Effects of an Integrated ‘Fast Track’ Rehabilitation Service for Multi-Trauma Patients: A Non-Randomized Clinical Trial in the Netherlands

Objectives The effects on health related outcomes of a newly-developed rehabilitation program, called ‘supported Fast Track multi-trauma rehabilitation service’ (Fast Track), were evaluated in comparison with conventional trauma rehabilitation service (Care as Usual). Methods Prospective, multi-center, non-randomized controlled study. Between 2009 and 2012, 132 adult multi-trauma patients were included: 65 Fast Track and 67 Care as Usual patients with an Injury Severity Score ≥16, complex multiple injuries in several extremities or complex pelvic and/or acetabulum fractures. The Fast Track program involved: integrated coordination between trauma surgeon and rehabilitation physician, shorter stay in hospital with faster transfer to a specialized trauma rehabilitation unit, earlier start of multidisciplinary treatment and ‘non-weight bearing’ mobilization. Primary outcomes were functional status (FIM) and quality of life (SF-36) measured through questionnaires at baseline, 3, 6, 9 and 12 months post-trauma. Outcomes were analyzed using a linear mixed-effects regression model. Results The FIM scores significantly increased between 0 and 3 months (p<0.001) for both groups showing that they had improved overall, and continued to improve between 3 and 6 months for Fast Track (p = 0.04) and between 3 and 9 months for Care as Usual (p = 0.03). SF-36 scores significantly improved in both groups between 3 and 6 months (Fast Track, p<0.001; Care as Usual, p = 0.01). At 12 months, SF-36 scores were still below (self-reported) baseline measurements of patient health prior to the accident. However, the FIM and SF-36 scores differed little between the groups at any of the measured time points. Conclusion Both Fast Track and Care as Usual rehabilitation programs were effective in that multi-trauma patients improved their functional status and quality of life. A faster (maximum) recovery in functional status was observed for Fast Track at 6 months compared to 9 months for Care as Usual. At twelve months follow-up no differential effects between treatment conditions were found. Trial Registration ISRCTN68246661

-Zo n M

Inhoud / Content
Probleemstelling / Problem definition -Health care problem: Annually +1-99.000 people are admitted to hospital after an accident. 880.000 people visit the accident & emergency department (A&E) after an accident (Stg. Consument & Veiligheid, 2005). These accidents lead to considerable societal costs. Direct medical costs are estimated at 1 billion euro annually, i.e. 3-4% of the total Dutch health care budget. Production losses due to acute trauma are estimated at 4 billion euro, thus widely surpassing costs associated with chronic illness like cardiovascular diseases and cancer (Mulder, 2002;Weseman, 2000;Beeck, 1997). Based on the 1997 bill "Met Zorg Verbonden" (Ministries of VWS and Internal Affairs) and the bill "Beleidsvisie Traumazorg" in 1999, 10 trauma centres were allocated nation-wide (van Vree et al., 2001). Medical care for trauma victims is a combined responsibility of hospitals, ambulance services, trauma centres, rehabilitation clinics, and GHOR (geneeskundige hulpverlening bij ongevallen en rampen). This co-operation is called trauma care chain (TCC). The azM is the trauma centre in Limburg (Hammacher, 2004). In conventional multi-trauma care service (CTCS) each of the partners has its own more or less autonomous treatment perspective, depending on the professionals individual treatment views and experience. Clinical evidence, however, suggests that an integrated multi-trauma rehabilitation service approach or 'Supported Fast track multi-Trauma Rehabilitation Service' (SFTRS), featuring: 1) shorter stay in hospital and earlier transfer of multi-trauma patients to a specialised trauma rehabilitation unit 2) an earlier start of both specific 'non-weight bearing' rehab training and multidisciplinary treatment 3) early individual goal setting 4) an integrated co-ordination of treatment between trauma surgeon and rehabilitation physician 5) shorter stay in trauma rehab unit may be more (cost-)effective. Conceptually, an analogy can be drawn between the SFTRS approach and the concept of 'stroke units' which have proven to be cost-effective (EDISSE project, ZONMW;Exel, 2003;Launois, 2004). Costs for using a hospital bed are higher than for a bed in a rehab centre (Oostenbrink, 2004).
-Disease/condition: Multi-trauma is defined as having at least 2 or more traumatic injuries, of which at least one is life threatening (Zelle, 2005a). Several tools for rating trauma severity have been designed (Cooper, 2004). The Injury Severity Score (]SS) (Baker, 1974) is used most. Major causes of multi-trauma are traffic accidents, accidents at work, (extreme) sports, falls, blasts, etc. (Erli, 2000). Several studies report that the legs (incl. pelvis) are most frequently injured in multi-trauma (Holbrook, 2001a,b).
-(Sub)group of patients: Multi-trauma occurs more often in males and in younger adults (Erli, 2000;www.prismant.nl, 2004). Many patients are at an age where they have a paid job.
-Usual care in the Netherlands for this (sub)group of patients: In conventional multi-trauma care service (CTCS) patients are admitted to hospital via the A&E. After I-Zo n M

