Injury, Int. J. Care Injured (2005) 36, 842—850

www.elsevier.com/locate/injury

Predictors of functional outcome followingintracapsular hip fracture in elderly womenA one-year prospective cohort study

P. Haentjens a,*, Ph. Autier b,c, M. Barette d, S. Boonen e

On behalf of the Belgian Hip Fracture Study Group1

aDepartment of Orthopaedics and Traumatology, Academisch Ziekenhuis V.U.B.,Vrije Universiteit Brussel, Laarbeeklaan 101, B-1090 Brussels, BelgiumbDivision of Epidemiology and Biostatistics, European Institute of Oncology, Milan, ItalycCenter for Research in Epidemiology and Health Information Systems Luxemburg,Luxemburg, Grand Duchy of LuxemburgdUnit of Epidemiology and Prevention of Cancer, Jules Bordet Institute, Brussels, Belgiume Leuven University Center for Metabolic Bone Diseases and Division of Geriatric Medicine,Katholieke Universiteit Leuven, Leuven, Belgium

Accepted 2 February 2005

KEYWORDSHip;Fracture;Arthroplasty;Predictors;Function

* Corresponding author. Tel.: +32 247E-mail address: orthsp@az.vub.ac.

1 All members of the study group are

0020–1383/$ — see front matter # 20doi:10.1016/j.injury.2005.02.002

Summary

Objectives: To explore potential predictors of functional outcome one year after theinjury in elderly women who sustained a displaced intracapsular hip fracture and whowere treated with internal fixation, hemiarthroplasty, or total hip arthroplasty.Participants and methods: Eighty-four women aged �50 years were enrolled on aconsecutive basis in this one-year prospective cohort study reflecting standard day-to-day clinical practice. The main outcome measure was the rapid disability rating scaleversion-2 (RDRS-2) applied at hospital discharge and one year later.Results: At hospital discharge, the total hip arthroplasty group was younger and had abetter functional status than the internal fixation or hemiarthroplasty groups. Oneyear later, the best function was still observed in the total arthroplasty group, but thedifferences were small and failed to achieve the level of statistical significance.During that one-year period, walking ability or mobility did not change significantlyafter total hip arthroplasty, but a significant proportion of the women developedcognitive impairment, including mental confusion, uncooperativeness, and depres-sion. Overall, the most significant predictors of poor functional status one year afterfracture were increasing age, living in an institution at time of injury, and poorfunctional status at discharge.

76537; fax: +32 24778689.be (P. Haentjens).listed in the acknowledgements.

05 Elsevier Ltd. All rights reserved.

Predictors of functional outcome after intracapsular hip fracture 843

Conclusions: In elderly women with a displaced intracapsular hip fracture, total hiparthroplasty is associated with a functional benefit within the first months aftersurgery. However, the extent to which this functional benefit is maintained over time,is less clear. These results support the need for randomised studies to quantify theextent to which, in elderly women, the early functional benefit of total hip arthro-plasty is maintained in the long run or compromised by progressive cognitiveimpairment and other negative determinants of functional outcome.# 2005 Elsevier Ltd. All rights reserved.

Introduction

In the European Union an estimated 400,000 womenexperience a hip fracture each year. These figuresare expected to double by the year 2040 because ofthe increasing number of elderly people.8 Becauseof the high incidence, the dramatic decline in phy-sical functioning following the fracture, and its higheconomic costs, reducing hip-fracture related dis-ability should be a healthcare priority. To targetpreventive strategies, research is needed to identifyoutcome after hip fracture.1,32

Several studieshaveattempted to identify deter-minants of functional outcome after hip fracture inthe elderly.6,12,14,19—22,28,29,34,38,41,42 In general,most evaluated populations were heterogenouswith respect to prefracture health status, livingsituation, and mental health. Moreover, nearly allreports were based on combined analyses of bothintracapsular (femoral neck) and extracapsular(intertrochanteric) fractures.12,14,21,22,28,29,38,41,42

Despite these limitations, several predictors offunctionaloutcomehavebeendelineated, includingage, gender, prefracture residence, and functionalstatus at hospital discharge. Interestingly, a numberof studies found no relationship between the typeof implant and functional recovery at oneyear.2,6,9,12,29,38 These studies, however, did notassess the impact of total hip replacement, eitherbecause this treatmentoptionwasnotconsideredatthat time, or because there were too few patientsreceiving total hip arthroplasty to allow comparisonwith internal fixation or hemiarthroplasty.12

The objective of our prospective study was todocument functional outcome one year after intra-capsular hip fracture, and to explore determinantsof functional outcome, including surgical treat-ment type. In this paper, we developed statisticalmodels to predict functional outcome at one yearin women with an intracapsular hip fracture. To ourknowledge, our predictive models are the first toinclude total hip arthroplasty, internal fixation andhemiarthroplasty, currently three widely usedtreatment options in standard day-to-day clinicalpractice, worldwide.5,17,23,25,26,27,31,33,39

Participants and methods

Study design, source of study population,and surgical procedures

Detailed descriptions of the study design, recruit-ment strategy, participants’ characteristics, out-come assessment, and statistical analysis havebeen reported.1,3,15,16 Briefly, at four Belgian hos-pitals, consecutive women who sustained a first hipfracture (intracapsular or intertrochanteric) werecontacted for inclusion in this one-year prospectivecohort study. Exclusion criteria were a history ofprevious hip fracture and a pathologic fractureresulting from metastatic disease. All patients wereidentified at the time of admission and were fol-lowed up for one year. All patients had operativefracture treatment. At each of the four participat-ing hospitals, the type of surgical repair was decidedby the operating surgeon according to his/her cur-rent day-to-day clinical practice.16

For the current report, only patients with anintracapsular hip fracture were considered, as thetwo main anatomical types of hip fracture (intra-capsular or intertrochanteric) have differentpatient characteristics and treatment patterns.31,39

The focus of the current report is functional out-come of patients with intracapsular hip fractureswho are alive one year after hospital discharge.To facilitate presentation of the results, threemutually exclusive groups were defined accordingto whether the patients had internal fixation, hemi-arthroplasty, or total hip arthroplasty.

Assessments of functional outcome

Functional outcome of hip fracture patients wasassessed using the rapid disability rating scale ver-sion-2.26 This scale comprises 18 items grouped intothree domains: activities of daily living (eight items),degree of disability (seven items), and cognitiveimpairment (three items). The eighteen items arerankedona fourpoint scale,with onepoint indicatingthe best function and four points, the worst. There-fore, total scores can range from 18 (no assistance

844 P. Haentjens et al.

required) to 72 (if the responses indicating the mostseveredisabilities are chosen for all items). The scorecanbe reported as the score for each item, the sumofthe scores for each domain, or as an aggregate score,with higher scores always indicating poorer function.

Four trained interviewers (one per participatinghospital) assessed the RDRS-2 score both at hospitaldischarge and 12 months later. Only patients whocompleted the RDRS-2 questionnaires both at dis-charge and at 12 months after discharge wereincluded in the analyses.

Statistical analyses

We analysed the differences in functional statusamong the three groups and the changes in functionalstatusduring theoneyearevaluationperiod.AsRDRS-2 data were positively skewed, a log transformationwas applied before performing statistical tests. Ana-lysis of variance was used to determine the signifi-cance of the overall difference in the outcomemeasurementsamong the threegroups ataparticularpoint in time.TheStudent’s t-test for paireddatawasused to test for changes in the RDRS-2 scores athospital discharge and one year later.

Multiple regression analyses were used to esti-mate adjusted effects of the explanatory variableson total RDRS-2 scores at one year. Explanatoryvariables were included in regression models if therewas statistical evidence of, or a theoretical basis for,a relationship to our main outcome of interest.

All statistical tests were two-sided.

Table 1 Selected baseline characteristics of the patients,

Prefracture characteristics Internal fixationgroup (n = 14)

Age at time of injuryMean (years)b 77Median (years) 80Range (years) 52—91

Neuropsychiatric disordersc 2 (14%)

Number of comorbidities per womanc

None or one 3 (21%)Two or more 11 (79%)

Prefracture residencec

Living in own house 4 (29%)Living with relatives 4 (29%)Living in nursing home 6 (43%)

a Data are presented as the number of hip-fracture women, with ttable includes data on the 74 patients having suffered a displacedhospital discharge and who completed the entire one year testingb Analysis of variance among the three groups, p = 0.001.c Chi-square for heterogeneity: p > 0.10.

Results

Participants’ characteristics

From November 1995 through July 1996, 184 womenadmitted consecutively with a hip fracture werecontacted; 170 (92.4%) agreed to participate inthe study. Patients or their relatives who declinedto participate typically did so because they did notwish to participate in a long-term study. Of the 170women originally enrolled, 84 women (49%) had adisplaced intracapsular hip fracture that was trea-ted operatively. Of these 84 women, one died duringthe initial hospital stay, and eight died during theone year after discharge. Of the nine women whodied, three had internal fixation, and six had hemi-arthroplasty. In addition, one woman who had ahemiarthroplasty was withdrawn from the studybecause her RDRS-2 questionnaires were not com-pleted properly. For the remaining 74 women (med-ian age 78 years) complete RDRS-2 data wereavailable. In 14 women the displaced intracapsularhip fracture was reduced and stabilised with multi-ple, cannulated, cancellous bone screws, 44patients were treated with a hemiarthroplasty(modular or bipolar) inserted with cement, and 16patients were treated with a cementless total hiparthroplasty.

