ingredients of this scheme are quite similar to those of HBR, with the exception of the different setting. A multidisciplinary team has a central role in the management of the HF elderly. 55, 56, Since their implementation about 30 years ago, 55, 56 these units demonstrated to produce, in well designed RCTs, short- and long-term better outcomes compared to traditional rehabilitation In particular, patients who underwent rehabilitation in a Geriatric Orthopaedic Rehabilitation Unit demonstrated a significantly greater recovery of their functional status (e.g., basic activities of daily living, ambulation), a lower risk of institutionalization and a higher rate of survival up to one year, compared to those treated in the traditional rehabilitation ward. 55, 56, Consistently, these favourable outcomes were reached with a reduction in length of stay in the rehabilitation setting. Finally, it should be noted that this rehabilitation approach demonstrated to be really successful even in patients with moderate to severe dementia. 67 Comments When looking across all the different post-acute rehabilitative interventions for HF older adults, 69, 70 HBR and Geriatric Orthopaedic Rehabilitation Units demonstrated to be valuable alternatives to the tradi-

12 tional in-patient rehabilitation. Both these programs allow reducing acute LOS, but do not improve longterm outcomes (particularly mortality) with respect to the traditional services (skilled nursing facilities, in-patient rehabilitation). Therefore, they should be considered always in tandem with the models of care described in this chapter, since they interest only a part of the overall care of HF patients. Indeed, the implementation of these services without a model including an early intervention may not produce the expected benefits. Special issues related to the setting and care organization There are a number of issues related to the setting and care organization not covered in the description of the models of care, which deserve to be highlighted since they have been raised in some recent study and, most important, should be considered for future studies on the topic. 12, Timing of surgery and immediate rehabilitation Evidence-based guidelines and RCTs recommend early surgery (as soon as the medical conditions allow), immediate weight-bearing and early ambulation (within 48 hours after surgery) in elderly patients presenting with hip fracture. 4, 8, 75 Indeed, these practices need a complex and coordinated organization. In recent years, it has been suggested that to reach those objectives are required an adequate setting, an operating room working daily with extra trauma lists, and a multidisciplinary team available 7 days a week. Although these assumptions are probably correct, they are not evidence-based and may be reasonably recommended only on the basis of expert opinion advices and common sense. In fact, studies undertook to evaluate alternative approaches to reduce the risk of operative delays and to ensure immediate rehabilitation in patients presenting with HF, were able only in part to produce evidences to be used in the organization of a service dedicated to older adults with HF. For example, it has been suggested that the use of extra orthopaedic trauma lists (>10 per week) may help to reduce the risk of operative delay after HF (>24 hours), although in an observational study, the beneficial effect of these additional trauma lists did not reach significance. 72 In a pre-post controlled intervention, Adie et al. studied the effect of scheduling surgery for HF with a nonemergency orthopaedic list or with an emergency list. 74 Basically, they found fewer cancellation and after-hours operation with a non-emergency management, when extra time slots for operating theatres where guaranteed. In a recent observational study, published by our group, it was demonstrated that the only factor affecting adherence to a protocol of immediate weight-bearing and early ambulation after hip fracture was an organizational variable. 12 We found that in a population of more than four hundred HF older adults, the most important factor associated with a delayed time to ambulation was having surgery before a public holiday (e.g., Sundays or Christmas), mainly due to the reduction in available resources during the weekend or public holidays. In conclusion, it seems clear that organizational aspects may be responsible for delays in time to surgery or in the early mobilization of HF patient. Therefore, further researches are warranted to identify optimal models of care able to overcome these issues and to maintain the same standard of daytime care every day of the week. 12 Moreover, the economic sustainability of these new programs should be evaluated accurately. Hospital volumes Kates et al. have suggested that there should be a reasonable volume of hip fractures per year to develop sufficient expertise in managing HF patients and to adopt one of the innovative models of care described in this review. 23 Nowadays, there are no studies to support a precise number, although it is believed that approximately 100 cases per year could be a reasonable volume to make these programs effective. A positive relationship between provider volume and outcomes has been shown for a wide range of surgical procedures across a variety of specialities Particularly, higher surgeon and hospital procedure volumes have been associated with lower mortality rates, fewer complications and shorter LOS. In the case of hip fracture, current literature led 292 EUROPEAN JOURNAL OF PHYSICAL AND REHABILITATION MEDICINE June 2011