Subsidieaanvraagformulier / Grant Application Forrn
Dossier nummer/ Dossier number: possible surgery, they are transferred to the IC-unit, followed by the general surgery ward, where the patient may stay for several days/weeks. The trauma surgeon seeks the advice of the rehabilitation physician, 1f riecessary. Ensuing treatment takes place in a hospital's outpatient clinic, in a (more distant) rehab centre, in a nursing home or with a local general practitiorier or physiotherapist. Van Vree, (2001) reported that, typically, each of the 'stations' in the CTCS may have its own more or less autonomous treatment perspective, depending on the professional's individual treatment views and experience.
-Disciplines involved in the usual care in the Netherlands for this (sub)group of patients: General surgeons, orthopaedic surgeons, anaesthetists, rehabilitation physicians, physiotherapists, occupational therapists, nurses -Disciplines in the usual care and their participation in this proposal: Genera] surgeons, orthopaedic surgeons, anaesthetists, rehabilitation physicians, physiotherapists, occupational therapists, nurses, psychologists, social workers are involved in CTCS and/or SFTRS.
-Motivation for intervention and the effectiveness of the intervention: Ample evidence is available confirming that each of the separate treatment elements within the TCC is effective. However, ambiguity as to general outcome exists. Dirnopoulou, (2004) report that the majority of multi-trauma patients still suffer from considerable levels of disability and impaired quality of life 1 year post-trauma. Van der Sluis, (1995) reported moderate to poor functional outcome in 26% of patients after 2 years. It is likely that earlier tra nsfer of multi-trauma patients from hospital to a specialised trauma rehab unit, leading to a faster discharge from rehabilitation, will lead to better functioning of patients, both somatically and psycho-socially. The proposed study airns at investigating which of 2 services (CTCS or SFTRS) is more (cost-)effective.

Relevantie / Relevance
-Motivation for chosen therne The study fits therne 8: organisation of care. It compares the (cost-)effectiveness of 2 multi-trauma services, ie. conventional multi-trauma care service (CTCS) and SFTRS, which differ as to organisation. The effects of the underlying therapy / care components have been established in nurnerous studies. However, as reported in the systematic review, contrasts in cost-effectiveness between multi-trauma rehab services have not been studied. The SFTRS features: a) shorter lengths of stay (in hospital and in the specialised multi-trauma rehab unit) through irnproved logistics within the TCC b) closer co-operation between trauma surgeon, rehab physician and the rehab team c) earlier start with specific corn prehensive rehab training modules early post-trauma d) earlier integrated multidisciplinary treatment, and rehab goal setting leading to: * optirnisation of treatment * reduction of secondary corn plications * reduction of function loss associated with prolonged bed rest (eg. muscle atrophy, endurance loss, contractures) * achievement of an optima] level of functioning, participation and quality of life -Contribution of results The study will yield results on the efficiency of an adapted care service for multi-traurna patients (SFTRS) featuring earlier (and condensed) involvernent of specialised rehab treatment. Results will show whether improved SFTRS logistics, cornbined with shorter stays in hospital and rehab clinic and specialised early rehab training modules are more (cost-)effective, relative to CTCS. Quality of life and functional recovery not only depend on injury severity, but also on prevention of secondary -o n Mw
-Similar studies In stroke rehab the development of stroke units proved to be cost-effective (EDISSE project, ZONMW;Exel, 2003;Launois, 2004). In the EDISSE study, stroke patients in 3 experimental stroke service settings were compared with patients receiving usual stroke care in a 6 months follow-up. Results showed that integrating services for acute stroke lead to organisational improvements, higher efficiency and better outcome (e.g. reduced hospital Iength of stay, less inappropriate hospita] days, more satisfied patients). The EDISSE study also provided criteria for optimising the quality of stroke services. The SFTRS is similar in its aims (i.e. higher efficiency, better patient's outcome), its approach (i.e. integrating services for multi-trauma patients, especially in the early phases post-trauma) and its design (ie. prospective, non-randomised clinical trial). -Potential effects on health Earlier start of specific rehab treatment after multi-trauma includes a.o. non-weight bearing physical training, multidisciplinary treatment of the patient, personalised goal setting and early start of psychological and social counselling. The SFTRS approach will lead to Iess secondary complications associated with bed rest, which would negatively influence recovery and quality of life. Early personalised goal setting and early treatment of depression are known to positively affect outcome. SFTRS will lead to faster reintegration into society. Early return to work and active support from the multidisciplinary rehab team will lead to a more stable social network, and the patient becoming Iess reliant on professional care in the long term.
-Potential effects on costs SFTRS will reduce the Iength of stay of multi-trauma patients in a hospita]. Earlier rehab treatment in a specialised rehab unit will also reduce the Iength of stay in the rehab clinic, thus reducing costs of hospital/clinic consumption. At this moment it is not possible to make a precise calculation of these savings. Since earlier discharge also means that patients take part in society and work earlier, costs related to production losses and patient&family costs are expected to be lower.