Table 1 shows the demographic characteristicsand coexisting illnesses of the three groups ofelderly patients. Patients in the screw fixation groupand the hemiarthroplasty group were significantlyolder.

stratified by surgical procedurea

Hemiarthroplastygroup (n = 44)

Total hip arthroplastygroup (n = 16)

81 7181 7164—95 56—83

3 (7%) 0 (0%)

14 (32%) 9 (56%)30 (68%) 7 (44%)

21 (48%) 9 (56%)12 (27%) 4 (25%)11 (25%) 3 (19%)

he percentages in parentheses, unless otherwise indicated. Thisintracapsular hip fracture, who were still alive one year afterprotocol.

Predictors of functional outcome after intracapsular hip fracture 845

Table

2RDRS-2resultsobtainedat

hospital

disch

arge

andoneye

arafterhospital

disch

arge

inthepatients

stratifiedbysurgical

proce

dure

a

Outcomemeasure

aInternal

fixa

tiongroup(n

=14

)Hemiarthroplastygroup(n

=44

)Totalhip

arthroplastygroup(n

=16

)

Score

athospital

disch

arge

(95%

CI)

One-year

score

(95%

CI)

p-valuecScore

athospital

disch

arge

(95%

CI)

One-year

score

(95%

CI)

p-ValuecScore

athospital

disch

arge

(95%

CI)

One-year

score

(95%

CI)

p-valuec

Assistance

withac

tivities

ofdaily

living(ADL)

23.7

b(21.0—26

.9)16

.7(12.9—21

.7)0.01

118

.7b(16.5—21

.1)15

.9(14.0—18

.2)0.02

014

.8b(11.9—18

.5)16

.0(13.3—19

.4)0.09

7

Walking

2.6(2.1—3.3)

1.7(1.2—2.3)

0.01

42.4(2.1—2.8)

2.1(1.8—2.5)

0.43

52.0(1.6—2.6)

1.8(1.4—2.4)

0.27

2Mobility

3.8(3.4—4.2)

2.5

(1.8—3.4)

0.00

62.8(2.4—3.3)

2.4(2.1—2.7)

0.78

42.6(1.9—3.6)

2.4(1.9—3.0)

0.30

8

Degreeofdisab

ility

10.4

(8.8—12

.2)

9.5

(7.7—11

.8)

0.30

510

.2(9.3—11

.2)

9.8(8.9—10

.8)

0.15

38.8(7.8—9.9)

9.5(8.3—11

.0)

0.02

0Degreeofco

gnitive

impairm

ent

4.4(3.6—5.4)

3.8

(3.2—4.5)

0.14

73.8(3.4—4.2)

3.7(3.4—4.1)

0.25

03.3(2.9—3.7)

4.0(3.4—4.6)

0.00

1

TotalRDRS-2score

38.9

b(34.5—43

.8)30

.5(24.6—37

.7)0.02

133

.7b(30.0—36

.5)29

.8(26.7—33

.2)0.03

527

.2b(23.3—31

.8)29

.7(25.5—34

.7)0.00

1aOutcomemeasure:rapid

disab

ilityratingscaleve

rsion-2

(RDRS-2)

score;2

1seealso

Table

A.1

(Appendix

A).

Higherscoresalways

indicate

poorerfunction.Thebest

possible

score

is18

points;theworstscore

possible

is72

points.95

%CI:ninety-five

percentco

nfidence

interval.

bp-value<

0.05

foran

alysis

ofva

rian

ce,testingthedifference

inscoresbetw

eentheinternal

fixa

tion,hemiarthroplasty,

andtotalhip

arthroplastygroups.

cp-valueforpairedStudent’st-test,testingthedifference

betw

eenscore

athospital

disch

arge

andoneye

arlaterin

onegroup.

Functional outcome

Table 2 shows the RDRS-2 measurements at hospitaldischarge and one year after hospital discharge.

At hospital discharge, the women in the total hiparthroplasty group had a significantly better totalRDRS-2 score at hospital discharge than those in thehemiarthroplasty group, and the women in thehemiarthroplasty group also had a significantly bet-ter total RDRS-2 score at hospital discharge thanthose in the internal fixation group. Similarly, theactivities of daily living (ADL) among the patientswho had total hip arthroplasty were significantlybetter than that among those who had internalfixation or hemiarthroplasty.

One year after hospital discharge, the best func-tion was still observed in the arthroplasty groups,but the differences were small, and with the num-bers available, not statistically significant.

