Embed Size (px)
Citation preview
UMEÅ UNIVERSITY MEDICAL DISSERTATIONSNew Series No. 1040—ISSN 0346-6612—ISBN 91-7264-133-9
From the Department of Community Medicine and RehabilitationGeriatric Medicine and Physiotherapy, Umeå University, Sweden
Hip fractures among old people
Their prevalence, consequences and complications, and the
evaluation of a multi-factorial intervention program designed
to prevent falls and injuries and enhance performance of
activities of daily living
Michael Stenvall
Umeå 2006
Department of Community Medicine and RehabilitationGeriatric Medicine and Physiotherapy, Umeå University
SE-901 87 Umeå Sweden
Copyright © Michael StenvallNew Series No. 1040—ISSN 0346-6612—ISBN 91-7264-133-9Printed in Sweden by Larsson & Co:s Tryckeri AB, Umeå 2006
To my wonderful family,
Lisette, Linn, Nellie & Emil
CONTENTS
ABSTRACT 7
SVENSK SAMMANFATTNING (SUMMARY IN SWEDISH) 9
ABBREVIATIONS 11
LIST OF ORIGINAL PAPERS 12
INTRODUCTION 13
Hip fracture epidemiology and surgical treatment 13
Morbidities and risk factors 13Hip fracture aetiology; osteoporosis and falls 14Consequences of a hip fracture 15Fall definition 16
Fall prevention 16Geriatric rehabilitation 17
Hip fracture rehabilitation 18
Rationale for this thesis 25
AIMS OF THE THESIS 26
METHODS 27Settings and participants 27Ethical approval 29
Procedure 31Methods of data collection 33Intervention 38
Statistical analyses 41
RESULTS 43Paper I 43
Paper II 47Paper III 50Paper IV 52
DISCUSSION 55
Prevalence of hip fracture and its association with dependency among 56the very old
Falls, fall-related injuries and risk factors 56
Fall prevention 58Improved performance in activities of daily living 59Ethical considerations 60Methodological considerations 60
Clinical implications 62Implications for future research 63
GENERAL CONCLUSIONS 64
ACKNOWLEDGMENTS 65
REFERENCES 67
PAPERS I-IV
LIST OF DISSERTATIONS
7
ABSTRACT
The number of old people is growing and will increase future demands on healthcare services
for old people. Hip fracture is one of the diagnoses that increases with age and it has becomea major problem, both for those suffering a fracture and for society due to the large numbersinvolved, the morbidity with complications such as falls, functional decline, and the high
mortality rate among those affected. The main purposes of this thesis were, to study theimpact of previous hip fractures on their life among the very old, to study in-patient falls, fall-related injuries and fall-risk factors, and to evaluate a multidisciplinary, multi-factorialintervention program designed to reduce in-patient falls and to enhance functional
performance among old people who have sustained a femoral neck fracture.The impact of a hip fracture was examined in a cross-sectional population-based study,
among the very old (Umeå 85+). After adjustment for potential covariates, participants with a
history of hip fracture were found to be more dependent in the performance ofPersonal/Primary Activities of Daily Living (P-ADL) (p=0.024), walked less independently(p=0.040) and used a wheelchair more frequently (p=0.017). Most of the participants withearlier hip fractures who had moved to institutional care or begun using mobility aids, as
compared to before the fracture, had started to do so permanently in connection with thefracture incident.
In-patient falls, fall-related injuries and fall-risk factors were studied in 97 participants,
aged 70 or more, treated for a femoral neck fracture. There were 60 postoperative fallsoccurring among 26/97 participants (27%). Thirty-two percent of the falls resulted in injuries,25 % were minor, and 7 % were serious. Delirium after day seven, (Hazard Rate Ratio (HRR)with a 95% Confidence Interval (CI)), 4.62 (1.30-16.37), male sex 3.92 (1.58-9.73), and
sleeping disturbances 3.49 (1.24-9.86), were associated with in-patient falls. Forty-fivepercent of the participants were delirious on the day they fell.
The effects of a multidisciplinary, multi-factorial intervention program on in-hospital
falls and injuries as well as the short- and long-term effects on living conditions, walkingability and performance of activities of daily living were evaluated in a randomised controlledtrial among 199 participants with femoral neck fracture, aged �70 years. Participants were
randomised to care in a geriatric ward (intervention, n=102) or to conventional postoperativeroutines (control, n=97). The intervention consisted of staff education, individualized careplanning and rehabilitation, systematic assessment and treatment of fall-risk factors, activeprevention, and detection and treatment of postoperative complications and an intervention
follow up at four-months. The staff worked in teams to apply comprehensive geriatricassessment, management and rehabilitation.
Twelve participants fell a total 18 times in the intervention group compared to 26
participants suffering a total 60 falls in the control group. Only one participant with dementiafell in the intervention group compared to 11 participants with dementia in the control group.The fall incidence rate was 6.29/1000 days vs. 16.28/1000 for the intervention and control
8
groups respectively. The Incidence Rate Ratio (IRR) was 0.38 (95% CI: 0.20-0.76, p=0.006)
for the total sample and 0.07 (95% CI: 0.01-0.57, p=0.013) among participants with dementia.No new fractures were incurred in the intervention group but there were four in the controlgroup. In addition, despite shorter hospitalization, significantly more people from the
intervention group had regained independence in P-ADL performance at the four- and twelve-month follow ups, Odds Ratios (OR), with 95% CI, were 2.51 (1.00-6.30) and 3.49 (1.31-9.23) respectively. More participants in the intervention group had also regained the ability towalk independently without walking aids indoors, at the end of the study period, 3.01 (1.18-
7.61).In conclusion, hip fracture among the very old seems to be associated with poorer P-
ADL performance and poorer mobility. Falls and injuries are common during in-patient
rehabilitation after a femoral neck fracture, delirium and sleep disturbances and male genderare factors associated with in-patient falls. Having a team apply comprehensive geriatricassessments and rehabilitation, including the prevention, detection and treatment of fall-riskfactors, can successfully prevent in-patient falls and fall-related injuries, even among
participants with dementia, and can also enhance the performance of Activities of DailyLiving (ADL) and mobility after a hip fracture, in both short- and long-term perspectives.
Key words: accidental falls, activities of daily living, aged, geriatric medicine, hip fracture,
in-hospital, intervention, physiotherapy, randomised controlled trial, rehabilitation
9
SVENSK SAMMANFATTNING
Andelen äldre i befolkningen blir allt större. Med stigande ålder ökar risken för fallolyckor
och att drabbas av frakturer. Detta kommer att leda till en allt större belastning på vård ochomsorg i framtiden. Fall med efterföljande höftfraktur är ett stort problem för såväl dedrabbade som för samhället. I Sverige drabbas årligen närmare 19 000 personer av en
höftfraktur, de flesta på grund av en fallolycka. En stor andel av de drabbade återfår aldrigsina tidigare funktioner, som att klara sin personliga vård och att kunna gå självständigt.Många tvingas även byta boende efter en genomgången höftfrakturoperation.
Syftet med denna avhandling har varit att studera hur en höftfraktur påverkat de allra
äldsta senare i livet, beträffande boende och vardagsaktiviteter som personlig vård och gång.Dessutom var syftet att studera förekomsten av nya fall och fallskador, samt att utvärdera omett tvärvetenskapligt, multifaktoriellt interventionsprogram kunde minska fall och skador,
samt förbättra funktionell återhämtning efter en fraktur på lårbenshalsen.Konsekvenserna av att drabbas av en höftfraktur studerades i en populationsbaserad
tvärsnittsstudie bland de allra äldsta i Umeå kommun. Ungefär 1/4 av dessa äldre hadedrabbats av minst en höftfraktur. De som tidigare i livet hade haft en höftfraktur var mer
beroende av hjälp för att klara sin personliga vård, var mer beroende av hjälp vid gång, samtanvände rullstol i större utsträckning än de som inte haft någon höftfraktur. Många av demsom bytt boende eller börjat använda gånghjälpmedel efter frakturen hade gjort det i direkt
anslutning till skadan.Bland 97 stycken äldre (70 år eller mer) med brott på lårbenshalsen inträffade totalt 60
nya fallolyckor under sjukhusvistelsen efter höftoperationen. Trettitvå procent av fallenresulterade i någon form av skada, varav 7 procent i nya frakturer. Långvarig förvirring och
sömnstörningar var faktorer som var associerade med fallen, samt att det var vanligare attmännen föll. Av deltagarna som föll under vårdtiden var nästan hälften förvirrade någon gångunder den dag då olyckan skedde.
I en studie bland 199 äldre med brott på lårbenshalsen utvärderades ettinterventionsprogram avseende nya fall och fallrelaterade skador under sjukhusvistelsen, samtkort och långsiktiga effekter avseende boendesituation, aktiviteter i det dagliga livet ochgångförmåga. Deltagarna blev lottade till en geriatrisk avdelning (intervention = 102 stycken)
eller till traditionell vård och rehabilitering på ortopeden (kontroll = 97 stycken).Interventionsprogrammet innehöll personalutbildning, individuell vård och rehabilitering,systematisk bedömning och behandling av fallriskfaktorer, samt aktiva åtgärder för att
förebygga, upptäcka och behandla postoperativa komplikationer. Personalen arbetade ivårdlag och använde ett noggrant geriatriskt utrednings-, behandlings- ochrehabiliteringsprogram.
Tolv deltagare i interventionsgruppen föll totalt 18 gånger jämfört med 26 deltagare
som föll totalt 60 gånger i kontrollgruppen. Bland deltagare med en demenssjukdom var deten person i interventionsgruppen som föll vid ett tillfälle jämfört med 11 deltagare i
10
kontrollgruppen som föll totalt 34 gånger. Fallincidensen var 6,29/1 000 vårddagar jämfört
med 16,28/1 000 vårddagar för interventions- respektive kontrollgrupp. Inga nya fraktureruppstod i interventionsgruppen medan fyra nya fallrelaterade frakturer uppstod ikontrollgruppen. Trots en kortare vårdtid hade fler personer i interventionsgruppen återfått
förmågan att klara sina vardagsaktiviteter självständigt vid 4 och 12 månader. De hadedessutom i större utsträckning återfått sin förmåga att kunna gå inomhus utangånghjälpmedel.
Sammanfattningsvis kan sägas att en höftfraktur bland de allra äldsta förefaller vara
associerad med en nedsatt förmåga att klara sina vardagsaktiviteter som personlig vård ochgång. Fall och nya skador under vårdtiden efter brott på lårbenshalsen är vanligtförekommande. Förvirring, sömnstörningar samt manligt kön ökar risken för fall. Ett
vårdprogram med en noggrann geriatrisk utredning, behandling och rehabilitering med fokuspå att förebygga, upptäcka och behandla fallrisker kan effektivt förebygga nya fall ochfallrelaterade skador, även för patienter med demenssjukdom. Vårdprogrammet förbättrardessutom deltagarnas förmåga att klara sina vardagsaktiviteter, som personlig vård och gång,
på både kort och lång sikt.
11
ABBREVIATIONS
ADL Activities of Daily Living
P-ADL Personal/Primary Activities of Daily LivingI-ADL Instrumental Activities of Daily Living
AIS Abbreviated Injury Scale
BBS Berg Balance ScaleBMD Bone Mineral DensityBMI Body Mass IndexCAM Confusion Assessment Method
CI Confidence IntervalCGA Comprehensive Geriatric AssessmentDSM-IV Diagnostic and Statistical Manual of mental disorders, fourth edition
ESD Early Supported DischargeGDS Geriatric Depression ScaleGHFP Geriatric Hip Fracture ProgramGORU Geriatric Orthopaedic Rehabilitation Unit
HRR Hazard Rate RatioIQR InterQuartile RangeIRR Incidence Rate Ratio
LOS Length Of StayMADRS Montgomery-Åsberg Depression Rating ScaleMARU Mixed Assessment Rehabilitation UnitMMSE Mini Mental State Examination
MNA Mini Nutritional AssessmentOBS-scale Organic Brain Syndrome ScaleOR Odds Ratio
OT Occupational TherapistPT PhysiotherapistRCT Randomised Controlled Trial
SD Standard Deviation
12
ORIGINAL PAPERS
The thesis is based on the following papers, which will be referred to in the text by their
Roman numerals:
I. Stenvall M, Elinge E, von Heideken Wågert P, Lundström M, Gustafson Y, Nyberg L.
Having had a hip fracture – association with dependency among the oldest old. Age andAgeing 2005; 34: 294-297.
II. Stenvall M, Olofsson B, Lundström M, Svensson O, Nyberg L, Gustafson Y. Inpatient
falls and injuries in older patients treated for femoral neck fracture. In press forArchives of Gerontology and Geriatrics.
III. Stenvall M, Olofsson B, Lundström M, Englund U, Borssén B, Svensson O, Nyberg L,Gustafson Y. A multidisciplinary, multifactorial intervention program reducespostoperative falls and injuries after femoral neck fracture – a randomised controlledtrial, Submitted.
IV. Stenvall M, Olofsson B, Nyberg L, Lundström M, Gustafson Y. Improved performancein activities of daily living and mobility after a multidisciplinary postoperative
intervention program in older people with femoral neck fracture: a randomisedcontrolled trial with a one-year follow up, Submitted.
The articles have been reprinted with the kind permission of the respective publishers.
13
INTRODUCTION
Hip fracture epidemiology and surgical treatment
The number of older people in the population is growing. In Sweden more than 17% of the
people are aged 65 years or older and a large proportion of them are more than 80 years old.The prognosis is that this group will be even higher in the future (1). Advanced age isassociated with an increased prevalence of diseases and impairments (2). Hip fractures are
common and the incidence increases exponential with age (3). In Sweden today, with apopulation of approximately 9 million, about 19 000 individuals sustain a hip fracture everyyear (4) with much suffering for the individual and incurring a heavy cost for society during
the first year after the fracture (5).Globally the high and increasing incidence mean that hip fractures have become a major
health problem for society with a significant cause of morbidity and mortality (6, 7).Assuming no change in age- and sex- specific incidence, it is estimated that the number of hip
fractures will approximately double worldwide, from 1.26 million in 1990 to 2.6 million 2025(8). In Sweden, the overall incidence of hip fractures (not age standardised) among womenhas increased from 3.2/1000 women/year in 1960 to 9.1/1000 in 1990 with a corresponding
increase for men but starting from a much lower incidence (4). The lifetime risk of suffering ahip fracture among women is estimated to be 20% in Sweden (4).
Hip fractures, or proximal femoral fractures, are divided into two main types: fracturesof the femoral neck, called cervical fractures or intracapsular fractures, and fractures through
the muscle insertions below the femoral neck, called trochanteric fractures (per- inter- orsubtrochanteric fractures) or extracapsular fractures. Intermediate types are calledbasocervical fractures (9). Cervical fractures can be further subdivided into four types
according to Garden, (Garden I-IV) depending on the degree of dislocation (10). These foursubgroups are classified into two groups, in-complete or un-displaced, fractures (Garden I-II),and displaced fractures (Garden III-IV). The usual surgical treatments are primaryosteosynthesis (with screws or pins) or replacement with an artificial hip joint (11).
Basocervical and trochanteric fractures are usually treated with dynamic hip screws (9). InSweden the two main groups, cervical or trochanteric fractures, are almost equalproportionally with the former having a slight predominance (4).
Morbidities and risk factors
The mean age of hip fracture patients in Sweden is about 81 years and almost three quartersare women (4). Suffering a hip fracture indicates frailty and those affected have higher levels
of morbidities e.g. impaired cognition, Parkinson’s disease, stroke, heart failure, impairedvision (12-14), and lower Body Mass Index (BMI) (13, 14) compared with controls. Theyalso need more help with the Activities of Daily Living (ADL) and are less mobile in their
14
community before fracture (12-14). A large proportion live in institutions (14, 15), many are
depressed, and they use medications extensively (16, 17).There are several risk factors for sustaining a hip fracture apart from a tendency to fall.
Both low Bone Mineral Density (BMD) and advanced age are mentioned as risk factors (18-
20). These studies were performed on women but a research team in the Netherlands foundsimilar risks regarding age and BMD in men (21). Other risk factors mentioned are previousfractures, drug treatments, inactivity, physical impairments such as walking difficulties andADL limitations (12, 18-20, 22, 23), low weight (19, 20) and living alone (12, 19). Those
living in institutions have a much increased risk of sustaining a hip fracture (24, 25) and thosewith several risk factors run an especially high risk (23).
Several risk factors are the same for both types of hip fractures (cervical or trochanteric)
but it seems that those suffering from a trochanteric fracture have a more severe andgeneralized bone loss, and a poorer health status (19, 26). Women with a trochanteric fracturehave been reported to be older and shorter compared to those with a cervical fracture (27, 28).
Hip fracture aetiology; osteoporosis and falls
Hip fractures in old age are due to a combination of skeletal fragility (osteoporosis) and aforce exerted on the bone, usually due to a fall (3). Hip fractures comprise almost 20% of allosteoporotic fractures (29) and have become a barometer for osteoporosis since they are
strongly related to low BMD (30).Osteoporosis is clinically diagnosed by the presence of fragility fractures after a
relatively mild trauma, but it is defined as a skeletal disorder characterized by compromised
bone strength predisposing a person to an increased risk of fracture. The World HealthOrganization (WHO) defines osteoporosis as BMD, 2.5 Standard Deviations (SD) below themean for young white adult women (29).
Osteoporosis can be either primary or secondary. Primary osteoporosis can occur in
both sexes at all ages but occurs mostly after the menopause in women and later on in men.Secondary osteoporosis can be a result of medications, diseases or other factors (29). InSweden almost 38% of women aged 60 years or older has been estimated to have
osteoporosis using the WHO definition (31).The second cause of hip fractures is falls, since almost all hip fractures occur as a result
of a fall (32, 33). Most of the falls occur indoors during daily activities such as walking orrising from sitting to standing or vice versa (22), so-called low energy trauma. Falls are a
major problem among older people, and the risk increases with age (34) although they are nota part of normal aging. Among older people living in the community, almost one third falleach year (34, 35) and among those living in institutions the annual figure is about two thirds
(36-38). Approximately 2,5-10% of the falls lead to a fracture, and about 1-2% of the fallsresult in a hip fracture (35, 38-40). A recent study (41) found that 7% of all hip fracturesoccur during hospitalization.
The aetiology of falls in older people is complex and multi-factorial with several fall-
risk factors and usually combinations of predisposing and precipitating factors. Predisposing
15
factors can be morbidity, functional disability, previous falls, intake of medication (35, 36,
42-49) and aging (34, 50), and among the oldest old, male sex (40). The fall risk increaseswith the number of risk factors (35, 51). Acute diseases such as infections and delirium havebeen identified as important precipitating factors for falls (38, 42, 45-48). Delirium, which is
very common after hip fracture surgery, especially among demented patients or those incognitive decline (52-54), has been found to be one of the most important precipitating factorsfor falls among older people living in residential care facilities (48) and it has previously beenreported that delirium is associated with in-patient falls (42, 46, 47). In a recent Danish study
(41) more then half of those who suffered a hip fracture during hospitalization were eitherdemented or delirious before the fracture.
Consequences of a hip fracture
Hip fracture is one of the most serious consequences of a fall; the fracture is a common reasonfor being institutionalised (13, 15, 55) as is associated with having difficulties in performingADL (56-58). A subsequent decline in mobility after a hip fracture is also well documented in
the literature (14, 55, 57, 58), one study reported that only 54% of those who were communitymobile before the fracture were still mobile two years after the fracture (14). Another studyfound that new dependency at 12 months post-fracture, for those who did not requireassistance pre-fracture, varied between 20% for putting on pants, to 90% for climbing five
steps (58). Advanced age is described as a negative factor for rehabilitation outcome (59-61)as are cognitive impairment and male sex (59, 60, 62).
There is also high mortality rate after hip fracture surgery; varying in the ensuring 12
months between 12-31% in different studies (63-66). The survival rate 5 years after a hipfracture is about 80% of that expected among people of the same age without a fracture (67).Several studies report a higher mortality rate than controls (63, 66, 68, 69) which persists formany years (13, 70) and cannot be explained by pre-fracture health status (68). Men have an
even higher mortality rate after hip-fractures than women (13, 63, 69).As mentioned above almost all hip fractures occur as a result of a fall (32, 33), and
those suffering a hip fracture will be at increased risk of sustaining more falls and fractures
(57, 71). The risk of a subsequent fracture after an osteoporotic fracture is high soon after thefirst event (72). A first hip fracture is associated with a 2.5-fold increased risk of a subsequentfracture (73). The hip fracture patient is thus a person with high fall risk; a fact that cannotentirely be explained by pre-fracture risk factors (73). A study of relatively healthy
individuals with hip fractures living independently before the fracture found that 12% fellduring the in-patient rehabilitation (74). The highest fall risks were in the second week afteradmission to a sub-acute rehabilitation ward.
Among hip fracture patients, delirium is a common complication, and occurs in 28-60%of cases (52-54, 75, 76). Males in particular run an increased risk of developing delirium afterhip fracture surgery (16, 77, 78). Other common complications after a hip fracture surgery arewound infections, anaemia, urinary tract infections, urinary retention, decubital ulcers, and
depression (53, 79, 80).
16
Several strategies must be combined to prevent hip fractures; both falls and osteoporosis must
be prevented to reduce the incidence. In treating osteoporosis adequate calcium and vitamin Dintake is crucial. Regular exercise is necessary to develop optimal peak bone mass and topreserve bone mass throughout life (29). Later on other therapies may be necessary such as
bisphosphonates, hormones and/or selective oestrogen-receptor modulators (29).
Fall definition
The literature contains several definitions of falls. An international group working on
prevention of falls (Kellog International Group) defined a fall as an event which results in aperson coming to rest inadvertently on the ground or an other lower level, other than as aconsequence of the following: sustaining a violent blow; loss of consciousness; sudden onsetof paralysis; a stroke; or an epileptic seizure (81). The Prevention of Falls Network Europe
(ProFaNe) recommends that a fall should be defined as an unexpected event in which thesubject comes to rest on the ground, floor, or a lower level (82). Jensen et al defined a fall asan event in which the resident unintentionally came to rest on the ground or floor, regardless
of whether or not an injury was sustained. This definition includes falls resulting from acuteillness or epileptic seizure or if the person fell and was found on the floor by staff or someoneelse (83). In this thesis the last mentioned definition is used.
An injury due to a fall may be classified according to the Abbreviated Injury Scale,
1990 (AIS) (84), which categorizes injuries according to a hierarchical level of severity, withscores ranging from one (minor) to six (maximum). Code 1, denotes minor injury, such aswounds or superficial lacerations; code two, denotes moderate, intermediate injuries, such as
major lacerations, wrist fractures and joint dislocations; code three denotes a serious injurysuch as major fractures including hip fractures. Codes four and five denote severe and critical,injuries respectively, while code six is reserved for the most serious injuries (currentlyuntreatable). AIS 4-6 are life threatening at all ages. Fall injuries in older people are seldom
ranked higher than three. The Maximum AIS (MAIS), which is the highest AIS code for aperson with multiple injuries, has been used by investigators to describe overall severity (84).
Fall prevention
Multi-factorial intervention strategies among older people living in the community canprevent falls (85-88) and are recommended in fall-prevention interventions nowadays (51).The recommendations in fall-prevention programs is that they should include gait training,
advice on use of assistive devices, medication reviews, exercise programs including balancetraining, treatment of hypotension, environmental modification and treatment ofcardiovascular disorders. In long-term care program recommendations should also includestaff education (51). Most fall-prevention studies are performed in the community, but
multidisciplinary and multi-factorial interventions have also been shown to be beneficial inresidential care facilities (83). It has also been shown that multifaceted interventions can
17
prevent falls even in nursing home residents, including both cognitively impaired and lucid
residents but the effect in the sub-group with cognitive impairment was not analysedseparately (89). Among people with cognitive impairment and dementia, a multi-factorialprogram seems not to be effective (90, 91). There are also reports of effective single-
intervention programs, for example exercise programs, home safety assessment and amodification program, and vitamin D supplementation (85, 87, 92-94).
Few randomised controlled fall-prevention studies have been carried out in hospitals; afew studies involving single interventions among older people in rehabilitation units have not
produced any significant effects (95-97). Recently two studies, one using multipleinterventions (98), and one using a multidisciplinary fall-prevention approach (99), havedemonstrated reductions in falls. The multiple intervention study (98), showed a 30%
reduction in falls and a trend towards a reduction in the number of patients who fell and fall-related injuries on sub-acute rehabilitation wards, using fall-risk cards, education, exerciseand hip protectors. The difference was most obvious after 45 days of observation. The otherstudy (99) was able to show a reduction in the number of falls and in the patients who fell as
well as in injuries on a geriatric ward using a multidisciplinary fall-prevention program withmedication and environmental reviews, safe transfers and the use of wristband to identifythose with a high fall risk. The effect however, disappeared when the statistics were adjusted
for number of days on the ward. None of these fall-prevention studies has focused on hip-fracture patients or included an effort to reduce postoperative complications as a fall-prevention measure together with a multidisciplinary effort. No fall-prevention trial hasmentioned the prevention and treatment of delirium as an intervention. Healey et al (100)
showed in a cluster randomised trial a reduction of relative risk of falls with 30% using a fall-risk screen and related interventions in the form of a pre-printed care plan. They mentiondelirium as a factor with an increased fall risk but the care plan included only checks for
urinary tract infections. One non-randomised delirium intervention study found that reducingthe delirium incidence and duration, led to fewer serious injurious falls occurring in theintervention group (77). This may indicate that delirium is an important risk factor forpostoperative falls and injuries in hip-fracture patients and that prevention and treatment of
delirium should thus be included in an effective fall-prevention program.
Geriatric rehabilitation
With a growing aged population, geriatric medicine will become increasingly important.
Geriatrics involves knowledge of the interaction between diseases, psychosocial andenvironmental factors and aging and such knowledge is necessary when working with frailold people. In general, rehabilitation is understood as restoration to a previous state or to the
highest possible level of function. In addition, the goals in geriatric rehabilitation can also beseen as maintaining the current level of function and preventing further decline. To improvetheir health situation, Comprehensive Geriatric Assessment (CGA) can be used to determineelderly people’s medical, functional and psychosocial needs. CGA involves a
multidimensional team approach that determines an older person’s pathology, impairments,
18
functional limitations, and disability (101, 102). Checklists and assessments scales are very
useful to geriatric teams when performing assessments. Different scales covering differentareas for different team members are recommended for use in the clinic so that all availablemedical knowledge can be called in to afford the frail old person adequate and proper
treatment (101). Those checklists and scales should include daily activities, communicationproblems, visual and hearing disabilities, incontinence, falls, gait, balance and immobility,nutrition, polypharmacy, cognitive impairment, affective disorders, and sleep disturbancesetc. Several scales useful in different areas have been specially developed for older people. A
multidisciplinary geriatric team usually consists of a physician, a registered nurse andlicensed practical nurse, a social worker, occupational therapist, physiotherapist, dietician,and psychologist (101). The whole person must be considered so that appropriate treatment,
rehabilitation, and follow ups can be initiated. A meta-analysis has found that a CGA programis effective in enhancing survival and function in older people (103). Similarly, a workinggroup of Nordic teachers in geriatric medicine have recently summarized recent researchshowing that geriatric rehabilitation is complicated but effective when properly performed
(104). Finally, Brocklehurst’s textbook of Geriatric Medicine and Gerontology concludes thatthe efficacy of CGA has been confirmed in many settings. While there is no single optimalblueprint for geriatric assessment, the participation of a multidisciplinary team and the focus
on functional status and quality of life as major clinical goals are common to all settings.Further the authors in the textbook mention that the greatest benefits have been found inprograms targeted on frail, elderly people (102).
Hip fracture rehabilitation
The numbers of individuals suffering from hip fracture have encouraged the development ofcare following hip fracture surgery. Wells et al, and Cameron recommend in recent reviews(105, 106) that elderly hip-fracture patients shall be offered multidisciplinary geriatric
rehabilitation.There are several different approaches used in Randomised Controlled Trials (RCT) on
hip-fracture patients (107, 108). So-called Geriatric Hip Fracture Programs (GHFP) is one
such which includes coordinated, multidisciplinary care and is usually located at the alreadyexisting orthopaedic wards with a geriatrician or a geriatric team included in the care of thepatients soon after admission. RCT with GHFP are presented in Table 1. In a review (107),the authors concluded that GHFP, and Early Supported Discharge programs (ESD), with early
discharge and rehabilitation in home settings, are probably effective, since they shorten theLength Of hospital Stay (LOS). This latter approach has also showed improved ADLperformance in two RCT (109, 110).
Tab
le 1
. An
over
view
of
rand
omis
ed c
ontr
olle
d m
ultid
isci
plin
ary
tria
ls a
mon
g hi
p-fr
actu
re p
atie
nts
with
Ger
iatr
ic H
ip F
ract
ure
Prog
ram
(G
HFP
) ap
proa
ch
Tri
alSe
tting
s an
dpa
rtic
ipan
tsSe
lect
ion
Inte
rven
tion
Out
com
es a
nd r
esul
tsC
omm
ents
Jette
et a
l.U
SA19
87(6
5)
Ort
hopa
edic
war
dsn=
75 (
35 I
, 40
C)
67%
wom
enm
ean
age:
78
yrs.
Qua
si r
ando
mis
ed(b
ased
on
on-c
all
rost
er)
Incl
usio
n>
54 y
rs w
ith p
roxi
mal
fem
oral
fra
ctur
e
Fol
low
up
at d
isch
arge
, 3, 6
and
12
mon
ths.
Ort
hopa
edic
war
d w
ith s
tand
ard
post
sur
gica
l reh
abili
tatio
n pl
usin
tens
ive
reha
bilit
atio
n in
clud
ing,
educ
atio
n, C
GA
, wee
kly
team
mee
tings
, hom
e vi
sits
, tel
epho
neca
llsC
ontr
olO
rtho
paed
ic w
ard
with
stan
dard
pos
t-su
rgic
alre
habi
litat
ion.
NS
M
orta
lity
NS
D
isch
arge
des
tinat
ion
NS
Fu
nctio
nal r
ecov
ery
(Fun
ctio
nal S
tatu
s In
dex,
FSI)
Smal
l sam
ple
Swan
son
et a
l.A
ustr
alia
1998
(111
)
Tea
chin
g ho
spita
lw
ith e
arly
dis
char
gen=
71 (
38 I
, 33
C)
78%
wom
enm
ean
age
78 y
rs
Ran
dom
ised
by
a tr
ial
coor
dina
tor
Incl
usio
n�5
5 yr
sN
on p
atho
logi
cal
prox
imal
fem
ur f
ract
ure
Inde
pend
ently
mob
ileA
ble
to g
ive
info
rmed
cons
ent
Exc
lusi
onD
emen
tia w
ithin
adeq
uate
Eng
lish
Liv
ing
in n
ursi
ng h
ome
Fol
low
up
6 an
d 12
mon
ths
for
som
e ou
tcom
esIn
terv
entio
n w
ith m
ulti-
disc
iplin
ary
team
, ger
iatr
icia
n, e
arly
mob
iliza
tion,
hom
e vi
sit,
follo
w u
psC
ontr
olSt
anda
rd o
rtho
paed
ic m
anag
emen
t
Pos
LO
SP
os
Func
tiona
l rec
over
y(m
od B
arth
el)
at d
isch
arge
NS
M
orta
lity
NS
D
isch
arge
dest
inat
ion,
(a
tren
d)
19
Tri
alSe
tting
s an
dpa
rtic
ipan
tsSe
lect
ion
Inte
rven
tion
Out
com
es a
nd r
esul
tsC
omm
ents
Vid
an e
t al .
Spai
n20
05(1
12)
Ort
hopa
edic
war
dsn=
319
(155
I, 1
64C
)m
ean
age
82 y
rs82
% w
omen
Incl
usio
n>
65 y
rsE
xclu
sion
Inab
ility
to w
alk
Dep
ende
nt in
all
P-A
DL
Path
olog
ical
fra
ctur
esT
erm
inal
illn
ess
Fol
low
up
at d
isch
arge
, 3, 6
and
12
mon
ths
CG
A, d
aily
vis
it by
ger
iatr
icia
n,te
amw
ork,
wee
kly
mee
tings
Con
trol
Sam
e or
thop
aedi
c w
ards
but
with
stan
dard
car
e
NS
LO
SP
osM
orta
lity
Pos
Med
ical
pro
blem
sP
osFu
nctio
nal r
ecov
ery,
part
ial r
ecov
ery
but n
ot a
tth
e en
d of
stu
dy
I =
Int
erve
ntio
n gr
oup
C =
Con
trol
gro
upN
S =
no
sign
ific
ant d
iffe
renc
ePo
s =
sig
nifi
cant
eff
ect o
f in
terv
entio
nL
OS=
leng
th o
f st
ayC
GA
= C
ompr
ehen
sive
Ger
iatr
ic A
sses
smen
t
20
21
Another way to rehabilitate old people with hip fractures is to use geriatric Mixed
Assessment and Rehabilitation Units (MARU). These are units with coordinated,multidisciplinary care for older people with various diagnoses. Another model forrehabilitation of old hip-fracture patients are Geriatric Orthopaedic Rehabilitation Units
(GORU) with geriatric or rehabilitation teams located in specialized wards for old,orthopaedic patients.
This thesis focuses mainly on GORU and MARU randomised controlled trials. Suchprograms, with coordinated multidisciplinary geriatric care after hip fracture surgery, do not
have a homogeneous outcome, and there is no conclusive evidence of their effectivenesscompared with conventional care, usually in orthopaedic units (108). The authors in aCochrane review conclude that combined measures tend to be better for those who receive
coordinated rehabilitation, but the results were not statistically significant (108). The resultsobtained in multidisciplinary rehabilitation trials are conflicting, and as the programs havedifferent study designs, inclusion criteria and methods of assessments it is difficult to makecomparisons (64, 112-119).
Regarding hospital stay, the multidisciplinary units have showed shorter LOS in somestudies (113, 116, 117) and longer in others (114, 118), three studies showed equal LOS (115,120, 121).
Concerning functional recovery, some studies show significant effects early on after thehospital stay (113, 116), but from a longer perspective there is only one study by Kennie et al,presented in two papers, (117, 119) that has produced significant effects. While this trialdemonstrates increased P-ADL performance in the intervention group 12 months after
surgery, it included only 108 women and there was a tendency towards more cognitivelyimpaired patients in the control group.
Table 2 presents an overview of RCT among hip-fracture patients using
multidisciplinary care (GORU/MARU).
Tab
le 2
. An
over
view
of
rand
omis
ed c
ontr
olle
d m
ultid
isci
plin
ary
tria
ls a
mon
g hi
p-fr
actu
re p
atie
nts
in G
eria
tric
Ort
hopa
edic
Reh
abili
tatio
n U
nits
/Mix
ed A
sses
smen
ts a
nd
Reh
abili
tatio
n U
nits
(G
OR
U/M
AR
U)
Tri
alSe
tting
s an
d pa
rtic
ipan
tsSe
lect
ion
Inte
rven
tion
Out
com
es a
nd r
esul
tsC
omm
ents
Ford
ham
et a
l,U
K19
86(1
20)
Ger
iatr
ic-o
rtho
paed
icco
mbi
ned
unit
vs.
orth
opae
dic
unit
n=50
I, 5
8 C
all w
omen
Ran
dom
isat
ion
uncl
ear
Incl
usio
n:Fe
mor
al n
eck
frac
ture
>65
yrs
Exc
lusi
on:
Adm
itted
fro
m o
ther
dis
tric
tsT
erm
inal
illn
ess
Dem
entia
for
mor
e th
an s
ixm
onth
s
Fol
low
up
at d
isch
arge
Com
bine
d ge
riat
ric-
orth
opae
dic
man
agem
ent
Con
trol
sta
ndar
d ca
re
NS
LO
SN
S
AD
LN
S
Dis
char
gede
stin
atio
n
Inte
rven
tion
grou
ptr
ansf
erre
d to
com
mun
ity h
ospi
tal
Lon
g w
aitin
g tim
ebe
fore
tran
sfer
red
toin
terv
entio
n w
ard,
hal
fof
thos
e ra
ndom
ised
toin
terv
entio
n gr
oup
neve
r w
ent t
o th
eco
mbi
ned
stud
y w
ard
Gilc
hris
t et a
l.U
K19
88(1
15)
Ort
hopa
edic
ger
iatr
icun
it vs
. ort
hopa
edic
war
dsn=
97 I
, n=
125
Cal
l wom
enm
ean
age:
~ 8
1 yr
s
Stra
tifie
d (i
ntra
- or
ext
ra-
caps
ular
)In
clus
ion
Prox
imal
fem
ur f
ract
ure
>65 Exc
lusi
onPa
tient
s fr
om n
earb
y ho
spita
lsan
d pa
tient
s w
ith r
apid
pro
gres
sw
ere
sent
bac
k di
rect
ly.
Fol
low
up
atdi
scha
rge
and
3,6
mon
ths
for
mor
talit
yO
rtho
paed
ic g
eria
tric
uni
t with
aw
eekl
y co
mbi
ned
roun
d,m
edic
al a
dvic
e by
ger
iatr
icia
n.C
ontr
olst
anda
rd c
are.
NS
M
orta
lity
NS
L
OS
NS
D
isch
arge
dest
inat
ion
Mor
e m
edic
al c
ondi
tions
wer
e de
tect
ed a
nd tr
eate
din
the
inte
rven
tion
grou
p
Wea
k in
terv
entio
nR
ando
mis
atio
n un
clea
rN
o as
sess
or b
lindi
ngIn
tent
ion
to tr
eat
Bot
h gr
oups
tran
sfer
red
to a
per
iphe
ral h
ospi
tal
Ken
nie
et a
l.U
K19
88(1
17, 1
19)
Reh
abili
tatio
n vs
.O
rtho
paed
ic w
ard
n=10
8 (5
4 ea
ch g
rp)
all w
omen
Seal
ed e
nvel
opes
Incl
usio
nPr
oxim
al f
emur
fra
ctur
e�6
5 yr
sE
xclu
sion
Die
d be
fore
bec
omin
g fi
t to
ente
r th
e tr
ial
Path
olog
ical
fra
ctur
esL
ikel
y to
be
disc
harg
ed w
ithin
seve
n da
ysU
nfit
to tr
ansf
er to
a p
erip
hera
lho
spita
ln=
36
Fol
low
up
at d
isch
arge
, and
12
mon
ths
Tre
atm
ent g
roup
wer
etr
ansf
erre
d to
an
orth
opae
dic
reha
bilit
atio
n un
it at
a p
erip
hera
lho
spita
l aft
er s
urge
ryG
ener
al p
ract
ition
er p
rovi
ded
med
ical
atte
ntio
nG
eria
tric
ian
roun
dsM
ultid
isci
plin
ary
care
with
team
conf
eren
ce e
ach
wee
kC
ontr
olth
eor
thop
aedi
cad
mis
sion
war
d w
ith s
tand
ard
care
Pos
P
hysi
cal
inde
pend
ence
at d
isch
arge
(Kat
z) a
nd a
t 12
mon
ths
Pos
R
esid
ence
aft
erdi
scha
rge
and
at 1
2m
onth
sP
os
LO
S
No
asse
ssor
blin
ding
Inte
ntio
n to
trea
tT
ende
ncy
to h
ave
mor
eco
gniti
vely
impa
ired
inth
e co
ntro
l gro
up
22
Tri
alSe
tting
s an
d pa
rtic
ipan
tsSe
lect
ion
Inte
rven
tion
Out
com
es a
nd r
esul
tsC
omm
ents
Cam
eron
et a
l.A
ustr
alia
1993
(113
, 122
)
Gen
eral
hos
pita
ln=
252
(127
I/1
25 C
)83
% w
omen
mea
n ag
e 84
yrs
39%
fro
m n
ursi
ngho
mes
Stra
tifie
d ra
ndom
isat
ion
in th
ree
grou
ps a
ccor
ding
to li
ving
arra
ngem
ents
and
dis
abili
tyIn
clus
ion
Prox
imal
fem
ur f
ract
ure
>50
yrs
Exc
lusi
onPa
thol
ogic
al f
ract
ure
Add
ition
al f
ract
ures
Wai
ting
time
over
7 d
ays
for
surg
ery
n=77
In-p
atie
nt f
ract
ure
or tr
ansf
erre
dto
ano
ther
hos
pita
ln=
20
Fol
low
up
until
dea
th o
r 4
mon
ths
Acc
eler
ated
reh
abili
tatio
n w
ithea
rly
mob
ilisa
tion,
ear
lyge
riat
ric
asse
ssm
ent a
nddi
scha
rge
plan
ning
, fam
ilyco
ntac
ts, o
ut-p
atie
ntre
habi
litat
ion
Con
trol
stan
dard
trea
tmen
t
Pos
LO
SP
osPh
ysic
alin
depe
nden
ce,e
ffec
tive
for
thos
e w
ith li
mite
ddi
sabi
lity
prio
r to
inju
ry(B
arth
el in
dex)
A tr
end
whe
n in
clud
ing
all p
atie
nts
at 2
wee
ks a
nd1
mon
th (
Bar
thel
)N
Sph
ysic
alin
depe
nden
ce a
t 4 m
onth
sP
osre
side
nce
afte
rdi
scha
rge
amon
g th
ose
not l
ivin
g in
nur
sing
hom
esN
Sre
adm
issi
ons,
com
plic
atio
ns a
t 4 m
onth
sP
osre
duce
d co
sts
up to
4 m
onth
s
No
asse
ssor
blin
ding
A la
rge
prop
ortio
n of
the
cont
rol g
roup
rece
ived
inte
rdis
cipl
inar
y ca
reeq
ual t
o th
ein
terv
entio
n gr
oup
Gal
vard
et a
l.Sw
eden
1995
(114
)
Ger
iatr
ic v
s. o
rtho
paed
icde
part
men
tsn=
371
(179
I/ 1
92 C
)74
% w
omen
mea
n ag
e ~7
9 yr
s
Ran
dom
ised
usi
ng r
ando
mnu
mbe
r ge
nera
tor
afte
r su
rger
yIn
clus
ion
Hip
fra
ctur
e pa
tient
s liv
ing
inde
pend
ently
Exc
lusi
onN
ursi
ng-h
ome
patie
nts
Hos
pita
lized
Fol
low
up
12 m
onth
sIn
terv
entio
n no
t des
crib
edG
eria
tric
reh
abili
tatio
n w
ithw
eekl
y vi
sits
by
an o
rtho
paed
icsu
rgeo
nC
ontr
ol:
Stan
dard
ort
hopa
edic
car
e
NS
mor
talit
yN
Sdi
scha
rge
dest
inat
ion
NS
w
alki
ng a
bilit
y at
12
mon
ths
Neg
L
OS
Pos
r
eadm
issi
ons
No
asse
ssor
blin
ding
Few
wer
e fo
llow
edre
gard
ing
wal
king
abili
ty
Sanc
hez
Ferr
in e
t al.
Spai
n19
99 (
121)
Ger
iatr
ic f
unct
iona
l uni
tvs
. con
vent
iona
l car
en=
206
(103
in e
ach
grou
p)76
% w
omen
, mea
n ag
e82
yrs
Ran
dom
num
ber
gene
rato
rIn
clus
ion
>64
yrs
Fol
low
up
6 m
onth
sIn
itial
ger
iatr
ic a
sses
smen
t for
thos
e in
inte
rven
tion
grou
p vs
.st
anda
rd o
rtho
paed
ic c
are
for
cont
rol g
roup
NS
LO
SN
SFu
nctio
n (A
DL
)N
SN
ursi
ng-h
ome
plac
emen
tN
SM
orta
lity
Pos
C
ompl
icat
ions
No
asse
ssor
blin
ding
Inte
ntio
n to
trea
tB
oth
grou
ps s
hare
dro
oms
and
staf
f
23
Tri
alSe
tting
s an
d pa
rtic
ipan
tsSe
lect
ion
Inte
rven
tion
Out
com
es a
nd r
esul
tsC
omm
ents
Huu
sko
et a
l.Fi
nlan
d20
00, 2
002
(64,
116
)
Ger
iatr
ic w
ard
vs. l
ocal
hosp
ital
n=24
3 (1
20 I
/123
C)
72%
wom
en
Com
pute
r-ge
nera
ted
and
seal
edin
num
bere
d en
velo
pes
Incl
usio
nC
omm
unity
dw
ellin
g>
64 y
rsE
xclu
sion
Path
olog
ical
or
mul
tiple
frac
ture
sD
epen
dent
wal
ker
Ter
min
ally
ill
Inst
itutio
naliz
ed
Fol
low
up
at d
isch
arge
, 2w
eeks
, 3 a
nd 1
2 m
onth
sA
bout
two
wee
ks o
f in
tens
ive
reha
bilit
atio
n w
ith e
arly
mob
iliza
tion,
sel
f-m
otiv
atio
nan
d fu
nctio
n, d
isch
arge
pla
nnin
gan
d fo
llow
up,
exe
rcis
es in
thei
rho
mes
for
two
mon
ths,
wee
kly
team
mee
tings
Con
trol
:st
anda
rd c
are
NS
M
orta
lity
NS
D
isch
arge
dest
inat
ion
Pos
L
OS
Pos
F
unct
iona
l rec
over
y
Shor
t-te
rm e
ffec
t but
not
at th
e en
d of
the
stud
y
Inte
ntio
n to
trea
tN
o as
sess
or b
lindi
ngSu
bgro
up a
naly
ses
ofpa
tient
s w
ith d
emen
tiapr
esen
ted
in a
sep
arat
epa
per.
Tho
se w
ith m
ild o
rm
oder
ate
dem
entia
coul
d re
turn
to th
eco
mm
unity
as
succ
essf
ull
y a
s p
atie
nts
wit
h n
orm
al c
og
nit
ive
fun
ctio
n
Nag
lie e
t al.
Can
ada
2002
(118
)
GO
RU
vs.
ort
hopa
edic
unit
n=27
9 (1
41 I
, 138
C)
mea
n ag
e 84
yrs
80%
wom
en
Stra
tifie
d (a
ge a
nd r
esid
ence
),co
mpu
ter-
gene
rate
dra
ndom
isat
ion
Incl
usio
n>
70 y
rsC
omm
unity
and
nur
sing
hom
esE
xclu
sion
Frac
ture
d du
ring
acu
te c
are
Path
olog
ical
fra
ctur
eE
xpec
ted
surv
ival
less
than
6m
onth
sD
epen
dent
wal
ker
Fol
low
up
3, 6
mon
ths
Tea
mw
ork,
CG
A, e
arly
mob
iliza
tion,
team
mee
tings
,di
scha
rge
plan
ning
, hom
e vi
sit
NS
M
orta
lity
NS
Fu
nctio
nN
eg
LO
S
Inte
ntio
n to
trea
tA
sses
sor
blin
ding
Tho
se w
ith m
ild-
mod
erat
e co
gniti
vede
clin
e be
nefi
t fro
mth
e pr
ogra
m
I =
Int
erve
ntio
n gr
oup
CG
A =
Com
preh
ensi
ve G
eria
tric
Ass
essm
ent
C =
Con
trol
gro
upG
OR
U =
Ger
iatr
ic O
rtho
paed
ic R
ehab
ilita
tion
Uni
tN
S =
no
sign
ific
ant d
iffe
renc
eM
AR
U =
Mix
ed A
sses
smen
t Reh
abili
tatio
n U
nit
Pos
= s
igni
fica
nt e
ffec
t of
inte
rven
tion
Neg
= n
egat
ive
effe
ct o
f in
terv
entio
nL
OS=
leng
th o
f st
ay
24
25
Rationale for this thesis
There is a lack of knowledge about the very old who sustain hip fractures, despite the fact thatthe incidence of such fractures increases with age and entail much suffering for those affected.Many longitudinal studies have demonstrated declines in mobility, performance of activitiesof daily living and institutionalization for those affected but none have investigated the impact
of a hip fracture in a population of the very old.Almost all hip fractures are due to falls, and those suffering from a fracture will
probably fall again soon after the injury. Knowledge is lacking about falls during hospital
stays following hip-fracture surgery. There is also no fall-prevention study carried out in ahospital among hip-fracture patients only and, generally, there are few fall-prevention trials ofany kind in hospitals. The literature recommends multidisciplinary and multi-factorial fallinterventions among old people in the community and in residential care facilities. So far, no
intervention has been reported to be effective among people with cognitive impairments ordementia. It is, therefore, of interest to study in-patient falls and evaluate fall-preventioninterventions among the elderly with hip fractures during their rehabilitation including those
with dementia.In addition, postoperative multidisciplinary, multi-factorial intervention programs using
comprehensive geriatric assessments can be used in the rehabilitation of old people after hip-fracture surgery. Some studies have demonstrated partial functional benefits but, from a
longer perspective, there is only one study, which has produced significant effects. Therefore,it is of great importance to develop and evaluate effective, intervention programs to enhancethe long-term functional performance in old people who have suffered a hip fracture.
26
AIMS OF THE THESIS
The overall aim of this thesis was to study the impact of earlier hip fractures on life in the
very old, to study fall-risk factors and to evaluate a multidisciplinary, multi-factorialintervention program to reduce in-patient falls and to enhance the functional performance inold people with a femoral neck fracture.
Specific aims
• to study the prevalence and impact of earlier hip fractures on life in a population of thevery old, and to investigate associations between having had a hip fracture and housingconditions, the performance of P-ADL, mobility, balance and falls. Paper I
• to study in-patient falls, fall-related injuries, and risk factors for falls following femoralneck fracture. Paper II
• to evaluate whether a postoperative multidisciplinary, multi-factorial intervention programcan reduce in-patient falls and fall-related injuries in patients with femoral neck fractures.
Paper III
• to investigate the short- and long-term effects of a multidisciplinary, multi-factorialintervention program in participants with femoral neck fracture regarding livingconditions, walking ability and performance of ADL. Paper IV
27
METHODS
Settings and participants
This thesis is based on two samples, one from the municipality of Umeå and one from
Umeå University Hospital. The first sample (Paper I) comprised 253 very old participants inUmeå 85+ (2), the second sample comprised 199 participants aged 70 years or over, eligiblefor an intervention study (Papers II-IV) (Figure 1). Paper I includes elderly people with all
types of proximal femoral fractures given various treatments and Papers II-IV are based onparticipants suffering femoral neck fractures and basocervical fractures.
Figure 1. The two samples studied in Papers I-IV.
Paper I
A sample, including half the population born in 1915 (85-year-olds), the total populationsborn in 1910 (90-year-olds) and from 1905-1897 (�95-year-olds), living in the municipality of
Umeå, Sweden on the 1 January 2000, were selected for participation (n=348).Figure 2 describes the inclusion procedure. Twenty-nine of the 348 (8.3%) died before
they were invited to participate; there were no sex differences among these compared to the
remaining 319 individuals. During recruitment 66 individuals, out of 319, (20.7%) declined toparticipate. These 66 individuals were more likely to be younger (p=0.008), married
Sample 1253 participants in themunicipality of Umeå
Sample 2199 participants atUmeå University
Hospital
Paper I253
participants
Paper II97
participants
Paper III199
participants
Paper IV199
participants
28
(p<0.001) and to live in ordinary housing (p<0.001), but there were no sex differences
compared to the study sample. The final study sample consisted of 253 (79.3%) participants,58 (22.9%) individuals with a previous hip fracture (cervical or trochanteric femoral fracture)and 195 (77.1%) without any history of a hip fracture (Table 3).
Figure 2. Flow chart of the inclusion procedure for the sample studied in Paper I.
Eligible participantsn=348
Died before request n=29
Asked toparticipate
n=319
Declined to participaten=66
Umeå 85+Study sample
n=253
Without earlier hipfracturen=195
With earlier hipfracture
n=58
29
Papers II-IV
The study sample in Papers II-IV consists of 199 participants (Table 3) with femoral neckfractures aged 70 years or older, consecutively admitted to the orthopaedic department atUmeå University Hospital, Sweden, between May 2000 and December 2002. The exclusioncriteria were: rheumatoid arthritis, severe hip osteoarthritis, severe renal failure, pathological
fracture, and being bedridden before the fracture occurred. These exclusion criteria wereapplied because of the operation methods used in the study.
Two hundred and fifty eight individuals met the inclusion criteria, 11 of those declined
to participate and 48 individuals were not invited to participate because they had sustained thefracture in the hospital or the inclusion routines failed (Figure 3). These 59 individuals weremore likely to be men (p=0.033), and living in own house/apartment (p=0.009) but there wasno difference in age (p=0.354) compared to those who agreed to participate.
In Paper II, the 97 participants randomised to the control group are analysed separately.
Ethical approval
The Ethics Committee of the Medical Faculty of Umeå University approved the studies (§ 99-
326) and (§ 00-137). All participants in the papers, or their relatives, received written and oralinformation and gave their informed consent to participation.
30
Figure 3. Flow chart for the randomised trial analysed in Papers II-IV. In Paper II only the 97 participants in the
control group are analysed. In Papers III-IV all 199 participants are analysed.
95 failed to meet inclusion critera
59 excluded:11 refused to participate
21 suffered the fracture in the hospital27 missing due to failure of inclusion routines
102 included in primary analysis
84 assessed at 12 months7 participants died between 4-12 months
1 participant declined to continue
92 assessed at four months3 participants died after discharge1 participant declined to continue
102 assigned to intervention group6 participants died during hospitalization
97 included in primary analysis
76 assessed at 12 months5 participants died between 4-12 months
2 participants declined to continue
83 assessed at four months6 participants died after discharge
1 participant moved to another city
97 assigned to control group7 participants died during hospitalization
199 randomised
258 eligible for inclusion
353 participants assessed for eligibility
31
Procedure
Paper I
In this population-based cross-sectional study, Umeå 85+ (2), names, addresses and civilregistration numbers were collected from the National Tax Board. The quasi-randomisation of85-year-olds was conducted by selecting every second person from the lists received from the
National Tax Board, on which individuals are listed according to date of birth.As a first step, the participants received a letter with information about the study and then,about two weeks later, they were informed about the home visit procedure through atelephone call and gave their informed consent. For people living in institutions, the staff were
contacted and asked about the elderly persons’ cognitive status, and whether the researchgroup should call them personally or their next-of-kin, who then decided whether they wouldparticipate. The oldest participants were asked first and the total data collection was
performed over a period of 18 months. Assessments were performed during two to threehome visits by one of the four investigators (two medical students, one nurse and onephysiotherapist). Data were collected from the participants, relatives, caregivers, and frommedical charts kept at the hospital, by general practitioners and/or the institutional care
facility. For the participants with a previous history of a hip fracture medical charts werereviewed before and after the fracture event in order to verify the fracture.
Papers II-IV
In the emergency room individuals with femoral neck fractures were asked both in writingand orally if they were willing to participate in the study. The next of kin were also asked inthose cases where there was cognitive impairment. The participants were randomised, to
postoperative care in a geriatric ward (n=102) with a special intervention program or toconventional care in an orthopaedic ward (n=97), using opaque sealed envelopes. Allparticipants received an envelope while in the emergency room but it was not opened untilimmediately before surgery to ensure that all received similar preoperative treatment. People
not involved in the study carried out these selection procedures. All participants received thesame preoperative treatment.
The randomisation was stratified according to the operation methods used in the study.
Depending on the degree of dislocation the participants were treated with two hook-pins(Swemac Ortopedica®, Linköping, Sweden) (n= 38 intervention vs. n= 31 control) or withbipolar hemiarthroplasty (Link®, Hamburg, Germany) (n=57 vs. 54). Basocervical fractures(n=7 vs. 10) were operated on using a Dynamic Hip Screw (DHS, Stratec Medical®, Oberdorf,
Switzerland). One participant had a resection of the femoral head due to deterioration inmedical status and one died before surgery (both in the control group).
32
Medical, social and functional data were collected from the participants, relatives, staff,
and medical records during hospitalization (Papers II-IV) and at four and twelve monthspostoperatively (Paper IV).
Two registered nurses were employed halftime in the study of whom the one from the
orthopaedic department carried out the assessments in the intervention group, and the onefrom the geriatric department carried out the same assessments in the control group. APhysiotherapist (PT), an Occupational Therapist (OT) and physicians were also employed tocollect data during the project. The nurses employed by the study performed the assessments
during the hospital stay and interviewed the participants three to five days after surgery, ondischarge the PT and OT, from the orthopaedic and geriatric departments respectively,measured functional ability. At four months (±2 weeks) and twelve months (±1 month)
postoperatively the survivors in both groups were followed up in their homes by a nurse, anda PT or an OT working in the study.
33
Methods of data collection
Housing. In all papers the type of housing was seen as institutional living when theparticipants were living in some form of institution, including residential care facilities,nursing homes and group dwellings for people with dementia. Having their own house orapartment was classified as independent living. Among the participants with a history of a hip
fracture in Paper I, medical charts were reviewed before and after the fracture event and thehousing situation was classified as permanently changed if participants still lived in a placeother than the one they had lived in before the fracture.
Diagnoses and medications. In Paper I, a geriatrician evaluated all the documentationconcerning diagnoses, drug treatments and assessments for completion of the final diagnose.
In Papers II-IV a geriatrician, unaware of study-group allocation, analyzed all assessments
and documentation after the study was finished, to complete the final diagnoses according tothe same criteria for all participants.
Cognitive impairment. Dementia was diagnosed if the participant had a previous dementiadiagnosis based on an earlier dementia assessment, and/or if the Mini Mental StateExamination (MMSE) (123) and the Organic Brain Syndrome (OBS) scale (124) showedsignificant cognitive impairment, indicating dementia. The MMSE (123) is a screening test
used to assess cognitive status among older people, scoring from 0 to 30 with a score of lessthan 24 indicating cognitive impairment (125). When the history of cognitive impairment wasuncertain, the participants were referred to the geriatric clinic for a complete dementia
assessment (Paper I). In Papers II-IV a geriatrician blindly analyzed all assessments, includingMMSE (123) and the OBS scale (124) and documentations including all the participants’medical and nursing records after the twelve-month follow up to decide whether theparticipant met the Diagnostic and Statistical Manual for mental disorder, fourth edition
(DSM-IV), criteria (126) for dementia.Delirium, is a neuropsychiatric syndrome characterised by disturbed attention and
cognition with a fluctuating course developed over a short period of time and by definition
always due to an underlying cause (126). In Papers II-IV delirium was assessed using themodified OBS scale (124), consisting of two main parts: the disorientation subscale, aquestionnaire containing 12 items, and the confusions subscale, an observation schedulecovering 21 clinical features. The disorientation subscale measures the person’s orientation to
time, place, and their own identity with a maximum score of 36 (higher score = poorer testresult). The confusion subscale includes various assessments of cognitive, perceptual,emotional, and personality changes, as well as fluctuations in the clinical states. It is based on
observations and interviews with the participants as well as interviews with the caregivers.The OBS Scale has been compared with other assessment scales and has shown goodconcurrent validity (124). It has also been compared to the Confusion Assessment Method(CAM) and showed 100% agreement regarding the diagnosis of delirium (127). Changes in
mental state and numbers of delirious days for each participant were also registered from the
34
medical records. The symptoms of delirium during the first 8 postoperative hours were
viewed as immediate effects of pre-medication and/or of other anaesthetic agents, and werenot registered as delirium. A geriatrician blindly analyzed the assessments anddocumentations and all participants’ medical and nursing records after the twelve-month
follow up to decide whether the participants met the DSM-IV criteria (126) for delirium.
Depression. Depression was diagnosed after an evaluation of earlier diagnoses documented inthe charts, current treatment with antidepressants, and depression screened for using the
Geriatric Depression Scale (GDS-15) (128). The participants who scored 5 or more on theGDS were further assessed by a specialist in geriatric medicine using the Montgomery-Åsberg Depression Rating Scale (MADRS) (129), see Paper I.
In Papers II-IV, pre-injury depression was diagnosed following an evaluation of earlierdiagnoses documented in the records, and current treatment with antidepressants. Depressionduring hospitalization was diagnosed according to current treatment with antidepressants andscreened for using the GDS-15 (128) in combination with depressive symptoms observed and
registered by the OBS Scale. A geriatrician blindly analyzed the assessments anddocumentation, including all participants’ medical and nursing records, after the twelve-month follow up to decide whether the patients met the DSM-IV criteria (126) for depression.
Nutrition. Nutrition was assessed in Paper I using the Mini Nutritional Assessment (MNA)(130). MNA is a screening instrument for nutritional status and has a maximum score of 30,which indicates good nutritional status. Scores between 23.5 and 17 indicate risk of
malnutrition and scores below 17 indicate the presence of malnutrition. Body Mass Index(BMI) was also calculated (kg/m2).
Vision and hearing. In Papers II-IV, the participants’ vision and hearing were assessed bytheir ability to read three-millimetre block letters with or without glasses, and their ability tohear a normal speaking voice from a distance of one meter.
Activities of daily living and walking ability. In Paper I, the Barthel ADL index (131) wasused to assess dependency in P-ADL. It consists of 10 items (bowel continence, bladdercontinence, grooming, toilet use, feeding, transfer, mobility, dressing, stairs, and bathing) and
each individual item score is summarized into a total score, where the maximum score of 20indicates P-ADL independence.
In Papers II-IV activities of daily living prior to the fracture were measuredretrospectively using the Staircase of ADL including the Katz ADL index (132, 133). The
scale measures both P-ADL (bathing, dressing, toileting, transfer, continence, and feeding)
and Instrumental ADL (I-ADL) (cleaning, shopping, transportation, and cooking). The ADLscore consists of ten steps; categorized from zero to ten, with a higher figure indicating more
ADL dependence.Information about the use of walking aids or a wheelchair in Paper I was collected from
the participants, relatives and staff and from medical charts at the time of the home visit.
35
Among the participants with a history of a hip fracture, medical charts were reviewed before
and after the fracture event up to the home visit. Dependency on a walking aid or a wheelchairwere classified as permanently changed if participants still used a mobility aid 6-12 monthsafter the accident that they had not used before the fracture and still used it at the home-visit.
One item from the Swedish version of Clinical Outcome Variables (134), (S-COVS)(135) was used in Papers II-IV to assess the participants’ walking ability. The item has sevenlevels, one indicating no functional walking ability and seven indicating normal function,including outdoor obstacles and gait speed. Use of a walking aid was also registered.
Balance. Balance was assessed in Paper I, using the Swedish version (136, 137) of Berg’sBalance Scale (BBS) (138). The scale consists of 14 static and dynamic balance tasks. Each
task is common in everyday life, e.g. sitting, standing, turning and reaching, and is scored 0 to4, where 0 represents inability to perform and 4 ability to perform the task safely. Themaximum score is 56, which reflects the ability to perform everyday tasks in a safe andcontrolled manner. A zero score means that the participant cannot sit unsupported.
Postoperative complications. Complications during hospitalization, morbidity and mortalitywere systematically registered from the medical and nursing records in Papers II-IV.
Falls and fall-related injuries. In Paper I the participants were asked if they had sustained anyfalls during the previous year, further fall information was collected from relatives, staff andmedical charts at the time of the home visit.
In Papers II-III postoperative falls were systematically registered from the records.Nurses are obliged by law to document all falls in the records (the Swedish Code of Statutes1985) (139). A fall was defined as an incident when the participant unintentionally came to
rest on the floor and included syncopal falls. Numbers of falls and time lapse to first fall afteradmission were calculated. A physician assessed the participants soon after every fall if anyinjuries were suspected. The Abbreviated Injury Scale (AIS) (84) was used to categorize theinjuries resulting from a fall into a hierarchical structure of severity, with AIS scores ranging
from one (minor) to six (maximum). The Maximum Injury (MAIS) connected with eachincident was recorded.
Out-patient rehabilitation. In connection with the follow up (Paper IV), the number of out-patient contacts with a PT and/or OT was registered. In the analyses more than five contactswas seen as an out-patient rehabilitation period. Readmissions and in-hospital days afterdischarge were also registered at four and twelve months respectively.
Tab
le 3
. Bas
elin
e ch
arac
teri
stic
s of
the
part
icip
ants
in th
e fo
ur p
aper
s, n
umbe
rs p
rese
nted
and
per
cent
age,
mea
n ±
stan
dard
dev
iatio
ns, (
SD),
or
med
ian
with
inte
rqua
rtile
rang
e, (
IQR
)
Pape
r I
Pape
r II
Pape
rs I
II-I
V
Prev
ious
hip
fra
ctur
e, n
=58
Non
-hip
fra
ctur
e, n
=19
5Fa
llers
, n=
26N
on f
alle
rs, n
=71
Inte
rven
tion,
n=
102
Con
trol
, n=
97
Soci
odem
ogra
phic
Age
91.5
±5.0
89.6
±4.7
81.9
±5.0
82.1
±6.2
82.3
±6.6
82.0
±5.9
Fem
ales
54 (
93%
)13
7 (7
0%)
15 (
58%
)59
(83
%)
74 (
73%
)74
(76
%)
Liv
ing
Inde
pend
ently
22 (
38%
)11
9 (6
1%)
18 (
69%
)42
(59
%)
66 (
65%
)60
(62
%)
Hea
lth
and
med
ical
pro
blem
s
Stro
ke14
(24
%)
42 (
22%
)6
(25%
)14
(20
%)
29 (
28%
)20
(21
%)
Prev
ious
hip
fra
ctur
e-
-1
(4%
)13
(19
%)
16 (
16%
)14
(14
%)
Hea
rt f
ailu
re17
(29
%)
47 (
24%
)-
--
-
Car
diov
ascu
lar
dise
ase
--
16 (
67%
)37
(54
%)
57 (
56%
)53
(55
%)
Dia
bete
s4
(7%
)27
(14
%)
4 (1
5%)
13 (
18%
)23
(23
%)
17 (
18%
)
Dem
entia
27 (
47%
)48
(25
%)
11 (
42%
)25
(35
%)
28 (
27%
)36
(37
%)
Dep
ress
ion
18 (
32%
)47
(25
%)
13 (
50%
)32
(46
%)
33 (
32%
)45
(46
%)
*Fal
ls d
urin
g pr
eced
ing
mon
th-
-9
(39%
)16
(24
%)
24 (
24%
)25
(26
%)
Sens
ory
impa
irm
ents
Impa
ired
hea
ring
--
8 (3
3%)
26 (
45%
)42
(41
%)
34 (
35%
)
Impa
ired
vis
ion
--
7 (3
7%)
20 (
36%
)37
(36
%)
27 (
28%
)
Fun
ctio
nal p
erfo
rman
ce
Bar
thel
AD
L in
dex
12.1
±7.9
16.4
±5.6
--
--
Stai
rcas
e of
AD
L-
-5
(4-7
)4
(0-8
)5
(1-7
.75)
5 (0
.25-
7)
Wal
ked
inde
pend
ently
out
side
--
11 (
42%
)36
(51
%)
55 (
54%
)47
(48
%)
Wal
ked
inde
pend
ently
, at l
east
indo
ors
33 (
57%
)16
3 (8
4%)
85 (
83%
)85
(88
%)
Wal
ked
with
out w
alki
ng a
ids
indo
ors
--
16 (
62%
)40
(56
%)
47 (
46%
)55
(57
%)
36
Pape
r I
Pape
r II
Pape
rs I
II-I
V
Prev
ious
hip
fra
ctur
e, n
=58
Non
-hip
fra
ctur
e, n
=19
5Fa
llers
, n=
26N
on f
alle
rs, n
=71
Inte
rven
tion,
n=
102
Con
trol
, n=
97
Med
icat
ions
Ant
idep
ress
ants
14 (
54%
)31
(44
%)
29 (
28%
)45
(46
%)
Ben
zodi
azep
ines
5 (1
9%)
20 (
28%
)
Neu
role
ptic
s1
(4%
)7
(10%
)
Ass
essm
ents
MM
SE s
core
17.5
±11.
122
.1±7
.414
.6±7
.216
.1±9
.817
.4±8
.215
.7±9
.1
GD
S5.
2±3.
84.
2±3.
45.
2±3.
64.
5±3.
5
OB
S sc
ale
13.2
±9.4
12.2
±12.
110
.1±1
0.8
12.5
±11.
4
MN
A19
.7±6
.423
.1±4
.422
.5±3
.621
.8±4
.2
BM
I22
.7±4
.924
.7±4
.4
* E
xcep
t for
the
fall
that
res
ulte
d in
the
hip
frac
ture
.
37
38
Intervention
The intervention program used in Papers III-IV was based on and developed from previousresearch in Umeå University Hospital and Piteå River Valley Hospital (17, 53, 140, 141). Inthe program carried out as part of this thesis the rehabilitation part has been elaborated toinclude follow ups and a focus on fall prevention, compared to the program used in Piteå
River Valley Hospital (17).The intervention ward in Papers III-IV, was a geriatric unit specializing in geriatric
orthopaedic patients. The staff worked in teams to apply comprehensive geriatric assessments,
management and rehabilitation (101, 104). Active prevention, detection and treatment ofpostoperative complications such as falls, delirium, pain, decubital ulcers, and malnutritionwere systematically implemented. Early mobilization, with daily training was provided byPTs, OTs and care staff during the hospital stay. A geriatric team, including a physician,
assessed the participants four months postoperatively to detect and treat any complicatingdisorders and to determine further rehabilitation needs. The staffing at the intervention wardwere 1.07 nurses/aides per bed.
The control ward was a specialist orthopaedic unit following conventional postoperativeroutines. A geriatric unit, specializing in general geriatric patients, was used for those whoneeded longer rehabilitation (n=40), but such participants were not admitted to the same wardas that used for the intervention. The staffing at the orthopaedic unit was 1.01 nurses/aides per
bed and 1.07 for the geriatric control ward. The main content of both the intervention programand the conventional care is described in Table 4.
The staff on the intervention ward were partly aware of the nature of the study, and the
staff working with the control group were told that a new care program was beingimplemented and that it was being evaluated on the geriatric ward.
Tab
le 4
. Mai
n co
nten
t of
the
post
oper
ativ
e pr
ogra
m a
nd d
iffe
renc
es b
etw
een
the
two
grou
ps in
Pap
ers
III
and
IV
Inte
rven
tion
grou
pC
ontr
ol g
roup
War
d la
yout
- Si
ngle
and
dou
ble
room
s.
- 24
-bed
war
d, e
xtra
bed
s w
hen
need
ed.
- Si
ngle
, dou
ble
and
four
-bed
roo
m.
- 27
-bed
war
d, e
xtra
bed
s w
hen
need
ed.
- T
he g
eria
tric
con
trol
war
d w
as s
imila
r to
the
inte
rven
tion
war
d.
Staf
fing
- 1.
07 n
urse
s/ai
des
per
bed.
- T
wo
full-
time
phys
ioth
erap
ists
.
- T
wo
full-
time
occu
patio
nal t
hera
pist
s.
- 0.
2, d
ietic
ian.
- 1.
01 n
urse
s pe
r be
d.
- T
wo
full-
time
phys
ioth
erap
ists
.
- 0.
5, o
ccup
atio
nal t
hera
pist
.
- N
o di
etic
ian.
- St
affi
ng o
n th
e ge
riat
ric
cont
rol w
ard
sim
ilar
to th
at o
n th
e
inte
rven
tion
war
d.
Staf
f ed
ucat
ion
- A
fou
r-da
y co
urse
on
cari
ng, r
ehab
ilita
tion,
team
wor
k an
d m
edic
al k
now
ledg
e
incl
udin
g se
ssio
ns a
bout
how
to p
reve
nt, d
etec
t and
trea
t var
ious
pos
tope
rativ
eco
mpl
icat
ions
suc
h as
pos
tope
rativ
e de
liriu
m a
nd f
alls
.
- N
o sp
ecif
ic e
duca
tion
befo
re o
r du
ring
the
proj
ect.
Tea
mw
ork
- T
eam
incl
uded
Reg
iste
red
Nur
ses
(RN
), L
icen
sed
Prac
tical
Nur
ses
(LPN
),
Phys
ioth
erap
ists
(PT
), O
ccup
atio
nal T
hera
pist
s (O
T),
a d
ietic
ian
and
geri
atri
cian
s.
- C
lose
co-
oper
atio
n be
twee
n or
thop
aedi
c su
rgeo
ns a
nd g
eria
tric
ians
in th
e m
edic
al c
are
of th
e pa
tient
s.
- N
o co
rres
pond
ing
team
wor
k at
the
orth
opae
dic
unit.
- T
he g
eria
tric
war
d, w
here
som
e of
the
cont
rol g
roup
pat
ient
s
wer
e ca
red
for,
use
d te
amw
ork
sim
ilar
to th
at in
the
inte
rven
tion
war
d.
Indi
vidu
al c
are
plan
ning
- A
ll te
am m
embe
rs a
sses
sed
each
pat
ient
as
soon
as
poss
ible
, usu
ally
with
in 2
4 ho
urs,
to b
e ab
le to
sta
rt th
e in
divi
dual
ized
car
e pl
anni
ng.
- T
eam
pla
nnin
g of
the
patie
nt’s
indi
vidu
al r
ehab
ilita
tion
proc
ess
and
goal
s tw
ice
a
wee
k.
- In
divi
dual
ized
car
e pl
anni
ng w
as u
sed
in th
e or
thop
aedi
c un
it
but n
ot r
outin
ely,
as
in th
e in
terv
entio
n w
ard.
- A
t the
ger
iatr
ic r
ehab
ilita
tion
unit
ther
e w
as w
eekl
y in
divi
dual
care
pla
nnin
g.
Pre
vent
ion
and
trea
tmen
t of
com
plic
atio
ns
- In
vest
igat
ion
as f
ar a
s po
ssib
le r
egar
ding
how
and
why
the
part
icip
ants
sus
tain
ed th
e
hip
frac
ture
, thr
ough
ana
lysi
ng e
xter
nal a
nd in
tern
al f
all-
risk
fac
tors
.
- A
ctio
n to
pre
vent
fur
ther
fal
ls a
nd f
ract
ures
impl
emen
ted
incl
udin
g gl
obal
rat
ings
of
the
part
icip
ants
fal
l ris
k ev
ery
wee
k du
ring
team
mee
tings
.-
Cal
cium
and
Vita
min
-D a
nd o
ther
pha
rmac
olog
ical
trea
tmen
ts f
or o
steo
poro
sis
wer
e
adm
inis
tere
d w
hen
indi
cate
d.
-Act
ive
prev
entio
n, d
etec
tion
and
trea
tmen
t of
post
oper
ativ
e co
mpl
icat
ions
suc
h as
delir
ium
, pai
n, d
ecub
ital u
lcer
s, w
as s
yste
mat
ic.
-Oxy
gen-
enri
ched
air
dur
ing
the
firs
t pos
tope
rativ
e da
y an
d lo
nger
if n
eces
sary
unt
il th
e
- N
o ro
utin
e an
alys
is o
f w
hy th
e pa
tient
s ha
d fr
actu
red
thei
r
hips
; no
atte
mpt
was
mad
e to
sys
tem
atic
ally
pre
vent
fur
ther
falls
nor
was
ther
e an
y ro
utin
e pr
escr
iptio
n of
Cal
cium
and
Vita
min
-D.
- A
sses
smen
ts f
or p
osto
pera
tive
com
plic
atio
ns w
ere
mad
e by
chec
king
e.g
. sat
urat
ion,
hae
mog
lobi
n, n
utri
tion,
bla
dder
and
bow
el f
unct
ion,
hom
e si
tuat
ion
etc.
but
thes
e ch
eck-
ups
wer
e
not c
arri
ed o
ut s
yste
mat
ical
ly a
s in
the
inte
rven
tion
grou
p.
39
mea
sure
d ox
ygen
sat
urat
ion
was
sta
ble.
-Uri
nary
trac
t inf
ectio
ns a
nd o
ther
infe
ctio
ns w
ere
scre
ened
for
and
trea
ted.
-If
a ur
inar
y ca
thet
er w
as u
sed
it w
as d
isco
ntin
ued
with
in 2
4 ho
urs
post
oper
ativ
ely.
-Reg
ular
scr
eeni
ng f
or u
rina
ry r
eten
tion,
and
pre
vent
ion
and
trea
tmen
t of
cons
tipat
ion.
-Blo
od tr
ansf
usio
n w
as p
resc
ribe
d if
B-h
aem
oglo
bin,
g/l,
<10
0 an
d <
110
for
thos
e at
risk
of
delir
ium
or
thos
e al
read
y de
lirio
us.
-If
the
patie
nt s
lept
bad
ly, t
he r
easo
n w
as in
vest
igat
ed a
nd th
e ai
m w
as th
en to
trea
t the
caus
e.
Nut
ritio
n-F
ood
and
liqui
d re
gist
ratio
n w
as s
yste
mat
ical
ly p
erfo
rmed
and
pro
tein
-enr
iche
d m
eals
wer
e se
rved
to a
ll pa
tient
s du
ring
the
firs
t fou
r po
stop
erat
ive
days
and
long
er if
nece
ssar
y.
-Nut
ritio
nal a
nd p
rote
in d
rink
s w
ere
serv
ed e
very
day
.
-No
diet
icia
n w
as a
vaila
ble
at th
e or
thop
aedi
c un
it.
-No
rout
ine
nutr
ition
reg
istr
atio
n or
pro
tein
-enr
iche
d m
eals
for
the
patie
nts.
Reh
abili
tatio
n-
Mob
iliza
tion
with
in th
e fi
rst 2
4 po
stop
erat
ive
hour
s.
- T
he tr
aini
ng in
clud
ed b
oth
spec
ific
exe
rcis
e an
d ot
her
reha
bilit
atio
n pr
oced
ures
deliv
ered
by
a PT
, OT
as
wel
l as
basi
c da
ily A
DL
per
form
ance
trai
ning
, by
cari
ng s
taff
.
- T
he p
atie
nts
alw
ays
do a
s m
uch
as th
ey c
an b
y th
emse
lves
bef
ore
they
are
hel
ped.
- T
he r
ehab
ilita
tion
was
bas
ed o
n fu
nctio
nal r
etra
inin
g w
ith a
spe
cial
foc
us o
n fa
ll-ri
sk
fact
ors.
- H
ome
visi
t by
an O
T a
nd/o
r a
PT
- T
he P
T/O
T c
o-op
erat
ed w
ith c
olle
ague
s w
orki
ng in
com
mun
ity s
ervi
ce f
or f
urth
erco
nsul
tatio
n af
ter
the
patie
nt w
as d
isch
arge
d fr
om h
ospi
tal.
- A
ll pa
tient
s w
ere
offe
red
furt
her
out-
patie
nt r
ehab
ilita
tion
afte
r di
scha
rge.
- T
he P
T o
r O
T f
ollo
wed
up
all p
atie
nts
with
a p
hone
cal
l tw
o w
eeks
aft
er d
isch
arge
and
a ho
me
visi
t fou
r m
onth
s po
stop
erat
ivel
y.-
A p
hysi
cian
met
the
patie
nts
four
mon
ths
post
oper
ativ
ely
to d
etec
t and
pre
vent
com
plic
atio
ns.
- M
obili
zatio
n us
ually
with
in th
e fi
rst 2
4 ho
urs.
- T
he P
T o
n th
e w
ard
mob
ilize
d th
e pa
tient
s to
geth
er w
ith th
e
cari
ng s
taff
.
- T
he P
T a
imed
to m
eet t
he lu
cid
patie
nts
ever
y da
y.-
Func
tiona
l ret
rain
ing
in A
DL
situ
atio
ns w
as n
ot a
lway
s gi
ven.
- T
he O
T a
t the
ort
hopa
edic
uni
t onl
y m
et th
e pa
tient
s fo
r
cons
ulta
tion.
- N
o ho
me
visi
ts w
ere
mad
e by
sta
ff f
rom
the
orth
opae
dic
unit.
- T
he g
eria
tric
con
trol
war
d ha
d bo
th s
peci
fic
exer
cise
and
othe
r re
habi
litat
ion
proc
edur
es d
eliv
ered
by
a PT
and
OT
,
sim
ilar
to th
at g
iven
at t
he in
terv
entio
n w
ard.
- N
o fo
llow
up
by a
phy
sici
an a
t fou
r m
onth
s.
40
41
Statistical analyses
All calculations were made using the statistical software package SPSS® (SPSS Inc., Chicago,Illinois) version 10.0 (Paper I) and version 11.0 (Papers II-IV), as well as Stata software,version 9.1 (StataCorp, College Station, Texas) in Paper III. All statistical tests were 2-tailedand a p-value of <0.05 was considered statistically significant.
Paper I
Descriptive data are presented as numbers and percentages, means with Standard Deviations(SD), and medians with InterQuartile Ranges (IQR). The proportions of those who had
sustained an earlier hip fracture are presented for each age group and for women and men,respectively, with 95% Confidence Intervals (CI) (142).
Unadjusted comparisons between participants with and without previous hip fractures
were analysed for statistical significance using Student’s t-test, Pearson’s chi-square test, andFisher’s exact test.
The associations between hip fracture and the outcome variables were adjusted forpossible confounding variables (e.g. sex, age) by multiple logistic and linear regression
analyses. The selections of variables used in the adjustments were made in two stages. 1)Possible confounding variables for which the differences between participants with andwithout hip fractures were statistically significant were selected. 2) Selected variables were
put in a factor analysis in order to extract a number of variables with low inter-correlations.The variables finally chosen for adjustments were those with the highest factor loading onseparate factors. In cases where two (or more) variables loaded similarly on the same factor,the one which was considered to have the greatest confounding effect on the dependent
variables was chosen. Varimax rotation was used for the factor analyse, eigenvalue was presetat >1.0, and a factor loading of >0.6 was seen as significant.
Paper II
Univariate Cox regression analyses, including Hazard Rate Ratio (HRR) and 95% CI, werecalculated between known and potential fall-risk variables and the time lapse to theoccurrence of first fall. Multiple regression analyses were performed including those variables
that were significantly associated with falls in the univariate analyses using the Coxregression forward stepwise (Ward) function. The Mann-Whitney U-test was used to analysedifferences in hospital stay and Pearson’s chi-square test was used to describe differencesbetween falls during a day with delirium and the total number of delirious days in all the
postoperative days.
42
Paper III
Student’s t-test, Pearson’s chi-square test and the Mann-Whitney U-test were performed toanalyse group differences regarding basic characteristics and postoperative complications.We analysed outcomes on an intention-to-treat basis. The incidence of falls betweenintervention and control groups was compared in three ways: First an unadjusted comparison
was made using Pearson’s chi-square and Fisher’s exact test regarding number of participantswho fell and injuries: Secondly, the fall-incidence rate was compared between interventionand control groups by calculating the fall Incidence Rate Ratio (IRR) using a negative
binomial regression, with adjustment for observation time and for over-dispersion. Negativebinomial regression (Nbreg) is a generalisation of the Poisson regression model and isrecommended for evaluating the efficacy of fall-prevention programs (143). Thirdly, a Coxregression was used to compare the time lapse to first fall between groups (Hazard Rate
Ratio; HRR). The difference in fall risk between groups was further illustrated by a Kaplan-Meier graph.
Basic characteristics that differed between the intervention and the control groups,
corresponding to a p-value <0.15 (depression, antidepressants and dementia) were consideredas covariates in the Poisson (Nbreg) and the Cox regression models. However, the inclusionof these variables had only marginal effects on the log likelihood values of the models as wellas on the IRR and HRR values and standard errors for the group allocation variable
(intervention or control). In addition, none of the variables showed significant effects on thedependent variable and are therefore not included in the Poisson (Nbreg) and Cox regressionanalyses.
Pearson’s chi-square test and Fisher’s exact test were also used to analyse theassociations between falls and days with delirium between the groups.
Paper IV
Student’s t-test, Pearson’s chi-square test and the Mann-Whitney U-test were used to analysegroup differences regarding basic characteristics and in some of the primary and secondaryoutcome variables. A binary logistic regression method was used to analyse the Odds Ratio(OR) of living situation, walking ability and use of walking aids as well as ADL performance
related to intervention or control group allocation. These regressions were adjusted fordepression and dementia diagnoses before injury (p-value <0.15) as well as for the baselinesituation of every outcome variable respectively.
43
RESULTS
Paper I
The proportions of participants with earlier hip fractures are presented in Table 5. Of the total
sample (n=253), 58 participants had had a hip fracture with the proportion varying between16-33% in the three age groups. In total, there were 70 hip fractures among the 58participants. Twelve had had two hip fractures (data not shown) and seven had had both a
cervical and a trochanteric fracture (Table 6). The four men had all had trochanteric fracturesand none of them had had two fractures. During the last five years, 28 participants (11%) hadhad a hip fracture. The time since their first hip fracture was in median 4-8 years in the three
age groups.
Table 5. Prevalence of previous hip fracture among the oldest old
85-year-olds
n=95
90-year-olds
n=86
�95-year-olds
n=72
Hip fracture, n (%) 15 (16) 19 (22) 24 (33)
95% CI 9%-23% 13%-31% 22%-44%
Female, n 69 63 59
Hip fracture, n (%) 14 (20) 18 (29) 22 (37)
95% CI 11%-30% 17%-40% 25%-50%
Male, n 26 23 13
Hip fracture, n (%) 1 (4) 1 (4) 2 (15)
95% CI 0%-11% 0%-13% 0%-35%
44
Table 6. Distribution of numbers of hip fractures among the age groups
85-year-olds
n=95
90-year-olds
n=86
�95-year-olds
n=72
Cervical fracture 11 6 10
Trochanteric fracture 4 11 9
Cervical and trochanteric
fractures
0 2 5
Years since first fracture,
median, (IQR)
Min-max
4.0 (2.0-10.0)
1-28
8.0 (3.0-16.0)
0-30
6.5 (3.0-16.75)
1-28
Hip fracture in last 5 years 10 7 11
Mean (±SD) age at first hip
fracture
78±7 80±9 87±8
45
As can be seen from Table 7, participants with hip fractures differed from those without
regarding age, sex, living alone, dementia, MMSE score, nutrition, BMI score, and urinarytract infections during the preceding year. A rotated factor analysis resulted in three separatefactors. Dementia, MMSE, and urinary tract infections loaded significantly (factor loading of
>0.6) on the first factor. Age, BMI and MNA loaded significantly on the second, and sex onthe third factor. To adjust further analyses dementia, age, and sex were selected as covariates.
Table 7. Basic characteristics among participants with and without an earlier hip fracture
Hip fracturen=58
No hip fracturen=195
p-value*
Mean age, mean ± SD 91.5±5.0 89.6±4.7 0.012
Female sex 54 (93%) 137 (70%) <0.001
Living alone 56 (97%) 166 (85%) 0.020
Stroke 14 (24%) 42 (22%) 0.675
Heart failure 17 (29%) 47 (24%) 0.423
Diabetes 4 (7%) 27 (14%) 0.156
Dementia 27 (47%) 48 (25%) 0.001
Depression (n=56/186) 18 (32%) 47 (25%) 0.309
Urinary tract infection
preceding year
23 (40%) 47 (24%) 0.020
MMSE, (n=54/181) 17.5±11.1 22.1±7.4 0.001
MNA (n=57/185) 19.7±6.4 23.1±4.4 <0.001
BMI (n=56/183) 22.7±4.9 24.7±4.4 0.007
Unadjusted t-tests and Chi square tests
46
Table 8 demonstrates large unadjusted differences between the groups in the outcome
variables. After adjusting for the confounding effects of the selected covariates, havingsuffered a hip fracture was independently associated with P-ADL performance and mobility,including wheelchair use. The association with BBS bordered on statistical significance.
However, no associations were seen with type of housing, use of walking aids or falls.
Table 8. Outcome variables between those with and without an earlier hip fracture, unadjusted and adjusted
models
Hip fracturen=58
No hip fracturen=195
Unadjusted p-value*
Adjustedp-value**
Institutional care 36 (62%) 76 (39%) 0.002 0.396
Barthel ADL, total index 12.1±7.9 16.4±5.6 <0.001 0.024
Barthel mobility item,
dichotomised to independent
walking
33 (57%) 163 (84%) <0.001 0.040
Use of walking aids 53 (91%) 146 (75%) 0.007 0.106
Use of wheelchair 26 (45%) 36 (18%) <0.001 0.017
Berg’s balance scale (BBS)
(n=46/167)
25.2±20.6 38.0±18.0 <0.001 0.053
Fall during preceding year(n=57/193)
32 (56%) 83 (43%) 0.080 0.293
* Unadjusted t-tests and Chi square tests
** Adjusted by multiple logistic and multiple linear regression, respectively. Factors selected for adjustment
were, age, sex and dementia. All regressions had a p-value of <0.001.
As can be seen from Table 9, 28% (16) of the participants with previous hip fractureshad been living in institutional care before the fracture incident, compared to 62% (36) at thetime of home visits for data collection. Of those 20 who had moved to an institution between
these time-points, 60% (12) had done so permanently after the fracture. Those who hadchanged their use of walking aids and wheelchairs did so in connection with the hip fracturein 62% and 32% of cases, respectively.
Table 9. Institutional care and use of walking aids and wheelchairs among old people with a history of hipfracture (n=58)
Before hip
fracture
Within 6-12 months after hip
fracture
At home visit
Institutional care 16 28 36
Use of walking aids 16 39 53
Use of wheelchair 1 9 26
47
Paper II
Falls and fall-related injuriesDuring hospitalization there were 60 postoperative falls among 26/97 participants
(27%). The postoperative fall incidence rate was 16.3/1 000 days (CI 12.2-20.4). Sixteen fellmore than once, mean 2.3 ± 2.1 times (range 1-11). Time lapse to first fall varied between 2 to
79 days, median 18 days after surgery. The falls were most common in the second and fourthweek (Figure 4), and between 12:00 and 22:00, with a peak between 20:00-22.00. Themajority (67%) of the falls took place in the participant’s room or in the bathroom.
Injuries were reported in 32% of the falls and four of them (7%) were serious injuries(AIS 3), including two hip fractures, one rib fracture with pneumothorax, and one withmultiple skull fractures. Finally, there were 15 minor injuries (AIS 1) (25 %) such as bruises,contusions, and wounds, according to the AIS injury scale.
The median hospital stay was 27.0 (IQR 11.0-55.0) with a range of 2 to 208 days. Therewas a significant difference in hospital stay between fallers and non-fallers (p=<0.001).Among those ten with the longest postoperative in-hospital stays (93-206 days) there were
eight fallers, two of whom had had new fractures. The mean in-hospital stay among those fourwith new fractures was 101 days.
Figure 4. Fall incidence /1 000 days and number of falls each postoperative week.
Postoperative week number
48
Fall-risk factorsUnivariate Cox regressions showed a significant association between time lapse to first
fall and males, history of falls, postoperative delirium after day seven, number of deliriousdays, and sleeping disturbances during hospital stay (Table 11). In the multiple Cox
regression analyses delirium after day seven, male sex, and sleeping disturbances remainedsignificant (Table 12). We did not find any associations between any of the functionalvariables or use of any medications.
Participants were registered as delirious during 746 out of the 3 685 observation days
(20%) but 27/60 (45%) of all falls occurred during a day when the participants were delirious(p=<0.001).
Table 10. Complications during in hospital stay
Total
n=97
Fallers
n=26
Non-fallers
n=71
Delirium postoperatively 73 (75%) 25 (96%) 48 (68%)
Delirium after day seven, (n=26/61) 45 (52%) 23 (88%) 22 (36%)
Number of delirious days, mean±SD 7.7±12.3 15.1±20.0 5.0±6.2
Sleeping disturbances 44 (45%) 21 (81%) 23 (32%)
Depression during hospital stay 53 (55%) 15 (58%) 38 (54%)
Nutritional problems 37 (38%) 14 (54%) 23 (32%)
Urinary tract infection, (n=26/70) 49 (51%) 16 (62%) 33 (47%)
Urinary retention 18 (19%) 5 (19%) 13 (18%)
Decubital ulcers, (n=26/69) 21 (22%) 5 (19%) 16 (23%)
Anaemia, (n=26/70) 79 (82%) 24 (92%) 55 (79%)
In-patient mortality 7 (7%) 2 (8%) 5 (7%)
49
Table 11. Potential fall-risk factors analyzed by univariate Cox regression, dependent variable time lapse to first
fall
Variable Hazard Rate Ratio 95% CI
Delirium after day seven, n=26/61 6.77 1.97-23.24
Delirium postoperatively 6.22 0.83-46.42 Sleeping disturbances 4.05 1.51-10.85
Males 3.57 1.53-8.31
Dementia 2.08 0.90-4.86 Falls during last month, n=23/67 2.04 1.01-4.15
Nutritional problems 1.42 0.64-3.16
Cardiovascular disease, n=24/69 1.36 0.57-3.21
Anaemia, n=26/70 1.29 0.30-5.52 Diabetes 1.17 0.40-3.44
Stroke, n=24/69 1.16 0.43-3.13
Impaired vision, n=19/55 1.05 0.39-2.82
Use of antidepressants 1.02 0.46-2.26 Number of delirious days, mean±SD 1.02 1.01-1.04
Depression, n=26/69 1.00 0.46-2.21
Age, mean±SD 0.98 0.91-1.06
Walking with walking aids indoors 0.95 0.43-2.12 Urinary tract infection, n=26/70 0.87 0.38-2.01
Walking independently outside 0.65 0.29-1.48
Table 12. Multiple Cox regression analyses using forward stepwise (Wald) between significant Hazard Rate
Ratios in the univariate analyses, dependent variable was time lapse to first fall
Variable Hazard Rate Ratio 95% CI
Delirium after day seven 4.62 1.30-16.37*
Males 3.92 1.58-9.73
Sleeping disturbances 3.49 1.24-9.86
* corrected 95% CI compared with the published paper
50
Paper III
During hospitalization 12 participants in the intervention group sustained a total of 18 falls(range 1-3) and in the control group 26 participants sustained a total of 60 falls (30 falls in theorthopaedic unit and 30 in the geriatric control unit) (range 1-11). Among individuals withdementia one participant sustained a single fall in the intervention group and 11 participants
with dementia sustained a total of 34 falls in the control group (Table 13).The crude postoperative fall-incidence rate was 6.29/1000 days in the intervention
group vs. 16.28/1000 days in the control group. Using a negative binomial regression, the
IRR was significantly lower in the intervention group, 0.38 (95% CI: 0.20–0.76, p=0.006) andamong participants with dementia, 0.07 (95% CI: 0.01-0.57, p=0.013) (Table 13). In Figure 5,a Kaplan-Meier survival analysis of time lapse to first fall illustrates the difference betweenthe two groups, with a significantly reduced fall rate in the intervention group (log rank p-
value 0.008).The difference in fall risk, expressed as time lapse to first fall, is compared between
intervention and control groups in a Cox regression (HRR) analysis. Including all participants
in the calculation, the fall risk was significantly lower in the intervention group, HRR 0.41(95% CI 0.20-0.82, p=0.012).
There were in total three minor or moderate injuries (MAIS 1-2) in the interventiongroup compared to 15 in the control group, according to the AIS injury scale. The serious
injuries (MAIS 3) were new fractures of which four, two hip fractures, one rib fracture withpneumothorax, and one with multiple skull fractures, all occurred in the control group withnone in the intervention group (Fisher’s Exact test: p=0.055).
Three of the participants who fell in the intervention group (25%) and 12 in the controlgroup (46%) did so during a day when they were delirious (p=0.294). Analysing the numberof falls, revealed that four out of 18 (22%) falls in the intervention group and 27 out of 60(45%) in the control group occurred on a day when the participant was delirious, p=0.083.
Apart from the falls there were fewer other postoperative complications in theintervention group, such as fewer participants with postoperative delirium (p=0.003) andfewer delirious days (p=<0.001), urinary tract infections (p=0.005), sleeping disturbances
(p=0.009), nutritional problems (p=0.038), and decubital ulcers (p=0.010). The postoperativein-hospital stay was shorter in the intervention group, 28.0±17.9 days vs. 38.0±40.6 days,p=0.028. Among those ten with the longest postoperative in-hospital stays in the controlgroup there were eight participants with any fall and two with new fractures.
51
Table 13. Falls during hospitalization
Intervention
n=102
Control
n=97
p-value
Number of falls 18 60
Postoperative in-hospital days 2860 3685
Crude fall incidence rate (number of falls/1000 days) 6.29 16.28
IRR (95% CI) 0.38 (0.20-0.76)* 1.00 (Ref) 0.006
Number of fallers 12 26 0.007
Number of fallers with injuries due to falls 3 15 0.002
Number of fallers with fractures due to falls 0 4 0.055
Number of falls among people with dementia 1 34
IRR (95% CI) among people with dementia 0.07 (0.01-0.57)* 1.00 (Ref) 0.013
Number of fallers among people with dementia, n=28/36 1 11 0.006
*Negative binomial regression analyses adjusted for over-dispersion and controlled for dementia, depression and
use of antidepressants.
CI = Confidence IntervalIRR = Incidence Rate Ratio
Kaplan-Meier survival graph
Log rank 0.008
time lapse to first fall, (days)
160140120100806040200
1,0
,9
,8
,7
,6
,5
,4
,3
,2
,1
0,0
Figure 5. Kaplan-Meyer survival graph of time lapse to first fall in days.
Intervention
Control
52
Paper IV
Similar proportions of participants in both groups returned to their pre-fracture livingsituation. Among participants in the intervention group, 81 (84%) were discharged fromhospital to the same place of residence as they had had before the fracture compared with 68(76%) in the control group, p=0.132. At the four- and twelve-month follow up there were
similar proportions of survivors in both groups who had returned to the same residentialsituation as they had had before the fracture (data not shown).
There was no difference regarding independent walking ability performance between
the groups during the follow-up period, Table 14. When analysing the whole seven-gradedwalking ability item according to the S-COVS, 57 participants (62%) in the interventiongroup and 40 participants (49%) in the control group had regained the same level of walkingability as they had had before the fracture or had improved their level by the four-month
follow up, p=0.081. Fifty-two participants (62%) vs. 40 participants (53%) in intervention andcontrol group respectively, had regained the same level of walking ability as they had hadbefore the fracture or had improved their ability at the twelve-month follow up, p=0.236.
More participants in the intervention group walked without walking aids indoors attwelve months (35/84 vs. 22/76, OR 3.01, 95% CI 1.18-7.61, adjusted for baselinedifferences, dementia, depression and baseline walking ability) (Table 14).
Significantly more participants in the intervention group had regained independent P-
ADL ability at four and twelve months, adjusted for baseline differences, 35/92 vs. 23/83, OR2.51 (95% CI 1.00-6.30) and 33/84 vs. 17/76, OR 3.49 (95% CI 1.31-9.23) respectively(Table 14).
On discharge, 47 out of 96 in the intervention group had returned to at least the sameADL performance level as before the fracture, according to the Katz ADL index, comparedwith 30/89 in the control group, p=0.036. At four months the figures were 56/92 in theintervention group and 39/82 in the control group, p=0.078 and at twelve months 49/84
compared to 27/76 in intervention and control group respectively, p=0.004.The out-patient rehabilitation consumption after discharge from hospital was similar for
the two groups, 37 participants in the intervention group and 31 in the control group were
given a rehabilitation period after their in-hospital stay with a PT and or OT, p=0.562. Ofthose who were given a rehabilitation period significantly more participants from theintervention group had received that training at an out-patient centre specializing inrehabilitation for the elderly, 21 compared to 7 in the control group, p=0.007.
The total length of in-hospital stay differed between the groups, the mean in-patient stayin the intervention group was 30.0±18.1 days compared with 40.0±40.6 days in the controlgroup, p=0.028. During the first postoperative year, the total mean in-hospital stay, (including
both the in-hospital stay in connection with the fracture itself and any in-hospital stays afterdischarge) was 37.0±28.2 days in the intervention group and 51.4±66.4 days in the controlgroup, p=0.051.
53
There was no difference in mortality between the groups, at discharge, at four months or
at the one-year follow up. By the one-year follow up 16 participants in the intervention grouphad died (16%) compared to 18 in the control group (18%), p=0.591.
Table 14. Numbers of participants living independently, independent walking ability, ADL performance and
Odds Ratio of being treated in the intervention group
Intervention
n=102
Control
n=97
OR* 95% CI
Living independently before fracture 66 60
On discharge 55 46 0.93 0.32-2.73
At four-month follow up 54 46 0.68 0.20-2.27
At twelve-month follow up 47 36 0.91 0.32-2.56
Independent walking ability, at least indoors,
before fracture
85 85
On discharge 51 45 0.75 0.34-1.63At four-month follow up 59 52 1.03 0.47-2.24
At twelve-month follow up 55 45 1.13 0.50-2.55
Walking without walking aid indoors beforefracture
47 55
On discharge 4 0 †
At four-month follow up 31 19 2.22 0.99-4.95
At twelve-month follow up 35 22 3.01 1.18-7.61
Independent in P-ADL before fracture 47 48
On discharge 30 20 1.81 0.74-4.37
At four-month follow up 35 23 2.51 1.00-6.30At twelve-month follow up 33 17 3.49 1.31-9.23
* Adjusted for baseline depression and dementia as well as baseline situation of the outcome variable.
ADL = Activities of Daily Living
OR = Odds RatioCI = Confidence Interval
P-ADL = Personal/Primary Activities of Daily Living
† Too few individuals for analysis
54
55
DISCUSSION
The studies for this thesis revealed that having had a previous hip fracture seems to have a
serious impact on dependency and living conditions in old people. Among very old people,previous hip fractures are common and are associated with dependency in P-ADL and for alarge proportion with permanently reduced mobility. The findings also showed that it was not
unusual to have sustained more than one hip fracture. Among a large proportion of the veryold it seems that the occurrence of a hip fracture earlier in their life had brought about apermanent change in their housing conditions and use of mobility aids.
In-patient falls and fall-related injuries were common occurrences during postoperativerehabilitation after femoral neck fracture. Postoperative delirium, male sex and sleepingdisturbances were associated with an increased fall risk during in-hospital rehabilitation.
Almost half the falls occurred during a day when the participants were recorded as beingdelirious.
The multidisciplinary, multi-factorial postoperative intervention program was shown to have
positive effects compared to standard postoperative routines after a femoral neck fracture. Aprogram with systematic assessment and treatment of fall-risk factors, active prevention,detection and treatment of other postoperative complications resulted in fewer participants
falling, a smaller total number of falls and fewer fall-related injuries. Participants withdementia benefited most from the intervention program. This program also resulted in moreparticipants recovering independence in P-ADL by the four- and twelve-month follow up,controlled for baseline differences and despite shorter hospitalization. More participants in the
intervention group had also regained walking ability without walking aids indoors at twelvemonths compared with controls treated according to standard postoperative routines.
56
Prevalence of hip fracture and its association with dependency among thevery old
In this cross-sectional population-based study among the very old, the prevalence of earlier
hip fractures (16% among 85-year-olds and 22% among 90-year-olds) was close to that foundby Österlind et al in a longitudinal population-based study which reported an 18% prevalenceamong the 85-year-olds and 29% among the 88-year-olds women in northern Sweden (144).
On the other hand, a lower figure (11%) was reported from mid-Sweden (145). However, thelatter study reported a relatively small proportion living in institutional care; among thoseolder than 85 years, only 18% lived in institutional care, compared with 44% in the presentstudy. Since the present cross-sectional study only included half the population in the 85-
year-old group and the results in the present study are not weighted, the 44% living ininstitutional care could be overestimated figures. Among people aged 85 years and older inthe USA, the prevalence of previous hip fractures was 13% (56).
The high proportion of old people who have sustained hip fractures earlier in life, withthe subsequent consequences, should be viewed against the expected increase in mortalityafter a hip fracture, which is described in many studies (13, 15, 63, 67, 146). The fact that thismortality is higher among men (13, 63, 69), could partly explain why there were so very few
men with previous hip fractures in the present study.The present study confirms the decline in function after a hip fracture, and that this
remains for many years. Having had a hip fracture is a factor that may be independently
associated with the ability to perform P-ADL and mobility. Hochberg et al (56) found that hipfracture is associated with difficulty in performing ADL among women not in institutionalcare and Norton et al (14) showed that the hip fracture cases have a lower ADL performanceand mobility two years after the fracture, compared with controls. The present study indicates
that this reduction in ability can be expected to remain until death. Norton et al also found anincreased use of walking aids and wheelchairs, which is similar to our findings. Unlikeprevious studies (55, 145) no significant effects could be found, from a long perspective, on
institutionalisation after adjusting for potential confounders, but many of the very old had hadto permanently change their type of housing after the fracture.
Falls, fall-related injuries and risk factors
The aim of Paper II was to study falls, fall-related injuries and risk factors for falls during thein-hospital stay after hip fracture surgery. A higher incidence of postoperative falls was found
among hip-fracture participants than previously reported (74). However, Pils et al excludedfor example those with dementia (MMSE <20), severe morbidity, and those living in nursinghomes. Our study included a large proportion of participants with morbidities such as
dementia, and many were admitted from institutions. When compared with non-hip-fracturesamples, the fall incidence remains high (50, 147). The fall incidence is as high as that foundat a psycho-geriatric ward among people with dementia (147). A large proportion (21-34%)
57
of those sustaining a hip fracture are demented (80, 116). In the present study 37% were
diagnosed with dementia.The only study found which analyzed in-patient falls in hip-fracture patients did not
present figures for injuries, apart from three new hip fractures occurring among the 935 hip-
fracture patients (74). The high proportion of falls resulting in injuries in the present study isin closer agreement with studies carried out in residential care facilities (36, 38, 147). In-hospital postoperative falls in hip-fracture patients is clearly a serious and common problem,which also appears to increase length of hospital stay and thereby the costs involved.
Males were at greater risk of falling during the in-hospital rehabilitation following a hipfracture, as has been reported earlier (74). One explanation could be that males are more illand frail than women when they suffer a hip fracture and have a lower threshold for
complications such as delirium. Male gender is also associated with poorer rehabilitationoutcome and higher mortality after a hip fracture (13, 59, 62, 63). In contrast to Pils et al (74),the present study found no increased fall risk with increasing age. On the other hand, thepresent study included only participants aged 70 years or older.
Delirium was associated with falls during the hospital stay, which agrees with otherstudies both in hospitals and in residential care facilities (42, 45, 48). Associations betweendelirium and males, and delirium and dementia have also been reported earlier (16, 52, 77,
78). There was a high prevalence of delirium (75%) in the present study compared with otherstudies with a prevalence of 28-60% (52-54, 75, 76). Nearly all fallers were delirious at somepoint during their hospital stay, and almost half the falls occurred during a day when it wasrecorded that the participants were delirious. Studies with a lower incidence of delirium have
often excluded people with dementia, or with signs of cognitive impairment or delirium onadmission (54, 76, 148) which inevitably results in a lower incidence of delirium. Deliriumduring hospitalization has also been reported to be associated with poor functional recovery
following a hip fracture (54), and new injuries, along with other complications, which maycontribute to the poor recovery.
Fewer participants fell during the first week despite the largest proportions beingdelirious during the first postoperative week. This is most likely because they are less mobile
and more strictly supervised immediately following surgery.As in other studies (42, 45-47), associations were seen between in-patient falls and
previous falls before the fracture. Those associations did not remain significant, however, in
the multiple regression analyses.The association between falls and sleeping disturbances is not clear, but since night-
time falls resulting in hip fractures are associated with hypoxemia (33), it is possible that thisassociation could be due to sleep-apnea syndrome, which results in hypoxemia, an entity
found to be associated with delirium in old stroke patients (149). In the present study, anyassociation between sleep disturbances and falls could at least partly be mediated by theprevalent occurrence of delirium. A study among community-dwelling older people (150),
found that sleeping disturbances are risk factors for falls even when controlling for otherknown fall-risk factors, but they had not controlled for delirium. The fall peak between 20:00
58
and 22:00 may be associated with the use of sleeping medication administered before the fall.
However, the exact time at which the participants took their sleeping pills in the present studywas not registered.
Fall prevention
The multidisciplinary, multi-factorial fall-prevention program resulted in fewer participantssustaining new falls during rehabilitation, as well as fewer total numbers of falls and fewerfall-related injuries. To my knowledge this is the first fall-intervention study among hip-
fracture patients, despite the fact that this group has a high fall risk. In general there are fewfall-prevention studies in hospital settings. Two studies (98, 99) with positive outcomes inother patient groups and on sub-acute wards have recently been published. The first one (98)reduced falls at three sub-acute rehabilitation wards, but the differences were most obvious
after 45 days of observation. Thus the results were not comparable with those from thepresent study, which included both the acute and rehabilitation hospital stay. The other study(99) resulted in fewer fallers, falls and injuries on a geriatric ward but the differences
disappeared when the results were adjusted for observation time. Those studies employed amultidisciplinary approach in their fall intervention similar to that used in our study but, inaddition, we have also focused on treating in-patient complications associated with falls, suchas delirium and urinary tract infections. Vassallo et al (99) tried to prevent delirious patients
from falling by using bedrails, alarms and changing the furniture arrangements but noinformation is given about any prevention and treatment of the underlying causes of delirium.The studies above used fall-risk assessment tools to distinguish those with a high fall risk. A
criticism of fall-risk assessment tools is that few have been tested for validity and reliability ina new independent sample. When fall-prediction tools are used in different clinical settingsthe specificity decreases (151). Instead, to reduce the falls, we used a CGA rehabilitation andcare program including assessment of risk factors for falls, for everyone, and every week
there were team meetings including global ratings of the participants’ fall risk. Global ratingsby staff have been shown to be an equally accurate fall-risk assessment tool for identifyingfalls among elderly patients (152). In the present study the intervention program did not
recommend the use of physical restraints, such as bedrails, belts and chairs to prevent theparticipants rising but neither was it forbidden.
One may speculate that the successful reduction in the numbers of falls in the presentstudy could be a result of the active prevention, detection and treatment of postoperative
complications after surgery. During the period of hospitalization there were differencesbetween the two groups regarding some complications associated with falls among olderpeople in residential care facilities and in hospitals, such as delirium and urinary tract
infections. The reduction of postoperative delirium can probably explain much of thedifference between the groups regarding the number of falls and the number of participantswho fell. There are studies which have found that delirium is an important risk factor for falls(45-48). Those with dementia especially are at high risk of developing delirium when they are
59
treated for femoral neck fractures (53, 54) and in the present study these participants seemed
to have benefited most from the intervention program regarding prevention of postoperativefalls. The present study supports the findings from an earlier, non-randomised study thatfewer injurious falls occur when the incidence and duration of delirium was reduced by a
geriatric-anaesthesiologic intervention program to counteract postoperative delirium (77).The investigation into why the participants had fractured their hip and why they fell
may also have influenced the results, as well as investigation and rehabilitation concerningexternal fall-risk factors such as the use of walking aids, safe transfers, balance and mobility
training. It appears that teamwork and individual care planning alone have no effect on falls,as half the falls in the control group occurred in the geriatric control ward, a ward specializingin geriatric patients where teamwork, as well as individual care planning are used.
Improved performance in activities of daily living
Among previous randomised controlled trials there is only one study by Reid et al, usinggeriatric rehabilitative care, that has shown positive effects from a long-term perspectiveregarding functional recovery (119). Their treatment group showed a significant effectregarding the Katz ADL index, almost 50% had the same or a better P-ADL performance at
12 months compared to 20% in the control group. In the present study almost 60% in theintervention group attained a better or at least the same level of P-ADL performance as beforethe fracture at 12 months. The differences between the groups regarding regaining P-ADL
performance level according to the Katz ADL index were significant at discharge as well as atthe end of the present study period.
Other intervention studies have failed to achieve functional recovery (65, 112-114, 116,118). Some of them showed an improved recovery after three months but not at the end of the
study at twelve months (112, 116). It is difficult to compare studies from different countriesbecause of the differences in healthcare organization and rehabilitation routines and, inaddition, the differing aims, interventions, observation times and outcome measures make
comparisons even harder.Despite a shorter in-hospital stay after surgery there were no further readmission days
during the first postoperative year, which supports the claim that the intervention program hada positive effect. The number of out-patient rehabilitation periods after hospital discharge
were similar in the two groups, but a larger proportion of the participants from theintervention group were given rehabilitation at a centre that specialized in out-patientrehabilitating for the elderly. That, in combination with the postoperative follow up four-
month by a geriatric team, focusing on the detection and prevention of complications as wellas on initiating further rehabilitation if needed, might have influenced the increaseddifferences in ADL performance and mobility at the 12-month follow up.
60
Ethical considerations
When studying old people the ethical considerations are crucial, many have cognitiveimpairment or dementia and might have difficulties consider the matter of participation.However they also need good and evidence-based care and it is important not to exclude thisgroup from research studies. A large number of older people suffering from hip fractures have
a dementia diagnosis and excluding them would limit the external validity and may preventthe targeting of those who need the interventions most. In all the papers in this thesis, allparticipants or relatives agreed to participate after receiving oral and written information. The
participants could withdraw from the study at any time or not complete an assessment theydid not like or simply if they get too tired. The assessors were also very sensitive to theparticipants’ body language or signs of fatigue. All assessors had many years of experience torely on, assessing old people with and without cognitive impairments.
Another ethical consideration in the present study concerned detecting complications inthe control group during the data collection and not treating them properly. Participants in thecontrol group with significant medical problems were recommended to contact the
orthopaedic department or their care centre for treatment.The Ethics Committee of the Faculty of Medicine at Umeå University also approved the
studies.
Methodological considerations
In Paper I, the cross-sectional population-based design, including retrospective chart review,results in some limitations. The increased mortality among people with hip fractures is likelyto influence the findings. Another limitation is that some of the hip fractures might have been
missed because some of the participants might have been operated on in a hospital in anothermunicipality and did not remember the incident when they were interviewed. These areprobably very few, because almost all the participants in the study had lived in the
municipality since they were young and there is only one hospital in the area. On the otherhand, the prevalence figures are perhaps a little too high due to the fact that those whodeclined to participate were more likely to be younger and living in ordinary housingcompared to the study sample. One may also speculate that they were healthier and had not
suffered from an earlier hip fracture. Nor are the results representative of the whole 85+population, just of the three age groups studied.
Dementia and functional dependence are two of the strongest factors for
institutionalisation among old people (55, 145), which confirms the decision to adjust thestatistics for dementia, as well as for age (60, 61) and sex (62). However, one of the mostimportant predictors for a positive outcome seems to be an active pre-fracture lifestyle orlevel of function (60, 61, 153), which not have been included in the analyses due to the cross-
sectional study design. It is therefore, impossible to declare definitively that the findings ondependency in P-ADL and mobility can be explained exclusively by the hip fracture.
61
However, the conclusion from the present study is that the fracture seemed to be associated
with the participants’ situation in the long term.The quality of the study reported in Papers III-IV is high with sealed randomisation,
intention-to-treat analyses and the use of a control group. Another of the study’s strengths was
that of all eligible participants only about 5% refused to participate. There was also a smalland similar drop-out rate over time between the two groups. Only two people in each grouprefused the follow-up visits and one in the control group had moved to another part of thecountry. Regarding the fall outcome in Papers II-III, some falls could have been missed, but
presumably very few as nurses are obliged to document falls in the records and hip-fracturesurgery patients can hardly get up by themselves so soon after surgery, and any such fallswould certainly be noted.
Limitations in the randomised controlled study are that the assessors were not blindedconcerning group allocation, either during the hospitalization or during the home visits, andbias cannot, therefore, be excluded. On the other hand the nurse who performed assessmentsin the intervention group was from the orthopaedic unit and the nurse who did the same
assessments in the control group was from the geriatric unit. This should have reducedassessor bias. Another limitation is that the outpatient rehabilitation after discharge was not asstandardized as during the in-hospital stay. In the intervention program the aim was to have a
well-planned discharge, followed up with a phone call and a home visit. The participants werealso offered further rehabilitation after discharge but the intensity and quality of thatoutpatient rehabilitation is unknown. Only data on the numbers of occasions rehabilitationwas given are available.
A further disadvantage is that the present study has no figures for cost effectiveness.The fact that the intervention group had a mean value of 14 fewer in-hospital days during theone-year follow up suggests that the intervention program was perhaps less expensive than
the standard care provided for the control group, despite the intervention group having oneextra visit from a geriatric team at four months. The in-hospital stay was shorter for theintervention group, but the staff rates on the wards were similar and both in-hospital days andoutpatient treatment were also similar after discharge from hospital.
62
Clinical implications
Old people with femoral neck fractures, with or without cognitive impairments can, andshould, be offered a multidisciplinary, multi-factorial care and rehabilitation program. Thisprogram should include a CGA program with systematic assessment and treatment of fall-riskfactors as well as active prevention, detection and treatment of other postoperative
complications. The care and rehabilitation after hip fracture surgery should concentrate on theregaining of physical ability, as well as the prevention of complications. The most importantcomplications to be prevented appear to be delirium and sleeping disturbances because these
seem to be connected with an increased risk of falling during the hospital stay, as shown inPaper II. Delirium, by definition, always has an underlying cause that should be treated ifpossible. A comprehensive assessment of the patient is often needed to detect suchcomplications. Sleeping disturbances should also be regarded as a symptom and should be
assessed for underlying causes and not only be treated with sleeping pills. There are studiesthat have shown successful results in preventing and treating delirium during the in-hospitalstay with multi-factorial, multidisciplinary efforts and geriatric consultations (17, 75). In
addition, one intervention study resulted in fewer in-patient injuries when delirium wasreduced after a femoral neck fracture surgery (77).
In the community and residential care facilities, interdisciplinary and multi-factorialinterventions have shown positive effects on the reduction of the number of falls, fallers, and
fall-related injuries (83, 87). Among those with cognitive decline or dementia there is noevidence that such strategies prevent falls (90, 91) but Paper III allowed the conclusion, thatat least during the in-hospital stay, this group of patients could benefit from such strategies.
The earlier recommendations concerning fall-prevention programs, reported in Guidelines forthe Prevention of Falls in Older Persons (51), should be completed with a systematicassessment and treatment of known complications among old people, such as delirium,urinary tract infections and sleeping disturbances, in accordance with the findings in this
thesis.The reduced number of falls and injuries also probably contributed to the shorter
hospitalization period seen in the intervention group. The program is easily applicable both in
acute, postoperative care as well as in the post-acute, rehabilitation settings and apart from thefour days of staff education there seemed to be no increased costs.
Regarding functional benefits, it has been reported earlier that the intervention programdescribed in Papers III-IV was effective during the in-hospital stay, regarding the prevention
of postoperative complications such as delirium, decubital ulcers, urinary tract infections,nutritional problems and sleep disturbances (154). These results together with those fromPaper IV, the Cochrane review (108) and the meta-analysis (103) support the view that a
CGA, management and rehabilitation approach is important for this group of old, people andmight be advantageous for both the old, hip-fracture patient and for the healthcare system. InBrocklehurst’s textbook of Geriatric Medicine and Gerontology (102) the author’s mentionthat the greatest benefits of CGA programs have been found when they are used on frail,
63
elderly people. Those who suffer a hip fracture must be classified as frail since many are
demented, live in institutions and suffer from many other diseases apart from the fracture.
Implications for future research
This thesis found that there are benefits from using a CGA management and rehabilitation
program including active prevention, detection and treatment of postoperative complicationson fall prevention and increasing ADL performance and mobility but more research isneeded. More fall-prevention studies are needed in hospital settings, since many fell during
hospitalization and a large proportion of hip fractures occurs in hospitals. Among those 199participants included in Papers III-IV there were in total 38 participants with 78 falls. Inaddition, 21 eligible cases had fractured their hip in the hospital and were not included in thepresent study for that reason. Far too many patients seem to fall while in hospital. Further
research needs to evaluate which intervention is most effective for this group of old people.The intervention program seemed to be most effective for those with cognitive impairments.To my knowledge, this is the first effective fall-prevention trial carried out on this group of
patients. However, more research in larger samples of people with dementia is needed toconfirm these findings. Falls and fractures after discharge should also be examined so thateffective fall-prevention programs can be initiated.
There were also benefits in functional recovery in conjunction with the intervention
program but there were still many who never regained their pre-fracture function. Futureresearch should include an even more comprehensive and intensive intervention to optimizethe situation for those who suffer a hip fracture; the outpatient rehabilitation especially should
be more individualized and performed at higher intensity. It has been reported that extendedoutpatient rehabilitation, including progressive resistance training, is effective in improvingphysical function among elderly community-dwelling, hip-fracture patients (155). Weight-bearing exercises at home have also been reported to be effective among this group of
patients. After one month of daily training, leg strength and mobility improved and those inthe intervention group rated their fall risk as lower than before (156). I believe that thestrength training should be performed at high intensity since this has more effect on old
people than low-intensity training (157). In Paper IV a larger proportion of the interventiongroup was given rehabilitation at a centre that specialized in rehabilitating the elderly and thatmight have influenced the results regarding functional recovery. It would be interesting toevaluate a more comprehensive and multidisciplinary outpatient program to improve the
rehabilitation outcome. This program should include individual care planning and intensiverehabilitation by a team with knowledge about old, frail people, as well as active prevention,detection and treatment of well-known complications among old people. I also believe that
one must be flexible about where the outpatient rehabilitation is delivered, not everyone hasthe possibility or the energy to travel to a rehabilitation centre thus home rehabilitation mustbe an alternative. It would therefore be interesting to develop and evaluate a randomised high-intensity home rehabilitation program.
64
GENERAL CONCLUSIONS
• A history of hip fracture is common among the very old and seems to be connected with
poorer P-ADL performance and mobility. Many of those who had had hip fractures hadpermanently changed their housing situation and use of walking aids after the fractureincident.
• Many falls and fall-related injuries occurred during hospitalization. The high incidencewas independently associated with postoperative delirium, male sex, and sleepingdisturbances. Falls and injuries cause suffering and prolong hospitalization. Interventionprograms, including prevention and treatment of complications such as delirium and
sleeping disturbances, as well as better supervision of male patients, are possible fallprevention strategies.
• A team applying comprehensive geriatric assessment and rehabilitation, including
prevention, detection and treatment of fall-risk factors, can successfully prevent in-patientfalls and fall-related injuries after treatment for hip fractures, even in old people withdementia.
• A multidisciplinary postoperative intervention program including CGA, and
rehabilitation, and a four-month follow up, resulted in a larger proportion of participantsregaining independence in P-ADL performance as well as a larger proportion walkingindoors without walking aids one-year after a femoral neck fracture.
ACKNOWLEDGEMENTS
This work was carried out at the Department of Community Medicine and Rehabilitation,
Geriatric Medicine and Physiotherapy, Umeå University, Umeå. I wish to thank everyonewho helped me and provided the opportunities that made it possible for me to complete thework and who have in various ways contributed to this thesis.
I want to direct my sincere and warmest gratitude particularly to:
Yngve Gustafson, my supervisor, for showing great interest in my work, for excellentguidance and unfailing support, for generously providing me with the opportunity to carry
through my studies and for always strengthening my confidence in my work;
Lars Nyberg, my co-supervisor for valuable advice, collaboration and co-authorship, for
introducing me to research and teaching me the importance of statistics in research;
Gösta Bucht and Helen Abrahamsson, for providing excellent working conditions;
Birgitta Olofsson, my friend, for all years you have supported me in my research and foralways being there when I needed help and advice, for always increasing my confidence, forall the discussions about life in good times and bad, for being a friend;
Maria Lundström and Eva Elinge, co-authors and co-data collectors, for your friendship, andsupport through all the years;
Bengt Borssén, Olle Svensson, and Undis Englund, co-authors, for valuable advice, andcollaboration and Petra von Heideken Wågert, co-author and friend, for your advice andsupport, good luck!
All the participants and staff at the orthopaedic and geriatric departments at the UniversityHospital in Umeå for their cooperation and participation, and the staff at all facilities I have
visited, who contributed to the data collection;
Patricia Shrimpton and Jamie Guerra, for careful language revision;
Larry Fredriksson for computer support;
My colleagues and friends at the Department of Geriatric Medicine, for valuable seminars,
exciting journeys and for all fun we had. Karin Gladh, for practical assistance and for creatinga pleasant working atmosphere;
65
66
My colleagues at the Department of Physiotherapy, particularly Staffan Eriksson, Jane
Jensen, Lillemor Lundin-Olsson, and Gunnevi Sundelin for all their valuable advice andsupport;
My colleagues at the Geriatric Centre in Umeå, for all fun we had, all the discussions and thesupport I always have felt you gave me; Åsa Karlsson, for all the years I had the opportunityto work with you; Anna-Lisa Lindkvist and Sofie Strandberg, former and current chief-physiotherapists, for always supporting me and all the other physiotherapists working in
research and development work;
My workmates at the Geriatric Centre, Ward 4, for all fun and co-operation. Börje
Hermansson, for all years I had the opportunity to work with you, your support andknowledge have been invaluable to me; Agneta Möller and Anita Persson, former and currentward sisters for all your support and for making the work on the ward enjoyable, Monika
Berggren who took part in the data collection in Paper IV, Anna Ovesson, Ann-Marie
Estensen, Anki Ivemon who took part in the development of the intervention program;
Tommy and Mona, for your friendship, for all the exciting journeys involving skiing, football,
and foreign travel; there is so much fun to be had outside the work, now I hope there will bemore time for football, skiing, wonderful dinners and talks;
My family and friends, for friendship and support;
My parents, Harriet and Håkan, for always giving me confidence and support;
Last but foremost, Lisette my other half, there are no words to describe my feelings for youand what you mean to me, I love you; Linn, Nellie and Emil, my wonderful children forreminding me about what is important in life and for challenging me about lots of things; youmake me proud every day, I love you;
This work was supported by grants from the ”Vårdal Foundation”, the Joint Committee of theNorthern Health Region of Sweden (Visare Norr), the JC Kempe Memorial Foundation, theDementia Fund, the Foundation of the Medical Faculty, the Borgerskapet of Umeå Research
Foundation, the Erik and Anne-Marie Detlof’s Foundation, Gun and Bertil Stohne’sFoundation, University of Umeå and the County Council of Västerbotten (“Dagmar”, “FoU” ,and “Äldre Centrum Västerbotten”) and the Swedish Research Council, Grants K2001-27VP-
14165-01A, K2002-27VP-14165-02B, K2002-27VX-14172-02B, K2005-27VX-15357-01A.
67
REFERENCES
1. The future population of Sweden. Projections for the years 2005-2050. Örebro:Statistics Sweden, 2005.
2. von Heideken Wågert P, Gustavsson J, Lundin-Olsson L, Kallin K, Nygren B,Lundman B, Norberg A, Gustafson Y. Health status in the oldest old: age and sex
differences in the Umeå 85+ study. Aging Clin Exp Res 2006;18:116-126.3. Thorngren KG. Fractures in older persons. Disabil Rehabil 1994;16(3):119-26.4. Thorngren KG, Hommel A, Norrman PO, Thorngren J, Wingstrand H. Epidemiology
of femoral neck fractures. Injury 2002;33 Suppl 3:C1-7.5. Zethraeus N, Strömberg L, Jönsson B, Svensson O, Öhlen G. The cost of a hip
fracture. Estimates for 1,709 patients in Sweden. Acta Orthop Scand 1997;68(1):13-7.
6. Johnell O, Kanis JA. An estimate of the worldwide prevalence, mortality anddisability associated with hip fracture. Osteoporos Int 2004;15(11):897-902.
7. Melton LJ, 3rd. Hip fractures: a worldwide problem today and tomorrow. Bone1993;14 Suppl 1:S1-8.
8. Gullberg B, Johnell O, Kanis JA. World-wide projections for hip fracture. OsteoporosInt 1997;7(5):407-13.
9. Essential Orthopaedics and Trauma. David J. Dandy. Churchill Livingstone Inc.
Edinburgh London Melbourne and New York 1989.10. Garden RS. Low-angle fixation in fractures of the femoral neck. J Bone Joint Surg.
1961;43:647-63.11. Masson M, Parker MJ, Fleischer S. Internal fixation versus arthroplasty for
intracapsular proximal femoral fractures in adults. Cochrane Database Syst Rev2003(2):CD001708.
12. Espino DV, Palmer RF, Miles TP, Mouton CP, Wood RC, Bayne NS, et al.
Prevalence, incidence, and risk factors associated with hip fractures in community-dwelling older Mexican Americans: results of the Hispanic EPESE study. EstablishPopulation for the Epidemiologic Study for the Elderly. J Am Geriatr Soc2000;48(10):1252-60.
13. Fransen M, Woodward M, Norton R, Robinson E, Butler M, Campbell AJ. Excessmortality or institutionalization after hip fracture: men are at greater risk than women.J Am Geriatr Soc 2002;50(4):685-90.
14. Norton R, Butler M, Robinson E, Lee-Joe T, Campbell AJ. Declines in physicalfunctioning attributable to hip fracture among older people: a follow-up study of case-control participants. Disabil Rehabil 2000;22(8):345-51.
15. Leibson CL, Tosteson AN, Gabriel SE, Ransom JE, Melton LJ. Mortality, disability,
and nursing home use for persons with and without hip fracture: a population-basedstudy. J Am Geriatr Soc 2002;50(10):1644-50.
16. Edlund A, Lundström M, Brännström B, Bucht G, Gustafson Y. Delirium before and
after operation for femoral neck fracture. J Am Geriatr Soc 2001;49(10):1335-40.
68
17. Lundström M, Edlund A, Lundström G, Gustafson Y. Reorganization of nursing and
medical care to reduce the incidence of postoperative delirium and improverehabilitation outcome in elderly patients treated for femoral neck fractures. Scand JCaring Sci 1998(13):193-200.
18. Dargent-Molina P, Favier F, Grandjean H, Baudoin C, Schott AM, Hausherr E, et al.Fall-related factors and risk of hip fracture: the EPIDOS prospective study. Lancet1996;348(9021):145-9.
19. Fox KM, Cummings SR, Williams E, Stone K. Femoral neck and intertrochanteric
fractures have different risk factors: a prospective study. Osteoporos Int2000;11(12):1018-23.
20. Taylor BC, Schreiner PJ, Stone KL, Fink HA, Cummings SR, Nevitt MC, et al. Long-
term prediction of incident hip fracture risk in elderly white women: study ofosteoporotic fractures. J Am Geriatr Soc 2004;52(9):1479-86.
21. De Laet CE, van Hout BA, Burger H, Hofman A, Pols HA. Bone density and risk ofhip fracture in men and women: cross sectional analysis. BMJ 1997;315(7102):221-5.
22. Jarnlo GB, Thorngren KG. Background factors to hip fractures. Clin Orthop1993(287):41-9.
23. Cummings SR, Nevitt MC, Browner WS, Stone K, Fox KM, Ensrud KE, et al. Risk
factors for hip fracture in white women. Study of Osteoporotic Fractures ResearchGroup. N Engl J Med 1995;332(12):767-73.
24. Butler M, Norton R, Lee-Joe T, Cheng A, Campbell AJ. The risks of hip fracture inolder people from private homes and institutions. Age Ageing 1996;25(5):381-5.
25. Norton R, Campbell AJ, Reid IR, Butler M, Currie R, Robinson E, et al. Residentialstatus and risk of hip fracture. Age Ageing 1999;28(2):135-9.
26. Mautalen CA, Vega EM, Einhorn TA. Are the etiologies of cervical and trochanteric
hip fractures different? Bone 1996;18(3 Suppl):133S-137S.27. Koval KJ, Aharonoff GB, Rokito AS, Lyon T, Zuckerman JD. Patients with femoral
neck and intertrochanteric fractures. Are they the same? Clin Orthop 1996(330):166-72.
28. Michaelsson K, Weiderpass E, Farahmand BY, Baron JA, Persson PG, Ziden L, et al.Differences in risk factor patterns between cervical and trochanteric hip fractures.Swedish Hip Fracture Study Group. Osteoporos Int 1999;10(6):487-94.
29. Osteoporosis prevention, diagnosis, and therapy. JAMA 2001;285(6):785-95.30. Cummings SR, Melton LJ. Epidemiology and outcomes of osteoporotic fractures.
Lancet 2002;359(9319):1761-7.31. SBU. The Swedish Council on Technology Assessment in Health Care 1995.
Measuring bone mass.32. Fuller GF. Falls in the elderly. Am Fam Physician 2000;61(7):2159-68, 2173-4.33. 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):156-60.
69
34. Luukinen H, Koski K, Hiltunen L, Kivela SL. Incidence rate of falls in an aged
population in northern Finland. J Clin Epidemiol 1994;47(8):843-50.35. Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly persons
living in the community. N Engl J Med 1988;319(26):1701-7.
36. Kallin K, Lundin-Olsson L, Jensen J, Nyberg L, Gustafson Y. Predisposing andprecipitating factors for falls among older people in residential care. Public Health2002;116(5):263-71.
37. Tinetti ME. Factors associated with serious injury during falls by ambulatory nursing
home residents. J Am Geriatr Soc 1987;35(7):644-8.38. 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.
39. O'Loughlin JL, Robitaille Y, Boivin JF, Suissa S. Incidence of and risk factors forfalls and injurious falls among the community-dwelling elderly. Am J Epidemiol1993;137(3):342-54.
40. Sadigh S, Reimers A, Andersson R, Laflamme L. Falls and fall-related injuries among
the elderly: a survey of residential-care facilities in a Swedish municipality. JCommunity Health 2004;29(2):129-40.
41. Foss NB, Palm H, Kehlet H. In-hospital hip fractures: prevalence, risk factors and
outcome. Age Ageing 2005;34(6):642-5.42. Vassallo M, Sharma JC, Briggs RS, Allen SC. Characteristics of early fallers on
elderly patient rehabilitation wards. Age Ageing 2003;32(3):338-42.43. van Doorn C, Gruber-Baldini AL, Zimmerman S, Hebel JR, Port CL, Baumgarten M,
et al. Dementia as a risk factor for falls and fall injuries among nursing homeresidents. J Am Geriatr Soc 2003;51(9):1213-8.
44. Lawlor DA, Patel R, Ebrahim S. Association between falls in elderly women and
chronic diseases and drug use: cross sectional study. BMJ 2003;327(7417):712-7.45. Gluck T, Wientjes HJ, Rai GS. An evaluation of risk factors for in-patient falls in
acute and rehabilitation elderly care wards. Gerontology 1996;42(2):104-7.46. Oliver D, Daly F, Martin FC, McMurdo ME. Risk factors and risk assessment tools
for falls in hospital in-patients: a systematic review. Age Ageing 2004;33(2):122-30.47. Salgado RI, Lord SR, Ehrlich F, Janji N, Rahman A. Predictors of falling in elderly
hospital patients. Arch Gerontol Geriatr 2004;38(3):213-9.
48. Kallin K, Jensen J, Lundin-Olsson L, Nyberg L, Gustafson Y. Why the elderly fall inresidential care facilities, and suggested remedies. J Fam Pract 2004;53(1):41-52.
49. Maurer MS, Burcham J, Cheng H. Diabetes mellitus is associated with an increasedrisk of falls in elderly residents of a long-term care facility. J Gerontol A Biol Sci Med
Sci 2005;60(9):1157-62.50. Halfon P, Eggli Y, Van Melle G, Vagnair A. Risk of falls for hospitalized patients: a
predictive model based on routinely available data. J Clin Epidemiol
2001;54(12):1258-66.
70
51. Guideline for the Prevention of Falls in Older Persons. American Geriatrics Society,
British Geriatrics Society, and American Academy of Orthopaedic Surgeons Panel onFalls Prevention. J Am Geriatr Soc 2001;49(5):664-72.
52. Edlund A, Lundström M, Lundström G, Hedqvist B, Gustafson Y. Clinical profile of
delirium in patients treated for femoral neck fractures. Dement Geriatr Cogn Disord1999;10(5):325-9.
53. Gustafson Y, Berggren D, Brännström B, Bucht G, Norberg A, Hansson LI, et al.Acute confusional states in elderly patients treated for femoral neck fracture. J Am
Geriatr Soc 1988;36(6):525-30.54. Marcantonio ER, Flacker JM, Michaels M, Resnick NM. Delirium is independently
associated with poor functional recovery after hip fracture. J Am Geriatr Soc
2000;48(6):618-624.55. Maggio D, Ubaldi E, Simonelli G, Cenci S, Pedone C, Cherubini A. Hip fracture in
nursing homes: an Italian study on prevalence, latency, risk factors, and impact onmobility. Calcif Tissue Int 2001;68(6):337-41.
56. Hochberg MC, Williamson J, Skinner EA, Guralnik J, Kasper JD, Fried LP. Theprevalence and impact of self-reported hip fracture in elderly community-dwellingwomen: the Women's Health and Aging Study. Osteoporos Int 1998;8(4):385-9.
57. Kirke PN, Sutton M, Burke H, Daly L. Outcome of hip fracture in older Irish women:a 2-year follow-up of subjects in a case-control study. Injury 2002;33(5):387-91.
58. Magaziner J, Hawkes W, Hebel JR, Zimmerman SI, Fox KM, Dolan M, et al.Recovery from hip fracture in eight areas of function. J Gerontol A Biol Sci Med Sci
2000;55(9):M498-507.59. Cree M, Soskolne CL, Belseck E, Hornig J, McElhaney JE, Brant R, et al. Mortality
and institutionalization following hip fracture. J Am Geriatr Soc 2000;48(3):283-8.
60. Parker MJ, Palmer CR. Prediction of rehabilitation after hip fracture. Age Ageing1995;24(2):96-8.
61. Thorngren KG, Ceder L, Svensson K. Predicting results of rehabilitation after hipfracture. A ten-year follow-up study. Clin Orthop 1993(287):76-81.
62. Lieberman D, Fried V, Castel H, Weitzmann S, Lowenthal MN, Galinsky D. Factorsrelated to successful rehabilitation after hip fracture: a case-control study. DisabilRehabil 1996;18(5):224-30.
63. Forsen L, Sogaard AJ, Meyer HE, Edna T, Kopjar B. Survival after hip fracture: short-and long-term excess mortality according to age and gender. Osteoporos Int1999;10(1):73-8.
64. Huusko TM, Karppi P, Avikainen V, Kautiainen H, Sulkava R. Randomised,
clinically controlled trial of intensive geriatric rehabilitation in patients with hipfracture: subgroup analysis of patients with dementia. BMJ 2000;321(7269):1107-11.
65. Jette AM, Harris BA, Cleary PD, Campion EW. Functional recovery after hip fracture.
Arch Phys Med Rehabil 1987;68(10):735-40.
71
66. Johnell O, Kanis JA, Oden A, Sernbo I, Redlund-Johnell I, Petterson C, et al.
Mortality after osteoporotic fractures. Osteoporos Int 2004;15(1):38-42.67. 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.
68. Katelaris AG, Cumming RG. Health status before and mortality after hip fracture. AmJ Public Health 1996;86(4):557-60.
69. Trombetti A, Herrmann F, Hoffmeyer P, Schurch MA, Bonjour JP, Rizzoli R.Survival and potential years of life lost after hip fracture in men and age-matched
women. Osteoporos Int 2002;13(9):731-7.70. Empana JP, Dargent-Molina P, Breart G. Effect of hip fracture on mortality in elderly
women: the EPIDOS prospective study. J Am Geriatr Soc 2004;52(5):685-90.
71. Shumway-Cook A, Ciol MA, Gruber W, Robinson C. Incidence of and risk factors forfalls following hip fracture in community-dwelling older adults. Phys Ther2005;85(7):648-55.
72. Johnell O, Kanis JA, Oden A, Sernbo I, Redlund-Johnell I, Petterson C, et al. Fracture
risk following an osteoporotic fracture. Osteoporos Int 2004;15(3):175-9.73. Colon-Emeric C, Kuchibhatla M, Pieper C, Hawkes W, Fredman L, Magaziner J, et
al. The contribution of hip fracture to risk of subsequent fractures: data from two
longitudinal studies. Osteoporos Int 2003;14(11):879-83.74. 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 atrisk of a second one? Z Gerontol Geriatr 2003;36(1):16-22.
75. Marcantonio ER, Flacker JM, Wright RJ, Resnick NM. Reducing delirium after hipfracture: a randomized trial. J Am Geriatr Soc 2001;49(5):516-22.
76. Williams MA, Campbell EB, Raynor WJ, Jr., Musholt MA, Mlynarczyk SM, Crane
LF. Predictors of acute confusional states in hospitalized elderly patients. Res NursHealth 1985;8(1):31-40.
77. Gustafson Y, Brännström B, Berggren D, Ragnarsson JI, Sigaard J, Bucht G, et al. Ageriatric-anesthesiologic program to reduce acute confusional states in elderly patients
treated for femoral neck fractures. J Am Geriatr Soc 1991;39(7):655-62.78. Edelstein DM, Aharonoff GB, Karp A, Capla EL, Zuckerman JD, Koval KJ. Effect of
postoperative delirium on outcome after hip fracture. Clin Orthop 2004(422):195-200.
79. Holmes JD, House AO. Psychiatric illness in hip fracture. Age Ageing2000;29(6):537-46.
80. van Balen R, Steyerberg EW, Polder JJ, Ribbers TL, Habbema JD, Cools HJ. Hipfracture in elderly patients: outcomes for function, quality of life, and type of
residence. Clin Orthop 2001(390):232-43.81. The prevention of falls in later life. A report of the Kellogg International Work Group
on the Prevention of Falls by the Elderly. Dan Med Bull 1987;34 Suppl 4:1-24.
72
82. Lamb SE, Jorstad-Stein EC, Hauer K, Becker C. Development of a common outcome
data set for fall injury prevention trials: the Prevention of Falls Network Europeconsensus. J Am Geriatr Soc 2005;53(9):1618-22.
83. 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 InternMed 2002;136(10):733-41.
84. The Abbreviated Injury Scale 1990 Revision. American Association for theAdvancement of Automotive Medicine, Des Plaines, IL.
85. Chang JT, Morton SC, Rubenstein LZ, Mojica WA, Maglione M, Suttorp MJ, et al.Interventions for the prevention of falls in older adults: systematic review and meta-analysis of randomised clinical trials. BMJ 2004;328(7441):680.
86. Davison J, Bond J, Dawson P, Steen IN, Kenny RA. Patients with recurrent fallsattending Accident & Emergency benefit from multifactorial intervention--arandomised controlled trial. Age Ageing 2005;34(2):162-8.
87. Gillespie LD, Gillespie WJ, Robertson MC, Lamb SE, Cumming RG, Rowe BH.
Interventions for preventing falls in elderly people. Cochrane Database Syst Rev2003(4):CD000340.
88. Tinetti ME, Baker DI, McAvay G, Claus EB, Garrett P, Gottschalk M, et al. A
multifactorial intervention to reduce the risk of falling among elderly people living inthe community. N Engl J Med 1994;331(13):821-7.
89. Becker C, Kron M, Lindemann U, Sturm E, Eichner B, Walter-Jung B, et al.Effectiveness of a multifaceted intervention on falls in nursing home residents. J Am
Geriatr Soc 2003;51(3):306-13.90. Shaw FE, Bond J, Richardson DA, Dawson P, Steen IN, McKeith IG, et al.
Multifactorial intervention after a fall in older people with cognitive impairment and
dementia presenting to the accident and emergency department: randomised controlledtrial. BMJ 2003;326(7380):73.
91. Jensen J, Nyberg L, Gustafson Y, Lundin-Olsson L. Fall and injury prevention inresidential care--effects in residents with higher and lower levels of cognition. J Am
Geriatr Soc 2003;51(5):627-35.92. Campbell AJ, Robertson MC, La Grow SJ, Kerse NM, Sanderson GF, Jacobs RJ, et al.
Randomised controlled trial of prevention of falls in people aged > or =75 with severe
visual impairment: the VIP trial. BMJ 2005;331(7520):817.93. Bischoff-Ferrari HA, Dawson-Hughes B, Willett WC, Staehelin HB, Bazemore MG,
Zee RY, et al. Effect of Vitamin D on falls: a meta-analysis. JAMA2004;291(16):1999-2006.
94. Flicker L, MacInnis RJ, Stein MS, Scherer SC, Mead KE, Nowson CA, et al. Shouldolder people in residential care receive vitamin D to prevent falls? Results of arandomized trial. J Am Geriatr Soc 2005;53(11):1881-8.
95. Donald IP, Pitt K, Armstrong E, Shuttleworth H. Preventing falls on an elderly carerehabilitation ward. Clin Rehabil 2000;14(2):178-85.
73
96. Mayo NE, Gloutney L, Levy AR. A randomized trial of identification bracelets to
prevent falls among patients in a rehabilitation hospital. Arch Phys Med Rehabil1994;75(12):1302-8.
97. Tideiksaar R, Feiner CF, Maby J. Falls prevention: the efficacy of a bed alarm system
in an acute-care setting. Mt Sinai J Med 1993;60(6):522-7.98. Haines TP, Bennell KL, Osborne RH, Hill KD. Effectiveness of targeted falls
prevention programme in subacute hospital setting: randomised controlled trial. BMJ2004;328(7441):676.
99. Vassallo M, Vignaraja R, Sharma JC, Hallam H, Binns K, Briggs R, et al. The effectof changing practice on fall prevention in a rehabilitative hospital: the hospital injuryprevention study. J Am Geriatr Soc 2004;52(3):335-9.
100. Healey F, Monro A, Cockram A, Adams V, Heseltine D. Using targeted risk factorreduction to prevent falls in older in-patients: a randomised controlled trial. AgeAgeing 2004;33(4):390-5.
101. Sletvold O, Tilvis R, Jonsson A, Schroll M, Snaedal J, Engedal K, et al. Geriatric
work-up in the Nordic countries. The Nordic approach to comprehensive geriatricassessment. Dan Med Bull 1996;43(4):350-9.
102. Tallis RC, Fillit HM. Brocklehurst's textbook of Geriatric Medicine and Gerontology,
6th edition. London: Churchill Livingstone 2003.103. Stuck AE, Siu AL, Wieland GD, Adams J, Rubenstein LZ. Comprehensive geriatric
assessment: a meta-analysis of controlled trials. Lancet 1993;342(8878):1032-6.104. Jonsson A, Gustafson Y, Schroll M, Hansen FR, Saarela M, Nygaard H, et al.
Geriatric rehabilitation as an integral part of geriatric medicine in the Nordiccountries. Dan Med Bull 2003;50(4):439-45.
105. 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.
106. Cameron ID. Coordinated multidisciplinary rehabilitation after hip fracture. DisabilRehabil 2005;27(18-19):1081-90.
107. Cameron I, Crotty M, Currie C, Finnegan T, Gillespie L, Gillespie W, et al. Geriatricrehabilitation following fractures in older people: a systematic review. Health TechnolAssess 2000;4(2):i-iv, 1-111.
108. Cameron ID, Handoll HH, Finnegan TP, Madhok R, Langhorne P. Co-ordinatedmultidisciplinary approaches for inpatient rehabilitation of older patients withproximal femoral fractures. Cochrane Database Syst Rev 2001(3):CD000106.
109. Crotty M, Whitehead CH, Gray S, Finucane PM. Early discharge and home
rehabilitation after hip fracture achieves functional improvements: a randomizedcontrolled trial. Clin Rehabil 2002;16(4):406-13.
110. Kuisma R. A randomized, controlled comparison of home versus institutional
rehabilitation of patients with hip fracture. Clin Rehabil 2002;16(5):553-61.
74
111. Swanson CE, Day GA, Yelland CE, Broome JR, Massey L, Richardson HR, et al. The
management of elderly patients with femoral fractures. A randomised controlled trialof early intervention versus standard care. Med J Aust 1998;169(10):515-8.
112. 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):1476-82.
113. Cameron ID, Lyle DM, Quine S. Accelerated rehabilitation after proximal femoralfracture: a randomized controlled trial. Disabil Rehabil 1993;15(1):29-34.
114. Galvard H, Samuelsson SM. Orthopedic or geriatric rehabilitation of hip fracturepatients: a prospective, randomized, clinically controlled study in Malmo, Sweden.Aging (Milano) 1995;7(1):11-6.
115. Gilchrist WJ, Newman RJ, Hamblen DL, Williams BO. Prospective randomised studyof an orthopaedic geriatric inpatient service. BMJ 1988;297(6656):1116-8.
116. Huusko TM, Karppi P, Avikainen V, Kautiainen H, Sulkava R. Intensive geriatricrehabilitation of hip fracture patients: a randomized, controlled trial. Acta Orthop
Scand 2002;73(4):425-31.117. Kennie DC, Reid J, Richardson IR, Kiamari AA, Kelt C. Effectiveness of geriatric
rehabilitative care after fractures of the proximal femur in elderly women: a
randomised clinical trial. BMJ 1988;297(6656):1083-6.118. Naglie G, Tansey C, Kirkland JL, Ogilvie-Harris DJ, Detsky AS, Etchells E, et al.
Interdisciplinary inpatient care for elderly people with hip fracture: a randomizedcontrolled trial. CMAJ 2002;167(1):25-32.
119. Reid J, Kennie DC. Geriatric rehabilitative care after fractures of the proximal femur:one year follow up of a randomised clinical trial. BMJ 1989;299(6690):25-6.
120. Fordham R, Thompson R, Holmes J, Hodkinson C. A cost-benefit study of geriatric-
orthopaedic management of patients with fractured neck of femur. Discussion paper14. York: Centre for Health Economics, University of York, 1986.
121. Sanchez Ferrin P, Manas Magana M, Asuncion Marques J, Dojoz Preciado M,Quintana Riera S, Gonzalez Ortega F. Geriatric assessment of elderly patients with
proximal fractures of the femur. Revista Espanola de Geriatria y Gerontologia1999;34(2):65-71.
122. Cameron ID, Lyle DM, Quine S. Cost effectiveness of accelerated rehabilitation after
proximal femoral fracture. J Clin Epidemiol 1994;47(11):1307-13.123. Folstein MF, Folstein SE, McHugh PR. "Mini-mental state". A practical method for
grading the cognitive state of patients for the clinician. J Psychiatr Res1975;12(3):189-98.
124. Jensen E, Dehlin O, Gustafson L. A comparison between three psychogeriatric ratingscales. Int J Geriatr Psychiatry 1993;8:215-229.
125. Tombaugh TN, McIntyre NJ. The mini-mental state examination: a comprehensive
review. J Am Geriatr Soc 1992;40(9):922-35.
75
126. American Psychiatric Association. Diagnostic and Statistical Manual of Mental
Disorders, 4th Ed. (DSM-IV). Washington, DC, APA 1994.127. Eriksson M, Samuelsson E, Gustafson Y, Åberg T, Engström KG. Delirium after
coronary bypass surgery evaluated by the organic brain syndrome protocol. Scand
Cardiovasc J 2002;36(4):250-5.128. Sheikh J, Yesavage J. Geriatric Depression Scale (GDS): recent evidence and
development of a shorter version. Clin Gerontol 1986;5(1/2):165-172.129. Montgomery SA, Asberg M. A new depression scale designed to be sensitive to
change. Br J Psychiatry 1979;134:382-9.130. Guigoz Y, Vellas B, Garry P. Mini Nutritional Assessment: A practical assessment
tool for grading the nutritional state of elderly patients. Facts Res Gerontol
1997;4(suppl 2):15-59.131. Wade DT. Measurement in neurological rehabilitation. Curr Opin Neurol Neurosurg
1992;5(5):682-6.132. 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 PsychosocialFunction. JAMA 1963;185:914-9.
133. Sonn U. Longitudinal studies of dependence in daily life activities among elderly
persons. Scand J Rehab Med 1996;34(Suppl):1-35.134. Seaby L TG. Reliability of a physiotherapy functional assessement used in a
rehabilitation setting. Physiotherapy Canada 1989;41(5):264-271.135. Hasselgren-Nyberg L, Omgren M, Nyberg L, Gustafson Y. S-COVS. Den svenska
versionen av Physiotherapy Clinical Outcome Variables. (In, Swedish, The Swedishversion of Physiotherapy Clinical Outcome Variables) Nordisk Fysioterapi1997(1):109-113.
136. Jensen J, Lundin-Olsson L, Lindmark B, Nillbrand A, Y G. Bergs balansskala:Prövning av interbedömarreliabilitet. (In Swedish, The Berg Balance Scale; inter-raterreliability testing) Nordisk Fysioterapi 1998;1:1-6.
137. Lundin-Olsson L, Jensen J, Waling K. Den svenska versionen av The Balance Scale.
Sjukgymnasten: (In Swedish, The Swedish version of The Berg Balance Scale)Vetenskapligt supplement 1996(1):16-9.
138. 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:S7-11.139. The Swedish Code of Statutes, Patient Records Act, 1985;562.140. Berggren D, Gustafson Y, Eriksson B, Bucht G, Hansson LI, Reiz S, et al.
Postoperative confusion after anesthesia in elderly patients with femoral neck
fractures. Anesth Analg 1987;66(6):497-504.141. Brännström B, Gustafson Y, Norberg A, Winblad B. Problems of basic nursing care in
acutely confused and non-confused hip- fracture patients. Scand J Caring Sci
1989;3(1):27-34.
76
142. Dawson B, Trapp RG. Basic & clinical biostatistics (3rd ed). New York, NY, USA:
Appleton & Lange; 2001.143. Robertson MC, Campbell AJ, Herbison P. Statistical analysis of efficacy in falls
prevention trials. J Gerontol A Biol Sci Med Sci 2005;60(4):530-4.
144. Österlind PO. Medical and social conditions in the elderly gender and age differences.The Umeå longitudinal study. [doctoral thesis]. Umeå, Sweden: University of Umeå;1993.
145. Agüero-Torres H, von Strauss E, Viitanen M, Winblad B, Fratiglioni L.
Institutionalization in the elderly: The role of chronic diseases and dementia. Cross-sectional and longitudinal data from a population-based study. Journal of ClinicalEpidemiology 2001;54:795-801.
146. Willig R, Keinanen-Kiukaaniemi S, Jalovaara P. Mortality and quality of life aftertrochanteric hip fracture. Public Health 2001;115(5):323-7.
147. Nyberg L, Gustafson Y, Janson A, Sandman PO, Eriksson S. Incidence of falls inthree different types of geriatric care. A Swedish prospective study. Scand J Soc Med
1997;25(1):8-13.148. Duppils GS, Wikblad K. Acute confusional states in patients undergoing hip surgery.
a prospective observation study. Gerontology 2000;46(1):36-43.
149. Sandberg O, Franklin KA, Bucht G, Gustafson Y. Sleep apnea, delirium, depressedmood, cognition, and ADL ability after stroke. J Am Geriatr Soc 2001;49(4):391-7.
150. Brassington GS, King AC, Bliwise DL. Sleep problems as a risk factor for falls in asample of community-dwelling adults aged 64-99 years. J Am Geriatr Soc
2000;48(10):1234-40.151. Myers H. Hospital fall risk assessment tools: a critique of the literature. Int J Nurs
Pract 2003;9(4):223-35.
152. Lundin-Olsson L, Jensen J, Nyberg L, Gustafson Y. Predicting falls in residential careby a risk assessment tool, staff judgement, and history of falls. Aging Clin Exp Res2003;15(1):51-9.
153. Svensson O, Stromberg L, Ohlen G, Lindgren U. Prediction of the outcome after hip
fracture in elderly patients. J Bone Joint Surg Br 1996;78(1):115-8.154. Lundström M, Olofsson B, Stenvall M, Karlsson S, Nyberg L, Englund U, Borssén B,
Svensson O, Gustafson Y. Postoperative delirium in old patients with femoral neck
fracture - a randomized intervention study. In press for Aging Clin Exp Res.155. Binder EF, Brown M, Sinacore DR, Steger-May K, Yarasheski KE, Schechtman KB.
Effects of extended outpatient rehabilitation after hip fracture: a randomizedcontrolled trial. JAMA 2004;292(7):837-46.
156. Sherrington C, Lord SR. Home exercise to improve strength and walking velocityafter hip fracture: a randomized controlled trial. Arch Phys Med Rehabil1997;78(2):208-12.
77
157. Latham N, Anderson C, Bennett D, Stretton C. Progressive resistance strength training
for physical disability in older people. Cochrane Database Syst Rev2003(2):CD002759.
