Research-Based-Decision-Making in Canadian Health Organizations: A Behavioural Approach

J Med Syst (2007) 31:185–196DOI 10.1007/s10916-007-9054-3

ORIGINAL PAPER

Research-Based-Decision-Making in Canadian HealthOrganizations: A Behavioural ApproachJalila Jbilou · Nabil Amara · Rejean Landry

Received: 27 November 2006 / Accepted: 22 January 2007 / Published online: 1 May 2007C© Springer Science+Business Media, LLC 2007

Abstract Decision making in Health sector is affectedby a several elements such as economic constraints,political agendas, epidemiologic events, managers’ valuesand environment . . . These competing elements create acomplex environment for decision making. Research-Based-Decision-Making (RBDM) offers an opportunityto reduce the generated uncertainty and to ensure efficacyand efficiency in health administrations. We assume thatRBDM is dependant on decision makers’ behaviour andthe identification of the determinants of this behaviour canhelp to enhance research results utilization in health sectordecision making. This paper explores the determinantsof RBDM as a personal behaviour among managers andprofessionals in health administrations in Canada. From thebehavioural theories and the existing literature, we build amodel measuring “RBDM” as an index based on five items.These items refer to the steps accomplished by a decisionmaker while developing a decision which is based on evi-dence. The determinants of RBDM behaviour are identifiedusing data collected from 942 health care decision makers inCanadian health organizations. Linear regression is used tomodel the behaviour RBDM. Determinants of this behaviourare derived from Triandis Theory and Bandura’s construct“self-efficacy.” The results suggest that to improve researchuse among managers in Canadian governmental health orga-nizations, strategies should focus on enhancing exposition toevidence through facilitating communication networks, part-nerships and links between researchers and decision makers,with the key long-term objective of developing a culture thatsupports and values the contribution that research can make

J. Jbilou (�) · N. Amara · R. LandryUniversite Laval,Quebec, Canadae-mail: [email protected]

to decision making in governmental health organizations.Nevertheless, depending on the organizational level, deter-minants of RBDM are different. This difference has to betaken into account if RBDM adoption is desired. Decisionmakers in Canadian health organizations (CHO) can help tobuild networks, develop partnerships between professionalslocally, regionally and nationally, and also act as changeagents in the dissemination and adoption of knowledgeand innovations in health services. However, the researchfocused on knowledge use as a support to decision-making,further research is needed to identify and evaluate effectiveincentives and strategies to implement so as to enhanceRBDM adoption among health decision makers and moretheoretical development are to complete in this perspective.

Keywords Research-Based-Decision-Making . Healthcaredecision-makers . Behavioural theories . Researchutilization . Empirical . Canada

Introduction

Research is developed to generate new knowledge and estab-lish evidences to provide information for decision makers touse to enhance efficacy and efficiency in the decision-makingprocess [1]. However, literature suggests that a gap betweenresearch and practice still exists [2]. The major part of factorsenumerated in literature are related to organizational dimen-sions for example uncertainty of evidence, competing results[3], trialability of evidence [4], researchers failure to commu-nicate with practitioners [5] complexity of decision makingcontext [6], lack of integration of theory into health servicesresearch [7], lack of relevant and timely evidence and lack ofculture of research in health organizations [8]. . . , but none

Springer

186 J Med Syst (2007) 31:185–196

has explored research utilisation by decision makers throughbehavioural theories.

The early part of the 1990s ushered in evidence-basedmedicine. Presently, there is resurgence in evidence-baseddecision making within health organizations and evidence-based management [4]. Evidence-based decision making in-volves a wide range of individuals from different organi-zational levels with different skills and different level ofpower on decisions they should take to achieve objectives.In the field of health care, turbulences are explored from twoangles: the structural reforms taking place within a givenhealth care system, and the crisis affecting public health, ofwhich epidemics- such as SARS and Avian Influenza—areof particular concern. In both cases, however, the recourseto scientific proofs remains the best option that would en-sure the most effective decision. In addition, decision mak-ing in public health care organizations confronts two typesof challenges which are intrinsically related to the system’sguiding principles. Essentially, each health care system’s de-cision must take into consideration three values: equity, effi-ciency, and autonomy. These three—at times oppositional—values could put the decision-making process into a cumber-some situation [9]. Likewise, studies effected in the field ofknowledge transfer have shown that there is a wide gap be-tween the existent knowledge, the taken decision, and prac-tice as well for clinic as for management [2, 4, 10]. Thevery fact that the decision-maker is a human being, it isour strong postulation, that he/she must be influenced byhis/her own subjective values, beliefs, and norms. In orderto explore the subjective factors affecting decision makers inthe Canadian health care system, we shall explore ResearchBased Decision Making (RBDM) through behaviouraltheories.

It has to be noted that in this study, the term decisionmakers will be used to designate individuals who work in alltype of Canadian Health organizations. The decision makerstargeted in this study are not mainly health care practitioners.Some of them are in charge, directly (managers) or indirectly(professionals), of developing decisions in Canadian healthorganizations. Professionals here are administrative profes-sionals not clinicians (MD, nurses . . .). The decisions thatare targeted in this study are managerial decisions (organiza-tional, financial, structural . . .) but not clinical or therapeuticones.

In the literature, the idea of evidence encompass, amongothers, different concepts such as experience, beliefs, val-ues, abilities, resources, legislations, and researches results[2, 11–14]. In the present study, we will focus solely on theconcept of “research results” as scientific evidence. Indeed,in the specialized literature, RBDM is defined as “ . . . theuse of the best possible evidence when dealing with “reallife” circumstances . . . ” [2, 15–17]. In this study, we do ex-

tend the definition of the concept of research results so asto include five stages: 1. the reception of the research result,2. the reading and understanding of the research result, 3.the citation of the research result, 4. the adaptation of theresearch result, 5. the efforts exerted in order to promotethe research result. The usage of the research result dependsgreatly on decision maker’s willing to, the benefits it providesto him/her, and the risks he/she is ready to take [18]. Thestudies of managers’ behaviours within health care presentbig challenges to the different actors in the public system(population, academics, and institutions). Through a betterunderstanding and prediction of the managers’ behaviours,such a research prospective provides an unmatched oppor-tunity for the development of strategies in favour of RBDM.Important to mention that several researches have exploredthis issue [19–24]. Accordingly, it is conspicuous that theseresearchers and because they have concentrated their re-searches on processes—information systems, developmentof capacities, structural reorganisation, organisational de-terminants, type of use · · · —they have not given the dueattention to the individuals’ behaviour [19, 25].

Important to mention that the results aimed at by thisresearch are inscribed within two different, but very com-plimentary, perspectives: 1. A conceptual perspective whoseobjective lies in the elaboration and the validation of anexplanatory model stemming from behavioural theories. 2.A practical perspective that consists in profiling a numberof specific strategies indispensable to the improvement ofthe use of research results in times of decision making inCanadian Health Organisations (CHO).

Theoretical framework

In the field of psychology, the bulk of theoretical works real-ized for the benefit of health care and whose main objectiveis behavioural prediction, has resulted in the developmentof a plethora of theories that make up what is commonlyknown as the Behavioural Approach. From the latter, wehave chosen two theories for two reasons: 1. since healthis a highly individual value, the concerned decision-makermust be unconsciously influenced by this fact while con-cerned making a decision. 2. In the behaviour in favour forhealth, the psychological weight of decision making in rela-tion to the environmental factors is comparable to the weighta decision-maker has to confront in relation to researchresults.

The Theory of Planned Behaviour (TPB) was proposedin 1975 by Fishbein and Ajzen [26]. It postulates that thebehavioural choices are well-thought of, reasoned, and actedupon. The intention of adopting such or such behaviourremains the most crucial factor. This model, it is to benoted, takes into consideration the enormous pressure of the

Springer

J Med Syst (2007) 31:185–196 187

Fig. 1 Theoretical modelassociating Triandis’ theory ofinterpersonal behaviours (1979)and Bandura’s perceivedeffectiveness (1986)

social environment in the adoption of particular behaviours.In the elaboration of intention, several factors are combined:1. cognitive (evaluating the advantages and the disadvantagesof the potential behaviours). 2. Affective (likes and dislikes).3. Social. 4. Moral. In order to elaborate his theory on inter-personal behaviours (TIB), Triandis [27] has relied heavilyon the researches of Ajzen and Fishbein. Triandis ‘start-ing point was that several behaviours manifest themselvesunconsciously. To consolidate this, he has integrated theconcept of habits and extrinsic—facilitating or inhibiting—factors to the TPB model. In addition, Trandis is credited withsetting the distinction between the beliefs linking emotionsto behaviour and those linking the identified consequencesand the intention of acting. This theoretical approach, it isto be noted, has been used to study the behaviour of indi-viduals in several fields. This is significantly true when itcomes to the adoption of the use of information and commu-nication technologies [28], enterprise creation [29] . . . Thebehavioural approach, however, has not been exploited inthe filed of the utilisation of research results as a support tothe decision- making process of managers in the health caresector.

Similar to other authors, and by introducing dimensionsrelatives to the individuals’ psychological characteristics, wewanted to give a larger perspective to our model [30]. There-fore, we have opted to use the concept of the perceivedself-efficacy developed by Bandura during his research onsocial and cognitive theory [31]. This concept refers to theperception that an individual has in relation to the controlthat he or she is able to exert in order to adopt the desired be-haviour. This concept possesses a double action [31, 32]. Infact, it acts not only through the arbitration of intention, butalso through direct action on behaviour. The perceived effec-tiveness has been used in several fields of research in generaland in the field of the study of behaviours in the health care

sector in particular [33–40]. Important to mention that thetheoretical articulation with Triandis’ TIB concepts was pre-viously tested and validated by other authors such as Caronand Coll [41].

In the present study, we have been inspired by the conceptsdeveloped by Triandis and Bandura in order to elaborate apredictive model of the adoption of the RBDM behaviour bythe managers and the professionals in the health care sector.Our contribution at this level is not only limited to the utili-sation of the concepts proposed by Triandis and Bandura inthe filed of knowledge transfer, but it includes also testingtheir direct impact on the RBDM behaviour without usingmotivation as a mediating variable. The underlying hypoth-esis is that the organisational behaviour in this context isdifferent from that in the health care sector. This is true sincemotivation is an independent component in which a directaction could be exerted without being obliged to go throughthe pressures of social/personal norms, cognition or attitude.The model (Fig. 1) which we propose to test, will allow us,in the first place, to identify the determinants of the RBDMbehaviour, and in the second place, to validate our theoreticalhypothesis.

The operational hypotheses—discerned from the previousmodel—which we will try to verify and are as follows, themeasures are derived from a review of the literature.

H.1 the perceived value of the research positively influencesthe RBDM adoption by health care decision makers(cognitive component) [4].

H.2 the personal convictions of the individual positively in-fluence the RBDM adoption by health care decisionmakers (personal normative component) [4, 10].

H.3 the beliefs in the existence of specific social roles andthe felt pressures positively influence the adoption ofthe RBDM behaviour by health care decision makers(social normative component) [42]

Springer

188 J Med Syst (2007) 31:185–196

Table 1 Dependant variable composing items’ distribution

Very Does notNever Rarely Sometimes Often often apply Average on 1–5(1) (2) (3) (4) (5) (6) Total scale (S.D.)a

During the last five years, on a Likert scale where 1 = never . . .

(in % of respondents)I have received research results concerning the

area for which I’m responsible3.5 8.6 29.8 36.6 20.3 1.2 100.0 3.62 (1.01)

I have read and understood the research reportsthat I have received

1.1 2.9 20.6 42.6 30.6 2.3 100.0 4.01 (0.86)

I have cited research evidence to colleagues orpatients as a reference in my work

4.8 6.5 29.2 38.7 19.1 1.7 100.0 3.62 (1.02)

I have adapted the format of the research resultsto provide information useful to our decision

5.8 11.0 35.4 31.0 14.5 2.2 100.0 3.38 (1.05)

I have made efforts to promote the adoption ofresearch evidence in my field

5.6 9.7 30.9 34.4 17.2 2.2 100.0 3.49 (1.07)

a(S.D.) refer to standard deviation.

H.4 the valuing of the RBDM by health care decision makers(emotional component) [42]

H.5 an organisational context favourable to research and tothe use of the research results positively influence theadoption of the RBDM behaviour by health care deci-sion makers (conditions facilitation adoption) [43, 44]

H.6 the recurrence of the behaviour positively influence theadoption of the RBDM behaviour by health care deci-sion makers (habit component) [44]

H.7 the intention positively influences the adoption of theRBDM behaviour by health care decision makers (mo-tivation component) [4, 10, 45]

H.8 the perceived effectiveness positively influences theadoption of the RBDM behaviour by health care de-cision makers [45]

H.9 the socio-demographic characteristics positively influ-ence the adoption of the RBDM behaviour by healthcare decision makers [44].

The aforementioned limitations of this behavioural ap-proach lie mainly in the fact that this theory “ . . . tended tobe particularly interested in understanding people’s motiva-tion to change behaviour rather than ability to change . . . ”[30]. In the present study, we aim at determining the ex-planatory factors of behaviours and it is important to men-tion that we are not interested in exploring the developmentof capacities supporting them. Indeed, the latter will be thesubject of a future work devoted to the subject we that weare developing about the use of research results from theperspective of individual behaviour. Form the latter, we willproceed to the study of the prediction of behaviour adoptionwhile taking into account the mediation of motivation andthrough which we will we use the methodology of the struc-

tural equations in order to determine the mediating effect ofmotivation.

Theoretical model

The objective of the present study is to identify the deter-minants of RBDM behaviour for health care managers. Assuch, we have chosen to use the quantitative methodologybased on a questionnaire elaborated at the end of a 1998study on the use of the university research results by healthcare managers and professionals in Canadian health careadministrations. The RBDM behaviour represents the de-pendent variable measured by an index of 5 items. In thefollowing lines, we will provide more precise informationon both the data that we have used and the characteristics ofthe studied sample. After that, we will delineate the processof the construction and present the validation of the indicesand the results of the realised analyses. Finally, we will fo-cus on the discussions and the implications of the presentwork.

Data

The data used in this research were collected during a pan-Canadian survey administered in 2001 by Infras.Inc de Que-bec. The target population was made up of managers andprofessionals belonging to the Canadian public administra-tions with a quota per province in proportion to the popu-lation. Three information sources were identified in orderto identify and match up the target population. Firstly, inthe case of the regional board of Quebec and the OntarianDistrict Health Council, it is the Health Evidence Appli-cations and Linkage Network (HEALNet). As for similar

Springer

J Med Syst (2007) 31:185–196 189

Table 2 Descriptive statistics

Variables Type of variables Minimum Maximum Mean Standard deviation Cronbach’s α

Continuous variableAcquisition effort Index: 6 items 1 5 3.40 0.82 0.8958Contributions of users Index: 4 items 1 5 3.12 0.66 0.7611Impact of research findings Index: 2 items 1 5 3.39 0.83 0.7460Relational capital Index: 5 items 1 5 2.81 0.75 0.7393Radicalness of research findings Index: 4 items 1.33 5 3.64 0.20 0.6387Log% of work time spent on research Continue: numeric 0 4.62 2.58 0.34

activitiesLog of size of the organization Continue: numeric 0 11.51 6.62 0.70Log% of time spent on acquisition of Continue: numeric 0 5 2.58 1.02

research evidence

Binary variablesPertinence of research in daily practice 69.7% has respond that research evidence is important or very important in their daily practiceCollaborative research 61.8% (YES)Most advanced university degree PhD = 10.1%, MSc = 46.9%, BSc = 25.7% = , college = 9.2% et autre = 9.2%Research culture 65.5% (YES)Gender 44.4% (Male)Status in the organization Manager = 45.1%, Professional = 54.6%, other = 0.3%

information on the decision makers working in the min-istries (chief executive officer, manager or division boss ineach ministry or governmental agency), were collected byconsulting the Scott’s Government Index [46]. Thirdly, in-formation on the addresses and telephone numbers of pro-fessionals and managers in the hospitals was provided by theCanadian Health Facilities Directory (2001).

The survey firm had as an objective of collecting 340usable questionnaires provided from each of these 3 groupsof decision makers targeted by the study. The number ofvalid questionnaire was 942. The rate of answer generated ishigher than 100%, what could be explained by the use of the“snow-ball” sampling procedure.

Results

Decriptive statistics

Among the 942 participants who replied to the questionnaire,10.1% had a PhD, 46.9% a MSc, 25.7% a BSc, 9.2% a colle-gial diploma and 8.1% other certificates. However, 44.4% ofthe respondents were men. Among the respondents, 45.1%are managers, 54.6% are professionals and 0.3% occupiesanother position in the organisation. On average, there re-spondents possess an occupational seniority of 11.96 yearsin a position while realising the interviews, with a stan-dard deviation of 14.52 years. The population repartition ofthe study by fields of activity reveals that 24.7% work inhospital or intensive care, 6.7% community care (1rst line,home care, . . .), 18.9% in public health or population health,6.5% in long-term care and 43.2% other organisations. How-

ever, the repartition by organisational level shows that 39%work in a ministry or provincial agency, 30.8% in a regionalboard (RHA for Anglophone Canada), 23.6% in a unit healthcare, 0.4% in a ministry or federal agency and 5.4% in an-other type of organisations. The geographic repartition of thissample, however, indicates that 24.5% are in Quebec, 16.7%in Ontario, 10.6% in Alberta, 5.2% in Nova Scotia, 9.2% inSaskatchewan, 9.8% in British Colombia, 7.5% in Manitoba,5.1% in New-Brunswick, 5.9% in Newfoundland, 3.5% inPrince Edward Island and 1.9% in Yukon/north Territories.

The dterminants of the RBDM behaviour

In order to study the impact of the explicatory variables on theadoption of the RBDM behaviour by managers and profes-sionals in the health care sector in Canada, we have estimatedby the linear regression method the following model:

RBDM = β0 + β1 PERTRE + β2 RECUL + β3 CONTU

+β4 LnTSA + β5 ACQEF + β6 RELAK

+β7 COLREC + β8 IMPRES + β9 GEND

+β10 MOSDEG + β11 STATOR

+β12 LnTRA + β13 LnUNS

+β14 NOVEL + ε (1)

where,β i (i = 0, . . . , 14) are the coefficients to estimate;ε is the error term.

Springer

190 J Med Syst (2007) 31:185–196

Tabl

e3

Cor

rela

tion

mat

rix

LnO

RS

GE

ND

PER

TR

EST

AT

OL

nTSA

RE

CU

LM

OSD

EG

IMPR

ES

CO

LR

EC

NO

VE

LA

CQ

EF

RE

LA

KC

ON

TU

LnT

RA

LnO

RS

1−

.056

.027

−.02

9−.

010

−.08

8−.

018

.053

.043

−.00

8.0

57.1

27.0

49.0

32G

EN

D1

−.03

6.0

36.0

92−.

053

.129

−.00

8.0

89.0

11−.

096

.046

−.07

8−.

069

PER

TR

E1

−.10

1.1

99.4

29.1

13.3

19.1

51.0

33.3

39.2

58.3

35.3

10ST

AT

O1

−.16

1−.

107

−.08

8−.

058

−.03

4−.

022

−.10

1−.

015

−.05

5−.

148

LnT

SA1

.165

.132

.168

.276

.044

.243

.270

.141

.465

RE

CU

L1

.136

.320

.204

.017

.394

.225

.302

.322

MO

SDE

G1

.108

.227

−.05

8.1

94.2

13.1

19.2

00IM

PRE

S1

.260

.077

.465

.364

.401

.364

CO

LR

EC

1.0

21.3

03.3

03.1

71.2

47N

OV

EL

1.1

23.0

85.1

16.0

87A

CQ

EF

1.4

31.3

71.4

83R

EL

AK

1.3

89.4

32C

ON

TU

1.3

30L

nTR

A1

This model has been elaborated by using information col-lected from 943 respondents from different organisationallevels in the health care sector in Canada. The dependentvariable RBDM is an index elaborated from 5 items whichcome from the definition that we have retained: 1) evidencereception, 2) reading and comprehension of the evidence, 3)citation of the evidence, 4) adaptation of the evidence and5) conceded efforts in order to promote the evidence adop-tion. For each of the steps, the respondents had to give theirbehaviour, by using a scale of 5 levels, ranging between 1(never) to 5 (very often). The attributed value to this indexvaries between 1 and 5 with an average of 3.6142 and a stan-dard deviation of 0.8194. Among the 942 individuals of thesample, 937 individuals have used the research results to sup-port the decision. The results of Table 1 revealed that nearlyhalf of the individuals have replied � often � or � veryoften, � during the 5 last years, for each of the 5 items. Thedependent variable RBDM is therefore the weighted aver-age of the different responses and can take the value rangingfrom 1 to 5. in order to evaluate the validity of the index,we have proceeded at a factorial analysis (principal compo-nents factorial analysis (PCFA)) with VARIMAX rotation,in order to evaluate the unidimensionality [47]. The resultsshow that one factor explains 61.95% of the initial varianceof the RBDM behaviour with an initial Eigenvalue of 3.098.Hence, we are able to confirm that the items constituting thedependent variable are associated and constitute a uniformand homogeneous concept.

Besides that, we have realised an evaluation of internalconsistence of the index, which has revealed a Cronbachalpha of 0.8457. We have, then, did in the same manner for allthe explicatory variables having necessity the constructionof an index for their measure namely ACQUEF, CONTU,IMPRES, NOVEL and RECAPO (Table 2).

In addition, we have tested the correlations betweenthe different independent variables retained in our model(Table 3). The highest coefficient of correlation is 0.483,improved for ACQUEF and LogTSA. This allows us to con-clude that there are no multi-colinearity problems betweenthe explicatory variables.

The results of the regression that correspond to the ex-plicatory model of the adoption of the RBDM behaviourby health care managers and professionals are delineated inTable 4. We can deduce that all of the independent vari-ables introduced in the model explain in a statistically il-lustrative manner the RBDM adoption behaviour. This issuggestive at levels varying between 1% and 5%. The totalvariance explained by this model is represented by an R2

adjusted by 0.58. Table 4 reveals that the adoption of theRBDM behaviour is favoured by: the positive evaluation ofthe pertinence of research results in daily practice, the adop-tion of a research culture that refers to the preference of

Springer

J Med Syst (2007) 31:185–196 191

Tabl

e4

Reg

ress

ion

mod

els

estim

atin

gde

term

inan

tsof

RB

DM

byde

cisi

onm

aker

sin

Can

adia

nhe

alth

orga

niza

tions

:Glo

bala

ndfo

rea

chor

gani

zatio

nall

evel

Dep

enda

ntva

riab

le:R

BD

Mby

deci

sion

mak

ers

inC

anad

ian

heal

thad

min

istr

atio

nsG

loba

lmod

elM

acro

leve

lM

ediu

mle

vel

Mic

role

vel

Inde

pend

entv

aria

ble

Coe

ffici

ents

(p-v

alue

)aT

ratio

sC

oeffi

cien

ts(p

-val

ue)a

Tra

tios

Coe

ffici

ents

(p-v

alue

)aT

ratio

sC

oeffi

cien

ts(p

-val

ue)a

Tra

tios

Inte

rcep

tR

esea

rch

Bas

edD

ecis

ion

Mak

ing

[RB

DM

].9

31(.

000)

∗∗∗

5.63

1.9

35(.

000)

∗∗∗

3.56

1.6

92(.

010)

∗∗∗

2.31

51.

179

(.00

0)∗∗

∗4.

089

Cog

nitiv

epe

rson

alco

mpo

nent

Res

earc

his

pert

inen

tfor

my

prof

essi

onal

prac

tice

[PE

RT

RE

].1

57(.

000)

∗∗∗

3.45

3.1

18(.

058)

∗1.

576

.172

(.01

7)∗∗

2.13

3.1

77(.

018)

∗∗2.

106

Aff

ectiv

eco

mpo

nent

Res

earc

hcu

lture

[RE

CU

L]

.100

(.01

4)∗∗

∗2.

205

.131

(.04

3)∗∗

1.72

4.1

23(.

052)

∗1.

630

.068

(.21

6)N

S.7

89Pe

rson

alno

rmat

ive

com

pone

ntC

olla

bora

tive

rese

arch

[CO

LR

EC

].1

25(.

001)

∗∗∗

3.05

0.2

29(.

000)

∗∗∗

3.30

7.0

64(.

186)

NS

.896

−.04

6(.

226)

NS

−.59

7So

cial

norm

ativ

eco

mpo

nent

Invo

lvem

ento

fre

sear

ch’

user

s[C

ON

TU

].0

52(.

049)

∗∗∗

1.65

2.0

41(.

217)

NS

.784

.095

(.05

9)∗

1.57

4−.

020

(358

)NS

−.36

6H

abits

Lnb

Tim

esp

ento

nre

sear

chac

tiviti

es[L

nTR

A]

.018

(.00

0)∗∗

∗3.

449

.020

(.00

7)∗∗

∗2.

482

.023

(.01

4)∗∗

2.22

9.0

12(.

101)

NS

1.28

3L

nbT

ime

spen

ton

evid

ence

acqu

isiti

on[L

nTSA

].1

20(.

000)

∗∗∗

6.52

4.1

15(.

001)

∗∗∗

2.95

5.1

27(.

001)

∗∗∗

3.16

8.1

09(.

013)

∗∗2.

243

Mot

ivat

ion

Acq

uisi

tion

effo

rts[

AC

QE

F]

.048

(.00

0)∗∗

∗5.

114

−.00

8(.

432)

NS

−.17

3.1

87(.

000)

∗∗∗

3.62

4.0

98(.

036)

∗∗1.

777

Faci

litat

ing

cond

ition

sN

ovel

tyof

evid

ence

[NO

VE

L]

−.06

1(.

020)

∗∗∗

−2.0

47−.

043

(.19

3)N

S−.

867

−.10

1(.

023)

∗∗−2

.006

−.06

9(1

25)N

S−1

.156

Lnb

Size

ofth

eor

gani

zatio

n[L

nOR

S].0

03(.

381)

NS

.303

.036

(.07

2)∗

1.46

4.0

11(.

228)

NS

.748

.008

(.31

2)N

S.4

90R

elat

iona

lcap

ital[

RE

LA

K]

.097

(.00

0)∗∗

∗3.

188

.166

(.00

1)∗∗

∗3.

148

.036

(.25

0)N

S.6

76.0

94(.

043)

∗∗1.

717

Perc

eive

dse

lf-e

ffica

cyIm

pact

ofre

sear

chon

prac

tice

[IM

PR

ES]

.384

(.00

0)∗∗

∗14

.704

.368

(.00

0)∗∗

∗8.

767

.386

(.00

0)∗∗

∗7.

539

.402

(.00

0)∗∗

∗8.

782

Soci

o-de

mog

raph

icch

arac

teri

stic

sG

ende

r[G

EN

DE

R]

.103

(.00

2)∗∗

∗2.

787

.031

(.30

1)N

S.5

21.1

95(.

001)

∗∗∗

2.97

2.0

67(.

173)

NS

.947

Stat

usin

the

orga

niza

tion

[STA

TO

]−.

060

(.05

2)∗∗

−1.6

28−.

060

(.16

1)N

S−.

993

−.07

6(.

116)

NS

−1.1

95−.

050

(.23

4)N

S−.

727

Uni

vers

ityde

gree

[MO

SDE

G]

.144

(.00

0)∗∗

∗3.

781

.074

(.12

1)N

S1.

175

.132

(.02

4)∗∗

1.99

9.2

71(.

000)

∗∗∗

3.64

4N

umbe

rof

resp

onde

nts

942

382

302

244

Adj

uste

dR

20.

560.

520.

560.

64C

alcu

late

dF

=75

.526

;T

heor

etic

alF

(13;

942)

=2.

05at

1%

Cal

cula

ted

F=

26.0

80;

The

oret

ical

F(1

3;38

2)=

2.15

at1%

Cal

cula

ted

F=

25.3

81;

The

oret

ical

F(1

3;30

2)=

2.16

at1%

Cal

cula

ted

F=

27.5

63;

The

oret

ical

F(1

3;24

4)=

2.17

at1%

Not

e.N

umbe

rsbe

twee

nbr

acke

tsin

dica

tep-

valu

e.a∗

;∗∗an

d∗∗

∗in

dica

teth

atva

riab

leis

sign

ifica

ntre

spec

tivel

yat

10%

,5%

and

1%.

NSno

nsi

gnifi

ant.

bL

nin

dica

tes

alo

gari

thm

ictr

ansf

orm

atio

nof

the

vari

able

.

Springer

192 J Med Syst (2007) 31:185–196

Table 5 Elasticities of determinants of RBDM by decision makers in Canadian health organizations: global and for each organizational level

Partial elasticity

% Global modelMacroorganizational level

Mediumorganizational level

Microorganizational level

(A) Continuous variablesAcquisition efforts (ACQEF) 0.47 NS 1.84 0.92Users’ contribution (CONTU) 0.46 NS 0.84 NSImpact of research (IMPRES) 3.67 −0.08 3.73 3.86Relational capital (RELAK) 0.78 −0.45 NS 0.73Novelty of research (NOVEL) −0.01 NS −1.08 NSLn Size of organization (LnORS) NS 1.34 NS NSLn%Time spent on evidence acquisition (LnTSA) 0.85 0.378 0.90 0.76Ln%Time spent on research activities (LnTRA) 0.1 3.45 0.02 NS

(B) Binary variablesRelevance of research (PERTRE) 3.33 1.81 5.02 5.16Collaborative research (COLREC) 2.70 3.67 NS NSMost advanced degree (MOSDEG) 3.14 NS 3.97 8.72Research culture (RESCUL) 2.10 1.73 3.54 NSGender (GEND) 2.16 NS 5.63 NSStatus in organization (STATO) −1.26 NS NS NS

aElasticity is reported for an improvement of 10% for each variable.bImpact of binary variables on EBDM is estimated through the following equation: Impacts = 100 × (Z1 − Z0)/Z0, where Z1 and Z0 is variablevalue, respectively when reference variable is equal to 1 and 0.

research as a source of information, the strong involvementof peers in activities related to research or to the use ofthese results, the percentage of time spent to acquire evi-dences, 5. the deployed efforts in this sense, the richness ofthe relational capital, by the participation to the collaborativeresearch with the universities, by the perceived effectivenesstowards change resulting form the utilisation of evidences,and the detention of a university diploma (PhD and Mas-ters). Moreover, it is important to stress the fact that thereexist two variables with a negative relationship. These re-sults have allowed us to notice that, in addition to the factthat it is women who adopt it, it is the professional’s sta-tus in the organisation that favours the adoption of RBDMbehaviour.

The results presented on Table 2 show the sings and sig-nificance of the coefficients of the explanatory variables.The presentation of these results does not take into accountthe scope of these coefficients since, the logistic functionalform upon which logit regression are based, the estimatedvalues of coefficients, can not be interpreted as elastici-ties or as coefficients reflecting the marginal impacts of theexplanatory variables (Landry, Amara and al., 2000) Thepartial elasticities for the variables that have been found tosignificantly explain the likelihood of adopting RBDM be-haviour by managers and professionals in Canadian healthorganizations were calculated using the equations 1 for thebinary variables and 2 for the continuous variables (Ap-pendix 1). The evaluated partial elasticities are presented onTable 5.

For the comprehensive model, the elasticity of the variableimpact of research results on daily practice is the highest onewith 36.74%. This indicate that a positive relative changeof 10% of the index of impact of research on daily practiceincreases the likelihood of RBDM behaviour’s adoption by36.74%. Likewise, a positive relative change of 10% in thepercentage of time spent on evidence acquisition and the in-dex of relational capital increases the likelihood of RBDMbehaviour’s adoption by, respectively, 8.5% and 7.85%. Con-versely, a positive relative change of 10% in the noveltyof research decreases the likelihood of RBDM behaviour’sadoption by 0.17%.

For the macro organizational level model, the elasticity ofthe variable percentage of time spent on research activitiesis the highest one with 34.50%. This indicates that a posi-tive relative change of 10% of the percentage of time spenton research activities increases the likelihood of RBDMbehaviour’s adoption by 34.50%. Likewise, a positive rel-ative change of 10% in the size of unit of work and thepercentage of time spent on evidence acquisition increasesthe likelihood of RBDM behaviour’s adoption by, respec-tively, 13.45% and 3.68%. Conversely, a positive relativechange of 10% in the indexes of relational capital and im-pact of research on daily practice decreases the likelihoodof RBDM behaviour’s adoption respectively by 4.48% and0.8%.

For the meso organizational level model, the elasticity ofthe variable impact of research on daily practice is the highestone with 37.34%. This indicate that a positive relative change

Springer

J Med Syst (2007) 31:185–196 193

of 10% of the percentage of time spent on research activi-ties increases the likelihood of RBDM behaviour’s adoptionby 37.34%. Likewise, a positive relative change of 10% inthe index of acquisition efforts, users’ contribution, and per-centage of time spent on evidence acquisition increases thelikelihood of RBDM behaviour’s adoption by, respectively,18.37%, 8.38% and 8.99%. Conversely, a positive relativechange of 10% in the variable degree of novelty of researchdecreases the likelihood of RBDM behaviour’s adoption by10.82%.

For the micro organizational level model, the elasticity ofthe variable impact of research on daily practice is the high-est one with 38.64%. This indicate that a positive relativechange of 10% of the percentage of impact of research ondaily practice increases the likelihood of RBDM behaviour’sadoption by 38.64%. Likewise, a positive relative change of10% in the variables acquisition efforts, percentage of timespent on evidence acquisition and relational capital increasethe likelihood of RBDM behaviour’s adoption by, respec-tively, 9.17%, 7.58% and 7.28%.

Conclusion and implications

The objectives of the present study were of two types: 1. con-ceptual to test the validity of the behavioural theory in orderto explain RBDM. 2. Operational in order to determine thekey factors favouring or hindering the adoption of this be-haviour by health care decision makers in Canada. In order toelaborate our RBDM explanatory model, we have opted forTriandis’s theory of interpersonal behaviour in associationwith Bandura’s concept of the perceived self-efficacy. Wehave assumed that RBDM for health care decision makers isa processual behavior that includes 5 phases: 1. the receptionof the evidence, 2. the reading and the understanding of theevidence, 3. the citation of the evidence, 4. the adaptationof the evidence, 5. the efforts exerted to promote the adop-tion of the evidence. These components have been the basisfor the elaboration of an RBDM index. Also, the data thatwe have used reveal that among the 943 respondents, morethan 50% have answered 4 (often) or 5 (very often) to eachitem constituting the index measuring the dependent variableRBDM.

The conceptual model that we have built has allowedus to confirm the fact that the RBDM behaviour could bemeasured within a behavioural perspective. This work hasstrongly approved the impact of the organizational and theindividual characteristics on the use of research results byCanadian health care managers. In the first place, this hasindeed allowed us to know the profile of the potential userof the evidences—let’ say a professional woman holding aPhD or a master’s degree (MSc) with a diversified network

and activity (research and administrative tasks). Also, andmore importantly, it has permitted us to confirm the fact thatin order to promote an individual RBDM behaviour, effortsmust be exerted to assure both individual and peer valuing ofresearch findings utilisation, to strive to create interfaces andopportunities so as to improve collaborative research withthe universities,to provide a modality of work organisationwhich leaves a room to the manager to manage his or hertime in order to be able to both collect and study the neces-sary evidences. These assertions show the importance of thedevelopment of a global strategy for favouring the RBDM inhealth care managers and professionals. This strategy targetsthe organisational structures (work time organisation and thedevelopment of collaborative research) as well as the be-liefs and values of the decision makers (individual and peervaluing of the RBDM). It is unquestionable that the RBDMbehaviour is intrinsically linked to the beliefs and values ofthe actors. In fact, this does strongly suggest the incumbentnecessity of profound reflection on the strategies to be devel-oped in order to reach a real paradigmatic change in favourof RBDM before even thinking on the structures and theprocesses. RBDM is an individual behavior that has to besupported by organizational structure and culture of KT/KU.More, determinants of RBDM are different depending onorganizational level; specific strategies and incentives haveto be developed for each level.

Moreover, the current context of crisis does strongly evokea state of emergency especially when it comes to the proac-tive organizations. Strategically speaking, the health caresector organisations should favour the recruitment of pivotalprofessionals adopting RBDM behaviour in order to dissem-inate such behaviour in the first place and, in the secondplace, to exert their full efforts in times of crises. However,the present work presents a limitation that is intrinsicallylinked to the administrative context of health care managersand professionals in Canada. This is particularly true dueto the fact that the Canadian health care system is publiclyfinanced and it is known for its hierarchic and bureaucraticmanagement which makes decision making extremely cen-tralized. These realities require us to assert the fact that re-sults generalization must be taken with precaution. In orderto enhance the knowledge and strength then the actions infavour of RBDM behaviour in health care, it is of paramountimportance to investigate a number of research issues andexplore other research avenues: 1) What are the strategieswhich are judged to be effective and efficient in order toachieve a pragmatic change in favour of RBDM ? 2) Whatare the effective and efficient organisational modalities to im-plement in order to provide a framework favouring RBDMin health care? 3) What are the challenges that are to beconfronted while striving to enhance the RBDM behaviourhealth care?

Springer

194 J Med Syst (2007) 31:185–196

Take

Hom

em

essa

ges

RB

DM

isa

beha

vior

that

isde

term

ined

bySe

lf-e

ffica

cyan

dT

IBco

mpo

nent

sIn

divi

dual

char

acte

rist

ics

•Edu

catio

nPh

D/M

Sc

•Gen

der

(Fem

ale)

•Res

earc

hac

tivity

and

cultu

re

•Net

wor

king

(div

ersi

fied)

Asu

ppor

tive

orga

niza

tion

with

Agl

obal

stra

tegy

for

RB

DM

impr

ovem

enta

mon

gD

ecis

ion

Mak

ers:

•An

inte

grat

ive

cultu

rals

trat

egy

(ind

ivid

uala

ndpe

ers

valu

atio

nof

rese

arch

and

RB

DM

,res

earc

hcu

lture

)•A

stru

ctur

alst

rate

gy(W

ork

time

man

agem

ent,

colla

bora

tive

rese

arch

,res

ourc

esfo

rK

T,im

pact

s’as

sess

men

t)T

rans

vers

alde

term

inan

tsT

ime

spen

ton

rese

arch

findi

ngs

acqu

isiti

onR

elev

ance

ofre

sear

chfo

rda

ilypr

actic

e(a

dequ

acy

and

read

ines

sto

use)

Perc

eive

dim

pact

ofre

sear

chon

prac

tice,

serv

ices

...

App

endi

x1

Var

iabl

eV

aria

ble

Des

crip

tion

Var

iabl

em

easu

re

PER

TR

ER

esea

rch

pert

inen

cefo

rda

ilypr

actic

e.It

mea

sure

sth

eco

gniti

veco

mpo

nent

ofth

ebe

havi

our

ofth

em

anag

erfa

cing

the

RB

DM

beha

viou

r.

Con

tinuo

usva

riab

lem

easu

red

byth

equ

estio

n“O

vera

ll,i

nth

efie

ldof

my

prof

essi

onal

prac

tice

,res

earc

his

pert

inen

tfor

my

prof

essi

onal

prac

tice

”co

ded

ona

Lik

erts

cale

betw

een

1(n

ever

)an

d5

(ver

yof

ten)

.

RE

CU

LR

esea

rch

cultu

re.I

tmea

sure

sth

eaf

fect

ive

com

pone

ntof

the

man

ager

’sat

titud

eto

war

dsth

ead

optio

nof

the

RB

DM

beha

viou

r.

Var

iabl

em

easu

red

byth

equ

estio

n“O

vera

ll,i

nth

efie

ldof

my

prof

essi

onal

prac

tice

,res

earc

his

my

pref

erre

dso

urce

ofin

form

atio

n,in

itial

lyco

ded

ona

Lik

erts

cale

betw

een

1(n

ever

)an

d5

(ver

yof

ten)

,the

ntr

ansf

orm

edin

tobi

nary

vari

able

or1

corr

espo

ndin

gto

the

choi

ceof

resp

onse

4an

d5

and

0w

hen

choo

sing

1,2,

3).

CO

NT

UIm

plic

atio

nof

rese

arch

user

s.It

mea

sure

sth

eno

rmat

ive

soci

alco

mpo

nent

cont

inuo

usva

riab

lere

pres

ente

dby

anin

dex

com

pose

dof

4ite

mm

easu

red

byth

equ

estio

ns1)

Inm

ypr

ofes

sion

alfie

ld,u

sers

ofre

sear

ch,c

reat

eev

ents

for

tran

sfer

ofre

sear

ch,2

)In

my

prof

essi

onal

field

,use

rsof

rese

arch

,org

aniz

etr

aini

ngac

tivi

ties

whi

chin

tegr

ate

rese

arch

resu

lts,

3)In

my

prof

essi

onal

field

,use

rsof

rese

arch

,inv

estm

ater

ialo

rfin

anci

alre

sour

ces

inkn

owle

dge

tran

sfer

et4)

Inm

ypr

ofes

sion

alfie

ld,u

sers

ofre

sear

ch,d

evel

opne

wap

proa

ches

,new

inte

rven

tion

sor

new

deli

very

syst

ems

base

don

rese

arch

resu

lts.

The

resp

onse

sar

eco

ded

ona

Lik

erts

cale

betw

een

1(n

ever

)an

d5

(ver

yof

ten)

ofea

chite

m.

Log

TSA

Perc

enta

geof

wor

ktim

esp

enti

nac

quir

ing

scie

ntifi

cev

iden

ces.

Itm

easu

res

the

habi

tof

real

isin

gth

eR

BD

Mbe

havi

our.

Con

tinuo

usva

riab

lem

easu

red

byth

equ

estio

n“W

hatp

erce

ntag

eof

your

wor

kti

me

doyo

usp

end

onac

quis

itio

nof

rese

arch

evid

ence

?”th

ere

spon

ses

vary

betw

een

0an

d10

0%.I

twas

cate

gori

sed

in5

quin

tiles

.Alo

gari

thm

ictr

ansf

orm

atio

nw

asre

alis

edin

orde

rto

besu

reof

ano

rmal

dist

ribu

tion.

Springer

J Med Syst (2007) 31:185–196 195

Con

tinue

d

Var

iabl

eV

aria

ble

Des

crip

tion

Var

iabl

em

easu

re

AC

QU

EF

Con

cede

def

fort

sin

acqu

irin

gth

ere

sear

chre

sults

.Itm

easu

res

the

mot

ivat

ion

toad

optt

heR

BD

Mbe

havi

our

cont

inuo

usva

riab

lere

pres

ente

dby

anin

dex

com

pose

dof

6ite

ms,

mea

sure

dby

the

ques

tions

:1)

Inte

rms

ofyo

urpr

ofes

sion

alin

volv

emen

t,w

hati

sth

eim

port

ance

ofac

tive

invo

lvem

enti

nre

sear

chpr

ojec

ts,2

)In

term

sof

your

prof

essi

onal

invo

lvem

ent,

wha

tis

the

impo

rtan

ceof

part

icip

atio

nin

prof

essi

onal

conf

eren

ces

and

wor

ksho

psin

volv

ing

rese

arch

ers,

3)In

term

sof

your

prof

essi

onal

invo

lvem

ent,

wha

tis

the

impo

rtan

ceof

mem

bers

hip

onex

pert

-pan

els

and

com

mit

tees

invo

lvin

gre

sear

cher

s,4)

Inte

rms

ofyo

urpr

ofes

sion

alin

volv

emen

t,w

hati

sth

eim

port

ance

ofne

wsl

ette

rsan

dre

sear

chin

form

atio

nde

live

red

dire

ctly

tom

e,5)

Inte

rms

ofyo

urpr

ofes

sion

alin

volv

emen

t,w

hati

sth

eim

port

ance

ofre

sear

chre

port

sse

ntto

me

byE

-mai

l,6)

Inte

rms

ofyo

urpr

ofes

sion

alin

volv

emen

t,w

hati

sth

eim

port

ance

ofre

sear

chof

evid

ence

avai

labl

eon

the

Inte

rnet

.The

resp

onse

sar

eco

ded

ona

Lik

erts

cale

betw

een

1(n

otim

port

ant)

and

5(v

ery

impo

rtan

t)fo

rea

chite

m.

RE

CA

POR

elat

iona

lcap

italw

ithre

sear

cher

s.It

mea

sure

sth

eco

nditi

ons

faci

litat

ing

the

adop

tion

ofth

eR

BD

Mbe

havi

our

Con

tinuo

usva

riab

lere

pres

ente

dby

the

inde

xco

mpo

sed

of5

item

sm

easu

red

byth

equ

estio

n.1)

how

freq

uent

lydo

you

have

pers

on-t

o-pe

rson

cont

actw

ith

rese

arch

ers

inun

iver

sity

,2)

how

freq

uent

lydo

you

have

pers

on-t

o-pe

rson

cont

actw

ith

rese

arch

ers

hosp

ital

s,3)

how

freq

uent

lydo

you

have

pers

on-t

o-pe

rson

cont

actw

ith

rese

arch

ers

inre

gion

alhe

alth

care

auth

orit

ies,

4)ho

wfr

eque

ntly

doyo

uha

vepe

rson

-to-

pers

onco

ntac

twit

hre

sear

cher

sin

gove

rnm

enta

genc

ies,

5)ho

wfr

eque

ntly

doyo

uha

vepe

rson

-to-

pers

onco

ntac

twit

hre

sear

cher

sin

priv

ate

firm

s.T

here

spon

ses

are

code

don

aL

iker

tsca

lebe

twee

n1

(nev

er)

and

5(v

ery

ofte

n)fo

rea

chite

m.

CO

LR

EC

Col

labo

rativ

ere

sear

chw

ithth

eun

iver

sitie

s.It

mea

sure

sth

epe

rson

alno

rmat

ive

com

pone

ntvi

s-a-

vis

the

RB

DM

beha

viou

r:di

chot

omou

sva

riab

lem

easu

red

byth

equ

estio

n“I

nth

ela

stfiv

eye

ars,

have

you

been

invo

lved

ina

rese

arch

stud

yth

atco

mbi

neun

iver

sity

and

non-

univ

ersi

tyin

vest

igat

ors?

”co

ded

1(y

es)

and

0(n

o).

IMPR

ES

Res

earc

him

pact

onth

epr

actic

e.It

mea

sure

the

perc

eive

def

fect

iven

ess

whe

nit

com

esto

the

real

isat

ion

ofth

eR

BD

Mbe

havi

our

cont

inuo

usva

riab

lere

pres

ente

dby

the

anin

dex

com

pose

dof

2ite

ms

mea

sure

dby

the

ques

tions

:1)

duri

ngth

ela

stfiv

eye

ars

rese

arch

evid

ence

has

led

me

tom

ake

prof

essi

onal

choi

ces

and

deci

sion

sth

atI

wou

ldno

thav

em

ade

othe

rwis

e,2)

duri

ngth

ela

stfiv

eye

ars

the

util

izat

ion

ofre

sear

chev

iden

ceha

sle

dto

conc

rete

chan

ges

inth

epr

ogra

ms

orse

rvic

espr

ovid

edby

my

orga

niza

tion

.The

resp

onse

sar

eco

ded

ona

Lik

erts

cale

betw

een

1(n

ever

)an

d5

(ver

yof

ten)

for

each

item

.G

EN

Dre

spon

dent

’sge

nder

Bin

ary

vari

able

code

d1

for

Man

and

0fo

rW

oman

.ST

AT

Oth

ere

spon

dent

’sst

atus

inth

eor

gani

satio

nB

inar

yva

riab

leco

ded

1fo

rm

anag

ers

and

0fo

rth

epr

ofes

sion

als.

MO

SDE

Ghi

ghes

tedu

catio

nald

egre

eB

inar

yva

riab

leco

ded

1fo

rth

ose

havi

nga

PhD

ora

mas

ters

(MSc

)an

d0

for

othe

rL

nTR

APe

rcen

tage

oftim

eof

wor

ksp

ento

nre

sear

chac

tiviti

es.I

tmea

sure

sth

eha

bitt

ore

aliz

eth

ebe

havi

orR

BD

M.I

tre

flect

sth

ero

utin

isat

ion

ofth

ere

sear

chac

tiviti

es.

Con

tinuo

usva

riab

lem

easu

red

byth

equ

estio

n“W

hatp

erce

ntag

eof

your

wor

kti

me

doyo

usp

end

one

rese

arch

acti

viti

es?”

the

answ

ers

vary

from

0to

100%

.Alo

gari

thm

ictr

ansf

orm

atio

nw

asre

aliz

edto

mak

esu

reof

ano

rmal

dist

ribu

tion.

LnO

RS

Num

ber

ofpe

rson

sw

orki

ngin

the

orga

niza

tion

whe

rere

spon

dent

wor

ks.

Itre

flect

sth

esi

zeof

the

orga

niza

tion

whe

reth

ere

spon

dent

wor

ks.

Con

tinuo

usva

riab

lem

easu

red

byth

equ

estio

n“W

hati

sth

eap

prox

imat

enu

mbe

rof

empl

oyee

sin

your

adm

inis

trat

ive

orga

niza

tion

?”th

ean

swer

sva

ryfr

om0

to65

00.A

loga

rith

mic

tran

sfor

mat

ion

was

real

ized

tom

ake

sure

ofa

norm

aldi

stri

butio

n.

NO

VE

LT

here

quir

edef

fort

san

dre

sour

ces

tous

eev

iden

ce.I

tmea

sure

sth

eef

fort

san

dre

sour

ces

that

resp

onde

ntha

sto

inve

stto

use

evid

ence

sw

hen

head

opts

RB

DM

;ext

ensi

vely

itre

fers

toth

eno

velty

ofev

iden

ce.

Con

tinuo

usva

riab

lere

pres

ente

dby

anin

dex

cons

titut

edof

4ite

ms,

mea

sure

dby

the

ques

tions

:In

your

field

ofpr

ofes

sion

alpr

actic

e,th

eut

iliz

atio

nof

rese

arch

resu

lts

tode

velo

pne

wor

impr

oved

heal

thse

rvic

esw

ould

requ

ire:

1)th

ecu

stom

izat

ion

ofre

sear

chre

sult

sfo

rea

chse

rvic

e,2)

the

deve

lopm

ento

fane

wte

chno

logy

,3)

sign

ifica

ntfin

anci

alin

vest

men

ts,4

)si

gnifi

cant

inve

stm

ents

inth

etr

aini

ngof

pers

onne

l.T

hean

swer

sar

eco

ded

ona

Lik

erts

cale

rang

ing

betw

een

1(s

tron

gly

disa

gree

)an

d5

(str

ongl

yag

ree)

for

each

ofth

eite

ms.

Springer

196 J Med Syst (2007) 31:185–196

References

1. Hemsley-Brown, J., Using research to support management de-cision making within the field of education. Manage. Decis.43(5):691–705, 2005.

2. Dobbins, M., et al., A framework for the dissemination and uti-lization of research for health-care policy and practice. Online J.Knowl. Synth. Nurs. 9:7, 2002.

3. Whynes, D., Policy forum: Health care reform: Towards anevidence-based national health service? Econ. J. 106(439):1702–1712, 1996.

4. Pfeffer, J., and Sutton, R., Evidence-based management. DecisionMaking, Harvard Business Review, pp. 63–74, 2006.

5. Kelemen, M., and Bansal, P., The conventions of management re-search and their relevance to management practice. Br. J. Manage.13:97–108, 2002.

6. Hewison, A., Evidence-based management in the NHS: Is it pos-sible? J. Health Org. Manage. 18(5):336–348, 2004.

7. Brazil, K., et al., From theory to practice: Improving the impact ofhealth services research. BMC Health Serv. Res. 5(1):1, 2005.

8. Kiefer, L., et al., Fostering evidence-based decision-making inCanada: examining the need for a Canadian population and publichealth evidence centre and research network. Can. J. Public Health96(3):I1–40, following 200, 2005.

9. Tavakoli, M., Davies, H. T., and Thomson, R., Decision analysisin evidence-based decision making. J. Eval. Clin. Pract. 6(2):111–120, 2000.

10. Learmonth, M., and Harding, N., Evidence-based management:The very idea. Public Administration 84(2):245–266, 2006.

11. Dobbins, M., et al., Use of systematic reviews in the developmentof new provincial public health policies in Ontario. Int. J. Technol.Assess. Health Care 20(4):399–404, 2004.

12. Estabrooks, C., et al., Decision aids: Are they worth it? A system-atic review. J. Health Serv. Res. Policy 6(3):170–182, 2001.

13. Estabrooks, C.A., Will evidence-based nursing practice make prac-tice perfect? Can. J. Nurs. Res. 30(1):15–36, 1998.

14. Sibbald, B., and Roland, M., Getting research into practice. J. Eval.Clin. Pract. 3(1):15–21, 1997.

15. Hayward, R. S., et al., Practice guidelines. What are internistslooking for? J. Gen. Intern. Med. 11(3):176–178, 1996.

16. Nutbeam, D., Improving the fit between research and practice inhealth promotion: Overcoming structural barriers. Can. J. PublicHealth 87(Suppl. 2):S18–23, 1996.

17. