consequences of a hip fracture among old...
TRANSCRIPT
Consequences of a hip fracture among old people
Monica Långström Berggren
Department of Community Medicine and Rehabilitation, Geriatric Medicine
Umeå 2017
Responsible publisher under Swedish law: the Dean of the Medical Faculty This work is protected by the Swedish Copyright Legislation (Act 1960:729) Dissertation for PhD ISBN: 978-91-7601-786-9 ISSN: 0346-6612 New Series No: 1923 Cover photo: Monica Långström Berggren Electronic version available at: http://umu.diva-portal.org/ Printed by UmU Print Service, Umeå University Umeå, Sweden 2017
“Be aware of what you’ve got. A grateful mind is a powerful mind.”
Nico & Vinz
i
Table of Contents
ABSTRACT ...................................................................................... iii SVENSK SAMMANFATTNING (SUMMARY IN SWEDISH) ............... v ABBREVATIONS ............................................................................ vii ORIGINAL PAPERS ........................................................................ ix INTRODUCTION .............................................................................. 1 BACKGROUND ................................................................................ 2
Hip fracture ...................................................................................................................... 2 Falls and fractures ............................................................................................................ 5 Complications after hip fracture ...................................................................................... 7 Orthogeriatric care models .............................................................................................. 9 Team-based home rehabilitation ................................................................................... 18
RATIONALE .................................................................................... 21 AIMS .............................................................................................. 22
Paper I. ........................................................................................................................... 22 Paper II. ......................................................................................................................... 22 Paper III. ........................................................................................................................ 22 Paper IV. ........................................................................................................................ 22
METHODS ..................................................................................... 23 Ethical approval ............................................................................................................. 24 Study 1 (Papers I and II) ................................................................................................ 24
Settings and participants ....................................................................................... 24 Procedure ................................................................................................................. 26 Data collection .......................................................................................................... 27 Assessments ............................................................................................................. 28 Intervention ............................................................................................................. 32 Statistical analyses ................................................................................................... 37
Study 2 (Papers III and IV) ........................................................................................... 38 Settings and participants ....................................................................................... 38 Procedure ................................................................................................................. 40 Data collection ......................................................................................................... 42 Assessments ............................................................................................................. 42 Intervention ............................................................................................................. 43 Statistical analyses .................................................................................................. 45
RESULTS ....................................................................................... 47 Paper I ............................................................................................................................ 49 Paper II ........................................................................................................................... 51 Paper III ......................................................................................................................... 58 Paper IV ......................................................................................................................... 60 Additional results .......................................................................................................... 62
ii
DISCUSSION ................................................................................. 63 Falls and fractures ......................................................................................................... 63 Complications, readmissions and mortality ................................................................. 66 Walking ability and LOS ................................................................................................ 70 Ethical considerations .................................................................................................... 72 Methodological considerations ...................................................................................... 73 Clinical implications ....................................................................................................... 75 Implications for future research .................................................................................... 76
GENERAL CONCLUSIONS ............................................................. 78 ACKNOWLEDGEMENTS ................................................................ 79 REFERENCES ................................................................................ 81 PAPERS I-IV .................................................................................. 99
iii
ABSTRACT This thesis has three related aims: to describe the comorbidities, complications and causes of death among a representative population of old people with hip fracture; to evaluate whether a successful care model using comprehensive geriatric assessment, which reduced in-hospital falls and complications such as delirium, decubital ulcers and infections, has an effect on falls after discharge; and to assess whether adding home rehabilitation to the existing care model has an effect on length of stay in hospital, walking ability, complications and readmissions after discharge. Previous research does not represent the entire hip-fracture population since those living in institutions and those with dementia/cognitive impairment are often excluded. Furthermore, orthogeriatric care models have had limited effect after discharge, mortality has remained high and knowledge is sparse about causes of death and complications after discharge. In addition, there are few randomized controlled multi- or inter-disciplinary team-based home rehabilitation studies, and data concerning complications and readmissions after discharge in those studies are limited. Only half of the hip-fracture population regain their prefracture level of mobility, which calls for further rehabilitation studies. An orthogeriatric care model was evaluated, to assess whether it had any continuing effect on falls after discharge, in a randomized controlled study comprising 199 people with acute femoral neck fracture aged ≥70 years. Falls and new fractures as well as fall-incidence rate were compared between an intervention and a control group during a one-year follow-up. Poisson regression adjusted for overdispersion and observation time was used in the analyses. Comorbidities, complications and causes of death were described, and a multivariate analysis adjusted for potential confounders was used to analyze risk factors for all-cause mortality during a 3-year follow-up. After one year 44 participants had fallen and there were 138 falls in the intervention group compared with 55 participants who fell and 191 falls in the control group. The crude postoperative fall incidence was 4.16/1 000 days in the intervention group vs. 6.43/ 1,000 days in the control group. The incidence rate ratio was 0.64 (95% CI: 0.40–1.02, p=0.063). In total, 136 participants suffered at least one urinary tract infection; 114 suffered 542 falls and 37 sustained 56 new fractures, including 13 hip fractures, during a 3-year follow-up, and 79 out of 199 participants (40 %) died. Cardiovascular events (23 %), dementia (24 %), hip-fracture (19 %) and cancer (13 %) were the most common primary causes of death. Multivariate analyses revealed that cancer, dependence in personal activities of daily living, cardiovascular disease and dementia at baseline, pulmonary emboli and cardiac failure during hospitalization were all independent predictors of 3-year mortality.
iv
Individually designed Geriatric Interdisciplinary Home Rehabilitation (GIHR) aimed at early hospital discharge was compared to conventional geriatric care in a randomized controlled study including 205 people with an acute hip fracture, aged ≥70 years. The use of walking devices and walking ability was assessed in interviews and gait speed was measured over 2.4 meters. Length of stay in hospital, complications, readmissions and days spent in hospital after initial discharge were compared between intervention and control groups. Binary logistic models adjusted for hypothetical confounders were used to analyze walking ability, the use of a walking device and the risk of falling after discharge. No significant differences were observed in walking ability, use of walking device, and gait speed at the 3- and 12-month follow-ups between the groups. At 12 months, 56.3% of the intervention group and 57.7% of the control group had regained or improved their prefracture walking ability. The median postoperative length of stay in the geriatric ward was 6 days shorter for the intervention group (p=0.003). Between discharge and the 12-month follow-up, comparisons between participants in the GIHR group and the control group were as follows; 46 (43.4%) vs. 38 (40.9%) fell (p=0.828); 13 (12.3%) vs. 6 (6.5%) suffered an additional fracture (p=0.250); 36 (34.0%) vs. 30 (32.3%) presented with an infection (p=0.917); 12 (11.3%) vs. 6 (6.5%) suffered a cardiovascular event (p=0.344); 38 (35.8%) vs. 27 (29.0%) were readmitted to hospital (p=0.383); and the median number of days spent in hospital in total was 11.5 vs. 11.0 (p=0.353). In conclusion, the successful care model, that reduced in-hospital falls and complications, did not have a prolonged effect on falls among older people with hip fracture, including people with cognitive impairment/dementia. This group of old people have multiple co-morbidities and suffer numerous complications. When home rehabilitation was added to the existing care model, regaining of walking ability in the short- and long-term was similar in both the GIHR participants and those receiving conventional geriatric care. In addition, the proportions of complications after discharge were no higher in the former group, even though their initial time in hospital was shorter. The results indicate that falls and complications after discharge are a major problem among old people with hip fracture. This thesis suggests that primary and secondary prevention of falls and fractures needs to become a part of routine care among old fall-prone people.
v
SVENSK SAMMANFATTNING (SUMMARY IN SWEDISH) Andelen äldre i befolkningen förväntas öka i framtiden och därmed också antalet personer som drabbas av en höftfraktur. En stor andel av de som drabbas har kognitiv nedsättning/demens och/eller bor på ett äldreboende och denna grupp har ofta exkluderats i tidigare forskning. Efter höftfrakturen är det en stor andel som inte återfår sina tidigare förmågor att klara sin personliga vård samt att kunna gå självständigt. Det akuta omhändertagandet vid en höftfraktur har förbättrats över tid men trots detta så finns det en ökad risk att dö i efterförloppet och denna risk har kvarstått oförändrat hög genom åren. Efter utskrivning från sjukhuset kan personer med höftfraktur drabbas av komplikationer som t.ex. infektioner och hjärtkärlsjukdomar. Dessa komplikationer samt orsakerna till varför vissa personer avlider är inte välstuderat. I och med minskade resurser och vårdplatser på sjukhusen så har hemrehabilitering utvecklats. Det finns dock få randomiserade kontrollerade studier med team-baserad hemrehabilitering och kunskap om komplikationer och återinläggningar i efterförloppet saknas. Denna avhandling har tre syften. Första syftet har varit att få en ökad kunskap om samsjuklighet, komplikationer samt orsaker till dödsfall bland äldre personer med höftfraktur. Andra syftet var att utvärdera om ett tvärvetenskapligt, multifaktoriellt vårdprogram som tidigare lyckats minska komplikationer och fallolyckor under vårdtiden på sjukhuset hade någon kvarstående effekt på fall efter utskrivning från sjukhuset. Tredje syftet var att utvärdera ett geriatriskt tvärvetenskapligt hem-rehabiliteringsprogram för personer med höftfraktur med avseende på vårdtid, gångförmåga, komplikationer och återinläggningar. Eventuell kvarstående effekt på fall upp till ett år efter fraktur utvärderades i en studie bland 199 personer 70 år eller äldre som drabbats av ett brott på lårbenshalsen. Deltagarna lottades till vård på en geriatrisk avdelning direkt efter operation eller sedvanlig vård på en ortopedisk avdelning. Vårdprogrammet på den geriatriska avdelningen fokuserade på att förebygga, identifiera och behandla komplikationer. I denna studie undersöktes också komplikationer samt orsaker till död upp till tre år efter fraktur i samma population. Under det första året efter frakturen föll 44 deltagare i interventionsgruppen 138 gånger jämfört med 55 personer med 191 fall i kontrollgruppen. Fallincidensen var 4,16/1000 vårddagar jämfört med 6,43/1000 vårddagar för interventions- respektive kontrollgrupp. Dessa skillnader var ej statistiskt säkerställda. Under de tre första åren efter fraktur drabbades 136 deltagare av minst en urinvägsinfektion, 114 personer föll 542 gånger och 37 personer fick 56
vi
nya frakturer varav 13 höftfrakturer. Efter tre år hade 79 personer avlidit och de vanligaste orsakerna till död var hjärtkärlhändelser, demens, höftfraktur och cancer. Cancer, beroende av hjälp i personlig vård, hjärtkärlsjukdom och demens samt att drabbas av propp i lungan och/eller att få hjärtsvikt var associerat till död. Ett individuellt anpassat geriatriskt tvärvetenskapligt hem-rehabiliteringsprogram utvärderades i en studie med 205 deltagare 70 år och äldre med en nytillkommen höftfraktur. Deltagarna lottades till delta i hemrehabilitering efter utskrivningen eller sedvanlig geriatrisk vård. Hemrehabiliteringsprogrammet syftade till att personen skulle skrivas hem från sjukhuset tidigare och få fortsatt vård och rehabilitering i hemmet. Gångförmågan och behovet av gånghjälpmedel undersöktes med intervjuer dessutom mättes gångförmågan med självvald och maximal gånghastighet över 2,4 meter. Gångförmåga, behovet av gånghjälpmedel och gånghastighet skilde sig inte åt mellan grupperna vid uppföljande bedömningar vid 3 och 12 månader efter fraktur. Vid 12 månader hade 56,3% i interventionsgruppen och 57,7% i kontrollgruppen återfått eller förbättrat sin gångförmåga jämfört med innan frakturen. Vårdtidens längd var 6 dagar kortare i interventionsgruppen jämfört med kontrollgruppen vid jämförelse av medianvårdtid. Inga skillnader avseende komplikationer, återinläggningar och antal dagar på sjukhuset efter utskrivningen kunde noteras mellan interventions- och kontrollgruppen. Sammanfattningsvis hade det tvärvetenskapliga, multifaktoriella vård-programmet som framgångsrikt minskat fallolyckor och komplikationer under tiden på sjukhuset inga kvarvarande effekter på fall efter utskrivningen. Ett geriatriskt tvärvetenskapligt hemrehabiliterings-program kan vara ett komplement till vård och rehabilitering på sjukhus. Deltagarna återfick sin tidigare gångförmåga på kort och lång sikt jämförbart med de som erhöll sedvanlig geriatrisk vård. Trots en kortare vårdtid initialt fick deltagarna inte fler komplikationer efter utskrivningen. Äldre personer med höftfraktur har många andra sjukdomar, drabbas av många fall och komplikationer samt återinläggs ofta på sjukhus. För att påverka detta bör fall och frakturförebyggande åtgärder bli en del av rutinvården av äldre fallbenägna personer.
vii
ABBREVATIONS ADL Activities of Daily Living
ASA American Society of Anesthesiologists
BBS Berg Balance Scale
CCS Charlson Comorbidity Score
CGA Comprehensive Geriatric Assessment
CI Confidence Interval
DHS Dynamic Hip Screw
DSM-IV Diagnostic and Statistical Manual of Mental Disorder, fourth edition
DRD Discharge-Ready Date
EpC Centre for Epidemiology
FNF Femoral Neck Fracture
FRAX Fracture Risk Assessment Tool
GDS Geriatric Depression Scale
GIHR Geriatric Interdisciplinary Home Rehabilitation
HIFE High Intensity Functional Exercise program
HR Hazard Ratios
HR Home Rehabilitation
I-ADL Instrumental Activities of Daily Living
IRR Incidence Rate Ratio
IQR Interquartile Range
LIH Lars Ingvar Hansson (hook pins)
LOS Length of Stay
MMSE Mini-Mental State Examination
NA Non-Assessed
Nbreg Negative binomial regression
NS Non-Significant
NSQIP National Surgical Quality Improvement Program
OBS Organic Brain Syndrome
viii
OR Odds Ratio
OT Occupational Therapist
P-ADL Personal/Primary Activities of Daily Living
PT Physiotherapist
PTF Pertrochanteric Fracture
RCT Randomized Controlled Trial
S Significant
S-COVS Swedish version of Physiotherapy Clinical Outcome Variables
SD Standard Deviation
STF Subtrochanteric Fracture
WHO World Health Organization
ix
ORIGINAL PAPERS The thesis is based on the following papers. In the text they will be referred to by their Roman numerals:
I. Evaluation of a fall-prevention program in older people after femoral neck fracture: a one-year follow-up. Berggren M, Stenvall M, Olofsson B, Gustafson Y. Osteoporos Int. 2008 Jun; 19(6):801-9
II. Co-morbidities, complications and causes of death among people with femoral neck fracture - a three-year follow-up study. Berggren M, Stenvall M, Englund U, Olofsson B, Gustafson Y. BMC Geriatr. 2016 Jun 3;16:120.
III. Effects of geriatric interdisciplinary home rehabilitation on walking ability and length of hospital stay after hip fracture: A randomized controlled trial. Karlsson Å, Berggren M, Gustafson Y, Olofsson B, Lindelöf N, Stenvall M. J Am Med Dir Assoc. 2016 May 1;17(5):464.e9-464.e15.
IV. Geriatric interdisciplinary home rehabilitation - effects on complications and readmissions after hip fracture: A randomized controlled trial. Berggren M, Karlsson Å, Lindelöf N, Englund U, Olofsson B, Nordström P, Gustafson Y, Stenvall M. Submitted.
The original articles have been reprinted with the kind permission of the respective publishers.
1
INTRODUCTION Among old people, a hip fracture is usually a low-energy trauma, that is, an injury sustained from a fall from standing height or less. It is a serious incident which may affect a person’s life in many ways. The typical hip fracture patient is a woman aged 82 years, living in her own home, with three or more comorbidities including cognitive impairment and/or a heart disease. She was able to walk independently indoors before the fracture, and the fracture was caused by a fall in her own home. After the fracture her prognosis is poor; she has a high risk of falling again, she might become dependent in Activities of Daily Living (ADL) and she might never regain her prefracture mobility level.
2
BACKGROUND
Hip fracture Hip fracture epidemiology Throughout the world the populations are “ageing”, due to declining mortality and to fertility rates; that is, the proportion of the population above a certain age is rising. In 2017 worldwide there were 962 million people aged 60 or over, comprising 13 per cent of the global population. This age group is increasing at a rate of 3.0 per cent per year.1 The structure of the Swedish population is also changing, with the number of older people growing as a result of increased life expectancy and the assumption is that it will increase further in the future. It is estimated that in 20452 more than one million people in Sweden will be aged 80 years or older and the annual number of hip fractures in the country is expected to almost double during the first half of this century.3 With these demographic changes in mind, epidemiological studies have been carried out which show that the age-specific incidence rate in hip fractures rose during the last half of the twentieth century, and later reached a plateau or declined.4-6 The reasons for these changes are not entirely clear. A recent study from Norway describes a continuing age-adjusted decline in hip fracture rates,7 although the prevalence is still expected to rise in the future due to changes in population size and age distribution.3,8 A Swedish population-based study shows an increased incidence among the oldest old, partly due to a larger number of older individuals.9 It is calculated that women in Sweden have a 20% lifetime risk of sustaining a hip fracture.10 Almost 18 000 people suffer a hip fracture each year in Sweden, at the mean age of 82 years and 68% are women. The proportion of men has increased from 28 % in 1996 to 32 % in 2011. Less than half of all those who sustain a hip fracture, (44 %) are living alone and 70% live in ordinary housing before the fracture.11 In addition, cognitive impairment/dementia is common among old people with hip fracture.12 In 2015, the mean length of stay in acute care in Sweden was 8.6 days and the median length of stay was 7 days in 2015.11 Risk factors for hip fracture A hip fracture in older individuals, caused by a “low energy trauma”, is usually defined as a fragility fracture and most of these individuals have a
3
low bone density, or osteoporosis, as an underlying risk factor for fracture. The fracture is a result of mechanical forces that would not ordinarily result in fracture (nice.org.uk). The Fracture Risk Assessment Tool (FRAX), which is used in clinical practice to calculate fracture risk, includes established risk factors such as; age, gender, body mass index, ethnicity, smoking, previous low trauma fracture, parental hip fracture, glucocorticoid use, rheumatoid arthritis, secondary osteoporosis, alcohol use and bone mineral density.13 However, FRAX does not include the individual’s risk of falling and the number of previous fractures. Having had a stroke14 or having dementia15 is also associated with a higher risk of hip fracture, probably due to the proneness to fall and, among individuals with stroke, bone loss on the paretic side.16 A large cohort study, including people ≥85 years, by Wiklund et al. also describes several risk factors for hip fracture.17 Age, Parkinson´s disease, currently smoking, underweight, delirium in the preceding month and help from no more than one person when walking indoors are all independent risk factors, whereas a bilateral hip prostheses seem to protect against hip fracture. Consequences of hip fracture Hip fracture is a significant cause of morbidity and mortality worldwide.18 Such a fracture might lead to considerable changes in a person’s life, with both personal and social consequences,19,20 and many become afraid of falling again.21 Only half of old individuals with hip fracture regain their pre-fracture level of mobility22 and only 40-70% regain their level of independence in basic activities, depending on population studied.23,24 These consequences might lead to old individuals losing their independence, and to some having to move into residential care facilities.25,26 Following acute care, 32% are discharged to ordinary housing, 18% to a residential care facility and 35% to a rehabilitation department, according to the Swedish National Hip Register.11 At four months after fracture 57% live in their own home.11 In addition, a hip fracture is also associated with increased mortality.27,28 It has been shown that older people have a 5- to 8-fold increased risk of dying during the first 3 months after a hip fracture,27 and most deaths occur within the first 6 months.29 In Sweden, 3.5% of those suffering a hip fracture die during acute hospital care.11 A review by Abrahamsen et al. reported in-hospital mortality ranging from 2.3% to 13.9 %.28 Five years after fracture the mortality is 20% above the expected level,29 a figure which has remained stable over the past 40 years.30 For society, hip fracture is already a huge economic burden25 and is predicted to continue to be so in the future.31
4
Hip fracture surgery ‘Hip fracture’ as a term includes fractures of the upper (proximal) part of the femur, and the two main types are the femoral neck fracture (cervical fracture or intracapsular) and fractures distal to the femoral neck (extracapsular or trochanteric). The intermediate types are called basocervical fractures.32 Cervical fractures can be divided into displaced and undisplaced fractures according to the Garden classification.33 Garden I-II include undisplaced fractures and Garden III-IV include displaced fractures. Almost 50% of hip fractures are cervical fractures and the remainder are trochanteric fractures, including 8% subtrochanteric fractures.11 This thesis is concerned with cervical and trochanteric fractures (Figure 1).
Figure 1. Undisplaced and displaced cervical fracture, and trochanteric fracture.
During the study periods undisplaced femoral neck fractures were operated on using internal fixation, and displaced fractures using mainly hemiarthroplasty, where only the femoral part is replaced, while a dynamic sliding-screw device was used for the basocervical and pertrochanteric fractures (Figure 2). Since 2000, a medullary nail11 has been used more often for trochanteric fractures with two or more
5
fragments, and some of the participants in Papers III and IV (n=18) were treated using this method.
Figure 2. Undisplaced cervical fracture operated on using internal fixation, two Hook pins (LIH), displaced cervical fracture operated on using hemiarthroplasty (HAP) and trochanteric fracture operated on using dynamic sliding-screw device (DHS).
Falls and fractures Fall epidemiology Approximately 30% of people above the age of 70 living in ordinary housing fall each year and around 4-6% of falls result in a fracture,34,35 with 1-2 % of the falls causing a hip fracture.34,36 In almost all cases a hip fracture is the result of a fall.37,38 The risk of falling again after a hip fracture is substantial.39,40 In a study by Colon-Emeric et al. the risk of sustaining a subsequent fracture after an initial hip fracture was increased 2.5-fold,41 and in a population-based study 21% of those who had a hip fracture had previously suffered 2 or more hip fractures.42 Around 7% of those who suffer a hip fracture fell in hospital.43 In Sweden, more than 70000 people were hospitalized and approximately 1600 died due to fall accidents in 2013.44
6
Risk factors for falls The majority of falls are due to the interacting of multiple risk factors, only a few have a single cause.34 In a study by Campbell et al. 15% of the falls resulted from an external event that would cause most people to fall, another 15% could be attributed to a single identifiable cause such as syncope, and the remaining falls were caused by multiple interacting factors.45 Risk factors for falls can be classified in a variety of ways. One is to differentiate between predisposing and precipitating factors, where the former represent “chronic factors” that increase the predisposition for falls, and the latter “acute factors” that trigger a fall. Another way of classifying risk factors for falls is to ascribe them to individually-related (intrinsic) or environmental factors (extrinsic).46 The category of intrinsic risk factors includes advanced age, previous falls, muscle weakness, gait and balance problems, poor vision and chronic diseases (for example; Parkinson’s, stroke, diabetes, dementia and incontinence), whereas the category extrinsic factors comprises environmental factors such as slippery or uneven surfaces, lack of hand rails, poor lighting, use of walking aids and poor footwear.47 There are also terms that have been used to describe both internal and external risk factors without any assessment of whether or not the risk factors are precipitating, “situational risk factors” or “circumstances that are hazardous regarding falls”.48,49 Previous research identifies several fall-risk factors, such as co-morbidity, cognitive impairment, functional disability, previous falls, use of drugs, fear of falling and aging.34,47,50,51 Acute disease or symptoms of disease, such as infections and delirium are examples of precipitating factors that are present in 23-39% of falls.36,52 Among in-patients 45% of falls are associated with delirium,40 and more than three out of four falls in people with dementia are caused by drug side-effects and acute disease or symptoms of disease.53 Only a small proportion of falls are witnessed by someone else.49,54 In general, the more risk factors that apply to a person, the greater their risk of falling.34 Fall prevention Previous research shows that both single and multifactorial fall-prevention interventions for older people are successful,55,56 though in some cases the results are conflicting. Interventions have been carried out with individuals in various settings, including those living in residential
7
care facilities, in the community, and in hospital. Current guidelines all recommend multifactorial interventions to prevent falls.57-59 A multifactorial intervention includes an assessment of the individual´s risk factors for falling, followed by specific interventions targeting those risk factors and, consequently, the same combination of interventions is not applied to everyone.56 For example, one person may receive supervised exercise and home-hazard modification whereas another may receive home-hazard modification and a medication review. According to a Cochrane report by Cameron et al., there is evidence that multifactorial interventions reduce the number of falls in hospitals. A possible benefit from multifactorial interventions in care facilities is suggested although the evidence is inconclusive. Exercise as a single intervention, appears effective in subacute hospital settings but, in contrast, the results from exercise interventions in care facilities are inconsistent and overall do not show any benefit.55 Fall interventions among individuals living in the community, such as exercise programs, home safety interventions, and multifactorial assessments and interventions, have reduced the rate of falls, according to a Cochrane report by Gillespie et al.56 Another approach to preventing fractures is to wear hip protectors. A multifactorial intervention program in residential care facilities resulted in a greater proportional reduction in hip fractures than in falls, suggesting that hip protectors contributed to the results, since no femoral fractures occurred among those wearing hip protectors.60 However, their effectiveness in preventing hip fractures in older people is questioned, and compliance is poor due to discomfort.61,62 Complications after hip fracture Complication epidemiology In general, adverse in-hospital events occur in approximately 9% of all people admitted, of these, operation- and drug-related events comprised approximately 50%, according to a review by Vries et al.63 Complications after a hip fracture can be divided into orthopedic (hip related) and medical (general), and in some studies into major and minor. The incidence of complications among people suffering a hip fracture varies from 12-33%.64-67 Common complications and the impact of complications Previous research has reported postoperative complications using a wide range of definitions, making comparison difficult.68,69 In a study by
8
Merchant et al. delirium and infection were the most common postoperative complications,66 while in another cohort study heart failure and chest infections were found to be the most common.65 In a study by Hansson et al. complications were reported until 6 months after fracture, with falls, fractures and pneumonia being the most common complications. They also found that complications were associated with loss of function.70 A study by Molina et al. describes the most common complications after hip fracture surgery as being myocardial infarction, sepsis, urinary tract infection and pneumonia. They reported data from the American College of Surgeons National Surgical Quality Improvement Program (NSQIP), a large database including complications up to 30 days post-surgery.71 In addition, complications have been found to be associated with a prolonged Length of Stay (LOS),66 and increased mortality.72 Risk factors for complications Several risk factors for postoperative complications have been described, for example: age, female gender, dependence in mobility,66 poor nutrition,73 and ≥3 comorbidities.65,66 Belmont et al. reviewed data from the National Trauma Data Bank in the US. They reported risk factors such as shock, dialysis, obesity, respiratory disease, cardiac disease, diabetes and ≥2 days’ delay in carrying out the procedure, were associated with postoperative complications.64 Age as a risk factor for complications after hip fracture, was confirmed by Jameson et al. who conclude that people ≥85 years had a higher risk of postoperative complications and a longer LOS than younger individuals.74 It has also been shown that the American Society of Anesthesiologists (ASA) classification of medical co-morbidities is associated with complications, whereby people in a higher ASA class have a greater risk of complications compared to those in a lower ASA class.71,75,76 Furthermore, Sathiyakumar et al., analyzed complications reported in NSQIP and described different combinations of demographic and clinical risk factors as being associated with complications, depending on type of surgery.67 In addition, a cohort study by Roche et al. shows that cardiovascular disease is a strong predictor of post-operative cardiac failure and that respiratory disease predicts chest infection.65 A more recent study by Sathiykumar et al., also using data from NSQIP, reports that several risk factors are significantly associated with adverse postoperative cardiac events: history of cardiac disease, stroke, chronic obstructive pulmonary disease, renal failure and peripheral vascular disease.77
9
Risk factors for death and causes of death Several risk factors have been associated with death following hip fracture: older age and male sex; severe systemic disease; pre-fracture functional impairment; cognitive decline; coronary heart disease and the number of co-morbidities.78-83 In an earlier Swedish study, a high ASA score was also shown to be associated with mortality.81 A large cohort study by Castronuovo et al.,84 reported that heart disease was a strong risk factor for mortality within 30 days but decreased beyond that. In addition, a study analyzing medications, found that the use of statins had a positive association with survival while the use of diuretics and a history of coronary heart disease were associated with death.83 The causes of death are described in only a few studies.85-89 In one autopsy study of old people with hip fracture by Perez et al. the most common causes of death were chest infection, cardiac failure, myocardial infarction and pulmonary embolism.89 Among nursing-home residents with hip fracture the most common causes of death were infection, dementia and cardiac events86 and, in more recent studies, cardiac and infectious diseases.87,88 The causes of death vary depending on the time point. In a study examining post-mortem reports, among individuals who died within 30 days after hip fracture, respiratory infections and cardiovascular disease were found to be the main causes of death.88 These results are in line with earlier studies.85,89 During follow-up, infections, cardiovascular events, dementia and cancer are the most common causes of death according to previous studies.86,87,90
Orthogeriatric care models The role of the geriatrician in the management of older individuals with hip fracture was described in the United Kingdom more than 50 years ago.91 In the 1980s in Sweden, a study in Malmö randomized hip fracture patients to rehabilitation on a geriatric ward or to conventional care on an orthopedic ward, and reported that the alternatives were comparable.92 During the same decade, an intervention study was performed in Umeå, including a geriatrician at the orthopedic ward. The intervention focused on prevention, detection and treatment of complications associated with postoperative delirium and the result reported was a significantly lower incidence of delirium and shorter LOS.93 Furthermore, an orthogeriatric care model by Lundström et al. significantly reduced delirium and complications post-operatively.94 Subsequently, during the late 20th and early 21st centuries several intervention studies attempting to improve the care and rehabilitation for individuals suffering a hip fracture were conducted and the consensus concerning preoperative management, time
10
to surgery, operative management, surgical technique and postoperative care has led to today’s recommendations/guidelines, including a multidisciplinary team approach and an orthogeriatric care model.57,95-101 Both multidisciplinary and interdisciplinary teams are described in the literature. A multidisciplinary team works together with regular meetings and common goals; the team leader formulates the goals and outlines the treatment and rehabilitation process.102 The team members work within their own professional boundaries and with little knowledge of the other team members’ professions.103 The differences between multidisciplinary and interdisciplinary teams lie mainly in the process of decision-making, as the team members in interdisciplinary teams are more involved in formulating the goals.102 The current literature supports an interdisciplinary team approach, using Comprehensive Geriatric Assessment (CGA), where each team member is respected and decisions are made together.97,104 Most care models include both multidisciplinary and interdisciplinary elements,102 although, the exact nature of the relationship between team members is not always specified in earlier studies. In this thesis the terms multidisciplinary and interdisciplinary are used interchangeably. CGA is defined as “a multidimensional, interdisciplinary diagnostic process focused on determining the medical, psychological, and functional capabilities of a frail elderly person to develop a coordinated and integrated plan for treatment and long-term follow-up” according to Rubenstein et al.105 Furthermore, CGA involves a geriatric team usually comprising a physician, a licensed practical nurse, a registered nurse, a physiotherapist, and an occupational therapist; a dietician, a social worker and a psychologist might also be included.102 According to Sletvold et al., the team work concerns checklists and scales used when performing assessments and also evaluation of the rehabilitation process. These instruments include a variety of areas, such as the individual’s performance in daily activities, gait, balance, cognitive function, nutrition, polypharmacy, vision and hearing, communication, incontinence, fall risk, sleep disturbances, affective disorders, oral health, abuse and social ties. During the rehabilitation process the team meet regularly to ensure appropriate treatment, rehabilitation and discharge planning for each individual. The goal of geriatric rehabilitation is to restore the individual’s previous level of function and autonomy and also, if possible, to prevent or postpone functional decline and the need for institutional care.102 CGA in hospital settings has improved function, reduced mortality and the need for institutional care,106 and it has proven successful in both medical and
11
orthogeriatric in-hospital units, according to a recent review by Pilotto et al.107 Furthermore, CGA is described as the most cost-effective orthogeriatric model of care.108 Orthogeriatric care can be defined as “medical care for older patients with orthopedic disorders that is provided collaboratively by orthopedic services and programs catering for older people”.57 A variety of different care models have been reported but 4 main orthogeriatric care models appear in the literature.69,109,110 The first, usual/standard care, is where the patient is admitted to the orthopedic ward and a geriatric review is performed when requested. In the second model, there is regular input from a geriatric team on the orthopedic ward. The third model is one of geriatric-led care with orthopedic expertise is consulted when needed. The fourth, is co-managed orthopedic-geriatric care. To date, the best model of care for older persons with a hip fracture cannot be determined, although a model involving joint care is endorsed in current guidelines.57,98,100,101 Among the reports on the third and fourth orthogeriatric care models, few have reported in-hospital falls as an outcome, one study reported a reduced fall rate111 but no effect was seen in two other studies.112,113 Falls after discharge have been reported in one study, but then with a significant risk reduction.114 Regarding complications, some studies show a significant reduction of in-hospital complications,112,115-120 while others do not,113,121-125 complications after discharge have been sparsely reported. The orthogeriatric care model in this thesis reduced in-hospital complications,126 including significantly fewer decubital ulcers and urinary tract infections, and a lower incidence of delirium. The number of readmissions after discharge was reduced in one study by Boddaert et al.,112 but several other studies have failed to confirm this.113,116,118,119,121,125,127-129 Regarding in-hospital mortality, some studies report a significant effect,112,118,123,130,131 while others do not.111,113,116,117,119,122,125,127-129,132,133 Few studies report on 30-day mortality,119,123,134, and only one shows a significant reduction.134 The results concerning significant differences in long-term mortality are inconsistent.112,115,118-121,129-131,135,136 Orthogeriatric care models are described in the literature under different names such as, orthogeriatric units, co-managed geriatric fracture centers and geriatric hip-fracture clinical care pathways. Likewise, the orthogeriatric care model presented in this thesis, using CGA in hospital and in participants’ homes, is described using different terms in the papers included, even though the underlying concept is the same. Table 1 presents an overview
12
of orthogeriatric care models among hip fracture patients, including models 3 and 4.
13
Tab
le 1
. An
ove
rvie
w o
f or
tho
geri
atri
c ca
re m
odel
s am
ong
hip
fra
ctu
re p
atie
nts
, in
clu
din
g m
odel
s 3
and
4.
Tri
al
Des
ign
S
elec
tion
s O
utc
omes
an
d r
esu
lts
Com
men
ts
3
Miu
ra
200
9
US
A
(132
)
Bef
ore-
afte
r pr
ospe
ctiv
e n=
163
>
55
Incl
usi
on: p
roxi
mal
fem
ur
frac
ture
S
LO
S, s
urge
ry <
24h,
tota
l cos
ts
N
S in
-hos
pita
l mor
talit
y
N
A c
ompl
icat
ions
, fal
ls, m
obili
ty,
“Hip
frac
ture
ser
vice
”
Pre-
and
pos
tope
rati
ve c
are
11%
rea
dmis
sion
s at
30
days
aft
er
disc
harg
e, o
nly
repo
rted
for
inte
rven
tion
gro
up
3
Ste
nva
ll
200
7 S
wed
en
(111
,126
,137
)
RC
T
n=19
9 >
70
Incl
usi
on: F
NF
Exc
lusi
on: s
ever
e os
teoa
rthr
itis
, rhe
umat
oid
arth
riti
s, r
enal
failu
re,
path
olog
ical
frac
ture
, be
drid
den
befo
re fr
actu
re
S L
OS,
in-h
ospi
tal c
ompl
icat
ions
, fa
lls, r
egai
ning
P-A
DL,
mob
ility
N
S m
orta
lity,
rea
dmis
sion
s
Preo
pera
tive
car
e no
t inc
lude
d A
sses
sors
not
blin
ded
Mor
e pa
tien
ts w
alke
d in
door
s w
itho
ut
wal
king
aid
s at
12
mon
ths
in
inte
rven
tion
gro
up
3
Ad
un
sky
200
3, 2
011
Is
rael
(1
35,1
38)
Pros
pect
ive
coho
rt
n 1=
320
Ret
rosp
ecti
ve
n 2=
3114
≥
65
Incl
usi
on: F
NF,
PTF
, st
able
med
ical
sta
tus
Exc
lusi
on: r
ehab
<7
days
S L
OS,
1-y
ear
mor
talit
y, m
obili
ty a
t di
scha
rge
NA
com
plic
atio
ns, f
alls
, re
adm
issi
ons
Adm
itte
d on
ava
ilabi
lity
on b
eds 1
A
sses
sors
not
blin
ded 1
Pa
rtic
ipan
ts o
lder
wit
h m
ore
com
orbi
diti
es a
t bas
elin
e in
in
terv
enti
on g
roup
s
3 Maz
zola
20
11
Ital
y
(124
)
Obs
erva
tion
alqu
asi
rand
omiz
ed
n=26
1 ≥
70
Incl
usi
on: h
ip fr
actu
re
S ti
me
to m
obili
zati
on
NS
com
plic
atio
ns, L
OS
NA
rea
dmis
sion
s, fa
lls, m
orta
lity,
m
obili
ty
Com
pari
son
of tw
o gr
oups
of
orth
oger
iatr
ic c
are,
one
wit
h pr
eope
rati
ve s
tart
, red
uced
tim
e to
m
obili
zati
on in
pre
oper
ativ
e gr
oup
3 Wag
ner
20
12
Ch
ile
(1
15)
Pros
pect
ive
coho
rt,
retr
ospe
ctiv
e co
ntro
l n=
275
>65
Incl
usi
on: F
NF
Exc
lusi
on: p
atho
logi
cal
and
peri
pros
thet
ic fr
actu
re
S c
ompl
icat
ions
N
S L
OS,
mor
talit
y,
NA
falls
, rea
dmis
sion
s, m
obili
ty
Diff
eren
ces
in d
iagn
osis
of d
elir
ium
an
d an
emia
pos
tope
rati
ve
14
Tab
le 1
. An
ove
rvie
w o
f or
tho
geri
atri
c ca
re m
odel
s am
ong
hip
fra
ctu
re p
atie
nts
, in
clu
din
g m
odel
s 3
and
4.
Tri
al
Des
ign
S
elec
tion
O
utc
omes
an
d r
esu
lts
Com
men
ts
3
Del
la
Roc
ca 2
013
U
SA
(1
27)
Bef
ore-
afte
r re
tros
pect
ive
n=
146
≥65
Incl
usi
on: h
ip fr
actu
re
S L
OS,
cos
ts
N
S m
orta
lity,
rea
dmis
sion
s
NA
falls
, com
plic
atio
ns, m
obili
ty
“Hip
frac
ture
pro
toco
l” S
mal
l con
trol
gr
oup
3
Bod
dae
rt
2014
F
ran
ce
(112
)
Pros
pect
ive
coho
rts,
ex
tern
al
valid
atio
n co
hort
n=
334
>
70
Incl
usi
on: h
ip fr
actu
re
Exc
lusi
on: i
n-ho
spit
al,
peri
pros
thet
ic a
nd
met
asta
tic
frac
ture
s,
pati
ents
tran
sfer
red
to
anot
her
hosp
ital
bef
ore
surg
ery
S L
OS,
mob
ility
, 6-m
onth
mor
talit
y,
30-d
ay r
eadm
issi
ons,
in-h
ospi
tal
com
plic
atio
ns
NS
in-h
ospi
tal f
alls
Uni
t for
Pos
t-O
pera
tive
Ger
iatr
ic C
are
O
rtho
pedi
c co
hort
had
less
co
mor
bidi
ties
3
Wat
ne
20
14
Nor
way
(1
13)
RC
T
n=32
9
N
o ag
e lim
it.
Incl
usi
on: F
NF,
PTF
, STF
. E
xclu
sion
: mor
ibun
d on
ad
mis
sion
S lo
nger
LO
S, m
obili
ty a
t 4 m
onth
s am
ong
thos
e liv
ing
in th
eir
own
hom
e
NS
mor
talit
y, r
eadm
issi
ons,
in-
hosp
ital
com
plic
atio
ns, i
n-ho
spit
al
falls
Stra
tifie
d ac
cord
ing
to h
ousi
ng
Pre-
and
pos
tope
rati
ve c
are
3 Gu
pta
20
14
UK
(1
39)
Bef
ore-
afte
r Pr
ospe
ctiv
e N
=49
4
≥50
Incl
usi
on: h
ip fr
actu
re
Exc
lusi
on: p
erip
rost
heti
c fr
actu
re
S L
OS
NA
com
plic
atio
ns, m
obili
ty,
mor
talit
y, r
eadm
issi
ons,
falls
Pre-
and
pos
t-op
erat
ive
care
3
Pre
stm
o 20
15
Nor
way
(1
21,1
40)
RC
T
N=
397
≥
70
Incl
usi
on: h
ip fr
actu
re,
able
to w
alk
10 m
Exc
lusi
on: p
atho
logi
cal
frac
ture
, sho
rt li
fe
expe
ctan
cy, l
ivin
g in
nur
sing
ho
me
S lo
nger
LO
S, m
obili
ty a
nd A
DL
at 4
an
d 12
mon
ths
N
S in
-hos
pita
l com
plic
atio
ns, 1
2 m
onth
s m
orta
lity,
rea
dmis
sion
s,
tim
e to
sur
gery
N
A fa
lls
No
blin
ding
, onl
y pa
rtly
mas
ked
asse
ssm
ents
Pr
e- a
nd p
osto
pera
tive
car
e
15
Tab
le 1
. An
ove
rvie
w o
f or
tho
geri
atri
c ca
re m
odel
s am
ong
hip
fra
ctu
re p
atie
nts
, in
clu
din
g m
odel
s 3
and
4.
Tri
al
Des
ign
S
elec
tion
O
utc
omes
an
d r
esu
lts
Com
men
ts
4
Vid
an
200
5 S
pai
n
(118
)
RC
T
n=
319
≥65
Incl
usi
on: h
ip fr
actu
re
Exc
lusi
on: d
epen
denc
y in
al
l bas
ic A
DLs
, pat
holo
gic
frac
ture
, ter
min
al il
lnes
s,
inab
ility
to w
alk
befo
re
frac
ture
S in
-hos
pita
l mor
talit
y an
d m
edic
al
com
plic
atio
ns, m
obili
ty a
t 3 m
onth
s
NS
LO
S, lo
ng-t
erm
mob
ility
, re
adm
issi
ons,
1-y
ear
mor
talit
y
NA
falls
Stra
tifie
d by
pre
frac
ture
AD
L le
vel
4
Kh
asra
ghi
200
5 U
SA
(1
17)
Bef
ore-
afte
r re
tros
pect
ive
n=
510
>65
Incl
usi
on: F
NF,
PTF
, STF
E
xclu
sion
: pat
holo
gica
l fr
actu
re
S L
OS,
tim
e to
sur
gery
, in-
hosp
ital
m
edic
al c
ompl
icat
ions
NS
in-h
ospi
tal m
orta
lity
NA
falls
, rea
dmis
sion
s, m
obili
ty
4
Bar
one
20
06
It
aly
(130
)
Bef
ore-
afte
r pr
ospe
ctiv
e
n=81
9 ≥
70
Incl
usi
on: h
ip fr
actu
re
E
xclu
sion
: not
rep
orte
d S
in-h
ospi
tal a
nd lo
ng-t
erm
m
orta
lity
N
S L
OS
NA
com
plic
atio
ns, m
obili
ty,
read
mis
sion
s, fa
lls
One
con
trol
gro
up b
efor
e an
d on
e af
ter
the
inte
rven
tion
4
Sh
yu
200
5,20
08
, 20
10
Tai
wan
(1
14,1
29,1
41 )
RC
T
n=16
2 ≥
60
Incl
usi
on: h
ip fr
actu
re
E
xclu
sion
: sev
erel
y co
gnit
ivel
y im
pair
ed, w
eak
mus
cle
pow
er, t
erm
inal
ly il
l
S m
obili
ty a
nd P
-AD
L, r
isk
of fa
lls
NS
LO
S, m
orta
lity,
rea
dmis
sion
s
NA
com
plic
atio
ns
Follo
w-u
p in
hom
es b
y nu
rse
and
phys
ioth
erap
ist
4
Fri
edm
an
200
9
US
A
(116
)
Ret
rosp
ecti
ve
coho
rt
n=
314
≥
60
Incl
usi
on: p
roxi
mal
fem
ur
frac
ture
Exc
lusi
on: p
atho
logi
cal,
recu
rren
t, pe
ripr
osth
etic
, no
nope
rati
ve fr
actu
res
S L
OS,
tim
e to
sur
gery
, in-
hosp
ital
co
mpl
icat
ions
N
S in
-hos
pita
l mor
talit
y, 3
0-da
y re
adm
issi
ons
N
A fu
ncti
on, f
alls
Part
icip
ants
in G
eria
tric
Fra
ctur
e C
ente
r w
ere
olde
r, h
ad m
ore
com
orbi
d co
ndit
ions
and
dem
enti
a, a
nd m
ore
lived
in n
ursi
ng h
omes
4
Cog
an
2010
Ir
elan
d
(131
)
Bef
ore
afte
r R
etro
spec
tive
n=
201
>65
Incl
usi
on: h
ip fr
actu
re
S in
-hos
pita
l mor
talit
y, lo
nger
LO
S
N
S 1
-yea
r m
orta
lity
N
A fa
lls, c
ompl
icat
ions
, re
adm
issi
ons,
func
tion
Sign
ifica
nt d
iffer
ence
s in
sou
rces
of
adm
issi
on a
nd fu
ncti
onal
leve
ls a
t ba
selin
e
16
Tab
le 1
. An
ove
rvie
w o
f or
tho
geri
atri
c ca
re m
odel
s am
ong
hip
fra
ctu
re p
atie
nts
, in
clu
din
g m
odel
s 3
and
4.
Tri
al
Des
ign
S
elec
tion
O
utc
omes
an
d r
esu
lts
Com
men
ts
4
Gre
gers
en
2012
D
enm
ark
(128
)
Bef
ore-
afte
r re
tros
pect
ive
n=49
5 ≥
65
Incl
usi
on: F
NF,
PTF
S
LO
S
N
S in
-hos
pita
l and
3 m
onth
m
orta
lity,
3+
6 m
onth
s re
adm
issi
ons
N
A c
ompl
icat
ions
, fal
ls, f
unct
ion
Hig
her
rate
of r
eope
rati
ons
afte
r di
scha
rge
in in
terv
enti
on g
roup
4
Gon
zale
z-M
onta
lvo
2010
S
pai
n
(133
)
Pros
pect
ive
quas
i-ra
ndom
ized
n=
224
≥65
Incl
usi
on: h
ip fr
actu
re
S L
OS
N
S in
-hos
pita
l mor
talit
y, m
obili
ty a
t di
scha
rge,
dis
char
ge d
esti
nati
on
NA
falls
, rea
dmis
sion
s,
com
plic
atio
ns
Acu
te o
rtho
geri
atri
c un
it
Adm
issi
on to
war
d on
cal
l
4
Bib
er
2013
G
erm
any
(122
)
Bef
ore-
afte
r re
tros
pect
ive
n=28
3
>
60
Incl
usi
on: F
NF
wit
h H
PA
S L
OS,
tim
e to
sur
gery
N
S in
-hos
pita
l sur
gica
l co
mpl
icat
ions
and
mor
talit
y
NA
med
ical
com
plic
atio
ns, f
alls
, re
adm
issi
ons,
mob
ility
“Ger
iatr
ic F
ract
ure
Cen
ter”
Su
rgic
al c
ompl
icat
ions
pro
long
ed
leng
th o
f sta
y
4
Fli
kwee
rt
2014
N
eth
erla
nd
s
(123
)
Bef
ore-
afte
r pr
ospe
ctiv
e
n=40
1 ≥
60
Incl
usi
on: F
NF,
PTF
S
LO
S, in
-hos
pita
l mor
talit
y, fa
stin
g ti
me
NS
in-h
ospi
tal c
ompl
icat
ions
, 30-
day
mor
talit
y
NA
falls
, rea
dmis
sion
s, m
obili
ty
Com
preh
ensi
ve c
are
path
way
4
Su
hm
20
14
Sw
itze
rlan
d
(119
)
Bef
ore-
afte
r pr
ospe
ctiv
e n=
493
≥
65
Incl
usi
on: f
emor
al fr
actu
re
Exc
lusi
on: p
atho
logi
c fr
actu
re
S L
OS,
in-h
ospi
tal c
ompl
icat
ions
N
S m
orta
lity,
30-
day
and
1-ye
ar
read
mis
sion
s
NA
falls
, mob
ility
“Car
e pa
thw
ay”
Inte
rven
tion
gro
up
had
mor
e co
mor
bidi
ties
, a la
rger
pr
opor
tion
of t
hose
who
nee
ded
help
in
AD
L an
d of
thos
e liv
ing
in n
ursi
ng
hom
es
Com
plic
atio
ns g
rade
d
17
Tab
le 1
. An
ove
rvie
w o
f or
tho
geri
atri
c ca
re m
odel
s am
ong
hip
fra
ctu
re p
atie
nts
, in
clu
din
g m
odel
s 3
and
4.
Tri
al
Des
ign
S
elec
tion
O
utc
omes
an
d r
esu
lts
Com
men
ts
4
Fol
bert
20
17
Net
her
lan
ds
(120
)
Pros
pect
ive
coho
rt
n=13
85
≥70
Incl
usi
on: h
ip fr
actu
re
Exc
lusi
on: c
oxar
thro
sis,
pa
thol
ogic
al a
nd
peri
pros
thet
ic fr
actu
res,
S L
OS,
1-y
ear
mor
talit
y, in
-hos
pita
l co
mpl
icat
ions
NA
falls
, rea
dmis
sion
s, m
obili
ty
Part
icip
ants
in in
terv
enti
on w
ere
olde
r, h
ad m
ore
seve
re c
omor
bidi
ties
, an
d w
ere
mor
e in
stit
utio
naliz
ed
4
Hen
der
son
20
17
Irel
and
(1
36)
Bef
ore-
afte
r Pr
ospe
ctiv
e co
hort
n=45
4
Incl
usi
on: F
NF
Exc
lusi
on:
S L
OS,
1-y
ear
mor
talit
y
N
A c
ompl
icat
ions
, fal
ls,
read
mis
sion
s, m
obili
ty
4
Lam
b
2017
U
SA
(12
5 )
Bef
ore-
afte
r Pr
ospe
ctiv
e co
hort
n=
437
Incl
usi
on: h
ip fr
actu
re
N
S L
OS,
in-h
ospi
tal c
ompl
icat
ions
an
d m
orta
lity,
30-
day
read
mis
sion
s
NA
falls
, mob
ility
A c
linic
al p
athw
ay
S =
sig
nific
ant,
NS
= n
on-s
igni
fican
t, N
A =
non
-ass
esse
d, R
CT
= R
ando
miz
ed C
ontr
olle
d Tr
ial,
LOS
= L
engt
h of
Sta
y, C
CS
= C
harl
son
Com
orbi
dity
Sco
re, F
NF
= F
emor
al N
eck
Frac
ture
, PTF
= P
ertr
ocha
nter
ic F
ract
ure,
STF
= S
ubtr
ocha
nter
ic F
ract
ure,
HA
P =
Hem
iart
hrop
last
y.
18
Team-based home rehabilitation Since LOS in hospital has been shortened during the last few decades, accelerated discharge has been promoted and Home Rehabilitation (HR) teams have been set up. Evidence to support this organization of care is sparse.97,142,143 Most HR studies for individuals with hip fracture describe training supervised to varying degrees by a physiotherapist, but multi- or inter-disciplinary HR studies are few in the literature.97 Even though a substantial proportion of the hip fracture population are cognitively impaired and live in residential care facilities,12 they are underrepresented in earlier studies.96 Interdisciplinary team rehabilitation can improve performance in ADL and physical activity,97 a finding that is supported by Donohue et al. who found a trend towards successful outcomes in support of multidisciplinary HR after acute care.143 Studies have also found that older individuals, in ordinary housing and without severe cognitive impairment who participate in HR, can increase their independence and confidence in performing ADL without falling,144,145 and that the burden on their caregivers is reduced.146 Falls, mortality and readmissions after discharge have been reported in a few studies though none show significant effects.145,147-150 A significant effect on walking ability has been reported in some HR studies,144,149,151 while others found no effects.147,148,152 Furthermore, studies have reported that HR can reduce hospital stay,97,148,153 but some have not.144,147,151 According to a Swedish review no effect on mortality has been shown and scientific data concerning complications are deficient.97 Table 2 presents an overview of team-based HR studies.
19
T
able
2.A
n o
verv
iew
of
team
-bas
ed h
ome
reh
abil
itat
ion
stu
die
s.
Tri
al
Des
ign
an
d
sett
ings
Sel
ecti
on
Ou
tcom
es
Inte
rven
tion
C
oncl
usi
on
Tin
etti
19
99
U
SA
(1
47,
154)
RC
T, h
ome-
base
d vs
. us
ual c
are
n=
304
≥65
Incl
usi
on: h
ip fr
actu
re
Exc
lusi
on: d
emen
tia,
term
inal
ill
ness
, >25
mile
s fr
om h
ospi
tal,
>10
0 da
ys in
reh
ab
Stra
tifie
d by
func
tion
al le
vel a
nd
disc
harg
e lo
cati
on
S a
rm s
tren
gth
at 6
m
onth
s N
S L
OS,
falls
, re
adm
issi
ons,
in-
hosp
ital
com
plic
atio
ns,
AD
L, m
obili
ty, m
orta
lity
Inte
rven
tion
: Med
ian
leng
th
of in
terv
enti
on w
as 1
2 w
eeks
. 6
and
12 m
onth
s fo
llow
-ups
Syst
emat
ic
mul
tico
mpo
nent
re
hab
prog
ram
no
bett
er th
an u
sual
ca
re
Ku
ism
a 20
02
Hon
g K
ong
(151
)
RC
T, h
ome-
base
d vs
. in
stit
utio
nal
n=81
>
50
Incl
usi
on: h
ip fr
actu
re
Exc
lusi
on: c
onco
mit
ant s
erio
us
cond
itio
n, li
ving
alo
ne, >
4 ho
urs
alon
e/da
y
S m
obili
ty
NS
LO
S N
A fa
lls, r
eadm
issi
on,
com
plic
atio
ns, A
DL,
m
orta
lity
Inte
rven
tion
: Mea
n 4.
6 vi
sits
by
PT, 1
.5 v
isit
s by
nur
se
Con
trol
gro
up m
ean
36.2
day
s in
reh
ab. 4
, 8 a
nd 1
2-m
onth
fo
llow
-up
by te
leph
one
inte
rvie
ws
Bet
ter
com
mun
ity
ambu
lati
on a
t 1-
year
follo
w-u
p
Cro
tty
200
2-0
3
Au
stra
lia
(14
6,1
48)
RC
T, h
ome-
base
d vs
. co
nven
tion
al
hosp
ital
car
e n=
66
≥65
Incl
usi
on: h
ip fr
actu
re, m
edic
al
stab
le, m
enta
l and
phy
sica
l ca
paci
ty to
par
tici
pate
, sui
tabl
e ho
me
envi
ronm
ent
Exc
lusi
on: i
nade
quat
e so
cial
su
ppor
t, no
tele
phon
e, li
ved
to fa
r aw
ay
S L
OS,
AD
L, fa
lls
effic
acy
NS
falls
, rea
dmis
sion
s,
mob
ility
N
A c
ompl
icat
ions
, m
orta
lity
Inte
rven
tion
: Med
ian
13.6
vi
sits
. 4 a
nd 1
2 m
onth
follo
w-
up
Impr
ovem
ent i
n A
DL
and
grea
ter
conf
iden
ce in
av
oidi
ng fa
lls a
t 4
mon
ths
Red
uced
ca
regi
ver
burd
en
at 1
2 m
onth
s
Rya
n
200
6
UK
(1
52)
RC
T, h
ome-
base
d vs
. in
tens
ive
hom
e-ba
sed
n=16
0(71
) ≥
65
Incl
usi
on: h
ip fr
actu
re a
nd
stro
ke
Exc
lusi
on: d
emen
tia,
Pa
rkin
son´
s
NS
AD
L, m
obili
ty
NA
rea
dmis
sion
s,
mor
talit
y, fa
lls,
com
plic
atio
ns, L
OS
Inte
rven
tion
: 6 v
isit
s/w
eek,
m
ean
visi
ts 2
4.4,
max
12
wee
ks
Mea
n 42
day
s in
hos
pita
l pri
or
to r
ehab
ilita
tion
at h
ome,
3
mon
th fo
llow
-up
No
sign
ifica
nt
diff
eren
ces
for
hip
pati
ents
20
Tab
le 2
.An
ove
rvie
w o
f te
am-b
ased
hom
e re
hab
ilit
atio
n s
tud
ies.
Tri
al
Des
ign
an
d
sett
ings
Sel
ecti
on
Ou
tcom
es
Inte
rven
tion
C
oncl
usi
on
Zid
en
200
8-1
0
Sw
eden
(1
44
,14
5 )
RC
T,
hom
e-ba
sed
vs.
conv
enti
ona
l car
e n=
102
≥
65
Incl
usi
on: h
ip fr
actu
re
Exc
lusi
on: s
ever
e m
edic
al
illne
ss, e
xpec
ted
surv
ival
<1
year
, dr
ug o
r al
coho
l abu
se, m
enta
l ill
ness
, sev
ere
cogn
itiv
e im
pair
men
t
S A
DL,
mob
ility
N
S L
OS,
falls
, mor
talit
y N
A r
eadm
issi
ons,
co
mpl
icat
ions
Inte
rven
tion
: Mea
n 4.
9 vi
sits
, max
3 w
eeks
, 1-,
6- a
nd
12-m
onth
follo
w-u
ps
Impr
ovem
ent i
n A
DL,
gre
ater
ba
lanc
e co
nfid
ence
and
ph
ysic
al fu
ncti
on
Sal
pak
oski
20
14-1
5 F
inla
nd
(1
49
,150
)
RC
T,
hom
e-ba
sed
vs.
conv
enti
ona
l car
e n=
81
>60
Incl
usi
on: F
NF,
PTF
E
xclu
sion
: <18
on
MM
SE,
alco
holis
m, s
ever
e ca
rdio
vasc
ular
-,
pulm
onar
y- o
r ot
her
prog
ress
ive
dise
ase,
sev
ere
depr
essi
on
S m
obili
ty
NS
falls
, AD
L, m
orta
lity
NA
LO
S, r
eadm
issi
on
Inte
rven
tion
: 1-y
ear
long
pr
ogra
m, p
arti
cipa
nts
assi
gned
mea
n 42
day
s af
ter
disc
harg
e, 5
-6 P
T vi
sits
, 3-
, 6-
and
12-m
onth
follo
w-u
ps
Red
uced
per
ceiv
ed
diff
icul
ties
in
nego
tiat
ing
stai
rs
S =
sig
nific
ant,
NS
= n
on-s
igni
fican
t, N
A =
non
-ass
esse
d, R
CT
= R
ando
miz
ed C
ontr
olle
d Tr
ial,
LOS
= L
engt
h of
Sta
y, A
DL
= A
ctiv
itie
s of
Dai
ly L
ivin
g, P
T =
Ph
ysio
ther
apis
t, FN
F =
Fem
oral
Nec
k Fr
actu
re, P
TF =
Per
troc
hant
eric
Fra
ctur
e.
21
RATIONALE As the population grows older, there will be an increasing number of people at risk of falls and consequently an increased number of hip fractures. Thus, there is an urgent need to develop preventive strategies for falls. People with cognitive impairment/dementia as well as people living in residential care facilities have been underrepresented in earlier studies, even though a substantial and increasingly proportion of individuals with hip fracture have dementia or cognitive impairment. Their inclusion is important to ensure that the sample studied will be representative of the group of older people with hip fracture today. Previous orthogeriatric care models for older individuals with hip fracture report a shortened LOS, reduced morbidity and mortality in the short term, but effects after discharge were limited, and mortality remained high. The prognosis for individuals with hip fracture is still poor, and knowledge about falls, complications and readmissions in the long-term, as well as causes of death is sparse. In this thesis we have tried to gain deeper knowledge by further investigating these outcomes among old people. The care for older people with hip fracture has changed over the last few decades, LOS has shortened and HR teams have been developed to meet the need for rehabilitation. However, the evidence to support HR is limited, and there are few randomized controlled multi- or inter-disciplinary HR-studies in the literature. Furthermore, data in team-based HR studies concerning complications and readmissions after discharge have only been provided in a few studies. The goal of rehabilitation is for the person to regain their ability to walk as they could prior to the fracture, even so, only half of the hip fracture population achieved this goal, which implies the need for further research.
22
AIMS This thesis has three aims. The primary aim is to describe the comorbidities, complications and causes of death among a representative population of old people with hip fracture. A secondary aim is to evaluate whether a successful care model, that reduced in-hospital falls and complications, had any effect on falls after discharge. The third aim is to evaluate whether adding home rehabilitation to the existing care model has an effect on LOS, walking ability, complications, and readmissions after discharge.
The specific aims are:
Paper I. The aim is to evaluate whether an orthogeriatric care model using CGA, that reduced inpatient falls and injuries, had any continuing effect after discharge.
Paper II. The aim is to describe the prevalence of co-morbidities, complications and the causes of death over three years among old people with femoral neck fracture.
Paper III. The aim is to investigate whether Geriatric Interdisciplinary Home Rehabilitation (GIHR) could improve walking ability for older people with hip fracture and shorten the LOS in hospital.
Paper IV. The aim is to evaluate whether GIHR could affect the number of complications, readmissions and total days spent in hospital after discharge.
23
METHODS
Table 3. Thesis at a glance. Trial Design
and setting Sample Data
collection Statistics Outcome
Study 1 Paper I (published)
RCT, Orthopedic and Geriatric clinic, University Hospital in Umeå, Sweden 00-02
199 people with femoral neck fractures aged 70 years or older.
Structured interviews, assessments, medical charts, follow-ups at 4 and 12 months
Poisson regression, Negative binomial regression (Nbreg), Student’s t-test, Pearson’s chi-square test, Fisher’s exact test, Mann-Whitney U test.
Postoperative fall incidence and fall incidence rate ratio
Paper II (published)
Data collection as in Paper I with the addition of 36 months follow-up.
Cox regression, Student’s t-test, Pearson’s chi-square test, Fisher’s exact test, Mann-Whitney U test.
Prevalence of co-morbidities, complications and causes of death
Study 2 Paper III (published)
RCT, Geriatric clinic, University Hospital in Umeå, Sweden 08-11
205 people with hip fracture, aged 70 or older.
Structured interviews, assessments, medical charts, follow-ups at 3 and 12 months.
Binary logistic regression, Student’s t-test, Pearson’s chi-square test, Mann-Whitney U test.
Walking ability, length of hospital stay
Paper IV (submitted)
Prevalence of complications, readmissions and number of days in hospital after discharge.
RCT = Randomized Controlled Trial
This thesis is based on data from two randomized controlled intervention studies including older people with hip fracture, conducted at Umeå University Hospital. The first included 199 persons with cervical hip fracture (study 1) and the second 205 persons with cervical or trochanteric fracture (study 2).
24
Ethical approval All participants were given oral and written information and in those cases where they were not able to answer themselves their next of kin was also asked. Written informed consent was required for participation in the studies. The studies were approved by the Ethical Committee of the Faculty of Medicine at Umeå University, DNR 00-137 and DNR 08-053 M. Study 2 was registered at Current Controlled Trials Ltd (ISRCTN 15738119).
Study 1 (Papers I and II)
Settings and participants Papers I and II are based on a randomized controlled trial which included 199 participants with femoral neck fracture aged 70 years or older consecutively admitted to the Orthopedic Department at the University Hospital in Umeå, Sweden. The participants were analyzed as one group in Paper II. The inclusion period started in May 2000 and ended in December 2002. The exclusion criteria were: severe rheumatoid arthritis, severe hip osteoarthritis and pathological fractures due to the surgical method planned in the study protocol. The surgeon on duty decided whether the operation planned according to the study protocol was appropriate for the participant. The anesthesiologist assessed the participant’s medical status and excluded those with severe renal failure. Participants who were bedridden before the fracture were also excluded. The inclusion criteria were met by 258 patients, 11 of whom declined to participate and 48 were not invited to participate due to failures in the inclusion routines or because they had sustained the fracture in hospital. These 59 individuals were more likely to be men (p=0,033), and live in their own house/apartment (p=0,009) but there was no difference in age (p=0,354) compared to the participants who were included.
25
Figure 3. Flow chart for Paper I.
26
Figure 4. Flow chart for Paper II.
Procedure In the emergency room the potential participants were asked orally and in writing whether they were willing to participate in the study. If they were cognitively impaired their next of kin were also asked to consent. While the participants were still in the emergency room, sealed opaque envelopes were used to randomly assign them to conventional postoperative care in an orthopedic ward or to a postoperative
27
intervention program conducted on a geriatric ward. To ensure that all participants were subject to the same preoperative routines, the envelopes were not opened until immediately before surgery. A nurse on duty on the orthopedic ward who was not involved in the study, opened the envelopes. The participants were stratified according to operation method. Undisplaced femoral neck fractures were treated with two hook-pins (Swemac Orthopedica®, Linköping, Sweden), displaced femoral neck fractures were treated with bipolar hemiarthroplasty (Link®, Hamburg, Germany) and the basocervical fractures were operated on using a Dynamic Hip Screw (DHS) (Stratec Medical®, Oberdorf, Switzerland). One participant died before surgery and one had a resection of the femoral head due to deterioration in medical status (they were both in the control group). In most cases (174/198, 88%) the operation was performed under spinal anesthesia. The stratification in study 1 is shown in Table 4.
Table 4. Stratification in study 1. Operation methods
Study 1 Intervention group n=102
Control group n=95
Two hook-pins 38 31 Bipolar hemiarthroplasty 57 54 Dynamic hip screw 7 10
Data collection Within three to five days after the operation a structured interview covering various assessments was performed by two registered nurses employed half-time in the study. The participants were interviewed concerning their medical history, social and functional status before the fracture. Data, including morbidity and complications, were also collected from relatives, staff and medical records. One nurse from the orthopedic ward carried out the assessments and interviews in the intervention group while the nurse from the geriatric department assessed and interviewed the participants in the control group. At discharge the functional ability was measured by the Physiotherapist (PT) and Occupational Therapist (OT) on duty. At 4, 12 and 36 months after surgery the participants were followed up and similar assessments were performed. A PT and an OT accompanied the nurses at the follow-ups. At the four-month follow-up a physician collected data and assessed the participants in the intervention group; this was not done in the control group. An overview of all assessments in study 1 are presented in Table 5.
28
Table 5. Assessments in study 1. Assessments
Study 1 Paper I Paper II
Staircase of ADL X X S-COVS X X ASA X Vision and hearing X MMSE X X OBS X X GDS X X BBS X Chair stand test X
Assessments Diagnoses, complications and medications A specialist in geriatric medicine, who was not involved in the patient care and was unaware of group allocation, analyzed the documentation concerning diagnoses, medications and the assessments performed by the OT, PT and the nurse for final diagnosis. The geriatrician also determined whether the participants met the Diagnostic and Statistical Manual of Mental Disorder, fourth edition (DSM-IV)155 criteria for dementia, delirium and depression. A geriatrician not blinded to group allocation and employed on the ward, registered and analyzed complications by reviewing the participant’s assessments and charts after each study was finished. The definitions of complications were chosen after reviewing the literature and a predefined list of complication criteria was used for the classification. Complications included both orthopaedic complications and medical incidents. The complications chosen were classified dichotomously (present or absent). The total number of times a complication occurred was also counted. If the participant had a cardiac failure at baseline and was treated by a doctor during the follow-up period due to an exacerbation/aggravation of the disease he/she was judged to have cardiac failure as a complication. Myocardial infarction and cardiac failure are referred to in the study as cardiovascular events. Five groups of infections were defined; urinary tract infections, chest infections, superficial and deep wound infections and other infections. For those participants who died during the study, medical records were reviewed and all complications and the “history of death” were noted. If a
29
participant died between follow-ups, complications that occurred between last follow-up and death were forwarded, meaning they were registered at the next follow-up. Reoperation is defined as an operation in the same area as the index fracture, due to infection, luxation, and failure to heal or material failure. An operation caused by a new fracture during follow-up is not considered to be a reoperation. The medications were classified according to the Anatomical Therapeutic Chemical classification system. The prescriptions were recorded as ‘yes’ or ‘no’, but no doses were registered, and pro re nata drugs were not included. Falls Falls and number of falls were registered from participants’ charts during hospitalization. At the assessments, the participants were asked if they had sustained any falls since discharge, or their last assessment. Information about falls was also collected from next of kin, staff and medical charts. The incident was defined as a fall when the participant unintentionally came to rest on the floor or ground and syncopal falls were included.60 Causes of death
Every citizen in Sweden has a unique 10-digit personal identification number which makes it possible to obtain date and cause of death from the Centre for Epidemiology (EpC), National Board of Health and Welfare, Sweden. Diagnoses are coded according to the International Classification of Diseases, a validated and international standard set by the World Health Organization (WHO).156 According to the international standards, a death within 30 days after a fracture would be coded as related to the trauma, irrespective of the actual cause of death, and the fracture diagnosis would be given as the primary cause of death. As an example, a person suffering a myocardial infarction which led to death within 30 days of sustaining a hip fracture would have the myocardial infarction cited as a secondary cause of death. The primary and secondary causes of death were obtained from death certificates. Medical records concerning the person’s “history of death” were also reviewed. Two specialists in geriatric medicine assessed information retrieved from death certificates and medical records and then established the causes of death; disagreements were resolved through discussion, although inter-rater agreement was not tested. Fifteen out of 79 causes of death stated on the death certificates were
30
altered after the person’s medical records were analyzed. Two out of 15 were changed as a result of autopsy statements, three more were changed after information was retrieved from the medical records and another ten were changed from the first into the second cause on the death certificate. An example of a change to a second cause concerns terminal pneumonia in a person suffering from advanced dementia. This was stated as the primary cause of death on the death certificate but we judged the person to have died as a result of dementia rather than pneumonia. Readmissions and days in hospital Total number of days spent in hospital after discharge and the number of readmissions were registered up to the end of the study or until the participant declined to continue, died or left the study for other reasons. Activities of Daily Living Functional status of ADL performance prior to the fracture was measured retrospectively using the staircase of ADL157 including the Katz ADL index.158 This scale measures both Personal/Primary Activities of Daily Living (P-ADL) (bathing, dressing, toileting, transfer, continence, and feeding), and Instrumental Activities of Daily Living (I-ADL) (cleaning, shopping, transportation, and cooking). The score ranges from 0-10, with a high score indicating greater ADL dependence. Independence in P-ADL (bathing, dressing, toileting, transfer, continence and feeding) was presented as a binary variable. Walking ability Walking ability indoors and outdoors was assessed using the Swedish version of Physiotherapy Clinical Outcome Variables (S-COVS), a scale with seven levels, where 1 indicates no functional ability or the need for assistance from two people and 7 indicates normal function.159 Walking ability prior to fracture was dichotomized as capacity for independence indoors, irrespective of eventual need for a walking aid, using one item from S-COVS. The use of a walking device was also registered. ASA The ASA classification,160 is a risk assessment instrument. The general health of all participants was assessed by the attending anesthesiologist before surgery according to the ASA classification. ASA 1 indicates a healthy person; ASA 2, a person with a mild systemic disease; ASA 3, a person with severe systemic disease; ASA 4, a person with an
31
incapacitating disease that is a constant threat to life; ASA 5, a moribund person who is not expected to live 24 h with or without surgery. The ASA results were categorized as ASA 1-2 and ASA 3-4. None of the people included were classified as ASA 5. Vision and hearing The participants’ hearing and vision were tested by their ability to hear a normal speaking voice from a distance of one meter and to read five millimeter block letters with or without glasses. Cognitive impairment Two different assessment scales were used to validate the participants’ cognitive status. The Mini-Mental State Examination (MMSE)161 is a screening test used to assess cognition among older people. It has a score from 0-30 where a score of less than 24 indicates cognitive impairment.162 The other scale used was a modified version of the Organic Brain Syndrome Scale (OBS),163 adjusted for persons with hip fractures, where variables affected by the hip fracture per se are excluded. The scale can help establish a person’s orientation and awareness. It consists of two subscales, a disorientation subscale with 12 items, a questionnaire with a maximum score of 36 (a high score indicating disorientation) and a confusion subscale with 21 items based on clinical observations and interviews with participants and caregivers. The confusion subscale can describe a person’s cognitive, emotional, perceptual and personality changes and fluctuations in clinical state. Depression The Geriatric Depression Scale (GDS-15, a shorter version)164,165 is a screening tool for depressive disorders among older people. It has 15 items and a score of between five and nine indicates a mild depression while a score of ten or more indicates a moderate to severe depression. Depression before hospitalization was diagnosed based on current treatment and earlier diagnosis, documented in the participant’s records. During hospitalization depression was diagnosed if the participant received ongoing treatment with antidepressants or if depression was indicated by the results on the GDS-15 scale in combination with depressive symptoms registered on the OBS-scale.
32
Balance Balance was assessed using the Berg Balance Scale (BBS) which is well-established, reliable and valid in older people, including those with cognitive impairment.166,167 The scale covers 14 tasks common in everyday life, e.g. sitting, standing, turning and reaching. Each task is scored 0 to 4, with a maximum score of 56; a higher score indicates better balance/functional capacity. The ability to perform tasks in a safe and controlled manner gives the maximum score, whereas those who exceed a specified time limit or require supervision, physical or verbal guidance are given lower scores. Chair stand test A modified version of the Chair stand test168 was performed by testing the participant’s ability to rise three times from a straight-backed chair with the help of the arms.
Intervention Control group. The participants in the control group received their postoperative care at a specialized orthopedic unit following conventional postoperative routines. Those who were in need of a longer rehabilitation period were transferred to a geriatric unit, though not the one where the intervention was conducted. The staffing at the orthopedic unit was 1.01 nurses/aids per bed and at the geriatric unit it was 1.07. Intervention group. Before the study started an intervention program was constructed based on a literature review and previous studies concerning the care of old people with hip fracture. The program was implemented before the study started with a four-day course in caring, teamwork, rehabilitation and medical knowledge about prevention and treatment of postoperative complications. The participants were admitted to a geriatric unit specializing in geriatric orthopedic patients where this program had been implemented. It was a 24-bed unit and the staffing level was 1.07 nurses/aids per bed. The staff worked as a team applying CGA, and the intervention program focused on prevention, detection and treatment of postoperative complications, early
33
mobilization and daily training in personal ADL and ambulation. Special attention was given to preventing, detecting and treating delirium. The PTs supervised individualized functional strength- and balance training, for example ambulation, in-bed transfers and stair climbing. Assessment and training in performance of P-ADL and inquiries about the patients’ home environment were carried out by the OTs. The PT and the OT made home visits together with the patients before discharge, if it was considered necessary. Discharge was carefully planned, to meet the participants’ further needs. After discharge the participants could be referred to primary healthcare or to an out-patient rehabilitation unit connected to the Geriatric Department for additional training. For those living in residential care facilities the PTs and OTs in charge were contacted by phone before discharge and informed of the participants’ needs regarding help and further training. A home training program was offered to the participants who were not assigned to any other kind of rehabilitation/were able to train by themselves or with the help of their next of kin. The participants received a telephone call two weeks after discharge from a PT or an OT, checking for additional needs. The nurse and a PT or an OT assessed the need for rehabilitation, assistive devices, modifications of the home environment and they also checked for nutritional problems four months postoperatively. The follow-up took place either in the geriatric out-patient clinic or, if the participant’s condition was poor, a home visit was made. The doctor checked for fall-risk factors such as inappropriate medication, low blood pressure and other complications. The intervention study was known to the staff on the intervention ward and the staff working with the control group were aware that a new care program was being implemented and evaluated. Table 6 presents an overview of the intervention program and the conventional care in study 1.
34
Tab
le 6
. Ove
rvie
w o
f th
e in
terv
enti
on p
rogr
am a
nd
th
e co
nve
nti
onal
car
e in
stu
dy
1 (P
aper
s I
and
II)
. A m
odif
ied
ver
sion
fro
m B
O
lofs
son
´s a
nd
M S
ten
vall
´s t
hes
is.
In
terv
enti
on g
rou
p
Con
trol
gro
up
In
divi
dual
car
e pl
anni
ng
-The
par
tici
pant
was
ass
esse
d by
all
team
mem
bers
, usu
ally
w
ithi
n 24
hou
rs, t
o st
art a
n in
divi
dual
ized
car
e pl
an
-Tea
m m
eeti
ngs
twic
e a
wee
k to
eva
luat
e th
e re
habi
litat
ion
proc
ess
and
goal
s
- Ind
ivid
ualiz
ed c
are
plan
ning
was
use
d, th
ough
not
rou
tine
ly,
as in
the
inte
rven
tion
war
d - T
here
was
wee
kly
indi
vidu
al c
are
plan
ning
at t
he g
eria
tric
re
habi
litat
ion
unit
Pr
even
tion
, de
tect
ion
and
trea
tmen
t of
com
plic
atio
ns
-Pul
se, b
lood
pre
ssur
e, te
mpe
ratu
re a
nd s
atur
atio
n w
ere
mea
sure
d du
ring
the
first
pos
tope
rati
ve d
ays
-O
xyge
n-en
rich
ed a
ir in
the
first
pos
tope
rati
ve d
ay a
nd
long
er if
nec
essa
ry
-Blo
od te
st a
nd u
rina
ry s
ampl
e th
e fir
st d
ay a
fter
ope
rati
on to
sc
reen
for
anem
ia, s
igns
of k
idne
y fa
ilure
, inf
ecti
ons
etc.
-U
rina
ry c
athe
ters
wer
e di
scon
tinu
ed w
ithi
n 24
hou
rs
post
oper
ativ
ely
- R
isk
fact
ors
for
falls
wer
e an
alyz
ed a
nd g
loba
l rat
ings
of t
he
part
icip
ant’s
fall
risk
wer
e m
ade
ever
y w
eek
at te
am m
eeti
ngs
-Reg
ular
scr
eeni
ng fo
r ur
inar
y re
tent
ion,
and
pre
vent
ion
and
trea
tmen
t of c
onst
ipat
ion
-Blo
od tr
ansf
usio
n if
B-h
emog
lobi
n, g
/l, w
as <
100
or <
110
for
thos
e at
ris
k of
del
iriu
m/d
elir
ious
-C
alci
um a
nd v
itam
in D
, and
oth
er p
harm
acol
ogic
al
trea
tmen
ts fo
r os
teop
oros
is/r
educ
ed b
one
dens
ity
if in
dica
ted
-Ant
i-th
rom
bosi
s ph
arm
acol
ogic
al tr
eatm
ent a
nd a
nti-
thro
mbo
sis
stoc
king
s th
e fir
st tw
o po
stop
erat
ive
wee
ks, a
nd
long
er if
indi
cate
d
- No
rout
ine
anal
ysis
of w
hy th
e pa
tien
ts h
ad fr
actu
red
thei
r hi
ps; n
o at
tem
pt m
ade
to s
yste
mat
ical
ly p
reve
nt fu
rthe
r fa
lls
no r
outi
ne p
resc
ript
ion
of c
alci
um a
nd v
itam
in D
- A
sses
smen
ts fo
r po
stop
erat
ive
com
plic
atio
ns w
ere
mad
e by
ch
ecki
ng e
.g. s
atur
atio
n, h
emog
lobi
n, n
utri
tion
, bla
dder
and
bo
wel
func
tion
, hom
e si
tuat
ion
etc.
but
not
car
ried
out
sy
stem
atic
ally
as
in th
e in
terv
enti
on g
roup
35
-Ass
essm
ent,
anal
ysis
and
trea
tmen
t of s
leep
ing
prob
lem
s -S
yste
mat
ic s
cree
ning
for
delir
ium
, ass
essm
ent a
nd
trea
tmen
t of u
nder
lyin
g ca
uses
Nut
riti
on
-Foo
d an
d liq
uid
regi
stra
tion
, and
pro
tein
- and
ene
rgy-
enri
ched
mea
ls, t
he fi
rst f
our
days
pos
tope
rati
vely
, lon
ger
if ne
cess
ary
-Ser
ved
nutr
itio
nal a
nd p
rote
in d
rink
s da
ily
-Foo
d co
nsis
tenc
y ad
apte
d if
nece
ssar
y -P
robl
ems
that
eff
ecte
d th
e ap
peti
te, s
uch
as p
ain
and
bad
oral
hyg
iene
wer
e an
alyz
ed a
nd a
ppro
pria
te a
ctio
ns ta
ken
-Con
sult
atio
n w
ith
diet
icia
n if
nece
ssar
y
-No
diet
icia
n av
aila
ble
at th
e or
thop
edic
uni
t -N
o ro
utin
e nu
trit
ion
regi
stra
tion
or
prot
ein-
enri
ched
mea
ls
for
the
pati
ents
Reh
abili
tati
on
-Mob
iliza
tion
wit
hin
the
first
24
post
oper
ativ
e ho
urs
-Bas
ic A
DL
perf
orm
ance
trai
ning
aro
und
the
cloc
k by
car
ing
staf
f -S
peci
fic e
xerc
ises
and
reh
abili
tati
on p
roce
dure
s by
PT
and
OT
-Reh
abili
tati
on b
ased
on
func
tion
al r
etra
inin
g, w
ith
spec
ial
focu
s on
fall-
risk
fact
ors
-H
ome
visi
t by
OT
and/
or P
T - T
he P
T/O
T co
-ope
rate
d w
ith
colle
ague
s w
orki
ng in
co
mm
unit
y se
rvic
e fo
r fu
rthe
r co
nsul
tati
on a
fter
the
pati
ent
was
dis
char
ged
from
hos
pita
l - A
ll pa
tien
ts w
ere
offe
red
furt
her
out-
pati
ent r
ehab
ilita
tion
af
ter
disc
harg
e - P
atie
nts
wer
e fo
llow
ed u
p by
PT
or O
T w
ith
a ph
one
call
two
wee
ks a
fter
dis
char
ge a
nd a
hom
e vi
sit f
our
mon
ths
post
oper
ativ
ely
- Mob
iliza
tion
usu
ally
wit
hin
the
first
24
hour
s - T
he P
T on
the
war
d m
obili
zed
the
pati
ents
toge
ther
wit
h th
e ca
ring
sta
ff
- The
PT
aim
ed to
mee
t the
luci
d pa
tien
ts e
very
day
- F
unct
iona
l ret
rain
ing
in A
DL
situ
atio
ns w
as n
ot a
lway
s gi
ven
- The
OT
at th
e or
thop
edic
uni
t onl
y m
et th
e pa
tien
ts fo
r co
nsul
tati
on
- No
hom
e vi
sits
wer
e m
ade
by s
taff
from
the
orth
oped
ic u
nit
- The
ger
iatr
ic c
ontr
ol w
ard
had
both
spe
cific
exe
rcis
e an
d ot
her
reha
bilit
atio
n pr
oced
ures
del
iver
ed b
y a
PT a
nd O
T,
sim
ilar
to th
at g
iven
at t
he in
terv
enti
on w
ard
- No
follo
w-u
p by
a p
hysi
cian
at f
our
mon
ths
36
- Fou
r m
onth
s po
stop
erat
ivel
y a
phys
icia
n m
et th
e pa
tien
ts
to d
etec
t and
pre
vent
com
plic
atio
ns
Team
wor
k - T
eam
incl
uded
Reg
iste
red
Nur
ses,
Lic
ense
d Pr
acti
cal
Nur
ses,
Phy
siot
hera
pist
s, O
ccup
atio
nal T
hera
pist
s, a
di
etic
ian
and
geri
atri
cian
s - C
lose
coo
pera
tion
bet
wee
n or
thop
edic
sur
geon
s an
d ge
riat
rici
ans
in th
e m
edic
al c
are
of th
e pa
tien
ts
- No
corr
espo
ndin
g te
amw
ork
at th
e or
thop
edic
uni
t - T
he g
eria
tric
war
d, w
here
som
e of
the
cont
rol g
roup
pat
ient
s w
ere
care
d fo
r, u
sed
team
wor
k si
mila
r to
that
in th
e in
terv
enti
on w
ard
PT =
Phy
siot
hera
pist
s, O
T =
Occ
upat
iona
l The
rapi
sts,
AD
L =
Act
ivit
ies
of D
aily
Liv
ing.
37
Statistical analyses All data were analyzed using the statistical software package SPSS version 12.01 and 23.0 (SPSS Inc., Chicago, USA) and STATA 9. All tests were two-tailed and a p-value <0.05 was regarded as statistically significant in all analyses. The analyses were based on the intention-to-treat principle, including all randomized patients, according to their original allocation, and regardless of level of attendance. In Paper I data, including the one-year follow-up, are analyzed and in Paper II data up to three-year follow-up are analyzed. Paper I The power calculation was performed a posteriori, based on data from previous studies, assuming that 50% of the participants in the control group would fall. The sample size of 199 participants provided a power of 30% at α = 0.05 to detect a 20% reduction in falls between intervention and control group. Differences regarding baseline characteristics and mortality between control- and intervention groups were analyzed using; Student’s t-test for independent samples when comparing normally distributed continuous variables; Pearson’s chi-square test and Fisher’s exact test when dichotomous data were compared; and the Mann-Whitney U test for continuous variables not normally distributed. The fall incidence was compared in two ways between intervention and control groups; first an unadjusted comparison was made regarding the number of participants who fell and the numbers of fractures using Pearson’s chi-square test and Fisher’s exact test. Secondly, a comparison was performed by calculating the fall incidence rate in the intervention and control groups and then estimating the fall Incidence Rate Ratio (IRR) using Negative binomial regression (Nbreg), a generalization of the Poisson regression model, with adjustment for over-dispersion and observation time.169 The observation time was calculated from the day of admission until the day of final follow-up. As an example, if the participant had died or declined between the 4- and 12-month follow-ups, only observation days up to 4 months were calculated. In the Poisson regression model, baseline characteristics were considered as covariates if they had a p-value <0.15 (depression and dementia) when comparing intervention and control groups. Since these variables only had a marginal effect on the IRR values they were not included in the presentation of the data.
38
Paper II Pearson’s chi-square test, Fisher’s exact test and the Mann–Whitney U test were used to analyze differences between the intervention and control groups and between the deceased and the survivors regarding complications. Univariate and multivariate Cox proportional hazard regression was used to analyze associations between baseline variables and complications during hospitalization and all-cause mortality during the 3-year follow-up. All baseline variables and complications during hospitalization associated with time to death (p < 0.05) in univariate analyses were included in two separate multivariate models (Model A and B in Table 11). Correlations between baseline variables and among complications during hospitalization were tested using Pearson’s and Spearman’s coefficients; the covariate Living in residential care facilities before fracture was removed due to its strong correlation with Dependence in P-ADL (r >0.6). Step-wise backward deletion was performed manually until only significant variables remained in the models. The proportionality of hazards was tested using Schoenfeld residuals and time-dependent variables in extended Cox regression models.170 A final multivariate model adjusted for age, sex and remaining significant variables from the two separate multivariate analyses was performed (Model C in Table 11).
Study 2 (Papers III and IV)
Settings and participants Papers III and IV are based on a randomized controlled study including 205 persons with acute hip fracture (cervical- and trochanteric fracture) consecutively admitted to the Geriatric Department at the University Hospital in Umeå, Sweden. The inclusion started in May 2008 and ended in June 2011. The one-year follow-up ended in 2012.The participants were aged 70 years or older, living in the municipality of Umeå, in independent housing or in residential care facilities. Dementia and cognitive impairment were not exclusion criteria. Those who fractured their hip in hospital and those with pathological fractures, however, were not included. A total of 466 persons were screened for eligibility but 187 people did not meet the inclusion criteria, 37 declined to participate and 33 were missed due to failure in the inclusion process. The group that declined or were missed did not differ significantly in age or sex from those included in the study. Two participants died and two declined before the first assessment, leaving 205 people to be included in the study.
39
Figure 5. Flow chart for Paper III.
40
Figure 6. Flow chart for Paper IV.
Procedure Everyone eligible for inclusion was admitted to the orthopedic ward and operated on. In accordance with the usual clinical pathways in the hospital, those with cervical fractures were transferred to a geriatric ward specializing in orthopedic geriatric care directly after surgery and those with trochanteric fractures were given acute postoperative care on the orthopedic ward. Those with trochanteric fractures who needed a longer period of rehabilitation were transferred to the geriatric ward after being assessed by a geriatrician.
41
An independent researcher was responsible for the randomization and the participants were randomized to the control group, with conventional geriatric care, or to the intervention group, with conventional geriatric care and GIHR after early discharge. The randomization was stratified according to housing (living in residential care facilities or in community dwellings) and to type of fracture (cervical or trochanteric). The surgical method used depended on the type of fracture, hemiarthroplasty was used for displaced femoral neck fractures and internal fixation for the un-displaced cervical fractures. Trochanteric fractures were operated on using dynamic sliding screw devices such as DHS and Twinhook or with intramedullary nails. The stratification in study 2 is shown in Table 7.
Table 7. Stratification in study 2.
Study 2 Intervention group Control group
Type of fracture
Community dwelling
Residential care facility
Community dwelling
Residential care facility
Cervical 49 29 46 24 Trochanteric 22 7 25 3
All participants were asked orally and in writing if they were willing to participate in the study and in those cases when they were unable to answer their next of kin was asked. Written consent was required and they were informed that they could withdraw at any time from the study without prejudice. All participants were randomized before arrival on the geriatric ward; a concealed numbered envelope was drawn by the nurse on call when the ward was informed that a new patient was to be admitted. This means that a person with a cervical fracture was randomized before even being asked to participate in the study. This was done for logistical reasons. Those with trochanteric fractures were asked to participate when still on the orthopedic ward, before randomization. The providing of information and asking about willingness to participate was often done by a geriatrician while it was mostly the independent researcher who gave information about group allocation. The geriatric ward had two wings; the control group (conventional care) was placed in one and the intervention group (GIHR) in the other.
42
Data collection During hospitalization, within five days of randomization, the participants were assessed by two experienced researchers (one PT and one nurse) who were blinded to group allocation and had no contact with the ward during the study. On the ward, the assessments were made in a neutral place in order not to reveal group allocation. The assessments were also carried out at 3 months after randomization and 12 months postoperatively by the same researchers. At the three-month follow-up group allocation was exposed in 27 times but the assessments were completed and at the 12-month follow-up the assessor was changed. Medical data were collected from medical and nursing records and through asking participants, relatives and medical staff. An overview of all assessments in study 2 is presented in Table 8.
Table 8. Assessments in study 2. Assessments
Study 2 Paper III Paper IV
Barthel ADL index X X Staircase of ADL, Katz index X S-COVS X X ASA X MMSE X X OBS X X GDS X X Gait speed test X
Assessments In study 2 (Papers III and IV) the participants were assessed with regard to medical diagnosis, falls, complications and readmissions as in study 1. The assessment and scales (Staircase of ADL, Katz index, S-COVS, ASA, MMSE, OBS, and GDS) used are described under Assessments in the method section for study 1 (Papers I and II) (above). The assessments not described in the previous section are presented below. Activities of Daily Living The participant or, if there was cognitive impairment, the next of kin or a nurse’s aide, were asked about ADL prior to the fracture, using the Barthel Index.171,172 This is a 10-item index covering primary activities of daily living (eating, grooming, toileting etc.), adding up to a total score ranging from 0 to 20. The maximum score of 20 indicates independence in P-ADL.
43
LOS In Paper III postoperative LOS was recorded in 3 different ways. Total LOS includes the time spent in all departments in the hospital from after surgery until discharge. In addition, LOS was measured from admission to the geriatric ward until discharge, as well as LOS from admission to the geriatric ward until the discharge-ready date (DRD). Gait speed In Paper III self-chosen and maximum gait speed (m/s) over a distance of 2.4 meters, was recorded at the 3- and 12-month follow-up visits in the participants’ homes. These measurements were taken from a standing start and with the participant’s usual walking aid. The stopwatch was started on the command “Go” and was stopped when the first foot crossed the finishing line. The mean of 2 tries was used for self-chosen gait speed, and the faster of 2 tries was used for maximum gait speed.
Intervention Control group. An orthogeriatric care model, comprising a multidisciplinary and multi-factorial intervention program, was implemented on the ward in the year 2000 (Study 1.). Thus conventional care and rehabilitation on the geriatric ward included this program, which had been successful in reducing in-hospital falls and complications, such as delirium, infections and decubital ulcers,111,126 and had become part of the regular ward routine. The staff worked in teams to apply CGA, with regular meetings and individual care planning for each participant. The program included active prevention, detection and treatment of postoperative complications, early mobilization, daily training and a carefully planned discharge, as previously described in study 1. Participants in ordinary housing, in need of further rehabilitation were referred to primary healthcare after discharge, and 3 months after fracture, they could also receive rehabilitation at a geriatric out-patient rehabilitation unit. Those living in residential care facilities in need of further actions after discharge were referred to PTs and OTs in those facilities. Intervention group. The participants who were randomized to GIHR received care during hospitalization according to the same multidisciplinary and multi-factorial program as the control group, but the aim was early discharge
44
from hospital and continuation of the rehabilitation in the participants’ homes. The participants could be discharged from hospital when no medical obstacles existed and when they could manage basic transfers and had access to the care they required at home. A member of the GIHR team met the participant before discharge, giving information about the team, and inquiring whether one of the team members needed to accompany the participant home. The meeting between the team member and the participant before discharge was partly done to attempt to alleviate anxiety about leaving hospital. The GIHR team was made up of an OT, two PTs and a nurse who regularly visited the participants. A geriatrician had medical responsibility, and a social worker and a dietician could be consulted when needed. The team worked using CGA with regular meetings, individual care planning and rehabilitation individually designed according to the participants´ own goals. The prevention, detection and treatment of complications and falls, were given special priority. During the first few days at home the team members usually made daily visits to the participant, but later the number of visits per week was in accordance with the participants’ needs. The team also worked in close contact with relatives and social home service or staff in the residential care facilities. The maximum duration of GIHR was ten weeks. The physiotherapist trained the participant’s functional strength and balance according to the High-Intensity Functional Exercise program (HIFE).173,174 The goal was usually for the participants to regain their previous walking ability, both indoors and outdoors. Those who had the capacity to exercise on their own or with support from others were given individually designed exercise programs. The OT focused on independence in personal and instrumental ADL, trying out assistive devices and modifications of the home environment. The aim was to prevent new falls and to make everyday activities safer. The nurse and the geriatrician were responsible for medical issues after the hip fracture, for example treatment of pain, supervision of the operation wound and evaluation of the participants’ ability to manage their medicines safely. All team members reported symptoms, for example, delirium, pain and sleeping disturbances to the nurse and geriatrician who assessed and treated the participant in order to minimize further complications. The participants’ nutrition was also evaluated by the nurse with aspects that might have an effect on nutrition, such as constipation, pain, or oral problems, being considered. GIHR was ended, when the participants achieved their goals, or the team members could not contribute to progress in the participant’s further
45
rehabilitation, or the time limit was up. When GIHR ended, participants in need of further actions were referred to primary healthcare, staff in the residential care facilities or to a geriatric out-patient rehabilitation unit.
Statistical analyses All data were analyzed using the statistical software package SPSS version 22.0 and 23.0 (IBM SPSS Statistics, IBM Corporation, Chicago, IL)/ (SPSS Inc., Chicago, USA). All tests were two-tailed and a p-value <0.05 was regarded as statistically significant in all analyses. The analyses were based on the intention-to-treat principle, including all randomized patients according to their original allocation, and regardless of level of attendance. A group of seven people randomized to GIHR never received the team-based intervention. Six of these participants remained in hospital because of the unavailability of social services and were judged not to need GIHR once they were discharged. One participant was missed. All seven participants were, nevertheless, included in the analysis. A power calculation was performed with the number of days that patients with hip fracture spent in the hospital during a year from a previous study.111 Assuming a power of 80% and with a 24% reduction in hospital days, the total sample size was estimated to be 206 participants. Paper III The Student t test, Pearson´s chi-square test, or the Mann-Whitney U test were used to analyze group differences in prefracture characteristics and for outcomes, as appropriate. Data concerning physical assistance and walking devices were dichotomized, and a binary logistic regression method was used to analyze the odds ratio (OR) of walking ability and the use of walking devices for the groups. The regressions were adjusted for age, sex, and prefracture status of the outcome variable and for significant differences between the groups at baseline (antidepressants, analgesics). The Mann-Whitney U test was used for postoperative LOS as the data were not normally distributed and because of differences between the groups regarding the extreme outliers. Paper IV Baseline characteristics, complications, readmissions and days in hospital were compared between the GIHR and control groups. Student’s t-test for independent samples was used when comparing normally distributed continuous variables. Pearson’s chi-square test or Fisher’s exact test were used for dichotomous data. The Mann-Whitney U test was used for non-normally distributed continuous variables. Observation time used in the regression model was registered as time from discharge until the end of
46
the study or until the participant declined to continue, died or left the study for other reasons. A binary logistic model was used to analyse the OR of falling after discharge according to group allocation. Correlations between the covariates in the model were tested using Pearson’s and Spearman’s coefficients. The first model was adjusted for age and gender. The final model was adjusted for age, gender, observation time and significant differences between the GIHR and control groups at baseline (e.g., analgesics, antidepressants, Parkinson medication).
47
RESULTS
At baseline, most of the patients included in studies 1 and 2 were women, who were living alone in ordinary housing and walked independently indoors before the fracture, though only four out of ten were independent in P-ADL before the index hip fracture. Approximately half of the population had a heart disease, and an ASA score of ≥3 or higher at baseline. The participants in study 1 had more cancer at baseline, 15% vs. 6% in study 2. Whereas dementia (32% vs. 50%) and having ≥3 comorbidities (40% vs. 59%) were more common among participants in study 2. Depression and the use of antidepressants differed significantly between intervention and control groups at baseline in study 1, and the use of analgesics, antidepressants and Parkinson medications differed significantly between groups in study 2 (Table 9).
48
Table 9. Basic characteristics and assessments among participants in studies 1 and 2.
Study 1 Study 2
Intervention (n=102)
Control (n=97)
Intervention (n = 107)
Control (n = 98)
Mean age, mean ± SD 82.3±6.6 82.0±5.9 83.2±7.0 82.6±6.4 Females 74 74 79 68 Independent living 66 60 71 71 Independent in P-ADL 42 42 45 47 Independent walking 85(101) 85(94) 95 85 indoors Health and medical problems Cancer 15 14(92) 6 6 Cardiovascular disease 57(101) 53(93) 53 52 Dementia 28 36 57 46 Depression 33 45(95)* 47(106) 30(97) Diabetes 23 17(95) 17 15 Stroke 29 20(93) 21 24 Previous hip fracture 16 14(96) 20 15 Previous wrist fracture 16(101) 23(95) 9 14 ≥3 comorbidities 37 42(97) 66 54 Medications Number of drugs, mean ± 5.8±3.8 5.9±3.6 8.8±3.0 8.3±3.3 SD Analgesics 30 32 87* 90 Antidepressants 29 45* 49* 26 Benzodiazepines 21 25 15 12 Calcium/Vitamin D 12 16 Neuroleptics 13 8 10 13 Parkinson medications 10* 1 Assessments ASA grade 3-4 54 55(95) 61(104) 56(96) Barthel ADL index, median 18(13-20)(105) 18(12.5-20)(97) (IQR) GDS, median (IQR) 4.0 (2-7) 4.0 (2-6.8) 4.0(2-6)(92) 4.0(2-6.2)(82) MMSE, median (IQR) 19.0 (12.5-23) 17.0 (8-23.2) 18.0(11-25)(104) 19.0(11-25)(95) SCOVS, need of assistance, 6 (5-7) (101) 5.5 (5-7) (94) 6(5-7) 6(5-6) median (IQR) Staircase of ADL, median 5 (1-7.8) (92) 5 (0.3-7) (88) (IQR)
*p< 0,05 IQR = Interquartile Range, ADL = Activities of Daily Living, ASA = American Society of Anesthesiologists, S-COVS = Swedish version of Physiotherapy Clinical Outcome Variables, SD = standard deviation, P-ADL = Personal Activities of Daily Living, GDS = Geriatric Depression Scale, MMSE = Mini Mental State Examination. If details are not given for the complete group, the number of subjects is given in parenthesis.
49
Paper I No significant differences between intervention and control groups could be seen when comparing falls and new fractures from discharge to the 4-month follow-up, nor between the 4− and 12-month follow-ups, nor cumulatively at the 12-month follow-up. Analysis of the data from the 4-month follow-up showed that a total of 26/92 participants in the intervention group had sustained 58 falls and 26/82 participants in the control group had sustained 65 falls. The crude fall-incidence rate was 6.30/1 000 days in the intervention group vs. 9.07/1 000 days in the control group. Using a Poisson regression (Nbreg), adjusting for over-dispersion, the IRR was 0.55 (95% CI: 0.27–1.12, p=0.100). Among participants with dementia, 8 in the intervention group sustained 22 falls and 11 in the control group sustained 38 falls. Between four and twelve months the crude fall-incidence rate was 2.96/1 000 days in the intervention group vs. 3.54/1 000 days in the control group. The IRR was 0.85 (95% CI: 0.48–1.50, p=0.577) (Table 10). Between admission and the 12-month follow-up 44 participants in the intervention group sustained 138 falls (range 1–11) and 55 participants in the control group sustained 191 falls (range 1–31). The crude fall-incidence rate was 4.16/1 000 days in the intervention group vs. 6.43/ 1,000 days in the control group. IRR was 0.64 (95% CI: 0.40–1.02, p=0.063). Among participants with dementia (diagnosed at baseline) 12 sustained 41 falls in the intervention group and 22 sustained 104 falls in the control group during one year (Table 10). The crude fall-incidence rate was 5.16/1 000 days in the intervention group vs. 9.52/ 1 000 days in the control group. IRR was 0.48 (95% CI: 0.21–1.09, p=0.079). The new fractures at 4 months were one hip fracture, one nose fracture and one fracture of the scapula. All these fractures occurred in the control group. Between 4 and 12 months four participants in the control group sustained new fractures - one hip fracture, one pelvic fracture, one proximal humerus fracture and one vertebral fracture. In the intervention group, seven participants suffered two pelvic fractures, one proximal humerus fracture, three wrist fractures, one rib fracture, one proximal tibia fracture, one dens fracture and one face fracture. Three out of seven participants in the intervention group sustained two fractures each. The differences in fracture rate between control and intervention groups were not significant at the 4− and 12-month follow-ups. After one year, 16 out of 102 (16%) participants in the intervention group and 18 out of 97 (19%) participants in the control group had died (p=0.591). Prescription of calcium and vitamin D was 55% in intervention group compared to 20% in control group at 4-month follow-up.
50
Tab
le 1
0. F
alls
aft
er d
isch
arge
in s
tud
y 1,
at
4- a
nd
12-
mon
th f
ollo
w u
p a
nd
cu
mu
lati
vely
at
12 m
onth
s.
D
isch
arge
- 4
mon
ths
4 -
12 m
onth
s A
dm
issi
on -
12 m
onth
s
In
terv
enti
on
n=92
C
ontr
ol
n=83
**
Inte
rven
tion
n=
84
Con
trol
n=
76
Inte
rven
tion
n=10
2 C
ontr
ol
n=97
Num
ber
of fa
llers
26
26
(82)
31
30
44
55
Rec
urre
nt fa
llers
, ≥2
falls
13
12
14
14
29
35
Tota
l num
ber
of fa
lls
58
65
62
66
138
191
Num
ber
of fa
llers
am
ong
peop
le w
ith
dem
enti
a 8(
24)
11(3
0)
8(19
) 11
(28)
12
(28)
22
(36)
Rec
urre
nt fa
llers
am
ong
thos
e w
ith
dem
enti
a 5
6 5
6 7
14
Tota
l num
ber
of fa
lls a
mon
g pe
ople
wit
h de
men
tia
22
38
18
32
41
10
4
Num
ber
of fa
llers
wit
h fr
actu
res
due
to fa
lls
0 3
7 4
7 11
Num
ber
of d
emen
ted
falle
rs
wit
h fr
actu
res
due
to fa
lls
0 1
3 1
3 5
Obs
erva
tion
tim
e(da
ys)
9206
71
69
2093
2 18
653
3320
2 29
707
Cru
de fa
ll in
cide
nce
rate
6.
30/1
000
9.07
/100
0 2.
96/1
000
3.5
4/10
00
4.16
/100
0 6
.43/
1000
Inci
denc
e ra
te r
atio
wit
h 95
%
CI
IRR
0.5
5*
(0.2
7-1.
12)
IR
R 0
.85*
(0
.48-
1.50
)
IRR
0.6
4*
(0
.40-
1.02
)
* N
egat
ive
bino
mia
l reg
ress
ion
anal
yses
adj
uste
d fo
r ov
er-d
ispe
rsio
n an
d co
ntro
lled
for
dem
enti
a an
d de
pres
sion
. **
One
per
son
mis
sing
, dec
lined
at 4
-mon
th fo
llow
-up
but w
as fo
llow
ed u
p at
12
mon
ths.
If
det
ails
are
not
giv
en fo
r th
e co
mpl
ete
grou
p, th
e nu
mbe
r of
sub
ject
s is
giv
en in
par
enth
esis
.
51
Paper II A history of cardiovascular disease, cancer, dependence in P-ADL, dementia, having three or more co-morbidities, dependence in walking, having an ASA score of 3 or higher, living in residential care facilities, male gender, depression and pulmonary disease were all independently associated with time to death in univariate Cox regression analyses (Table 11). The results from multivariate Cox regression analyses are also displayed in the columns to the right of Table 11. The hazard ratios (HR) and 95 % confidence intervals (CI) are presented for the significant variables remaining in the multivariate models. Analysis of the associations between baseline variables and time to death shows that cancer, cardiovascular disease, dependence in P-ADL and dementia were associated with time to death in a Cox proportional hazard regression model adjusted for age and sex (Table 11. Model A). In a second model (Table 11. Model B.), complications during hospitalization were analyzed; pulmonary emboli, pneumonia, cardiac failure and delirium were all associated with time to death. In the final model (Table 11. Model C.) all significant variables in the two multivariate models above were combined; pulmonary emboli (HR 69.396, CI: 7.107–677.632), cancer (HR 3.393, CI: 1.959–5.877), cardiac failure (HR 2.221, CI: 1.148–4.294), cardiovascular disease (HR 2.026, CI: 1.160–3.539), dementia (HR 1.883, CI: 1.091–3.250) and dependence in P-ADL (HR 2.362, CI: 1.271–4.387) remained independently associated with all-cause mortality adjusted for age and gender. Mortality was 13/199 (6 %) at 30 days and 34/199 (17 %) at 1 year after the index hip fracture. Twenty-six out of 51 (51 %) men died compared to 53/148 (36 %) women. After discharge, a total of 65 participants died during the 3 years of follow-up. From admission until 3-year follow-up 79 out of 199 participants (40 %) died. (Table 12). In this study 13 participants died within 30 days of sustaining the index hip fracture. Ten of these 13 died during hospitalization: one before the operation, due to myocardial infarction; two on the day of operation, due to cardiovascular events; two the day after the operation (one due to gastrointestinal bleeding and one to myocardial infarction); one suffered a cerebrovascular event during the operation and never regained consciousness; one suffered a myocardial infarction after a week; two died of pneumonia and one died due to an intestinal infarction. The partici-pants who died during hospitalization due to a cardiovascular event all had a history of cardiovascular disease. The three participants who died after discharge but within 30 days of the index hip fracture did so due to
52
a cerebrovascular event, a myocardial infarction and a pulmonary emboli, respectively. The secondary cause of death within 30 days of the index fracture was due to a cardiovascular event in 6/13(46 %) cases (Table 13). In addition, four participants died within 30 days after sustaining a new fracture. One of these participants fell and suffered a new hip fracture while still in hospital and then died due to an infection. After discharge one participant suffered a hip fracture and died due to pneumonia, one sustained a head trauma with a face fracture and died due to intracerebral bleeding and one had a knee fracture and died due to rupture of the aorta (Table 13.). Cardiovascular events (18/65), dementia (19/65), cancer (10/65), fractures (6/65) and cerebrovascular events (4/65) were the most common primary causes of death after discharge (Table 12). In total, from admission until 3-year follow-up, 166 participants suffered from 583 infections, including 136 participants who suffered 363 urinary tract infections. Seventy-four participants suffered 111 cardiovascular events during the 3 years following the hip fracture. One-hundred and fourteen participants suffered 542 falls. Thirty-seven participants suffered 56 new fractures, including 13 new hip fractures, seven wrist fractures and seven rib fractures. Forty-nine participants suffered 60 decubital ulcers. During follow-up 96 participants had 234 hospital admissions with a total of 3984 days spent in hospital. Infections and cardiovascular events were more common after discharge among those who died. No differences in the incidence of complications between intervention and control group in study 1 was seen (data not shown).
53
Tab
le 1
1. B
asel
ine
char
acte
rist
ics
and
com
pli
cati
ons
amon
g p
arti
cip
ants
an
d d
iffe
ren
ces
amon
g d
ecea
sed
an
d s
urv
ivor
s. U
niv
aria
te
and
mu
ltiv
aria
te C
ox r
egre
ssio
n a
nal
yses
.
All
cas
es
Dec
ease
d
Su
rviv
ors
Un
ivar
iate
M
ult
ivar
iate
n=19
9(%
) n=
79
n=12
0 P
¹ P
² H
R
CI
Bas
elin
e ch
arac
teri
stic
s:
Mod
el A
.
A
ge, m
ean
± S
D
82.2
±6.
2 82
.9±
5.8
81.7
±6.
5 0.
233
0.43
5 1.
016
0.97
7-1.
056
Fem
ale
148(
74%
) 53
95
0.
029
0.05
2 0.
604
0.36
3-1.
005
Livi
ng in
res
iden
tial
car
e fa
cilit
ies
73(3
7%)
37
36
0.02
8
Livi
ng a
lone
14
6(74
%)
59
87
0.69
6
Cur
rent
sm
oker
(n=
169)
6(
4%)
2 4
0.95
5
Dep
ende
nce
in w
alki
ng (n
=19
5)
25(1
3%)
15
10
0.00
4
Dep
ende
nce
in P
-AD
L 11
5(58
%)
59
56
<0.
001
0.00
5 2.
379
1.30
0-4.
354
Can
cer(
n=19
4)
29(1
5%)
21
8 <
0.00
1 <
0.00
1 3.
449
2.02
0-5.
890
Car
diov
ascu
lar
dise
ase(
n=19
4)
110(
57%
) 55
55
<
0.00
1 0.
004
2.17
7 1.
287-
3.68
2 D
emen
tia
64(3
2%)
37
27
0.00
1 0.
016
1.92
1 1.
131-
3.26
2 D
epre
ssio
n(n=
197)
78
(40%
) 39
39
0.
040
D
iabe
tes(
n=19
7)
40(2
0%)
22
18
0.05
9
Kid
ney
dise
ase(
n=19
4)
21(1
1%)
10
11
0.32
1
Pulm
onar
y di
seas
e(n=
194)
33
(17%
) 17
16
0.
049
St
roke
(n=
195)
49
(25%
) 23
26
0.
184
A
SA g
rade
3-4
(n=
197)
10
9(55
%)
51
58
0.00
7
Inte
rnal
fixa
tion
69
(35%
) 33
36
0.
420
H
emia
rthr
opla
sty
111(
56%
) 37
74
0.
228
D
ynam
ic h
ip s
crew
17
(9%
) 7
10
0.99
8
Oth
er
1(0.
5%)
1 0
0.34
5
No
of. c
omor
bidi
ties
:
0 31
(16%
) 3
28
<0.
001
1
37(1
9%)
7 30
0.
260
54
Tab
le 1
1. B
asel
ine
char
acte
rist
ics
and
com
pli
cati
ons
amon
g p
arti
cip
ants
an
d d
iffe
ren
ces
amon
g d
ecea
sed
an
d s
urv
ivor
s. U
niv
aria
te
and
mu
ltiv
aria
te C
ox r
egre
ssio
n a
nal
yses
.
All
cas
es
Dec
ease
d
Su
rviv
ors
Un
ivar
iate
M
ult
ivar
iate
n=19
9(%
) n=
79
n=12
0 P
¹ P
² H
R
CI
No
of. c
omor
bidi
ties
:
2 52
(26%
) 25
27
0.
005
≥
3 79
(40%
) 44
(56%
) 35
(29%
) 0.
001
C
omp
lica
tion
s d
uri
ng
hos
pit
aliz
atio
n:
Mod
el B
.
Pneu
mon
ia/c
hest
infe
ctio
n 8
6 2
0.00
1 0.
021
2.91
6 1.
175-
7.23
7 U
rina
ry tr
act i
nfec
tion
82
38
44
0.
241
O
ther
infe
ctio
n 34
15
19
0.
528
W
ound
infe
ctio
n 0
0 0
Dee
p w
ound
infe
ctio
n 1
1 0
0.05
4
Car
diac
failu
re
17
12
5 0.
001
0.01
0 2.
457
1.24
2-4.
858
Myo
card
ial i
nfar
ctio
n 6
6 0
0.05
2
Dee
p ve
in th
rom
bosi
s 1
1 0
0.05
0
Pulm
onar
y em
bolis
m
1 1
0 0.
002
0.00
1 33
.219
3.
961-
278.
570
Stro
ke
3 2
1 0.
072
C
ance
r 3
2 1
0.12
0
Gas
tric
ulc
er
7 3
4 0.
538
D
elir
ium
12
9 63
70
0.
005
0.01
8 2.
006
1.12
7-3.
570
Falle
rs
38
19
19
0.26
7
Num
ber
of fa
lls
78
41
37
0.26
2
Frac
ture
4
2 2
0.60
9
Luxa
tion
6
3 3
0.41
6
Reo
pera
tion
8
3 5
0.92
6
Dec
ubit
al u
lcer
s 30
16
14
0.
044
55
Tab
le 1
1. B
asel
ine
char
acte
rist
ics
and
com
pli
cati
ons
amon
g p
arti
cip
ants
an
d d
iffe
ren
ces
amon
g d
ecea
sed
an
d s
urv
ivor
s. U
niv
aria
te
and
mu
ltiv
aria
te C
ox r
egre
ssio
n a
nal
yses
.
All
cas
es
Dec
ease
d
Su
rviv
ors
Un
ivar
iate
M
ult
ivar
iate
n=19
9(%
) n=
79
n=12
0 P
¹ P
² H
R
CI
Com
bin
ing
bas
elin
e an
d
com
pli
cati
on f
acto
rs:
Mod
el C
.
Age
0.
705
1.00
8 0.
968-
1.04
9 Fe
mal
e
0.
118
0.65
5 0.
386-
1.11
3 D
epen
denc
e in
P-A
DL
0.00
7 2.
362
1.27
1-4.
387
Can
cer
<0.
001
3.39
3 1.
959-
5.87
7 C
ardi
ovas
cula
r di
seas
e
0.
013
2.02
6 1.
160-
3.53
9 D
emen
tia
0.02
3 1.
883
1.09
1-3.
250
Pneu
mon
ia/c
hest
infe
ctio
n
Car
diac
failu
re
0.01
8 2.
221
1.14
8-4.
294
Pulm
onar
y em
bolis
m
<0.
001
69.3
96
7.10
7-67
7.63
2 D
elir
ium
¹ p
acco
rdin
g to
uni
vari
ate
Cox
reg
ress
ion
mod
el
² p
acco
rdin
g to
mul
tiva
riat
e C
ox r
egre
ssio
n m
odel
SD
= S
tand
ard
Dev
iati
on
P-A
DL
= P
erso
nal A
ctiv
itie
s of
Dai
ly L
ivin
g A
SA =
Am
eric
an S
ocie
ty o
f Ana
esth
esio
logi
sts
clas
sific
atio
n H
R =
Haz
ard
Rat
io
CI
= 9
5% C
onfid
ence
Int
erva
l
56
Tab
le 1
2. P
rim
ary
cau
ses
of d
eath
.
Cau
se o
f d
eath
D
uri
ng
hos
pit
aliz
atio
n
Dis
char
ge -
4 m
onth
s 4
- 12
m
onth
s 12
– 3
6
mon
ths
Tot
al
Car
diov
ascu
lar
0 1
2 15
18
Dem
enti
a 0
0 4
15
19
Hip
frac
ture
11
3
0 1
15
Can
cer
0 2
3 5
10
Cer
ebro
vasc
ular
0
0 1
3 4
Infe
ctio
n 1
0 0
1 2
Dee
p in
fect
ion
0 0
0 2
2
Ren
al fa
ilure
1
0 1
0 2
Frac
ture
0
1 1
0 2
Gas
troi
ntes
tina
l ble
edin
g 1
0 0
0 1
Rup
ture
d ao
rtic
ane
urys
m
0 0
0 1
1
Park
inso
n`s
dise
ase
0 0
0 1
1
Panc
reat
itis
0
0 0
1 1
Gan
gren
e 0
1 0
0 1
Sum
14
8
12
45
79
Aut
opsy
7
2 1
4 15
57
Tab
le 1
3. S
econ
dar
y ca
use
s o
f d
eath
.
Cau
se o
f d
eath
Du
rin
g h
osp
ital
izat
ion
D
isch
arge
- 4
m
onth
s 4
- 12
m
onth
s 12
- 36
m
onth
s T
otal
Dem
enti
a 0
0 4
15
19
Car
diov
ascu
lar
5 2
2 15
24
C
ance
r 0
2 3
5 10
C
ereb
rova
scul
ar
1 2
1 3
7 In
fect
ions
4
0 0
2 6
Dee
p in
fect
ion
0 0
0 2
2 R
enal
failu
re
1 0
1 0
2 G
astr
oint
esti
nal b
leed
ing
2 0
0 0
2 R
uptu
red
aort
ic a
neur
ysm
0
0 1
1 2
Park
inso
n’s
dise
ase
0 0
0 1
1 Pa
ncre
atit
is
0 0
0 1
1 G
angr
ene
0 1
0 0
1 in
test
inal
infa
rcti
on
1 0
0 0
1 pu
lmon
ary
embo
li 0
1 0
0 1
Sum
14
8
12
45
79
Aut
opsy
7
2 1
4 15
58
Paper III There were no significant differences between the GIHR and control groups in independent walking ability either indoors or outdoors at 3 and 12 months, or in use of walking devices (Table 14). Walking ability deteriorated in both groups. At the 3-month follow-up, 49 (51.6%) participants in GIHR group and 48 (54.5%) participants in the control group had regained or improved their pre-fracture walking ability level (p = 0.800). At 12 months, the totals were 45 (56.3%) and 45 (57.7%) in the GIHR and control groups, respectively (p = 0.982). Two participants in the GIHR group and 1 participant in the control group were not able to walk before the fracture. These numbers increased to 8 (8.4%) versus 3 (3.4%) at 3 months for the GIHR and control groups, respectively, and to 9 (11.3%) versus 8 (10.3%) at 12 months, but there were no significant differences between the groups. The use of a walker indoors did not differ between the groups. Before the fracture, 45.8% of participants in the GIHR group and 43.9% of participants in the control group walked with a walker on wheels indoors, and 12 months after the fracture the proportions were 51.2% and 57.7% for the GIHR and control groups, respectively. Gait speed, both self-chosen and maximum, were almost identical for the groups at the 3- and 12-month follow-up visits (Table 15). Postoperative LOS was significantly shorter for the GIHR group compared with the control group. LOS from admission to the geriatric ward to discharge was a median (Q1-Q3) of 17 days (12-26) versus 23 days (17-32) for the GIHR and control groups, respectively (p = 0.003). LOS from admission to the geriatric ward to DRD was a median (Q1-Q3) of 15 days (11-22) versus 21.5 days (16-29) for the GIHR and control groups, respectively (p< 0.001). Moreover, when total postoperative LOS after the hip fracture was analyzed, the GIHR group had a significantly shorter LOS, with a median (Q1-Q3) of 22 days (15-34) compared with 26.5 days (19-38) for the control group (p = 0.021). There were no differences between the groups in the 1-year mortality rate. The rates were 19.6% in the GIHR group and 16.3% in the control group (p = 0.666). The GIHR team made an average of 14.2 ± 10.5 visits in the participants’ homes (0-50). Number of days in the GIHR team was a median (Q1-Q3) of 21 days (11.0-35.5). One-third of the participants in the control group were followed-up in primary healthcare or in outpatient rehabilitation during the year after discharge. In the GIHR group, approximately 10% of participants received additional rehabilitation after the intervention ended.
59
Table 14. Walking ability and use of walking device for the 2 groups before fracture and at the 3- and 12-month follow-up visits in study 2.
GIHR* n = 107
Control n = 98
OR†
95% CI‡
Walking independently indoors, n (%) Before fracture At 3-month follow-up, (n = 95/88) At 12-month follow-up, (n = 80/78) Walking independently outdoors, n (%) Before fracture At 3-month follow-up, (n = 95/88) At 12-month follow-up, (n = 80/78) No walking device indoors, n (%) Before fracture At 3-month follow-up, (n = 95/88) At 12-month follow-up, (n = 80/78) No walking device outdoors, n (%) Before fracture At 3-month follow-up, (n = 95/88) At 12-month follow-up, (n = 80/78)
95 (88.8) 54 (56.8) 53 (66.3) 70 (65.4) 41 (43.2) 39 (48.8) 53 (49.5) 15 (15.8) 24 (30.0) 33 (30.8) 3 (3.2) 8 (10.0)
85 (86.7) 57 (64.8) 56 (71.8) 71 (72.4) 39 (44.3) 38 (48.7) 47 (48.0) 11 (12.5) 21 (26.9) 34 (34.7) 3 (3.4) 7 (9.0)
0.84 0.84 1.76 1.50 1.91 1.41 0.80 1.20
0.39–1.80 0.35–2.06 0.83–3.75 0.69–3.28 0.72–5.03 0.59–3.33 0.14–4.80 0.36–4.01
Binary logistic regression analysis adjusted for age, sex, pre-fracture status of the outcome variable, and significant differences between the groups at baseline (antidepressants, analgesics). * GIHR = Geriatric Interdisciplinary Home Rehabilitation † OR = Odds Ratio of being treated in the GIHR group‡ CI = Confidence Interval Table 15. Self-chosen and maximum gait speed over 2.4 m for the 2 groups at the 3- and 12-month follow-up visits in study 2.
GIHR* Control
p-value
Self-chosen gait speed, m/s ± SD At 3 months, (n = 80/76) At 12 months, (n = 68/70)
0.43 ± 0.19 0.49 ± 0.19
0.43 ± 0.20 0.48 ± 0.17
0.899 0.945
Maximum gait speed, m/s ± SD At 3 months, (n = 77/72) At 12 months, (n = 67/68)
0.70 ± 0.31 0.74 ± 0.30
0.69 ± 0.29 0.75 ± 0.27
0.845 0.846
* GIHR= Geriatric Interdisciplinary Home Rehabilitation
60
Paper IV There were no significant differences between GIHR (n=106) and the control group (n=93) in terms of complications or readmissions after discharge. Between discharge and the 12-month follow-up comparisons between the participants in GIHR and the control group are as follows: 46 (43.4%) vs. 38 (40.9%) fell (p=0.828); 13 (12.3%) vs. 6 (6.5%) suffered an additional fracture (p=0.250); 36 (34.0%) vs. 30 (32.3%) presented with infection (p=0.917); 12 (11.3%) vs. 6 (6.5%) suffered a cardiovascular event (p=0.344); 38 (35.8%) vs. 27 (29.0%) were readmitted to hospital (p=0.383); and the median number of days spent in hospital was 11.5 vs. 11.0 (p=0.353). (Table 16). Fifty-seven (53.8%) participants in the GIHR suffered a complication (including medical and surgical complications) after discharge vs. 44 (47.3%) in control group (p=0.443). After adjustment for age, gender, baseline differences and observation time, the risk of falling during the period from discharge to the 12-month follow-up did not differ between the GIHR and the control groups, 46/106 vs. 38/93, odds ratio 0.99 (95% CI 0.53-1.88). Subgroups analysis revealed no differences in number of complications, readmissions, or days spent in hospital when the intervention and control groups were compared -according to the stratification, type of housing, and type of fracture (data not shown).
61
Table 16. Complications during hospitalization and at 12-month follow-up.
Complications In hospital Discharge -12 months
GIHR Control P GIHR Control P n = 107 n = 98 n = 106(%) n = 93(%)
Infection 51 41 0.486 36(34.0) 30(32.3) 0.917
-Pneumonia/chest infection 13 9 0.646 11(10.4) 10(10.8) 1.000
-Urinary tract infection 38 28 0.361 28(26.4) 23(24.7) 0.913
-Superficial wound infection 4 4 1.000 0 1(1.1) 0.467
-Deep wound infection 2 1 1.000 1(0.9) 0 1.000
Cardiovascular event 10 8 0.959 12(11.3) 6(6.5) 0.344
-Cardiac failure 10 7 0.751 10(9.4) 4(4.3) 0.256
-Myocardial infarction 1 1 1.000 3(2.8) 2(2.2) 1.000
Deep vein thrombosis 0 1 0.478 0 0
Pulmonary emboli 2 0 0.499 0 1(1.1) 0.467
Stroke 1 1 1.000 4(3.8) 4(4.3) 1.000
Gastric ulcer 2 1 1.000 2(1.9) 1(1.1) 1.000
Decubital ulcers 27 20 0.513 13(12.3) 13(14.0) 0.883
Fallers 24 19 0.717 46(43.4) 38(40.9) 0.828
Falls 33 27 0.662 163 113 0.700
Additional fracture 1 0 1.000 13(12.3) 6(6.5) 0.250
Luxation 2 0 0.499 2(1.9) 0 0.500
Reoperation 5 4 1.000 8(7.5) 5(5.4) 0.741
Deceased 1 2 0.607 20(18.9) 14(15.1) 0.60
Delirium 84 69 0.242
Days with delirium, median, 3.0(1-7) 3.0(0-7) 0.745
LOS, median, IQR 17.0 23.0 0.003
Readmission after discharge 38(35.8) 27(29.0) 0.383
Number of readmissions after discharge 58 53 0.426
Days in hospital after discharge 560 502 0.353
Median number of days in hospital, IQR, after discharge 11.5(5-20) 11.0(3-36) 0.353 Note: Numbers in parentheses indicate that values are missing. P = Differences between control and GIHR groups according to Pearson´s chi-square, Student´s t-test, the Mann-Whitney U test or Fisher´s exact test as appropriate. GIHR: Geriatric Interdisciplinary Home Rehabilitation; IQR: Inter quartile Range; LOS: Length of stay
62
Additional results In study 2, no significant differences between the GIHR and control groups were found when comparing mortality and readmission at 30 days following discharge from hospital. Four individuals died in the GIHR group vs. 1 in the control group (p=0.374) and 10 individuals were readmitted to hospital in the GIHR group vs. 6 in the control group (p=0.610). No significant differences between the GIHR and control groups were found when comparing mortality and readmission at 30 days following admission to hospital. No one person died in the GIHR group vs. 1 in the control group (p=0.467) and 3 individuals from the GIHR group and one from the control group were readmitted (p=0.624). Among those who had a LOS of 10 days or less, 10 in the GIHR group vs. 2 in the control group, none were readmitted or died within 30 days of admission, nor within 30 days of discharge.