During the one-year period after hospital dis-charge, total RDRS-2 scores improved by 8.4 unitsafter internal fixation; in the same period, thefunctional status, as assessed by the ADL scores,improved by 7.0 units. Similarly, after hemiarthro-plasty, the functional status improved significantly,with a change in total RDRS-2 scores of 3.9 units anda change in ADL scores of 2.8 units, respectively.After total hip arthroplasty, on the other hand, totalRDRS-2 scores worsened by 2.2 units during the one-year period after hospital discharge. During thesame period, the scores for ADL, walking abilityor mobility did not change significantly among totalhip arthroplasty patients. However, these womendeveloped significant degrees of cognitive impair-ment, including mental confusion, uncooperative-ness, and depression.

Multiple regression analyses

Table 3 shows the association of a number of factorswith the total RDRS-2 score one year after dis-charge. Model 1 refers to the RDRS-2 score at oneyear as a function of age, prefracture health, andtreatment related factors. Model 2 refers to theRDRS-2 score at one year as a function of RDRS-2score at hospital discharge, age, and treatmentrelated factors. The RDRS-2 score at hospital dis-charge was not included in model 1 because theRDRS-2 score at hospital discharge was stronglycorrelated with the prefracture health factorsincluded in model 1. Model 1 identified two signifi-cant predictors of the total RDRS-2 score at oneyear: age at the time of injury and being institutio-nalised at the time of injury. In addition, there weretrends for an association between higher RDRS-2scores and neuropsychiatric illnesses, the presence

846 P. Haentjens et al.

Table 3 Multiple regression analyses of potential predictors of RDRS-2 score one year after hospital discharge

Potential predictora Regression Bcoefficient

Standard error ofB coefficient

p-value

Model 1: adjusted R2 = 0.306b

Increasing age (years) 0.006 0.002 0.005Prefracture neuropsychiatric

disorders (presence vs. absence)0.064 0.034 0.067

Number of prefracture comorbidities(�2 vs. none or one)

0.124 0.068 0.075

Prefracture residence (living ininstitution vs. living in own house)

0.087 0.040 0.034

Surgical procedure typeHemiarthroplasty vs. internal fixation 0.005 0.042 0.916Total hip arthroplasty vs. internal fixation 0.092 0.052 0.080

Model 2: adjusted R2 = 0.407c

Increasing age (years) 0.004 0.002 0.018RDRS-2 score at hospital discharge 0.599 0.114 0.001

Surgical procedure typeHemiarthroplasty vs. internal fixation 0.016 0.038 0.674Total hip arthroplasty vs. internal fixation 0.109 0.046 0.021

a Increasing age (years) and RDRS-2 score at hospital discharge are continuous variables, whereas all other variables aredichotomous variables.b Model 1 includes variables related to prefracture health status and treatment related factors.c Model 2 includes treatment related factors and RDRS-2 score at hospital discharge.

of two or more comorbidities, and managementwith total hip arthroplasty. In model 2, after con-trolling for the other variables in themodel, youngerage, and a better functional status at hospital dis-charge were associated with better function at oneyear; total hip arthroplasty was associated withworse function at one year. Similar multiple regres-sion analyses performed in relatively youngerwomen (i.e., one analysis restricted to women aged<78 years, and another analysis restricted towomen aged <71 years; data not shown) yieldedresults essentially identical to those for the fullcohort.

Discussion

In recent years, the popularity of total hip arthro-plasty for fracture repair has continued to grow,especially among surgeons using the same prosthe-tic components to reconstruct degenerative hipjoints. In these non-traumatic conditions the clinicalresults of total hip arthroplasty have met the expec-tations of their users. Likewise, for the treatment ofa displaced intracapsular hip fracture, it wasexpected that total hip arthroplasty could providebetter long-term results than hemiarthroplasty orinternal fixation with regard to functional status.That expectation was reinforced by the fact thattotal hip arthroplasty is mainly performed among

younger and more active patients. Our results sug-gest that women in the total hip arthroplasty grouphad a significantly better functional status at hos-pital discharge than those in the hemiarthroplastyand internal fixation groups. One year after hospitaldischarge, the best function was still observed in thetotal hip arthroplasty group, but the differenceswere small and no longer statistically significant.During the same period, the scores for ADL, walkingability or mobility did not change significantlyamong total hip arthroplasty patients, but a signifi-cant proportion of these women developed cogni-tive impairment, including mental confusion,uncooperativeness, and depression, suggestingthat, in elderly patients, the functional benefitprovided by the total hip arthroplasty may be com-promised in the long-term, because of progressivecognitive failure and related problems.

With regard to walking ability and mobility, manyreports have claimed promising results after pri-mary total hip arthroplasty in elderly patients witha displaced intracapsular hip fracture.4,7,10,11,13,15,24,30,35—37,40 In the early 1980s, for example, Dorret al.11 and Skinner et al.36 performed a randomallocation of their patients to different methodsof treatment. One year after surgery, these twoprospective studies did not reveal any significantdifference in mortality, pain or mobility betweenthe treatment options. Somewhat similarly, twodecades later, we did not find any significant differ-

Predictors of functional outcome after intracapsular hip fracture 847

ence in walking ability or mobility at twelvemonths.

To measure outcome, we used the RDRS-2 scale.This scale has been developed to appraise globalfunctional status, and not only those aspects relatedto walking ability or mobility. Compared to otheroutcome measures like the Harris hip score, theMerle d’Aubigne hip score, or the Charnley hipscore, a particular advantage of the RDRS-2 is thatit incorporates mental confusion, uncooperative-ness, and depression into one metric scale. Noneof these scales have been specifically developed toassess physical function of patients with hip frac-ture,18,43 but the RDRS-2 allows to assess functionalstatus and cognitive impairment and is thereforeparticularly useful in elderly individuals.

Our study documented current surgical treat-ment types, but it was not a randomised clinicaltrial and its primary objective was not to comparesurgical treatment techniques. Rather, our studywas designed to reflect standard day-to-day clinicalpractice, and to identify the factors which contri-bute to poor functional outcome. To this end, wealso included the type of surgical procedure. Ourresults fit the paradigm that hip fractures occur in analready compromised individual and suggest thatthe decline in physical ability after hip fracture isnot related to a decrease in walking ability ormobility, but to other items, such as cognitiveimpairment and increasing disability (for example,communication, hearing, sight). Hip fractures areoften associated with a decline in overall perfor-mance, including mental status. What is not clear,however, is whether hip fracture is the event thatinitiates this process or a marker showing that it hasalready begun. Overall, younger patients with hipfracture (and managed with total hip arthroplasty)might represent women with more advanced biolo-gical ageing, with their fracture being a manifesta-tion of generalised body failure.

In line with earlier reports,20,21,25,28,36 age wasfound to be predictive of failure to recover func-tionally at one year. Being institutionalised at thetime of injury was found to be an independentpredictor of poor functional recovery as well. Inprevious studies, Kitimura et al.20 also reported abetter functional recovery in patients residing intheir own home before injury, whereas others foundno difference.29,38 These surveys, however, did notfocus the same target population (unselected Japa-nese20 versus ambulatory US29 or Scandinavian38

patients). In addition to age and prefracture resi-dence, functional status at hospital dischargeappeared to be useful to predict functional outcomeone year later. This is consistent with reports show-ing that better walking ability two weeks after

surgery is associated with improved recovery atone year follow up.20,29 A number of studies foundno relationship between the type of implant andfunctional recovery at one year.2,6,9,12,29,38 Thesestudies, however, did not assess the impact of totalhip replacement, either because this treatmentoption was not considered at that time, or becausethere were too few patients receiving total hiparthroplasty to allow comparison with internal fixa-tion or hemiarthroplasty.12

In this prospective study, we developed statisticalmodels to predict functional outcome at one year inwomen with an intracapsular hip fracture. Wereport the first statistical models that include totalhip arthroplasty, internal fixation and hemiarthro-plasty, currently three widely considered treatmentoptions in standard day-to-day clinical practice,worldwide.26,31,39

Nevertheless, we realise that the treatment ofdisplaced intracapsular hip fractures is a difficultand controversial area and that the current study —despite its prospective design and a thorough func-tional outcome assessment — has a number of lim-itations. Our study was observational and not basedon a randomisation of the patients. Therefore, thegroups may not have been comparable a priori basedon surgeon preferences. As indicated, the type ofsurgical repair was decided by the operating surgeonat each of the four participating hospitals. Thedifferences in the baseline characteristics of thepatients in the three surgical treatment groupssuggest that surgeons take the clinical patient pro-file into account when selecting the type of surgicalprocedure. Clearly, total hip replacement was per-formed in younger patients. We also observed atrend towards internal fixation in patients whohad more comorbidities at the time of injury, a viewsupported by Rodriguez et al.33 Our findings are ingeneral agreement with other studies having shownmarked regional, national, and international varia-tion in the use of surgical techniques to treat anelderly woman with an intracapsular hip frac-ture.5,17,23

Failure to consider functional status prior tofracture might overestimate the impact of hip frac-tures on functional outcome. We did not collectretrospective data on pre-fracture functional statusas these are known to be subject to significant recallbias.25 Taking into account pre-fracture comorbid-ities, pre-fracture mental impairment and pre-frac-ture place of residence in our multiple regressionmodels should have allowed to adjust for differ-ences in prefracture level of functioning.

Also, the numbers recruited were relatively small(despite twelve months prospective collection fromfour hospitals and a 92% uptake), in particular the

848 P. Haentjens et al.

internal fixation and total hip arthroplasty groups.Even within these small groups, however, severalstatistically significant (and clinically relevant) dif-ferences were observed.

Our findings do not mean that, within the contextof a randomised controlled trial, total hip arthro-plasty would not be superior to other treatmentmodalities. As indicated, the limitations of the cur-rent study do not allow any recommendations withregard to the most appropriate use of total hiparthroplasty and our results do not question thevalue of total hip arthroplasty for the appropriateset of surgical criteria. Most importantly, our find-ings underscore the need to perform an adequatelypowered, randomised trial to address the criticallyimportant question whether differences in outcomefor the three groups, if any, are related to thepatient’s general condition or the result of thedifferent treatment regimens given.

Conclusions

We conclude that, in elderly womenwith a displacedintracapsular hip fracture, total hip arthroplasty isassociated with a functional benefit within the firstmonths after surgery. However, our findings areconsistent with the hypothesis that overall healthstatus may have more impact on one year functionaloutcome than the type of surgical procedure. Theseresults support the need for randomised studies toquantify the extent to which, in elderly women, theearly functional benefit of total hip arthroplasty ismaintained in the long run or compromised by pro-gressive cognitive impairment and other negativedeterminants of functional outcome.

Acknowledgements

We are indebted to the participating women andtheir families. We would also like to thank HealthManagement Creative (Brussels) for the professionalway they ensured the interviews and the follow-upof patients.

Table A.1 Rapid disability rating scale version-2 (RDRS-2),

Assistance with activities of daily livingEating None

Walking (with cane or walker if used) NoneMobility (going outside and getting aboutwith wheelchair, etc., if used)

None

Bathing (include getting supplies, supervising) None

This work was partly supported by a medicalresearch grant from Merck Sharp and Dohme BV,Belgium. S. Boonen is Senior Clinical Investigatorof the Fund for Scientific Research–—Flanders, Bel-gium (F.W.O.–—Vlaanderen), and holder of the Leu-ven University Chair in Metabolic Bone Diseases,founded and supported by Merck, Sharp and Dohme.

The Belgian Hip Fracture Study Group consists ofthe following investigators: P. Autier, M.D. (Divisionof Epidemiology and Biostatistics, European Insti-tute of Oncology, Milan, Italy, and Center forResearch in Epidemiology and Health InformationSystems Luxemburg, Grand Duchy of Luxemburg),J.M. Baillon, M.D. (Department of Orthopedics,Ixelles-Etterbeek Hospital, Brussels, Belgium), M.Barette, M.D. (Unit of Epidemiology and Preventionof Cancer, Jules Bordet Institute, Brussels, Bel-gium), J. Bentin, M.D. (Service of Rheumatology,Louis Cathy Hospital, Baudourt, Belgium), S. Boo-nen, M.D., Ph.D. (Leuven University Center forMetabolic Bone Diseases and Division of GeriatricMedicine, Katholieke Universiteit Leuven, Leuven,Belgium), R. Bouillon, M.D., Ph.D. (Leuven Univer-sity Center for Metabolic Bone Diseases and Divisionof Endocrinology, Katholieke Universiteit Leuven,Leuven, Belgium), P. Broos, M.D., Ph.D. (LeuvenUniversity Center for Metabolic Bone Diseases andDivision of Traumatology and Emergency Surgery,Katholieke Universiteit Leuven, Leuven, Belgium),M.C. Closon, Ph.D. (Interdisciplinary Center inHealth Economics, Universite Catholique de Lou-vain, Brussels, Belgium), A.R. Grivegnee, Ph.D.(Unit of Epidemiology and Prevention of Cancer,Jules Bordet Institute, Brussels, Belgium), P. Haent-jens, M.D., Ph.D. (Department of Orthopaedics andTraumatology, Vrije Universiteit Brussel, Brussels,Belgium), and P. Opdecam, M.D., Ph.D. (Departmentof Orthopaedics and Traumatology, Vrije Universi-teit Brussel, Brussels, Belgium).

Appendix A

See Table A.1.

adapted from Linn and Linn26

A little A lot Spoon-feed;intravenous tube

A little A lot Does not walkA little A lot Is housebound

A little A lot Must be bathed

Predictors of functional outcome after intracapsular hip fracture 849

Dressing (include help in selecting clothes) None A little A lot Must be dressedToileting (include help with clothing, cleaning,or help with ostomy, catheter)

None A little A lot Used bedpan or unableto care for ostomy/cathether

Grooming (shaving for men, hairdressing forwomen, nails, teeth)

None A little A lot Must be groomed

Adaptive tasks (managing money/possessions,telephoning, buying newspaper, toiletarticles, snacks)

None A little A lot Cannot manage

Degree of disabilityCommunication (expressing self) None A little A lot Does not communicateHearing (with aid, if used) None A little A lot Does not seem to hearSight (with glasses, if used) None A little A lot Does not seeDiet (deviation from normal) None A little A lot Fed by intravenous tubeIn bed during day (ordered or self-initiated) None A little (<3 h) A lot Most/all the timeIncontinence (urine/feces, with catheter orprosthesis, if used)

None Sometimes Frequently(weekly+)

Does not control

Medication None Sometimes Daily, takenorally

Daily; injection(+oral if used)

Degree of cognitive impairmentMental confusion None A little A lot ExtremeUncooperativeness (combat efforts to helpwith care)

None A little A lot Extreme

Depression None A little A lot Extreme

Directions: rate what the person does to reflect current behavior. Circle one of the four choices for each item. Consider rating withany aids or prostheses normally used. None: completely independent or normal behavior. Total: that person cannot, will not, or maynot (because of medical restriction) perform a behavior or has the most severe form of disability or problem.

Table A.1 (Continued )

References

1. Autier P, Haentjens P, Bentin J, et al., The Belgian HipFracture Study Group. Costs induced by hip fractures: aprospective controlled study in Belgium. Osteoporos Int2000;11:373—80.

2. Barnes B, Dunoven K. Functional outcomes after hip fracture.Phys Ther 1987;67:1675—9.

3. Boonen S, Autier Ph, Barette M, et al. Functional outcomeand quality of life following hip fracture in elderly women: aone-year prospective controlled study. Osteoporos Int2004;15:87—94.

4. Cartlidge IJ. Primary total hip replacement for displacedsubcapital femoral fractures. Injury 1982;13:132—5.

5. Chua D, Jaglal SB, Schatzker J. An orthopedic surgeon surveyon the treatment of displaced femoral neck fracture: oppos-ing views. Can J Surg 1997;40(4):271—7.

6. Clayer MT, Bauze RJ. Morbidity and mortality followingfractures of the femoral neck and intertrochantericregion: analysis of risk factors. J Trauma 1989;29(12):1673—8.

7. Coates RL, Armour P. Treatment of subcapital femoral frac-tures by primary total hip replacement. Injury 1979;11:132—5.

8. Cooper CC, Campion G, Melton III LJ. Hip fractures in theelderly: a world-wide projection. Osteoporos Int 1992;2:285—9.

9. Cummings SR, Philips SL, Wheat ME, et al. Recovery offunction after hip fracture. The role of social supports. JAm Geriatr Soc 1988;36:801—6.

10. Delamarter R, Moreland JR. Treatment of acute femoral neckfractures with total hip arthroplasty. Clin Orthop 1987;218:68—74.

11. Dorr LD, Glousman R, Hoy AL, et al. Treatment of femoralneck fractures with total hip replacement versus cementedand non-cemented hemiarthroplasty. J Arthroplasty 1986;1:21—8.

12. Fox KM, Magaziner J, Hebel JR, et al. Intertrochanteric versusfemoral neck hip fractures: differential characteristics,treatment, and sequelae. J Gerontol A Biol Sci Med Sci1999;54(12):M635—40.

13. Gebhard JS, Amstutz HC, Zinar DM, Dorey FJ. A comparison oftotal hip arthroplasty and hemiarthroplasty for treatment ofacute fracture of the femoral neck. Clin Orthop 1992;282:123—31.

14. Greendale GA, DeAmicis TA, Bucur A, et al. A prospectivestudy of the effects of fracture on measured physical per-formance: results from the MacArthur study–—MAC. J AmGeriatr Soc 2000;48:546—9.

15. Gregory RJ, Wood DJ, Stevens J. Treatment of displacedsubcapital femoral fractures with total hip replacement.Injury 1992;23:168—70.

16. Haentjens P, Autier Ph, Barette M, Boonen S. On behalf of theBelgian Hip Fracture Study Group. Costs of care after hospitaldischarge among women with a femoral neck fracture. ClinOrthop 2003;414:250—8.

17. Hudson JI, Kenzora JE, Hebel JR, et al. Eight-year outcomeassociated with clinical options in the management offemoral neck fractures. Clin Orthop 1998;(348):59—66.

18. Jagjal S, Lakhani Z, Schatzker ML. Reliability, validity, andresponsiveness of the lower extremity measure for patients

850 P. Haentjens et al.

with a hip fracture. J Bone Joint Surg Am 2000;82A:955—62.

19. Kenzora JE, Magaziner J, Hudson J, et al. Outcome afterhemiarthroplasty for femoral neck fractures in the elderly.Clin Orthop 1998;348:51—8.

20. Kitimura S, Hasegawa Y, Suzuki S, et al. Functional outcomeafter hip fracture in Japan. Clin Orthop 1998;348:29—36.

21. Koval KJ, Aharonoff GB, Rosenberg AD, et al. Functionaloutcome after hip fracture. Efffect of general versus regionalanesthesia. Clin Orthop 1998;348:37—41.

22. Koval KJ, Skovron ML, Aharonoff GB, Zuckerman JD. Predic-tors of functional recovery after hip fracture in the elderly.Clin Orthop 1998;348:22—8.

23. Laursen JO. Treatment of intracapsular fractures of thefemoral neck in Denmark: trends in indications over the pastdecade. Acta Orthop Belg 1999;65(4):478—84.

24. Lee BP, Berry DJ, Harmsen WS, Sim FH. Total hip arthroplastyfor the treatment of an acute fracture of the femoral neck:long-term results. J Bone Joint Surg Am 1998;80(1):70—5.

25. Lingard EA, Wright EA, Sledge CB. The Kinemax outcomesgroup. Pitfalls of using patient recall to derive preoperativestatus in outcome studies of total knee arthroplasty. J BoneJoint Surg Am 2001;83A:1149—56.

26. Linn MW, Linn BS. The rapid disabability rating scale-2. J AmGeriatr Soc 1982;30(6):378—82.

27. Lu-Yao GL, Keller RB, Littenberg B, Wennberg JE. Outcomesafter displaced fractures of the femoral neck. A meta-ana-lysis of one hundred and six published reports. J Bone JointSurg Am 1994;76(1):15—25.

28. Magaziner J, Simonsick EM, Kashner TM, Hebel JR, KenzoraJE. Predictors of functional recovery one year followinghospital discharge for hip fracture: a prospective study. JGerontol 1990;45:M101—7.

29. Mossey JM, Mutran E, Knott K, Craik R. Determinants ofrecovery 12 months after hip fracture: the importance ofpsychosocial factors. AJPH 1989;79(3):279—86.

30. Papandrea RF, Froimson MI. Total hip arthroplasty after acutedisplaced femoral neck fractures. Am J Orthop 1996;25:85—8.

31. Parker MJ. The management of intracapsular fractures of theproximal femur. J Bone Joint Surg Br 2000;82B:937—41.

32. Randell AG, Nguyen TV, Bhalerao N, et al. Deterioration inquality of life following hip fracture: a prospective study.Osteoporos Int 2000;11:460—6.

33. Rodriguez J, Herrara A, Canales V, Serrano S. Epidemiologicfactors, morbidity andmortality after femoral neck fracturesin the eldrly. A comparative study: internal fixation versushemiarthroplasty. Acta Orthop Belgica 1987;53:472—9.

34. Salai M, Rosenberg E. Multivariate analysis on the long-termresults of hemiarthroplasty after subcapital fracture of thefemur. J Am Geriatr Soc 1996;44:477—8.

35. Sim FH, Stauffer RN. Management of hip fractures by total hiparthroplasty. Clin Orthop 1980;152:292—7.

36. Skinner P, Rikey D, Ellery J, et al. Displaced subcapitalfractures of the femur: a prospective randomised comparisonof internal fixation, hemiarthroplasty and total hip arthro-plasty. Injury 1989;20:291—3.

37. Squires B, Bannister G. Displaced intracapsular neck of femurfractures in mobile independent patients: total hip replace-ment or hemiarthroplasty? Injury 1999;30:345—8.

38. Svensson O, Stromberg L, Ohlen G, Lindgren U. Prediction ofoutcome after hip fracture in elderly patients. J Bone JointSurg Br 1996;78B:115—8.

39. Swiontkowski MF. Current concepts review. Intracapsularfractures of the hip. J Bone Joint Surg Am 1994;76A:129—38.

40. Taine WH, Armour PC. Primary total hip replacement fordisplaced subcapital fractures of the femur. J Bone JointSurg 1985;67B(2):214—7.

41. Wolinsky FD, Fitzgerald JF, Stump TE. The effect of hipfracture on mortality, hospitalisation, and functionalstatus: a prospective study. Am J Public Health 1997;87:398—403.

42. Young Y, Brant L, German P, et al. A longitudinal examinationof fuctional recovery among older people with subcapital hipfracture. J Am Geriatr Soc 1997;45(3):294.

43. Zuckerman JD, Koval KJ, Aharonoff GB, et al. A functionalrecovery score for elderly hip fracture patients. II. Validityand reliability. J Orthop Trauma 2000;14(1):26—30.