llIl II;
multi-trauma care throughout the Netherlands about the results of the study and to train them on how to implement the SFTRS approach into the trauma care chain (TCC). An implementation strategy will be developed in the last phase of the project, depending on the results of the study.
-Results Results will be disseminated through publications in peer-reviewed journals, presentations at (international) congresses, symposia, the Dutch Societies for Traumatology (NVT) and Rehabilitation (VRA), and the 'werkgroep Traumarevalidatie (WTR)'. Furthermore, the SFTRS will lead to spin-off' regarding: a) collaboration among TCC members as to courses, training of staff and exchange of knowledge; b) further development of treatment protocols for the whole TCC; c) further elaboration of the role of traumatology in the medical curricula (CHIVO); d) intensifying research collaboration and multi-centered TCC research projects; 1f proven (cost-)effective, SFTRS will be implemented in other trauma centres / TCC's throughout the Netherlands. An implementation plan as to this purpose will be drawn in the last phase of the project, depending of the project results.
-Organisations who should integrate the results in their activities: Professionals involved in multi-trauma patient care in hospitals and rehabilitation centres in the Netherlands (specialistenverenigingen NVT, WTR, NVvH, NOV, Federatie Traumatologie i.o.) and abroad (international trauma / rehab associations), TCC specialists: Trauma surgeons, A&E specialists, registrars in rehabilitation medicine, physiotherapists, occupational therapists, psychologists, social workers, university hospitals, regional hospitals, rehabilitation centres/depts, trauma centres.

Main objective
The main objective of this study is to examine the effectiveness, costs and cost-effectiveness of an integrated care service for multi-trauma patients involving: a) faster transfer of patients in an earlier post-trauma phase from the traumatology dept. to a specialised rehabilitation clinic b) earlier involvement of registrar in rehabilitation medicine in the treatment of the patient c) earlier start of multidisciplinary rehabilitation training, tailored to individual patient's need, including a.o. earlier partial weight bearing training, individual goal setting, prevention of complications associated with prolonged inactivity (e.g. muscle athrophy, reduced mobility, reduced cardiovascular endurance) and psychological and social counselling d) earlier discharge of patients from the rehabilitation clinic, leading to faster and better participation in Sub-questions are: What are the effects of the SFTRS on generic quality of life in multi-trauma patients as compared to the CTCS?
What are the effects of the SFTRS on functional health status as compared to the CTCS?
What are the costs to health care and to society of the SFTRS as compared to the CTCS?
What is the cost-effectiveness of the SFTRS as compared to the CTCS?

Plan van aanpak / Strategy
Clinical study preliminary studies by applicants Henk A.M. Seelen, PhD, has performed (and led) several experimental and clinical studies into the effects of major trauma (eg. trauinatic spinal cord injury, 11mb amputation and stroke) on patients' performance and outcome of training, for which he has obtained several grants (ZONMW / NWO, EU, Industry). Furthermore, he is leading a project on optimisation of integrated post-rehab care in high cervical spinal cord injured persons requiring 24-hours ventilation.
Peter G.R. Brink, PhD, professor of Traumatology, has conducted numerous trauma-related clinical studies, focussing on intensive care medicine and trauma surgery. He is strongly involved in the organisation of trauma care in the Netherlands and especially in Limburg.
Bena Hemmen, MD, PhD, has conducted several studies on treatment outcome in stroke and orthopaedic trauma. Together with S. Evers and H. Seelen she investigated the (cost-)effectiveness of computer-controlled prostheses in leg amputees. She is starting a large multi-centre study into functional assessment and early prognostics in traumatic amputees, in collaboration with H. Seelen, H. van der Linde and Prof. J. Geertzen, MD, PhD (UMC Groningen). Together with Prof. Brink and Dr. Seelen she currently is performing a study on early prognostics in multi-trauma patients. The latter project is made possible because of the close co-operation between the trauma surgeons and rehabilitation physicians from the azM, the hospitals in Heerlen and Sittard and the Hoensbroeck Rehabilitation Centre.
Publications on the results of the project are in progress.
Silvia M.A.A. Evers, MA, PhD, has specific experience with trial based health technology assessment studies in a.o. stroke and pain. She is (and has been) involved as a HTA supervisor in several ZONMW studies.
Arie B. van Vugt, PhD, professor of Traumatology, is specialised in effectiveness and quality of trauma care research in severely injured patients. As to projects related to the current proposal: he has led many research projects on treatment and outcome in patients who have suffered accident injuries. He Ø-Zo n Mw

Subsidieaanvraagformulier / Grant Application Form
Dossier nummer/ Dossier number: ,]1 IIlI I has specific expertise in the area of hip, pelvis and heel bone fractures.
Harmen van der Linde, MD, PhD, has led several projects related to outcome of treatment in lower limb amputation and prosthetics. He is strongly involved in the development of guidelines for prosthetics prescription in the Netherlands.
design A prospective, multi-centre, non-randomised clinical trial will be performed. Two multi-trauma rehabilitation services will be contrasted, ie. Conventional Trauma Care Service' (CTCS) and 'Supported Fast track multi-Trauma Rehabilitation Service' (SFTRS). Duration of follow-up is 12 months, with measurements taken at 3, 6 and 9 months post-injury.
As soon as possible post-trauma (TO) the following data will be recorded: Outcome measures (see below) are collected at baseline, 3, 6, 9 and 12 months post-trauma, i.e. TO, TI, T2, T3 and T4.
study population Multi-trauma patients admitted to one of the accident & emergency departments (A&E) of the participating hospitals are included. Multi-trauma is defined as having at least 2 or more injuries of which at least 1 is life threatening. The contrasts with conventional multi-trauma care service (CTCS) are: a) The rehabilitation physician from the rehabilitation centre is routinely involved at a very early stage post-trauma. This allows an early start for multidisciplinary rehab treatment. b) Early transfer (within five days after being added to the waiting list from the rehabilitation centre) to a centralized, specialized trauma rehabilitation unit equipped with facilities for early training programs. c) Early individual rehabilitation goal setting. d) Close co-operation and exchange of views and experiences between the trauma surgeon and the rehabilitation team by, for example, monthly clinical sessions and individual patient visits by the trauma surgeon in the first weeks after discharge. e) WelI-documented treatment protocols for multi-trauma patients for both the hospita] and rehabilitation centre phases.
Three phases can be identified in the treatment of multi-trauma patients: 1 Early rehabilitation phase 2 Stage II rehabilitation phase 3 Discharge or post-discharge phase Phase 1: Early rehabilitation phase In the early rehabilitation phase, the patient is not allowed to mobilize weight bearing. Consequently, the physiotherapist is concerned with maintaining joint mobility, muscle strength, sitting balance, condition and training transfers as well as treatments with non-weight-bearing conditions such as hydrotherapy and non-weight-bearing gait training. There are 10 sessions per week of 30 minutes each. In addition, fitness, gymnastics, table tennis, swimming, bowling, hand bike, wheelchair training, and archery are given. There are 2-3 sessions per week for each treatment modality of 60 minutes each.
The occupational therapist advises on bed posture, mattress types, aids for independent daily self-care, wheelchair-dependency training and meaningful activities that can be performed while bedridden. In addition, the wheelchair accessibility and wheelchair friendliness of the patient's home are studied. 1f necessary, written and oral advice on temporary and long-term adaptations to the home is given and support is given and the patient is helped to apply for financial support so that the patient can return home as soon as possible. At first, this would be for a day or two at the weekend, supervised by an occupational therapist, but would later become permanent. With regard to work, the patient's job is analysed and the patient's workplace is visited. There are 4 sessions per week of 30 minutes each.
The social worker and the psychologist will see every multi trauma patient within the first week after admission. The social worker helps the patient to return home by dealing with the family and offering advice and support to the patient on financial matters, transport facilities. The social worker also contacts the employer and company doctor to look into the possibility of reintegrating the patient into their present job.
The psychologist will examine the patient with regard to such things as mood disorders, posttraumatic The occupational therapist continues with the treatment goals as mentioned for Phase 1 and trams the patient to perform household tasks, hobbies, etc in a home-like environment. There are three sessions per week of 30 minutes each. In addition, group therapies such as occupational therapy and recreational therapy are given 2-4 times per week each.
The social worker and the psychologist continue the work mentioned in Phase 1, depending on the individual needs of each patient.
Phase 3: (Post) discharge phase In the discharge phase, the patient is prepared for living at home and is referred to local physiotherapists, specialized sport clubs and mental health care professionals.

CTCS
Conventional multi-trauma care service (CTCS) is provided in several centres. Multi-trauma patients are admitted to hospital via the A&E department. After possible surgery, they are transferred to the IC-unit, followed by the hospital's nursing ward, where the patient may stay for several days or weeks. The trauma surgeon, as chief consultant, decides whether or not a rehabilitation physician will be consulted during hospitalization. Next, ensuing treatment may take place in the hospital's outpatient clinic, in a rehabilitation centre, in a nursing home or with a local GP or physiotherapist. Van Vree and co-authors (2001) reported that, typically, each of the CTCS "stages" might have its own more-or-less autonomous treatment perspective, depending on the professional's individual treatment views and experience.
The effectiveness of multi-trauma rehabilitation interventions and its constituting elements has been established in numerous studies. Even most recently, Holtslag (2007), in his PhD research, investigated the long term outcome after major trauma. Furthermore, effectiveness of multi-trauma care has been established by others (e.g. Cameron et al., 2004), as reported in the systematic review (see below). In the group of van Vugt, Frankema and colleagues (2002)  IIIlI ; -Cognitive functioning: MMSE Next to that costs will be assessed using the PRODISQ, a cost questionnaire and hospita] databases.
-Argumentation for the chosen measures. In the current study the primary outcome measures are quality of life and functional health status. In several studies it was found that in multi-trauma patients quality of life and functional recovery do not solely depend on injury severity and complications (Holbrook et al., 2001), but also on psychological and social factors (eg. Richmond et al. 1998;Erli, 2000;Halcomb et al. 2005;Zelle, 2005b) as well as the patient's cognitive status (Fernandez, 2001). The functional independence measure (FIM) is widely used in assessing functional health status in different groups of patients. Baldry-Currens (2000) recommended using the FIM in assessing trauma outcome, the FIM correlating high with measures of injury severity and demonstrating clinical and statistical significance. Similarly, Hetherington et al (1995) reported that in rehabilitation services,the FIM is a useful, practical and simple methodology, providing a measure for assessing the original disability, its progress and residual limitations. The National Trauma Data Bank collects data on trauma centre performance throughout the US. As to functional outcome assessment in trauma patients FIM data are used (Nirula et al., 2006). At an international and interdisciplinary consensus conference in 1999 about the assessment and application of quality of life (QoL) measures after multiple trauma, experts clinicians and methodologists agreed on the SF-36 as generic tools for QoL assessment across all trauma patients (Neugebauer et al., 2002). In the proposed study both generic QoL and utilities will be derived using the SF-36. An overall utility score for population based QoL can be obtained, which facilitates comparisons-with other interventions, i.e. the social tariff of the SF-36 (Brazier et al. 1992(Brazier et al. , 2002. The Canadian Occupational Performance Measure (COPM) is an individualised client-oriented measure to assess the evolution of self-perception of skilis in patients across time (McCoII et al., 2000;Kinebanian et al. 2006). The COPM was, for example, used by Trombly et al (2002) to investigate the association between participation in goal-specific outpatient occupational therapy and improvement in self-identified goals in adults with acquired brain injury. In our study the COPM will be used to assess the extent to which individual treatment aims of the multi-trauma patient, set during rehabilitation, are met. The Hospital Anxiety and Depression Scale (HADS) gives clinically meaningful results as a psychological screening tooi, in clinical group comparisons and in correlational studies with several aspects of disease and quality of life. It is sensitive to changes both during the course of diseases and in response to psychotherapeutic and psychopharmacological intervention. Finally, HADS scores predict psychosocial and possibly also physical outcome (Hermann, 1997). The HADS has been used by Kempen et al. (2003) to investigate the effect of depressive symptoms on the recovery of activities of daily living after faIl-related injuries to the extremities in older persons. As stated before, anxiety and depression a.o. may influence therapy outcome in multi-trauma rehabilitation. Therefore, in the present study the HADS is used to assess this aspect. The Mini-Mental State Examination (MMSE) is a test that briefly surveys global mental status in a wide range of cognitive domains (Jackson et al. 2007;Folstein, 1975). Jackson et al (2007) used the MMSE in trauma survivors without intracranial hemorrhage. Their findings corroborated earlier research stating that these patients display persistent cognitive impairment associated with functional defects, poor quality of life, and an inability to return to work (Jackson et al., 2007). In our study the MMSE will be used similarly, i.e. to assess global cognitive functioning of multi-trauma patients.  data analysis and presentation In non-randomised comparative studies, variations in case mix between centres can influence the interpretation of outcome data (Davenport et al. 1996). Therefore, for each of the data sets collected at TI through T4, differences in outcome variable between the 2 services will be tested using multiple MANCOVA's, entering various indicators of case mix as co-variates, i.e. age, gender, ISS, number of complications, pre-trauma psycho-social status. Results will presented in peer reviewed (inter)national papers and congresses.
When patients drop out of the study, the reason for their withdrawal will be recorded. Drop-outs may bias the treatment effect evaluations. Therefore, the following regime will be applied: -Missing T4 measurement: 'last-observation-carried-forward' principle will be applied.
-Missing T3 or T2 measurement: linear interpolation of data using data from adjacent time points (e.g. TI and T4)for imputation.
-Missing TO or TI measurement or more than 2 missing measurements: discarding of patient data and influx of additional patient in order to meet n=82 per group.
treatment credibility and expectancy In studies comparing the effectiveness of different treatment regimes, differences in treatment credibility and expectancy may influence the outcome. In the proposed study the credibility/expectancy questionnaire (CEQ) (Devilly & Borkovec, 2000) will be administered directly following the explanation of the study's rationale to patients, i.e. after informed consent has been obtained.
economic evaluation General considerations For the economic evaluation the main research question is: From the viewpoint of the society is another organisation of professional care service for trauma patients (i.e. SFTRS) compared to CTCS preferable in terms of costs, effects and utilities? Based on this main research question several sub-questions are relevant: 1) What are the costs of SFTRS compared to CTCS preferable in terms of costs, effects and utilities?
2) What are the extra effects (measured in quality of life, utilities, and saving by reducing inpatient hospital admissions of multi-trauma patient) of SFTRS compared to CTCS preferable in terms of costs, effects and utilities? We hypothesise that SFTRS is associated with a reduction in health care and patient costs, and an improvement in quality of life, compared to CTCS. We expect SFTRS to be cost-effective from a societal perspective. Assessments of the quality of life and costs will take place at TI through T4.

Cost-analysis
In the cost identification, the following costs are considered: Health care costs: cost of the intervention program and other health care resources both by the patient and the caregiver. • Patient and family costs: informal care, paid domestic help, transportation, over the counter medication, and other out-of-pocket expenses. • Production losses: absenteeism, presenteeism (loss of productivity while at work), and compensation mechanisms for both the patient and the caregiver, if relevant. Measurement of volumes: • Hours spent on the intervention program will be recorded on a pre-structured form by the acting health care professionals. • All other health care costs and patient & family costs will be recorded in a cost questionnaire • Production losses will be measured using the patient modules of the PRODISQ (Koopmanschap et al., 2004(Koopmanschap et al., , 2005 The PRODISQ will be used together with the costs questionnaire, every 3 months at baseline and TI through T4. For the valuation of health care costs and patient & family costs, an update of the Dutch manual for costing in economic evaluations (Oostenbrink et al. 2004) will be used. For care for which no costs-guidelines are available estimations of the costs will be made, based on the real costs and/or on population based estimates from literature. Valuation of production losses will be based on a modificatiori of the friction cost method which has been developed in the ZonMW project "Measuring & valuing productivity costs in HTA".
Patient outcome analysis Both generic Quality of life (Q0L) and utilities are derived from the SF-36. An overall utility score for population based QoL can be obtained, which facilitates comparisons with other interventions, ie. the social tariff of the SF-36 (Brazier 1992(Brazier , 2002. The primary outcome measure for the cost-effectiveness analysis will be HM. The primary outcomes measure for the cost-utility measure will be utilities based on the SF-36 social tariff. The time horizon is 12 months. Ratios will be determined, based on incremental costs and effects of SFTRS compared to CTCS. The cost-effectiveness ratio will be stated in terms of costs per improvement on the FIM. The cost-utility ratio will focus on the net cost per QALY gained. Bootstrap re-sampling techniques (Briggs et al., 1997;O'Hagan, 2003) are used to explore cost-effectiveness uncertainty. Sensitivity analyses will be performed for the costs that turn out to have the largest impact on the differences in total costs between SFTRS and CTCS. In these analyses both the variarice in volumes and prices will be considered. The range over which uncertain factors are thought to vary will be assessed by calculating a minimum and maximum (mean value of costs minus or plus the SID).
-Systematic review Selection procedure: Assessment of these manuscripts resulted in 17 relevant studies. lncluded were: systematic reviews, trials reporting group treatments and care outcome, including functional outcome, health status, quality of life and multidisciplinary approaches. Not included were studies on specific Accident & Emergency room (ME) and ICU protocols, A&E and ICU diagnostics, pre-hospital life support, nutritional programmes, metabolic assessments, ventilation-related problems.
Validity assessment: Two reviewers independently assessed all studies for relevance and quality.

Results:
The total number of selected papers was 17, i.e. 4 (systematic) reviews, 0 ROT's, and 12 papers on clinical trials / follow-up studies, 1 paper reporting a costs study. Summary and conclusion: Eastwood (1999) reviewed 23 repor-ts on standardised measures of functional status in rehabilitation. Such measures can be used in determining outcome of interventions, managing rehabilitation services, providing empirical data for formal accreditation of care centres and guiding reimbursement of services, based on the patient's specific needs and severity of impairment. The author concluded that there is a lack of agreement on appropriate tools for functional assessment, in contrast to the urgent need to better understand service efficacy based on outcome measures and accountability in rehabilitation. Elliott (1999) reviewed 31 papers reporting measures used for patient outcomes assessment in adult ICU care. Apart from mortality rates, functional status, health status and quality of life measures were assessed. It was concluded that, although in the majority of survivors health status and activity levels seem to be similar at follow-up time relative to pre-trauma time, methodological flaws as to control over exclusion, heterogeneity of groups and losses to follow-up seriously hampered comparisons of findings. A strong recommendation for consensus on reporting patients' characteristics data as well as outcome measures to be used was given. The review by Cameron et al. (2004), including 9 reports, aimed at examining the (cost-) effectiveness of specialised multidisciplinary inpatient rehabilitation supervised by a geriatrician or rehabilitation physician compared with usual (orthopaedic) care for older patients with proximal femoral fracture. Al] main outcome measures tended to be better for patients receiving co-ordinated inpatient rehabilitation. However, statistically, no differences between groups was found, due to heterogeneity in length of stay 8-7o n Mw

Subsidieaanvraagformulier / Grant Application Form
Dossier nummer/ Dossier number: II1iI4 and cost data. None of the studies reported quality of life measures. It was concluded that, although further evidence is necessary, there is some rationale to adopt multidisciplinary inpatient rehabilitation. Furthermore, it was concluded that future trials involving specialised inpatient rehabilitation should aim at a) standardising outcome measures (especially regarding functional status); b) clearly recording patient characteristics, including cognitive status; c) assessing (cost-) effectiveness of multidisciplinary rehabilitation integrally, rather than attempting to evaluate its components. In the (non-systematic) review by Halcomb et al. (2005) 89 studies related to outcome after severe trauma were assessed. Inadequate support by current trauma systems; referral to rehabilitation facility being affected by a.o. provider preferences; long-term loss of productivity; a high incidence of psychological problems; negative effects of alcohol and drug abuse on outcome and positive effects of social support on recovery after severe trauma are reported. It is recommended that research into trauma rehabilitation interventions aiming at patients' individual needs and individual treatment goals is performed, taking into account the sustainability and (cost-) effective follow-up of such interventions. Van der Sluis et al. (1995), in a retrospective medical records study, assessed the functional outcome in multiple trauma patients (ISS >=16) (n=723). Age, gender, type of accident, AISIISS, discharge destination, length of hospital stay and functional outcome (Glasgow Outcome Scale) were retrieved. Off all patients 25.7% died. Half of the survivors were discharged home and 29% were transferred to a rehabilitation centre. Although the functional outcome deteriorated linearly with increasing AIS/ISS, the final functional result was good. Two years post-injury, 68% had mild or no disabilities, 19% were moderately and 7% severely disabled. Holbrook et al. (1998Holbrook et al. ( , 1999Holbrook et al. ( , 2001, in a prospective study, investigated the effects of major trauma on functional outcome and quality of welI-being. They found that major complications, occurring in 10.1% of all in-hospital patients (n820), encompassing foremost pulmonary, gastrointestinal, musculoskeletal problems and infections, led to significant lower quality of life levels at discharge as well as at 6, 12 and 18 months post-discharge. High functional limitation levels were reported in multiple trauma patients (n826) at 6 months follow-up. At 12 and 18 months follow-up (n=806 and n780 resp.) approx. 80% of all patients stilI scored well below healthy norm values as to quality of welI-being. Depression, serious extremity injury and length of stay were significant predictors of quality of well-being at 6, 12 and 18 months post-discharge. A negative association between functional outcome and extremity injury was found, in contrast to chest or abdominal injury. Czyrny et al. (1998) performed a retrospective study (n=33) to assess functional outcome of acute in-hospita] rehabilitation in patients with multiple limb trauma. They reported significant increases in functional health status (FIM) after comprehensive intensive early inpatient rehabilitation, mean FIM motor scores differences reaching 28.7 between admission and discharge (p<0.05) after an average rehabilitation period of 27 days. The authors mention a.o. the small group size and the lack of a follow-up data set as limitations to their study. Richmond et al. (1998) investigated predictors of disability in persons after non-neurological trauma (n=109) using questionnaires and medical record data. They found that extremity injuries, high levels of posttraumatic psychological distress and educational level are risk factors for severe disability at 3 months post-discharge (odds ratios 2.9, 2.9 and 3.4 resp.). The study by Fern et al. (1998) compared resource utilisation and long-term outcome in a group of patients having multiple extremity injuries (n=54) and a control group of patients with other major (non-neurological) injuries (n=18) at a level 1 trauma centre. Results showed that average length of stay was 92% Ionger and resource usage doubled in the first group. After discharge the first group experienced greater long-term disability and higher productivity losses. The authors conclude that the trauma patients with multiple extremity injuries are a distinct group with special needs, that hospita] resources for this group are underestimated and that new instruments to predict health status and resource utilisation are necessary. Miller et al. (2000) investigated functional outcome in severely injured patients staying more than 3 weeks in an intensive care unit (ICU) (n=115). Approx. 47% of patients were discharged to a Ø-Zo n Mw

Subsidieaanvraagformulier / Grant Application Form
Dossier nummer/ Dossier number: II] Il iii ; rehabilitation centre with a mean FIM score of 52. After rehab discharge (+1-48 days) mean FIM score was 86 and improved to 101 at 3 month follow-up. For the rehab group, no dear differences in mean FIM related to age were found. Erli et al. (2000) investigated predictors of quality of life after multiple trauma (n=173). Results showed that quality of life not only dependent on injury severity, but also on ventilation time, age and psychosocial factors. Rehabilitation highly correlated with long-term quality of life. Shortening prolonged in-patient time also significantly improved quality of life in multiple trauma patients. Dimopoulou et al. (2004) evaluated health-related quality of life and disability in multiple trauma patients (n=87) up to 1 year post-ICU. Results indicate that, next to somatic complaints, emotional weIl-being was affected. Approx. 72% of patients experienced problems with daily activities, 47% reported an inability to work whereas 59% feit moderately to severely disabled. Vies et al (2005) studied the prevalence and determinants of disability and return to work after severe trauma. In the group of survivors (n=196) problems with mobility (34%), self-care (15%), daily activities (51%), pain & discomfort (58%), anxiety & depression (37%) and cognitive ability (57%) were reported. Quality of life scores were well below normative values. They concluded that a quarter of trauma patients of working age were unable to return to work and more than half of the trauma patient group experienced problems in their daily life. The study of Holtslag et al. (2007) aimed at assessing long-term functional health status in severely injured patients and comparing results to normative data (n=335). Furthermore, relations between functional health status and patient characteristics were assessed. Most problems reported related to work, ambulation, housekeeping, recreation and alertness. Deviation from norm data seemed inversely related to age. Important predictors of psychosocial and physical functioning were age, co-morbidity and injury type. Husler et al. (2006) investigated the direct costs and consequential costs of 63 major trauma survivors with an ISS>1 3. Disability is the most important factor determining trauma costs. 35% of all major trauma patients are permanently disabled and cost in excess of 8.7 times as much as those patients (65%) who recover and are rehabilitated. On average, 2/3 of trauma costs are a result of production loss and other consequential costs. Costs like primary hospital treatment costs represent a minor fraction of the total costs. The authors conclude that it is essential that medical and political decision-makers adopt a comprehensive socio-economic view of trauma costs.
In conciusion, most clinical studies compared single treatment outcome. Only few studies comparing outcome of different treatment regimes in multiple trauma care, either experimental or control regimes, were reported. No ROT on treatment effects in multiple trauma patients was found. Although many studies report the multidisciplinary character of the problems multiple trauma patients encounter, in none of the reviewed studies the multidisciplinary rehabilitation program was described in detail. One study suggests that reducing Iength of stay may positively influence quality of life. Both methodology and measures used in outcome measurement in multiple trauma studies differ considerably. In many cases reliable data on variation in group data are lacking due to either small group sizes or non-reporting. Several reviews point out the necessity of standardising outcome measures as wei] as assessing (cost-) effectiveness of multidisciplinary rehabilitation.

II] 1 Il 1 li1
international publications on economic evaluations on several subjects, a.o. on costing methods. She is an assistant professor at the department of Health Organisation Policy and Economics. A number of projects in which she was co-applicant were refunded mainly on international comparison (OECD Ageing Related Disease project, and the project Cross-national transferability of cost-effectiveness data) and on transferral of organizational care. She is a.o. a member of the Cochrane Health Economics Methods Group and the Mental Health Economics Europe.
Harmen vd Linde, PhD, is Consultant Rehabilitation at the UMC St. Radboud and SMK, Nijmegen. His research interest is in Prosthetics and multitrauma. He is strongly involved in the development of P&O guidelines. He is chairman of the Dutch board of the International Society for Prosthetics & Orthotics, charman of the committee for medical aids of the VRA, and member of the scientific committee of the VRA. He (co-)authored numerous papers on teh above mentioned subjects.