In study 1 83/185 (45%) and in study 2 84/199 (42%) fell between discharge and 12-month follow-up and 92/160 (57%) vs. 90/158 (57%) regained the same level of walking ability as before the fracture. The rate of readmissions was 62/185 (34%) in study 1 and 65/199 (33%) in study 2 within one year after fracture.
63
DISCUSSION The results in this thesis reveal that an in-hospital orthogeriatric care model using CGA, which reduced in—hospital falls and complications, had no significant effect on falls after discharge. Old individuals who sustain hip fracture have many co-morbidities and suffer from numerous complications. Co-morbidities at baseline and complications during hospitalization are both associated with mortality. The regaining of walking ability is similar for individuals that participated in GIHR, who have a significantly shorter LOS in hospital, and those who received conventional geriatric care. No significant differences were seen regarding complications, readmissions and total days spent in hospital after discharge, between participants in GIHR and those receiving conventional geriatric care.
Falls and fractures Although falls are one of the most hazardous complications, few studies investigating orthogeriatric care models have included falls as an outcome measure, either in-hospital or during follow-up. Falls after discharge were reported in one study of orthogeriatric care by Shyu et al.114,129 The risk of falling after discharge in the intervention group in study 1 was lower than in the control group, but not significant. A power calculation showed that a larger sample size would have been required to detect significant differences in fall rates. In contrast to our results Shyu et al. reports a significant risk reduction of subsequently falling between intervention and control groups after 12 months and up to 2 years after discharge.114 One explanation for the inconsistency with the results from study 1 might be that individuals with severe cognitive impairment and those with weak muscle power were excluded, resulting in a population more responsive to fall-prevention interventions.59 Furthermore, the participants in their intervention group received home visits from a nurse and a PT after discharge, whereas no home visits were made in study 1. In study 2, the in-hospital orthogeriatric care model in study 1 was combined with a GIHR team, trained in CGA. In GIHR fall prevention was given special priority, but despite this there was no significant reduction in number of falls compared to the control group. On the contrary, there was a tendency for the intervention group to have more falls and fractures, but the difference was not significant. One can only speculate that the intervention group may have been more fall-prone, perhaps related to more prescribing of antidepressants and Parkinson medications at baseline.
64
Two previous team-based HR studies report falls after discharge,147,148 neither showing significant differences between intervention and control groups. However, differences in lengths of follow-up and in methodology used for data analysis make comparison with our results difficult. In another team-based HR study by Ziden et al., the reported number of falls during the period from discharge to the 12-month follow-up is similar to that in study 2, although individuals with severe cognitive impairment and those living in care facilities were not included and the intervention was shorter.145 These findings are surprising since we expected to have a higher fall rate than in their population. However, they did not report any complications after discharge. It is known that complications might lead to falls, 50 and since we do not know whether the rate of complications in their study was higher than in ours, this might provide an explanation for the lack of difference. Another team-based HR study reports fall rates of 26-33% between assessments; even though their intervention lasted for one year no difference in fall rates between intervention and control groups was seen.150 The proportion of fallers after discharge up to the 1-year follow-up in study 1 (83/185, 45%) and study 2 (84/199, 42%) was high compared to the expected fall rate among individuals in the community,34,35 although not compared to those in residential care facilities.175,176 This was not unexpected since the participants were a mix of those living in residential care facilities and ordinary housing before fracture, and individuals with hip fracture are at high risk of falling again.39,40 The results regarding fallers and number of falls after discharge in studies 1 and 2 were similar, although the study populations differed. Study 2 included individuals with trochanteric fractures, which study 1 did not, and there was also a larger proportion of individuals with cognitive impairment and with ≥3 comorbidities in study 2. These findings might indicate that the participants in the second study were frailer individuals, but even so the number of falls were not higher. The orthogeriatric care model described in this thesis, using CGA and a systematic prevention, detection, and treatment of complications, reduced in-hospital falls and complications,111,126 such as delirium, urinary tract infection and decubital ulcers. Considering that no effect was seen on these outcomes after discharge, not even when GIHR was added, there is a need for reflection. Delirium was the main focus during hospitalization, since it is strongly associated with falls.40 The effect on falls in hospital might have been due primarily to a reduction in delirium. Moreover, urinary tract infections were common among people with hip fracture, during hospital stay and after discharge, in both studies in this thesis. Since urinary tract infection is an important fall risk factor,177 its reduction during hospitalization probably contributed to the decreased
65
in-hospital fall rate. Furthermore, the interdisciplinary team work, performed by specially trained personnel111 working around the clock on the geriatric ward as an important part of fall prevention, was not maintained after discharge. Knowing that falls are associated with several risk factors,34,47,50,51 it is possible to speculate that since the close focus on fall-risk factors during hospitalization was not continued after discharge, partly due to less supervision, there was less prospect of preventing falls. Consequently, in studies 1 and 2 the prevention of falls and complications was not as comprehensive and systematic after discharge as during the hospital stay. Furthermore, the intermediate groups regarding cognitive function and ADL functions have a high risk of becoming fallers,178 thus those who can rise from a chair but need help when walking and have less insight into their abilities have a high risk of falling. Our results show that 32% vs. 50% of the population in studies 1 and 2 had dementia, and only 57% (92/160)137 vs. 57% (90/158) regained the same level of walking ability as before the fracture. These findings might imply that a large proportion of participants in studies 1 and 2 were “at risk of falling”, which could have had an impact on the results. In addition, a hypothesis was put forward by Cameron et al. in a Cochrane report where they suggest that exercise programs might increase falls in frail residents and reduce falls in less frail residents.55 Since fall prevention among people with dementia has not been successful so far, according to a clinical practice guideline,59 better supervision might be needed for these individuals. Considering other aspects of fall prevention, Boonen et al. report in a review that vitamin D significantly improved body sway and lower extremity strength, reducing the risk of falling.179 When vitamin D is prescribed together with calcium the risk of fracture is somewhat reduced among old people.180,181 However, in Sweden the recommendation today is to only prescribe vitamin D for those who have a diagnosed deficiency.182,183 According to two Cochrane reports, prescription of vitamin D reduces the number of falls in care facilities, probably because residents have low vitamin D levels,55 and it might also have an effect among individuals living in the community who have low levels prior to treatment.56 In study 1 vitamin D and calcium seem to have had no effect on reducing falls, even though the prescription rate in the intervention group was higher (66% vs. 22%). However, no vitamin D concentrations were measured. The prescription of bisphosphonates was not reported in studies 1 and 2. There is under-treatment in this respect according to the Swedish National Board of Health and Welfare183 even though bisphosphonate treatment is recommended. The use of Bisphosphonates is supported by Nordström et al. who reports an association with reduced risk of hip
66
fracture, with comparable effects in people aged >80 years.184 Antidepressants were also associated with falls in a study by Kallin et al.52 However, there was no significant difference in the number of falls between intervention and control groups in either study 1 or 2, even though there was a significant difference in prescriptions of antidepressants. In conclusion, this thesis confirms that older individuals run a high risk of falling after hip fracture, since almost half of the population in studies 1 and 2 fell again during the period from discharge to the 12-month follow-up and the participants in study 1 suffered 464 falls and 56 new fractures during 3 years of follow-up. Previous research reports that multifactorial interventions can reduce falls,55,56 thus better adherence to national recommendations, including osteoporosis treatment, individualized fall prevention and supported physical training, is urgently needed to prevent falls.183 There is also a crucial need for much greater effort from society, since fall accidents are the most costly accidents even though they are the lowest funded compared to other accidents, such as car accidents.44 Considering that 95% of hip fractures are caused by a fall,37 primary and secondary fall and fracture prevention need to become a part of the routine care for older individuals.
Complications, readmissions and mortality Complications The two studies in this thesis highlight the situation that old individuals with hip fracture suffer from numerous complications after discharge from hospital. In study 2, 101/199 (51%) had ≥1 complication (medical and surgical) after discharge up to 1 year after surgery. Despite the importance of complications, several orthogeriatric studies,113,118,121 and cohort studies64-66 including old people with hip fractures, register only medical and surgical complications during acute hospital stays. Among team-based HR studies complications are also only sparsely reported after the initial hospital stay. Furthermore, in the published literature complications have a range of definitions and the time points for reporting differ, making comparison difficult. A review by Goldhahn et al. concludes that there is a lack of homogeneity among orthopedic clinical trials when complications are reported. They suggest that complications should be defined prior to start of any study, assessment should be blinded, the complications monitored by an independent data review board and the observation time should be standardized.68 Complications in studies 1 and 2 were predefined, but assessment of complications was not blinded and there was no independent monitoring. More recently, a recommendation on how to
67
report orthogeriatric care models has been published, covering definitions of complications, which to include and time points of assessment.185 All the complications recommended in this report, apart from renal failure and adverse drug reactions, were included in studies 1 and 2, although the definitions of complications and the time point for assessment did not comply with the recommendations in all aspects. Infections, cardiovascular events and decubital ulcers were the most common complications, after discharge up to the 1-year follow-up, found in studies 1 and 2, when medical and surgical complications were included. A cohort study by Hansson et al. reports general complications up to 6 months after fracture, describing new falls, pneumonia and new fractures as being the most common.70 Another cohort study by Flikweert et al. reports that 75% of patients suffered ≥1 complication up to 6 months after fracture, and describes delirium, pneumonia and heart failure as the most frequent complications.186 Both of these studies include complications during hospital stay and the reporting of complications is inconsistent. One team-based HR study by Tinetti et al. reports pain related to exercising during the 6 months of follow-up, though no significant differences between intervention and control groups was seen.147 Pain was not reported as a complication in our study, even though it is common among people who have suffered a hip fracture.11 The intervention group in study 2 had a tendency towards more complications, but it was not significant. The low power concerning complications in study 2 means that the results must be interpreted with caution. As a result of increasing healthcare costs and changes in healthcare payment systems complications have attracted increasing interest in the United States and the United Kingdom.98,187 Large cohort studies report complication rates up to 30 days after surgery of between 19-25%.67,72 Complications 30 days after surgery were not analyzed in studies 1 and 2. The findings in this thesis, and the reported association with both functional outcome70 and mortality,72 might imply that medical complications beyond the recommended 30 days after admission should be reported.185 Readmissions Study 1 confirms a high rate of readmissions among old people with hip fracture. During the 3-year follow-up 96 participants had 234 hospital admissions, a total of 3984 days were spent in hospital. The rate of readmissions was 62/185 (34%) in study 1 and 65/199 (33%) in study 2 within one year after fracture. This is comparable to a cohort study by Merchant et al. who present a readmission rate of 31.7%.66 There was a
68
higher proportion of readmissions in the GIHR than in the control group, 35.8% vs. 29.0%, but it did not reach significance. One explanation could be that those in the intervention group were frailer and suffered more complications leading to hospitalization. Two previous team-based HR studies for older people with hip fracture evaluated readmissions after discharge,147,148 Crotty et al. report means of admissions, and Tinetti et al. have an admission rate of 11% vs. 13% in an intervention and a control group in a 6-month follow-up. However, people with cognitive impairment and those living in residential care facilities were not included in these studies, and the inconsistency in methods of reporting makes comparison difficult. Early readmission within 30 days of discharge has become a mesure of performance for hip fracture care. The 30-day readmission rate in study 2 was 16/199 (8%), similar to a recent review by Ali et al. who report a median all-cause 30-day readmission rate after hip fracture of 10.1% (range = 4.5–23.1%).188 These reported rates include both 30 days following the first admission, and 30 days following discharge to readmission. Ali et al. also describe patient-related risks factors for readmission, such as age, ASA grade, co-morbidities and functional status, as stronger risk factors than hospital-related factors.188 The majority of readmissions after hip fracture are for medical reasons.188,189 According to Ali et al.188 pneumonia, cardio-vascular disease, renal dysfunction and gastrointestinal disorder were the most common reasons for readmission, and the most common surgical reasons were infection, dislocation and fixation failure. Risk factors and reasons for readmissions have not been analyzed in studies 1 and 2. Mortality The high mortality after hip fracture is confirmed in study 1, which shows that 79/199 (40%) had died by the 3-year follow-up. The in-hospital mortality in study 1 was 7% (14/199), which is higher than reported in the Swedish national hip register.11 However, these data were retrieved from orthopedic wards with a much shorter mean LOS. The 30-day mortality rate in study 1 was 6% (13/199), similar to that in earlier orthogeriatric models reporting mortality 30 days after fracture.119,120,123,134 Adunsky et al. report a mortality rate of 1.9 at 30 days. We found a 1-year mortality rate of 17% (34/199) in study 1, which is comparable to previously published reports,118,121,129,135 but lower and higher rates are also reported (9.7%-29%).119,120,127,130 Compared to the results in study 2, one-year mortality is lower in earlier team-based HR studies for older people with hip fracture.145,147,148
69
The inconsistency in mortality rates might be due to differences in LOS and in inclusion criteria for participation. The populations in studies 1 and 2 are presumably less healthy, with a poorer prognosis than those in studies which exclude people who have dementia/cognitive impairment or live in residential care facilities. Historically the LOS has decreased which might have resulted in lower in-hospital mortality rates in more recent studies due to a shorter observation time. In addition, differences between healthcare systems and the ability to transfer a person to end of life care in the community may also affect LOS and mortality rates. Therefore, a comparison of 30-day mortality is probably more relevant.185 The causes of excess death have been the subject of debate. Vestergard et al. suggest that the increased mortality is associated with postoperative complications,85 while others ascribe it to prefracture co-morbidities together with postoperative complications65,78 and yet others suggest that the co-morbidities are the underlying cause.190 Previous research reports several risk factors associated with death, although differences in methodology have probably contributed to the deviating results. We found that both comorbidities and post-operative complications were of significance, a finding that is in line with a recent review using data from a National Trauma Data Bank in USA.64 A cohort study by Roche et al. reports that male gender, cancer, chest infection, cardiac failure and stroke are the strongest predictors for 1-year mortality, although they do not report dementia or functional levels.65 Dementia was included in our model, since it is common among old individuals with hip fracture.12 In another prospective database cohort study, Pugely et al. find that the risk factors for mortality are age, male gender, reduced functional status, an ASA class >3, and cancer.191 These findings are only partly consistent with our findings. However, they included only baseline characteristics in their multivariate model and not complications as we did. In a recent study that examines post-mortem reports in people with hip fracture, respiratory infections and cardiovascular disease, are the main causes of death within 30 days after admission.88 These results are in line with earlier studies,85,89 and support the results of study 1, as 46 % (6/13) of early deaths were due to a cardiovascular event. Cardiovascular disease was the most common comorbidity and also a strong predictor of post-operative cardiac failure among people with hip fracture in a study by Roche et al.65 A large cohort study by Castronuovo et al.84 also cites heart disease as a risk factor for 30-day mortality. All participants in study 1 who died due to cardiovascular events during hospitalization, had a pre-fracture cardiovascular disease. This indicates that a hip fracture can destabilize an old individual with several co-morbidities thereby causing death. A prospective observational study by Juliebo et al. reports that treatment of cardiovascular disease among people with hip fracture is not
70
in accordance with current guidelines, thus, there might be room for improvement.83 During follow-up cardiovascular events, dementia and cancer were the most common causes of death in study 1, which is partly consistent with earlier studies.86,87,90 The difference in the prevalence of infection as a cause of death might be due to how the causes of death were determined in study 1, (see method section). There might also be an under diagnosis of dementia among many old people,192 which could affect the assessment of the cause of death. Participants in study 1 were tested for cognitive impairment during hospitalization and at 4, 12 and 36 months, thus increasing diagnostic validity.
Walking ability and LOS Walking ability Participants in the GIHR and control groups regained walking ability in both the short and long term in a similar way. Furthermore, no significant differences in walking ability, use of walking devices and gait speed were found when comparing participants in the intervention and control groups at follow-ups in study 2. In contrast, significant effects on walking ability are reported in three previous team-based HR studies for older people with hip fractures.145,149,151 However, the populations differed as mentioned previously, that is, individuals living in residential care facilities and those with severe co-morbidities were not included, individuals living alone151 and those with severe,145 and moderate149 cognitive impairment were also excluded. Despite the positive effects on walking ability reported, Kuisma et al.151 conclude that neither of the groups reached prefracture level and Ziden et al. report that only 29% vs. 9% in the groups thought that they had fully recovered.145 The decline in walking ability after hip fracture is confirmed in study 1 and 2, as only 57% regained or improved their walking ability by the 12-month follow-up compared to their prefracture status, according to S-COVS. This is consistent with previous studies.22,23 One reason why the effect on walking ability was similar in the GIHR and control groups might be that both groups received a multifactorial, multidisciplinary intervention program while in hospital. Compared to standard care, this intervention was successful in reducing postoperative complications, reducing LOS, and improving mobility and performance of ADL, both in the short and long term.111,137 In addition, the program was particularly successful in people with dementia.193 Other explanations for the lack of differences might be that the control group seems to have had
71
more PT and OT interventions than the GIHR group during the year after discharge/end of GIHR. The intervention may not have been comprehensive enough or the time period long enough to have an effect on walking ability.194 The duration, frequency and intensity of exercise are important to achieve improvements in physical function according to Liu et al.195 A previous HR study had a number of home visits similar to that in study 2.148 In a study by Ziden et al. there was only one third as many home visits, yet they report significant improvements in independence, balance confidence, and physical activity in the HR group. One possible explanation might be that these participants had the ability to exercise on their own.145 Our clinical experience is that cognitively impaired individuals and those living in residential care facilities have limited ability to exercise without supervision. Another important aspect to consider is the effect of complications on functional outcome. Hansson et al. report that general complications during the 6 months after fracture correlate to loss of function.70 A trend towards more complications, falls and fractures in the intervention group in study 2 could have affected walking ability. The numerous complications after discharge might also explain the low rate of regained walking ability among people with hip fracture. LOS The participants in GIHR had a significantly shorter postoperative LOS. In contrast, the earlier team-based HR studies described in this thesis show no reduction in the LOS in hospital, apart from Crotty et al. who report a mean LOS of 7.8 days vs. 14.3 days in a control group.148 Different settings and concepts in the team-based HR studies included might explain this discrepancy. Reduced LOS may reflect improved clinical outcome, however, during the last few decades LOS has become shorter, mostly due to lack of hospital beds and shrinking resources.11 As a result, the concept of using intermediate forms of care such as community facilities, skilled nursing homes and home-based rehabilitation have evolved. Furthermore, previous level of function, in-hospital complications,66 as well as the local healthcare organization can also affect the LOS.196 A shorter LOS might not be beneficial for older individuals with hip fracture. A recent study presents an association between a LOS shorter than 10 days and increased mortality among those discharged to short-term nursing homes.197 However, a post-hoc analysis in study 2 shows no significant differences in 30-day or 1-year mortality, when comparing individuals living in residential care facilities to those in ordinary housing. There were only 12 individuals with a LOS shorter than 10 days in study 2, and none of them died during follow-up.
72
Ethical considerations Medical research involving old individuals requires careful ethical consideration of whether the benefits outweigh the risks and burdens. A majority of the population sustaining hip fracture are old individuals with multiple comorbidities and many suffer from cognitive impairment. This group of people is less likely to receive rehabilitation following hip fracture, and their rehabilitation is shorter than that for older adults without dementia,198,199 although research indicates that they benefit from team rehabilitation.111,199-201 However, it is important to aim for evidence-based care for these individuals and they have been underrepresented in previous research. We considered it important to include these individuals, particularly as excluding them would also limit the external validity. GIHR can offer a task-specific training and individual support in the home environment, to help participants resume their pre-fracture activities. This might be especially valuable for people with dementia,(e.g. Alzheimer’s disease) because their ability to transfer skills is impaired and activities should be practiced in an environment that is similar to the one in which the skill will be used.202 People with cognitive impairment have a higher risk of falling than those without cognitive impairment, and this was an ethical consideration when aiming at early discharge for participants in GIHR group. This could mean putting people with cognitive impairment at a higher risk of falling because they had less supervision? Another aspect to consider was whether an early discharge would put an extra burden on the next of kin. These issues were taken into consideration when planning for discharge, which included planning for care required at home. Another ethical consideration in study 2 concerns the randomization; participants with cervical fractures were randomized before they had consented to participate for logistical reasons. However, the process of giving information and asking for consent was the same for all participants and it was emphasized that they could withdraw at any time without prejudice. Furthermore, all participants in the two intervention studies in this thesis were given oral and written information, and in those cases where the participants could not provide consent the next of kin were asked. All participants were informed that they could withdraw from the study at any time or stop an assessment. The assessors were experienced in assessing old people with cognitive impairment and were also sensitive to signs of discomfort or fatigue. Detection of medical issues during data collection in control groups was also an ethical consideration, however, all participants with medical problems were recommended to contact the appropriate healthcare service. Furthermore, the control group in study 2 did not receive rehabilitation at the geriatric out-patient clinic until 3 months after the fracture. This was done to minimize the
73
confounding effects at the 3-month follow-up. The second study in this thesis was registered at Current Controlled Trials to improve transparency in conduct and reporting, and to reduce the risk of bias.
Methodological considerations The major strength of studies 1 and 2, is the design, a randomized controlled study including a sealed randomization and an intention-to-treat analysis. Furthermore, only a small proportion of those eligible declined to participate in the studies. Incorporating people living in nursing homes and individuals with cognitive impairment/dementia in both study 1 and 2 improved the generalization of results. Another strength of our studies is that all complications were systematically analyzed. A protocol with definitions of each variable was drawn up prior to the chart review. The detailed medical information retrieved and the systematic analysis of the causes of death were another strength. In addition, all assessors in study 2 were blinded to group allocation. The fall definition used in this thesis is consistent with the recommended definition by The Prevention of Falls Network Europe Consensus.203 This study has some limitations, assessors were not blinded as to group allocation in the first study, or in hospital or during home visits. In order to reduce assessor bias the assessments in the intervention group were made by a nurse from the orthopedic unit and the assessments in control group were carried out by a nurse from the geriatric unit. The geriatricians working on the ward in the second study were occasionally responsible for both the intervention and control groups and, since they are responsible for discharges, this might unintentionally have influenced the LOS. One specialist in geriatric medicine who was not blinded to allocation, assessed all complication data retrospectively in both studies, which could have led to bias and also to a limited ability to ensure the accuracy of some of the diagnoses. For logistic reasons a selection bias became unavoidable in the second study, leading to the selection of individuals with pertrochanteric fractures needing a longer period of rehabilitation, presumably they were frailer. This reduces the external validity among the population of pertrochanteric fractures. In studies 1 and 2, the ASA risk score was used to assess the participants’ general health before surgery. This tool is simple, easy, inexpensive, and used routinely in our hospital, although it includes subjective variables. In study 1 only three assessments were performed after discharge, during the 3-year follow-up, which undoubtedly led to complications being missed, even though the participants’ medical charts were thoroughly reviewed. In addition, the fall outcome in Papers I, II and IV was only registered at follow-ups and when reviewing the medical charts. Some falls might have been missed, though presumably none that resulted in a fracture which
74
would have called for a medical assessment. Even so, the number of vertebral and rib fractures was probably underestimated as people with such fractures do not always seek medical care. Today, the use of prospective daily calendars with monthly reporting is the recommended method for recording falls.203 This was not done in the studies in this thesis. The number of outpatient rehabilitation periods were similar between the groups in study 1, however, more participants in the intervention group in study 1 were assigned to geriatric daytime rehabilitation care after discharge. The only data available in study 2 were the proportion of participants receiving an intervention from a PT or an OT after discharge from hospital/the end of GIHR. All fall-preventive interventions during follow-up might have affected the outcomes. Causes of death were obtained from death certificates. A comparison of cause-of-death data over time and across countries should be undertaken with caution. Although the intention of the International Classification of Diseases is to provide a standard means of recording underlying causes of death, the rules for selecting the underlying cause of death have been re-evaluated and sometimes changed. Incorrect or incomplete death certificates, misinterpretations of the rules of the International Classification of Diseases for selection of the underlying cause, and variations in the use of coding categories for unknown and ill-defined causes might all occur, according to WHO. Another limitation is that no gait speed tests were carried out at baseline because the in-hospital assessments took place soon after the participants had fractured their hips. Moreover, gait speed tests were performed with the participant’s usual walking device, which may limit the ability to detect initial gait and mobility deficiencies and changes over time.204 The sample size in study 1 was based on reduction of delirium, whereas in Paper I the power calculation was estimated retrospectively, based on fall assumptions. In study 2 the power calculation was based on assumed length of stay. The studies were not sufficiently powered to detect differences in low-incidence events, which increased the risk of making a Type 2 error, i.e. that there is a true difference that we cannot detect due to the smallness of the sample size. This must be considered when interpreting the results. Considering the age and morbidity of the participants, losses before follow-up were to be expected, which also resulted in missing values. No imputations were made in the statistical analyses. In Paper I, the Poisson regression model (Nbreg) was used for fall analysis since it takes into account observation time and frequent fallers, and the selection of adjusting variables in the analysis was based on known associations to falls. The multivariate analysis in Paper II should be interpreted with caution as only one person had pulmonary
75
emboli and died soon after the fracture. The studies included in this thesis were single-center trials, which should be considered when generalizing the results. It is not known what constitutes the optimal team-based HR model. The finding that there were no significant effects on outcomes, apart from a shorter LOS, in study 2, this calls for consideration. Perhaps the orthogeriatric care model in our study is of a higher quality than the average "usual care", reducing the possibility of finding any further benefit from GIHR. Another possible explanation lies in the study design; a longer and more comprehensive intervention might have yielded a better effect. A review by Littbrand et al. reported that, to have an effect on walking, exercise interventions among people with dementia should last for at least a few months, be task-specific and challenge an individual’s physical capacity.205 The home-based team and the in-hospital orthogeriatric care model had different preconditions. A dedicated personnel on a ward, focusing on the care of older people, can precipitate learning and the development of skills and expertise according to Ellis et al.106 Furthermore, working closely together facilitates more efficient interdisciplinary team work and team building. A mobile team, in contrast, often faces difficulties when trying to modify the behavior of other health professionals involved in the care of the individual.106 In a study of the effectiveness of a geriatric consultation team, Allen et al. report that compliance with the recommendations made by the team is sometimes minimal.206 In line with this Ellis et al. report that the positive effects of CGA were seen in dedicated wards and not in mobile teams in hospital.106
Clinical implications Since an orthogeriatric care model based on CGA can reduce falls and complications during hospital stay,111,126 compared to usual care, it should be offered to all older people suffering a hip fracture. In line with current recommendations, it should become a part of standard care. Furthermore, this model of care focusing on preventing, identifying and treating complications could, with some modifications, probably be used in all departments treating the old and vulnerable. It seems that GIHR is a complement to geriatric care and rehabilitation. Individuals with comorbidities, cognitive impairment and those living alone could participate in GIHR. The time spent in hospital could be reduced without any ensuing disadvantages, and walking ability could be regained to an extent comparable to those who received conventional care. Nevertheless, due to the low power of the study and a tendency for there to be more complications in the GIHR group, the results must be
76
interpreted with caution. In clinical practice, the GIHR team did not include participants from residential care facilities after the study ended. This decision is based on difficulties in interacting with staff leading to a lack of compliance with interventions recommended by the team. Earlier research reports positive effects on fall prevention by using multifactorial interventions both in residential care facilities and in the community.55,56 However, the research into how to prevent falls among people with dementia/cognitive impairment is inadequate.59 In contrast, the in-hospital orthogeriatric care model presented in this thesis reduced falls among those with dementia,193 by using CGA and a systematic prevention, detection and treatment of complications. Specific attention needs to be focused on precipitating factors, such as acute illness, urinary tract infections, acute drug side effects,177 and delirium,40 as well as on optimizing predisposing factors such as gait and balance; chronic diseases50 and a poor nutritional status73 should be heeded. A large proportion of people with hip fractures are cognitively impaired, 32-50% in our studies, and they have a low threshold for developing delirium. Delirium can develop within a short period of time and can fluctuate over the course of the day. Probably the most successful approach to preventing falls among these frail individuals is a sound understanding of delirium, the use of hip protectors where suitable60 and adequate supervision.177 Furthermore, considering that there is an under-treatment of osteoporosis, all clinicians meeting old individuals should be given further education regarding the risk factors for fractures.183 Looking to the future, the expected increase in the population of frail, old individuals,2 the reduction in the number of hospital beds and the limiting of access to residential care facilities, are bound to increase the incidence of falls and fractures. This situation can only be counteracted by putting in place interventions that adhere to guidelines and recommendations59,183 and by the provision of skilful personnel in care facilities and in social home services.
Implications for future research The studies reported in this thesis confirm the high risk of falling after hip fracture among older individuals, and also the occurrence of numerous complications after discharge. Considering that common complications, such as delirium, infections and aggravations of chronic diseases, such as cardiac failure, can cause falls and fractures,177 these should be the focus of future research. Fall prevention has to become a part of everyday care for frail, old individuals for it to be effective over time, and all personnel and next of kin living or working close to these individuals need to be
77
educated and informed about fall prevention. As many individuals with hip fracture need assistance in their daily life, future interventions among relatives, staff in social services working in the individual’s home and in residential care facilities, might be relevant. High quality methodological studies are needed in future research, including a representative hip fracture population, in order to find fall-preventive strategies that are effective over time. In addition, fragility fracture services need to be developed and evaluated in future research. In the future, the role of the geriatric team might be to collaborate with a variety of specialists, since old individuals with several comorbidities and highly complex circumstances will form a majority in the hospitals. They will need a geriatric assessment to optimize care and outcome. The co-managed care model is unlikely to cause harm to anyone and might become a standard concept in the care of frail, old individuals. This also calls for more clinical research. In the literature, the designs of orthogeriatric care models differ; they vary with respect to concepts of co-management, inclusion criteria, outcome variables, time points for evaluation and methods of assessment. The descriptions of the interventions are not always clear, and it may be unrealistic to provide explicit protocols. In more recent years, a standard set of outcome parameters has been identified, based on international guidelines98 and recommendations from an international expert group.185 Adherence to the recommendations will be important for future research as it will allow comparisons to be made between studies and results to be summarized. Moreover, large RCTs with interdisciplinary HR, including the entire group of people with hip fracture, are needed to help clinicians to determine which subgroup benefits the most from GIHR. In a world with limited resources economic analyses should accorded high priority. Furthermore, investigating the participants’ quality of life and experiences of GIHR would also be of value.
78
GENERAL CONCLUSIONS The orthogeriatric care model in this thesis successfully reduced in-hospital falls and complications, although it did not have a long-lasting effect on falls among older people with hip fracture, including people with cognitive impairment/dementia. Old people with femoral neck fracture have multiple co-morbidities, suffer numerous falls and complications, and are frequently readmitted to hospital. Co-morbidities at baseline and complications during hospitalization were associated with 3-year mortality. For older people with hip fractures, GIHR seems to complement conventional geriatric care and rehabilitation. Individuals with serious medical conditions, cognitive impairment and those living alone could receive GIHR. Participants in GIHR and those receiving conventional geriatric care regained their walking ability in the short- and long-term to a similar extent. The proportion of complications after discharge was no higher for those in GIHR, even though the initial time they spent in hospital was shorter. The results confirm that there is a deterioration in walking ability after hip fracture. It seems that the prevention of falls and complications has to be a part of everyday life in older people. This thesis suggests that primary and secondary prevention of falls and fractures needs to become a part of routine care.
79
ACKNOWLEDGEMENTS
I am finally here! I wish to take the opportunity to extend my gratitude to all who has contributed to this thesis in various ways. Thanks to you all it was possible for me to complete this work.
This work was carried out at the Department of Community Medicine and Rehabilitation, Geriatric Medicine, Umeå University.
My special gratitude goes to:
Yngve Gustafson, my supervisor, for valuable advice and unfailing support, for not giving up on me! Your passion for research into old people are inspirational. Thank you for giving me the opportunity to carry through my studies these years.
Michael Stenvall, my co-supervisor, for your support and advice, always listening and patient, pointing in the right direction, for discussions about life and being a friend.
Birgitta Olofsson, my co-supervisor, for excellent support and advice, encouragement to reach the final and being a friend.
Åsa Karlsson, my co-author, for support and advice, always a smile, and for being a friend.
Nina Lindelöf and Peter Nordström, my co-authors, for superb advice and guidance.
Undis Englund, my friend and co-author, for encouragement and support, thank you for believing in me, and for providing excellent working conditions.
Maria Lundström, for cooperation and useful opinions.
Henrik Holmberg, Annika Toots and Bodil Wiedung, for a great support and help in statistical analysis.
Chatarina Carlen, for assistance and support with practical matters.
Karina Sjögren and Elisabeth Johansson, for valuable advice and help with references.
80
All the participants and staff at the Orthopedic and Geriatric departments at the University Hospital in Umeå for their cooperation and participation. All the wonderful personal at the Ward 4, who took part in development, implementation and the ongoing work with the orthogeriatric care model and the Geriatric interdisciplinary home rehabilitation team. Börje Hermansson, for inspiring me to become a Geriatrician. Anita Persson, for support and engagement.
My colleagues and friends at the Department of Geriatric Medicine and at Geriatric Centre in Umeå, for valuable seminars, a pleasant working atmosphere and support in my work.
Patricia Shrimpton, for excellent language revision.
Larry Fredriksson, for fantastic computer support.
My family and friends, for making the world brighter.
Per, you are the love of my life, thank you for just being you! Andre, Caroline and Sandra, you are the best!
This work was supported by the the Vårdal Foundation, the Joint Committee of the Northern Health Region of Sweden (Visare Norr), the JC Kempe Memorial Foundation, the Foundation of the Medical Faculty, the Borgerskapet of Umeå Research Foundation, the Arneska Foundation, University of Umeå and the County Council of Västerbotten (“Dagmar”, “FoU” , and “Äldre Centrum Västerbotten”), the Swedish Research Council, Grant 2005/D1255-V and grant K 2014-99X-22610-01-6, the Strategic Research Programme in Care Sciences, Sweden, the Umeå University Foundations for Medical Research; Umeå University (ALF) and The Swedish Dementia Foundation.
81
REFERENCES
1. United Nations, Department of Economic and Social Affairs, Population Division (2017). World Population Prospects: The 2017 Revision, Key Findings and Advance Tables. Working Paper No. ESA/P/WP/248. 2017.
2. Statistics Sweden. The future population of Sweden 2015–2060
(publication in Swedish). Örebro, Sweden. 2015. 3. Rosengren BE, Karlsson MK. The annual number of hip fractures in
Sweden will double from year 2002 to 2050: projections based on local and nationwide data. Acta Orthop 2014;85(3):234-237.
4. Cooper C, Cole ZA, Holroyd CR, et al. Secular trends in the incidence of
hip and other osteoporotic fractures. Osteoporos Int 2011;22(5):1277-1288.
5. Rosengren BE, Ahlborg HG, Mellström D, Nilsson JA, Björk J, Karlsson
MK. Secular trends in Swedish hip fractures 1987-2002: birth cohort and period effects. Epidemiology 2012;23(4):623-630.
6. Kannus P, Niemi S, Parkkari J, Palvanen M, Vuori I, Jarvinen M.
Nationwide decline in incidence of hip fracture. J Bone Miner Res 2006;21(12):1836-1838.
7. Sogaard AJ, Holvik K, Meyer HE, et al. Continued decline in hip fracture
incidence in Norway: a NOREPOS study. Osteoporos Int 2016;27(7):2217-2222.
8. Melton LJ, 3rd. Hip fractures: a worldwide problem today and tomorrow.
Bone 1993;14 Suppl 1:S1-8. 9. Bergström U, Jonsson H, Gustafson Y, Pettersson U, Stenlund H,
Svensson O. The hip fracture incidence curve is shifting to the right. Acta Orthop 2009;80(5):520-524.
10. Thorngren KG, Hommel A, Norrman PO, Thorngren J, Wingstrand H.
Epidemiology of femoral neck fractures. Injury 2002;33 Suppl 3:C1-7. 11. RIKSHÖFT. Årsrapport. 2015. Avaliable from www.rikshoft.se. 12. Seitz DP, Adunuri N, Gill SS, Rochon PA. Prevalence of dementia and
cognitive impairment among older adults with hip fractures. J Am Med Dir Assoc 2011;12(8):556-564.
82
13. Kanis JA, Oden A, Johnell O, et al. The use of clinical risk factors enhances the performance of BMD in the prediction of hip and osteoporotic fractures in men and women. Osteoporos Int 2007;18(8):1033-1046.
14. Ramnemark A, Nyberg L, Borssen B, Olsson T, Gustafson Y. Fractures
after stroke. Osteoporos Int 1998;8(1):92-95. 15. Melton LJ, 3rd, Beard CM, Kokmen E, Atkinson EJ, O'Fallon WM.
Fracture risk in patients with Alzheimer's disease. J Am Geriatr Soc 1994;42(6):614-619.
16. Ramnemark A, Nyberg L, Lorentzon R, Englund U, Gustafson Y.
Progressive hemiosteoporosis on the paretic side and increased bone mineral density in the nonparetic arm the first year after severe stroke. Osteoporos Int 1999;9(3):269-275.
17. Wiklund R, Toots A, Conradsson M, et al. Risk factors for hip fracture in
very old people: a population-based study. Osteoporos Int 2016;27(3):923-931.
18. Johnell O, Kanis JA. An estimate of the worldwide prevalence, mortality
and disability associated with hip fracture. Osteoporos Int 2004;15(11):897-902.
19. Ziden L, Wenestam CG, Hansson-Scherman M. A life-breaking event:
early experiences of the consequences of a hip fracture for elderly people. Clin Rehabil 2008;22(9):801-811.
20. Ziden L, Scherman MH, Wenestam CG. The break remains - elderly
people's experiences of a hip fracture 1 year after discharge. Disabil Rehabil 2010;32(2):103-113.
21. Visschedijk J, van Balen R, Hertogh C, Achterberg W. Fear of falling in
patients with hip fractures: prevalence and related psychological factors. J Am Med Dir Assoc 2013;14(3):218-220.
22. Vochteloo AJ, Moerman S, Tuinebreijer WE, et al. More than half of hip
fracture patients do not regain mobility in the first postoperative year. Geriatr Gerontol Int 2013;13(2):334-341.
23. Bertram M, Norman R, Kemp L, Vos T. Review of the long-term disability
associated with hip fractures. Inj Prev 2011;17(6):365-370. 24. Dyer SM, Crotty M, Fairhall N, et al. A critical review of the long-term
disability outcomes following hip fracture. BMC Geriatr 2016;16:158.
83
25. Nikitovic M, Wodchis WP, Krahn MD, Cadarette SM. Direct health-care costs attributed to hip fractures among seniors: a matched cohort study. Osteoporos Int 2013;24(2):659-669.
26. Parker M, Johansen A. Hip fracture. Bmj 2006;333(7557):27-30. 27. Haentjens P, Magaziner J, Colon-Emeric CS, et al. Meta-analysis: excess
mortality after hip fracture among older women and men. Ann Intern Med 2010;152(6):380-390.
28. Abrahamsen B, van Staa T, Ariely R, Olson M, Cooper C. Excess mortality
following hip fracture: a systematic epidemiological review. Osteoporos Int 2009;20(10):1633-1650.
29. Cooper C. The crippling consequences of fractures and their impact on
quality of life. Am J Med 1997;103(2A):12S-17S; discussion 17S-19S. 30. Haleem S, Lutchman L, Mayahi R, Grice JE, Parker MJ. Mortality
following hip fracture: trends and geographical variations over the last 40 years. Injury 2008;39(10):1157-1163.
31. Borgström F, Sobocki P, Ström O, Jönsson B. The societal burden of
osteoporosis in Sweden. Bone 2007;40(6):1602-1609. 32. David JD. Essential orthopaedics and trauma. Edinburgh London
Melbourne and New York. Churchill Livingstone Inc. 1989. 33. Garden RS. Stability and Union in Subcapital Fractures of the Femur. J
Bone Joint Surg Br 1964;46(4):630-647. 34. Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly
persons living in the community. N Engl J Med 1988;319(26):1701-1707. 35. Campbell AJ, Borrie MJ, Spears GF, Jackson SL, Brown JS, Fitzgerald
JL. Circumstances and consequences of falls experienced by a community population 70 years and over during a prospective study. Age Ageing 1990;19(2):136-141.
36. Jensen J, Lundin-Olsson L, Nyberg L, Gustafson Y. Falls among frail
older people in residential care. Scand J Public Health 2002;30(1):54-61. 37. Nyberg L, Gustafson Y, Berggren D, Brännström B, Bucht G. Falls leading
to femoral neck fractures in lucid older people. J Am Geriatr Soc 1996;44(2):156-160.
38. Fuller GF. Falls in the elderly. Am Fam Physician 2000;61(7):2159-2168,
2173-2154.
84
39. Pils K, Neumann F, Meisner W, Schano W, Vavrovsky G, Van der Cammen TJ. Predictors of falls in elderly people during rehabilitation after hip fracture--who is at risk of a second one? Z Gerontol Geriatr 2003;36(1):16-22.
40. Stenvall M, Olofsson B, Lundström M, Svensson O, Nyberg L, Gustafson
Y. Inpatient falls and injuries in older patients treated for femoral neck fracture. Arch Gerontol Geriatr 2006;43(3):389-399.
41. Colon-Emeric C, Kuchibhatla M, Pieper C, et al. The contribution of hip
fracture to risk of subsequent fractures: data from two longitudinal studies. Osteoporos Int 2003;14(11):879-883.
42. Stenvall M, Elinge E, von Heideken Wagert P, Lundström M, Gustafson
Y, Nyberg L. Having had a hip fracture--association with dependency among the oldest old. Age Ageing 2005;34(3):294-297.
43. Foss NB, Palm H, Kehlet H. In-hospital hip fractures: prevalence, risk
factors and outcome. Age Ageing 2005;34(6):642-645. 44. MSB. Fallolyckor. 2014. Avaliable from www.msb.se. 45. Campbell AJ, Robertson MC. Implementation of multifactorial
interventions for fall and fracture prevention. Age Ageing 2006;35 Suppl 2:ii60-ii64.
46. Lach HW, Reed AT, Arfken CL, et al. Falls in the elderly: reliability of a
classification system. J Am Geriatr Soc 1991;39(2):197-202. 47. Todd C, Skelton D. (2004) What are the main risk factors for falls
among older people and what are the most effective interventions to prevent these falls? Copenhagen, WHO Regional Office for Europe (Health Evidence Network report; http://www.euro.who.int/document/E82552.pdf, accessed 5 April 2004).
48. Tinetti ME, Doucette JT, Claus EB. The contribution of predisposing and
situational risk factors to serious fall injuries. J Am Geriatr Soc 1995;43(11):1207-1213.
49. Eriksson S, Strandberg S, Gustafson Y, Lundin-Olsson L. Circumstances
surrounding falls in patients with dementia in a psychogeriatric ward. Arch Gerontol Geriatr 2009;49(1):80-87.
50. Kallin K, Lundin-Olsson L, Jensen J, Nyberg L, Gustafson Y.
Predisposing and precipitating factors for falls among older people in residential care. Public Health 2002;116(5):263-271.
85
51. Deandrea S, Lucenteforte E, Bravi F, Foschi R, La Vecchia C, Negri E. Risk factors for falls in community-dwelling older people: a systematic review and meta-analysis. Epidemiology 2010;21(5):658-668.
52. Kallin K, Gustafson Y, Sandman PO, Karlsson S. Drugs and falls in older
people in geriatric care settings. Aging Clin Exp Res 2004;16(4):270-276. 53. Tängman S, Eriksson S, Gustafson Y, Lundin-Olsson L. Precipitating
factors for falls among patients with dementia on a psychogeriatric ward. Int Psychogeriatr 2010;22(4):641-649.
54. Nyberg L, Gustafson Y, Janson A, Sandman PO, Eriksson S. Incidence of
falls in three different types of geriatric care. A Swedish prospective study. Scand J Soc Med 1997;25(1):8-13.
55. Cameron ID, Gillespie LD, Robertson MC, et al. Interventions for
preventing falls in older people in care facilities and hospitals. Cochrane Database Syst Rev 2012;12:CD005465.
56. Gillespie LD, Robertson MC, Gillespie WJ, et al. Interventions for
preventing falls in older people living in the community. Cochrane Database Syst Rev 2012(9):CD007146.
57. ANZSGM, Australian and New Zealand Society för Geriatric Medicine.
Orthogeriatric Care Revised 2010. Avaliable from www.anzsgm.org/posstate.asp.
58. NICE, National Institute for Health and Care Excellence. Falls in older
people. Updated 2017. Avaliable from www.nice.org.uk/guidance/qs86. 59. Summary of the Updated American Geriatrics Society/British Geriatrics
Society clinical practice guideline for prevention of falls in older persons. J Am Geriatr Soc 2011;59(1):148-157.
60. Jensen J, Lundin-Olsson L, Nyberg L, Gustafson Y. Fall and injury
prevention in older people living in residential care facilities. A cluster randomized trial. Ann Intern Med 2002;136(10):733-741.
61. Parker MJ, Gillespie WJ, Gillespie LD. Hip protectors for preventing hip
fractures in older people. Cochrane Database Syst Rev 2005(3):CD001255.
62. Santesso N, Carrasco-Labra A, Brignardello-Petersen R. Hip protectors
for preventing hip fractures in older people. Cochrane Database Syst Rev 2014(3):CD001255.
63. de Vries EN, Ramrattan MA, Smorenburg SM, Gouma DJ, Boermeester
MA. The incidence and nature of in-hospital adverse events: a systematic review. Qual Saf Health Care 2008;17(3):216-223.
86
64. Belmont PJ, Jr., Garcia EJ, Romano D, Bader JO, Nelson KJ, Schoenfeld
AJ. Risk factors for complications and in-hospital mortality following hip fractures: a study using the National Trauma Data Bank. Arch Orthop Trauma Surg 2014;134(5):597-604.
65. Roche JJ, Wenn RT, Sahota O, Moran CG. Effect of comorbidities and
postoperative complications on mortality after hip fracture in elderly people: prospective observational cohort study. Bmj 2005;331(7529):1374.
66. Merchant RA, Lui KL, Ismail NH, Wong HP, Sitoh YY. The relationship
between postoperative complications and outcomes after hip fracture surgery. Ann Acad Med Singapore 2005;34(2):163-168.
67. Sathiyakumar V, Greenberg SE, Molina CS, Thakore RV, Obremskey WT,
Sethi MK. Hip fractures are risky business: an analysis of the NSQIP data. Injury 2015;46(4):703-708.
68. Goldhahn S, Sawaguchi T, Audige L, et al. Complication reporting in
orthopaedic trials. A systematic review of randomized controlled trials. J Bone Joint Surg Am 2009;91(8):1847-1853.
69. Kammerlander C, Roth T, Friedman SM, et al. Ortho-geriatric service--a
literature review comparing different models. Osteoporos Int 2010;21(Suppl 4):S637-646.
70. Hansson S, Rolfson O, Åkesson K, Nemes S, Leonardsson O, Rogmark C.
Complications and patient-reported outcome after hip fracture. A consecutive annual cohort study of 664 patients. Injury 2015;46(11):2206-2211.
71. Molina CS, Thakore RV, Blumer A, Obremskey WT, Sethi MK. Use of the
National Surgical Quality Improvement Program in orthopaedic surgery. Clin Orthop Relat Res 2015;473(5):1574-1581.
72. Lawrence VA, Hilsenbeck SG, Noveck H, Poses RM, Carson JL. Medical
complications and outcomes after hip fracture repair. Arch Intern Med 2002;162(18):2053-2057.
73. Olofsson B, Stenvall M, Lundström M, Svensson O, Gustafson Y.
Malnutrition in hip fracture patients: an intervention study. J Clin Nurs 2007;16(11):2027-2038.
74. Jameson SS, Khan SK, Baker P, et al. A national analysis of complications
following hemiarthroplasty for hip fracture in older patients. QJM 2012;105(5):455-460.
87
75. Radcliff TA, Henderson WG, Stoner TJ, Khuri SF, Dohm M, Hutt E. Patient risk factors, operative care, and outcomes among older community-dwelling male veterans with hip fracture. J Bone Joint Surg Am 2008;90(1):34-42.
76. Donegan DJ, Gay AN, Baldwin K, Morales EE, Esterhai JL, Jr., Mehta S.
Use of medical comorbidities to predict complications after hip fracture surgery in the elderly. J Bone Joint Surg Am 2010;92(4):807-813.
77. Sathiyakumar V, Avilucea FR, Whiting PS, et al. Risk factors for adverse
cardiac events in hip fracture patients: an analysis of NSQIP data. Int Orthop 2016;40(3):439-445.
78. Kannegaard PN, van der Mark S, Eiken P, Abrahamsen B. Excess
mortality in men compared with women following a hip fracture. National analysis of comedications, comorbidity and survival. Age Ageing 2010;39(2):203-209.
79. Petersen MB, Jorgensen HL, Hansen K, Duus BR. Factors affecting
postoperative mortality of patients with displaced femoral neck fracture. Injury 2006;37(8):705-711.
80. Hershkovitz A, Polatov I, Beloosesky Y, Brill S. Factors affecting mortality
of frail hip-fractured elderly patients. Arch Gerontol Geriatr 2010;51(2):113-116.
81. Söderqvist A, Ekström W, Ponzer S, et al. Prediction of mortality in
elderly patients with hip fractures: a two-year prospective study of 1,944 patients. Gerontology 2009;55(5):496-504.
82. Meyer HE, Tverdal A, Falch JA, Pedersen JI. Factors associated with
mortality after hip fracture. Osteoporos Int 2000;11(3):228-232. 83. Juliebo V, Krogseth M, Skovlund E, Engedal K, Wyller TB. Medical
treatment predicts mortality after hip fracture. J Gerontol A Biol Sci Med Sci 2010;65(4):442-449.
84. Castronuovo E, Pezzotti P, Franzo A, Di Lallo D, Guasticchi G. Early and
late mortality in elderly patients after hip fracture: a cohort study using administrative health databases in the Lazio region, Italy. BMC geriatr 2011;11:37.
85. Vestergaard P, Rejnmark L, Mosekilde L. Increased mortality in patients
with a hip fracture-effect of pre-morbid conditions and post-fracture complications. Osteoporos Int 2007;18(12):1583-1593.
86. Berry SD, Samelson EJ, Bordes M, Broe K, Kiel DP. Survival of aged
nursing home residents with hip fracture. J Gerontol A Biol Sci Med Sci 2009;64(7):771-777.
88
87. Cameron ID, Chen JS, March LM, et al. Hip fracture causes excess
mortality owing to cardiovascular and infectious disease in institutionalized older people: a prospective 5-year study. J Bone Miner Res 2010;25(4):866-872.
88. Chatterton BD, Moores TS, Ahmad S, Cattell A, Roberts PJ. Cause of
death and factors associated with early in-hospital mortality after hip fracture. Bone Joint J 2015;97-B(2):246-251.
89. Perez JV, Warwick DJ, Case CP, Bannister GC. Death after proximal
femoral fracture--an autopsy study. Injury 1995;26(4):237-240. 90. Panula J, Pihlajamaki H, Mattila VM, et al. Mortality and cause of death
in hip fracture patients aged 65 or older: a population-based study. BMC Musculoskelet Disord 2011;12:105.
91. Devas M. Geriatric orthopaedics. Ann R Coll Surg Engl 1976;58(1):16-21. 92. Galvard H, Samuelsson SM. Orthopedic or geriatric rehabilitation of hip
fracture patients: a prospective, randomized, clinically controlled study in Malmö, Sweden. Aging 1995;7(1):11-16.
93. Gustafson Y, Brännström B, Berggren D, et al. A geriatric-anesthesiologic
program to reduce acute confusional states in elderly patients treated for femoral neck fractures. J Am Geriatr Soc 1991;39(7):655-662.
94. Lundström M, Edlund A, Lundström G, Gustafson Y. Reorganization of
nursing and medical care to reduce the incidence of postoperative delirium and improve rehabilitation outcome in elderly patients treated for femoral neck fractures. Scand J Caring Sci 1999;13(3):193-200.
95. Mak JC, Cameron ID, March LM. Evidence-based guidelines for the
management of hip fractures in older persons: an update. Med J Australia 2010;192(1):37-41.
96. Handoll HH, Cameron ID, Mak JC, Finnegan TP. Multidisciplinary
rehabilitation for older people with hip fractures. Cochrane Database Syst Rev 2009(4):CD007125.
97. SBU. Rehabilitation of older people with hip fracture-interdisciplinary teams. A systematic review. Stockholm: Swedish Council on Health Technology Assessment (SBU): 2015. Report number 235.
98. NICE, National Institute for Health and Care Excellence. Hip fracture:
management. Clinical guideline 2011. Updated May 2017. Avaliable from www.nice.org.uk/guidance/cg124.
89
99. Swift C, Ftouh S, Langford P, Chesser TS, Johanssen A. Interdisciplinary management of hip fracture. Clin Med 2016;16(6):541-544.
100. American Academy of Orthopaedic Surgeons. Management of Hip
Fractures in the Elderly. 2014. Avaliable from www.aaos.org. 101. British Geriatrics Society. Orthogeriatric models of care. 2007. Avaliable
from www.bgs.org.uk. 102. Sletvold O, Tilvis R, Jonsson A, et al. Geriatric work-up in the Nordic
countries. The Nordic approach to comprehensive geriatric assessment. Dan Med Bull 1996;43(4):350-359.
103. Momsen AM, Rasmussen JO, Nielsen CV, Iversen MD, Lund H.
Multidisciplinary team care in rehabilitation: an overview of reviews. J Rehabil Med 2012;44(11):901-912.
104. Wells JL, Seabrook JA, Stolee P, Borrie MJ, Knoefel F. State of the art in
geriatric rehabilitation. Part II: clinical challenges. Arch Phys Med Rehabil 2003;84(6):898-903.
105. Rubenstein LZ, Stuck AE, Siu AL, Wieland D. Impacts of geriatric
evaluation and management programs on defined outcomes: overview of the evidence. J Am Geriatr Soc 1991;39(9 Pt 2):8S-16S; discussion 17S-18S.
106. Ellis G, Whitehead MA, Robinson D, O'Neill D, Langhorne P.
Comprehensive geriatric assessment for older adults admitted to hospital: meta-analysis of randomised controlled trials. Bmj 2011;343:d6553.
107. Pilotto A, Cella A, Pilotto A, et al. Three Decades of Comprehensive
Geriatric Assessment: Evidence Coming From Different Healthcare Settings and Specific Clinical Conditions. J Am Med Dir Assoc 2017;18(2):192 e191-192 e111.
108. Eamer G, Saravana-Bawan B, van der Westhuizen B, Chambers T,
Ohinmaa A, Khadaroo RG. Economic evaluations of comprehensive geriatric assessment in surgical patients: a systematic review. J Surg Res 2017;218:9-17.
109. Sabharwal S, Wilson H. Orthogeriatrics in the management of frail older
patients with a fragility fracture. Osteoporos Int 2015;26(10):2387-2399. 110. Giusti A, Barone A, Razzano M, Pizzonia M, Pioli G. Optimal setting and
care organization in the management of older adults with hip fracture. Eur J Phys Rehabil Med 2011;47(2):281-296.
90
111. Stenvall M, Olofsson B, Lundström M, et al. A multidisciplinary, multifactorial intervention program reduces postoperative falls and injuries after femoral neck fracture. Osteoporos Int 2007;18(2):167-175.
112. Boddaert J, Cohen-Bittan J, Khiami F, et al. Postoperative admission to
a dedicated geriatric unit decreases mortality in elderly patients with hip fracture. PloS one 2014;9(1):e83795.
113. Watne LO, Torbergsen AC, Conroy S, et al. The effect of a pre- and
postoperative orthogeriatric service on cognitive function in patients with hip fracture: randomized controlled trial (Oslo Orthogeriatric Trial). BMC Med 2014;12:63.
114. Shyu YI, Liang J, Wu CC, et al. Two-year effects of interdisciplinary
intervention for hip fracture in older Taiwanese. J Am Geriatr Soc 2010;58(6):1081-1089.
115. Wagner P, Fuentes P, Diaz A, et al. Comparison of complications and
length of hospital stay between orthopedic and orthogeriatric treatment in elderly patients with a hip fracture. Geriatr Orthop Surg Rehabil 2012;3(2):55-58.
116. Friedman SM, Mendelson DA, Bingham KW, Kates SL. Impact of a
comanaged Geriatric Fracture Center on short-term hip fracture outcomes. Arch Intern Med 2009;169(18):1712-1717.
117. Khasraghi FA, Christmas C, Lee EJ, Mears SC, Wenz JF, Sr. Effectiveness
of a multidisciplinary team approach to hip fracture management. J Surg Orthop Adv 2005;14(1):27-31.
118. Vidan M, Serra JA, Moreno C, Riquelme G, Ortiz J. Efficacy of a
comprehensive geriatric intervention in older patients hospitalized for hip fracture: a randomized, controlled trial. J Am Geriatr Soc 2005;53(9):1476-1482.
119. Suhm N, Kaelin R, Studer P, et al. Orthogeriatric care pathway: a
prospective survey of impact on length of stay, mortality and institutionalisation. Arch Orthop Trauma Surg 2014;134(9):1261-1269.
120. Folbert EC, Hegeman JH, Vermeer M, et al. Improved 1-year mortality in
elderly patients with a hip fracture following integrated orthogeriatric treatment. Osteoporos Int 2017;28(1):269-277.
121. Prestmo A, Hagen G, Sletvold O, et al. Comprehensive geriatric care for
patients with hip fractures: a prospective, randomised, controlled trial. Lancet 2015;385(9978):1623-1633.
91
122. Biber R, Singler K, Curschmann-Horter M, Wicklein S, Sieber C, Bail HJ. Implementation of a co-managed Geriatric Fracture Center reduces hospital stay and time-to-operation in elderly femoral neck fracture patients. Arch Orthop Trauma Surg 2013;133(11):1527-1531.
123. Flikweert ER, Izaks GJ, Knobben BA, Stevens M, Wendt K. The
development of a comprehensive multidisciplinary care pathway for patients with a hip fracture: design and results of a clinical trial. BMC Musculoskelet Disord 2014;15:188.
124. Mazzola P, De Filippi F, Castoldi G, Galetti P, Zatti G, Annoni G. A
comparison between two co-managed geriatric programmes for hip fractured elderly patients. Aging Clin Exp Res 2011;23(5-6):431-436.
125. Lamb LC, Montgomery SC, Wong Won B, Harder S, Meter J, Feeney JM.
A multidisciplinary approach to improve the quality of care for patients with fragility fractures. J Orthop 2017;14(2):247-251.
126. Lundström M, Olofsson B, Stenvall M, et al. Postoperative delirium in old
patients with femoral neck fracture: a randomized intervention study. Aging Clin Exp Res 2007;19(3):178-186.
127. Della Rocca GJ, Moylan KC, Crist BD, Volgas DA, Stannard JP, Mehr DR.
Comanagement of geriatric patients with hip fractures: a retrospective, controlled, cohort study. Geriatr Orthop Surg Rehabil 2013;4(1):10-15.
128. Gregersen M, Morch MM, Hougaard K, Damsgaard EM. Geriatric
intervention in elderly patients with hip fracture in an orthopedic ward. J Inj Violence Res 2012;4(2):45-51.
129. Shyu YI, Liang J, Wu CC, et al. Interdisciplinary intervention for hip
fracture in older Taiwanese: benefits last for 1 year. J Gerontol A Biol Sci Med Sci 2008;63(1):92-97.
130. Barone A, Giusti A, Pizzonia M, Razzano M, Palummeri E, Pioli G. A
comprehensive geriatric intervention reduces short- and long-term mortality in older people with hip fracture. J Am Geriatr Soc 2006;54(7):1145-1147.
131. Cogan L, Martin AJ, Kelly LA, Duggan J, Hynes D, Power D. An audit of
hip fracture services in the Mater Hospital Dublin 2001 compared with 2006. Ir J Med Sci 2010;179(1):51-55.
132. Miura LN, DiPiero AR, Homer LD. Effects of a geriatrician-led hip
fracture program: improvements in clinical and economic outcomes. J Am Geriatr Soc 2009;57(1):159-167.
92
133. Gonzalez-Montalvo JI, Alarcon T, Mauleon JL, Gil-Garay E, Gotor P, Martin-Vega A. The orthogeriatric unit for acute patients: a new model of care that improves efficiency in the management of patients with hip fracture. Hip Int 2010;20(2):229-235.
134. Zeltzer J, Mitchell RJ, Toson B, Harris IA, Ahmad L, Close J.
Orthogeriatric services associated with lower 30-day mortality for older patients who undergo surgery for hip fracture. Med J Aust 2014;201(7):409-411.
135. Adunsky A, Lerner-Geva L, Blumstein T, Boyko V, Mizrahi E, Arad M.
Improved survival of hip fracture patients treated within a comprehensive geriatric hip fracture unit, compared with standard of care treatment. J Am Med Dir Assoc 2011;12(6):439-444.
136. Henderson CY, Shanahan E, Butler A, et al. Dedicated orthogeriatric
service reduces hip fracture mortality. Ir J Med Sci 2017;186(1):179-184. 137. Stenvall M, Olofsson B, Nyberg L, Lundström M, Gustafson Y. Improved
performance in activities of daily living and mobility after a multidisciplinary postoperative rehabilitation in older people with femoral neck fracture: a randomized controlled trial with 1-year follow-up. J Rehabil Med 2007;39(3):232-238.
138. Adunsky A, Lusky A, Arad M, Heruti RJ. A comparative study of
rehabilitation outcomes of elderly hip fracture patients: the advantage of a comprehensive orthogeriatric approach. J Gerontol A Biol Sci Med Sci 2003;58(6):542-547.
139. Gupta A. The effectiveness of geriatrician-led comprehensive hip fracture
collaborative care in a new acute hip unit based in a general hospital setting in the UK. J R Coll Physicians Edinb 2014;44(1):20-26.
140. Thingstad P, Taraldsen K, Saltvedt I, et al. The long-term effect of
comprehensive geriatric care on gait after hip fracture: the Trondheim Hip Fracture Trial--a randomised controlled trial. Osteoporos Int 2016;27(3):933-942.
141. Shyu YI, Liang J, Wu CC, et al. A pilot investigation of the short-term
effects of an interdisciplinary intervention program on elderly patients with hip fracture in Taiwan. J Am Geriatr Soc 2005;53(5):811-818.
142. Ward D, Drahota A, Gal D, Severs M, Dean TP. Care home versus hospital
and own home environments for rehabilitation of older people. Cochrane Database Syst Rev 2008(4):CD003164.
143. Donohue K, Hoevenaars R, McEachern J, Zeman E, Mehta S. Home-
Based Multidisciplinary Rehabilitation following Hip Fracture Surgery: What Is the Evidence? Rehabil Res Pract 2013;2013:875968.
93
144. Ziden L, Frändin K, Kreuter M. Home rehabilitation after hip fracture. A
randomized controlled study on balance confidence, physical function and everyday activities. Clin Rehabil 2008;22(12):1019-1033.
145. Ziden L, Kreuter M, Frändin K. Long-term effects of home rehabilitation
after hip fracture - 1-year follow-up of functioning, balance confidence, and health-related quality of life in elderly people. Disabil Rehabil 2010;32(1):18-32.
146. Crotty M, Whitehead C, Miller M, Gray S. Patient and caregiver outcomes
12 months after home-based therapy for hip fracture: a randomized controlled trial. Arch Phys Med Rehabil 2003;84(8):1237-1239.
147. Tinetti ME, Baker DI, Gottschalk M, et al. Home-based multicomponent
rehabilitation program for older persons after hip fracture: a randomized trial. Arch Phys Med Rehabil 1999;80(8):916-922.
148. Crotty M, Whitehead CH, Gray S, Finucane PM. Early discharge and
home rehabilitation after hip fracture achieves functional improvements: a randomized controlled trial. Clin Rehabil 2002;16(4):406-413.
149. Salpakoski A, Tormakangas T, Edgren J, et al. Effects of a
multicomponent home-based physical rehabilitation program on mobility recovery after hip fracture: a randomized controlled trial. J Am Med Dir Assoc 2014;15(5):361-368.
150. Edgren J, Salpakoski A, Sihvonen SE, et al. Effects of a home-based
physical rehabilitation program on physical disability after hip fracture: a randomized controlled trial. J Am Med Dir Assoc 2015;16(4):350 e351-357.
151. Kuisma R. A randomized, controlled comparison of home versus
institutional rehabilitation of patients with hip fracture. Clin Rehabil 2002;16(5):553-561.
152. Ryan T, Enderby P, Rigby AS. A randomized controlled trial to evaluate
intensity of community-based rehabilitation provision following stroke or hip fracture in old age. Clin Rehabil 2006;20(2):123-131.
153. SBU. Avancerad hemsjukvård och hemrehabilitering. Effekter och
kostnader. Statens beredning för medicinsk utvärdering. 1999.;rapport nr 146.
154. Tinetti ME, Baker DI, Gottschalk M, et al. Systematic home-based
physical and functional therapy for older persons after hip fracture. Arch Phys Med Rehabil 1997;78(11):1237-1247.
94
155. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. Washington, DC. 1994.
156. The National Board of Health and Welfare Sweden. Dödsorsaksstatistik.
Report number 2011-4-33. Avaliable at www.socialstyrelsen.se. 157. Sonn U. Longitudinal studies of dependence in daily life activities among
elderly persons. Scand J Rehabil Med Suppl 1996;34:1-35. 158. Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of
Illness in the Aged. The Index of Adl: A Standardized Measure of Biological and Psychosocial Function. Jama 1963;185:914-919.
159. Hasselgren-Nyberg L, Omgren M, Nyberg L et al. S-COVS. The Swedish
version of Physiotherapy Clinical Outcome Variables. Den svenska versionen av Physiotherapy Clinical Outcome Variables. Nordisk Fysioterapi 1997;1:1-15.
160. Owens WD, Felts JA, Spitznagel EL, Jr. ASA physical status
classifications: a study of consistency of ratings. Anesthesiology 1978;49(4):239-243.
161. Folstein MF, Folstein SE, McHugh PR. "Mini-mental state". A practical
method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12(3):189-198.
162. Tombaugh TN, McIntyre NJ. The mini-mental state examination: a
comprehensive review. J Am Geriatr Soc 1992;40(9):922-935. 163. Berggren D, Gustafson Y, Eriksson B, et al. Postoperative confusion after
anesthesia in elderly patients with femoral neck fractures. Anesth Analg 1987;66(6):497-504.
164. Sheikh JI. Geriatric Depression Scale (GDS): recent evidence and
development of a shorter version. Clin Gerontol 1986;5(1/2):165. 165. Yesavage JA, Brink TL, Rose TL, et al. Development and validation of a
geriatric depression screening scale: a preliminary report. J Psychiatr Res 1982;17(1):37-49.
166. Berg KO, Wood-Dauphinee SL, Williams JI, Maki B. Measuring balance
in the elderly: validation of an instrument. Can J Public Health 1992;83 Suppl 2:S7-11.
167. Conradsson M, Lundin-Olsson L, Lindelöf N, et al. Berg balance scale:
intrarater test-retest reliability among older people dependent in activities of daily living and living in residential care facilities. Phys Ther 2007;87(9):1155-1163.
95
168. Guralnik JM, Simonsick EM, Ferrucci L, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol 1994;49(2):M85-94.
169. Robertson MC, Campbell AJ, Herbison P. Statistical analysis of efficacy
in falls prevention trials. J Gerontol A Biol Sci Med Sci 2005;60(4):530-534.
170. Kleinbaum DG KM. Survival Analysis: A Self-Learning Text. 3rd ed.
2012.;New York, NY: Springer New York. 171. Mahoney FI, Barthel DW. Functional Evaluation: The Barthel Index. Md
State Med J 1965;14:61-65. 172. Wade DT. Measurement in neurological rehabilitation. Curr Opin Neurol
Neurosurg 1992;5(5):682-686. 173. Littbrand H, Rosendahl E, Lindelöf N, Lundin-Olsson L, Gustafson Y,
Nyberg L. A high-intensity functional weight-bearing exercise program for older people dependent in activities of daily living and living in residential care facilities: evaluation of the applicability with focus on cognitive function. Phys Ther 2006;86(4):489-498.
174. Littbrand H Lindelöf N, Rosendahl E. The HIFE Program: The High-
Intensity Functional Exercise Program. 2nd ed. Umeå: Umeå University, Department of Community Medicine and Rehabilitation, Geriatric Medicine; 2014. 2014.
175. Rubenstein LZ, Josephson KR, Robbins AS. Falls in the nursing home.
Ann Intern Med 1994;121(6):442-451. 176. Rapp K, Becker C, Cameron ID, Konig HH, Buchele G. Epidemiology of
falls in residential aged care: analysis of more than 70,000 falls from residents of Bavarian nursing homes. J Am Med Dir Assoc 2012;13(2):187 e181-186.
177. Kallin K, Jensen J, Olsson LL, Nyberg L, Gustafson Y. Why the elderly
fall in residential care facilities, and suggested remedies. J Fam Pract 2004;53(1):41-52.
178. Kallin K, Gustafson Y, Sandman PO, Karlsson S. Factors associated with
falls among older, cognitively impaired people in geriatric care settings: a population-based study. Am J Geriatr Psychiatry 2005;13(6):501-509.
179. Boonen S, Bischoff-Ferrari HA, Cooper C, et al. Addressing the
musculoskeletal components of fracture risk with calcium and vitamin D: a review of the evidence. Calcif Tissue Int 2006;78(5):257-270.
96
180. Avenell A, Mak JC, O'Connell D. Vitamin D and vitamin D analogues for preventing fractures in post-menopausal women and older men. Cochrane Database Syst Rev 2014(4):CD000227.
181. SBU. Vitamin D och kalcium för att förebygga frakturer. 2015. Avaliable
from www.sbu.se. 182. Läkartidningen. Läkartidningen. 2014;111:CW6C 183. Socialstyrelsen. Nationella riktlinjer – Utvärdering – Vård vid
rörelseorganens sjukdomar – Rekommendationer, bedömningar och sammanfattning 2014. Avaliable from www.socialstyrelsen.se.
184. Nordström P, Toots A, Gustafson Y, Thorngren KG, Hommel A,
Nordström A. Bisphosphonate Use After Hip Fracture in Older Adults: A Nationwide Retrospective Cohort Study. J Am Med Dir Assoc 2017;18(6):515-521.
185. Liem IS, Kammerlander C, Suhm N, et al. Identifying a standard set of
outcome parameters for the evaluation of orthogeriatric co-management for hip fractures. Injury 2013;44(11):1403-1412.
186. Flikweert ER, Wendt KW, Diercks RL, et al. Complications after hip
fracture surgery: are they preventable? Eur J Trauma Emerg Surg 2017. 187. Sathiyakumar V, Thakore RV, Greenberg SE, et al. Adverse Events in
Orthopaedics: Is Trauma More Risky? An Analysis of the NSQIP Data. J Orthop Trauma 2015;29(7):337-341.
188. Ali AM, Gibbons CE. Predictors of 30-day hospital readmission after hip
fracture: a systematic review. Injury 2017;48(2):243-252. 189. Boockvar KS, Halm EA, Litke A, et al. Hospital readmissions after
hospital discharge for hip fracture: surgical and nonsurgical causes and effect on outcomes. J Am Geriatr Soc 2003;51(3):399-403.
190. Asouhidou I, Asteri T, Sountoulides P, Natsis K, Georgiadis G. Early
postoperative mortality in the elderly: a pilot study. BMC Res Notes 2009;2:118.
191. Pugely AJ, Martin CT, Gao Y, Klocke NF, Callaghan JJ, Marsh JL. A risk
calculator for short-term morbidity and mortality after hip fracture surgery. J Orthop Trauma 2014;28(2):63-69.
192. Lithgow S, Jackson GA, Browne D. Estimating the prevalence of
dementia: cognitive screening in Glasgow nursing homes. Int J Geriatr Psychiatry 2012;27(8):785-791.
97
193. Stenvall M, Berggren M, Lundström M, Gustafson Y, Olofsson B. A multidisciplinary intervention program improved the outcome after hip fracture for people with dementia--subgroup analyses of a randomized controlled trial. Arch Gerontol Geriatr 2012;54(3):e284-289.
194. Sylliaas H, Brovold T, Wyller TB, Bergland A. Prolonged strength
training in older patients after hip fracture: a randomised controlled trial. Age Ageing 2012;41(2):206-212.
195. Liu CJ, Latham NK. Progressive resistance strength training for
improving physical function in older adults. Cochrane Database Syst Rev 2009(3):CD002759.
196. Grigoryan KV, Javedan H, Rudolph JL. Orthogeriatric care models and
outcomes in hip fracture patients: a systematic review and meta-analysis. J Orthop Trauma 2014;28(3):e49-55.
197. Nordström P, Michaelsson K, Hommel A, Norrman PO, Thorngren KG,
Nordström A. Geriatric Rehabilitation and Discharge Location After Hip Fracture in Relation to the Risks of Death and Readmission. J Am Med Dir Assoc 2016;17(1):91 e91-97.
198. Seitz DP, Gill SS, Gruneir A, et al. Effects of dementia on postoperative
outcomes of older adults with hip fractures: a population-based study. J Am Med Dir Assoc 2014;15(5):334-341.
199. Allen J, Koziak A, Buddingh S, Liang J, Buckingham J, Beaupre LA.
Rehabilitation in patients with dementia following hip fracture: a systematic review. Physiother Can 2012;64(2):190-201.
200. Resnick B, Beaupre L, McGilton KS, et al. Rehabilitation Interventions
for Older Individuals With Cognitive Impairment Post-Hip Fracture: A Systematic Review. J Am Med Dir Assoc 2016;17(3):200-205.
201. Seitz DP, Gill SS, Austin PC, et al. Rehabilitation of Older Adults with
Dementia After Hip Fracture. J Am Geriatr Soc 2016;64(1):47-54. 202. van Halteren-van Tilborg IA, Scherder EJ, Hulstijn W. Motor-skill
learning in Alzheimer's disease: a review with an eye to the clinical practice. Neuropsychol Rev 2007;17(3):203-212.
203. Lamb SE, Jorstad-Stein EC, Hauer K, Becker C, Prevention of Falls
Network E, Outcomes Consensus G. Development of a common outcome data set for fall injury prevention trials: the Prevention of Falls Network Europe consensus. J Am Geriatr Soc 2005;53(9):1618-1622.
204. Schwenk M, Schmidt M, Pfisterer M, Oster P, Hauer K. Rollator use
adversely impacts on assessment of gait and mobility during geriatric rehabilitation. J Rehabil Med 2011;43(5):424-429.
98
205. Littbrand H, Stenvall M, Rosendahl E. Applicability and effects of
physical exercise on physical and cognitive functions and activities of daily living among people with dementia: a systematic review. Am J Phys Med Rehabil 2011;90(6):495-518.
206. Allen CM, Becker PM, McVey LJ, Saltz C, Feussner JR, Cohen HJ. A
randomized, controlled clinical trial of a geriatric consultation team. Compliance with recommendations. Jama 1986;255(19):2617-2621.
99
PAPERS I-IV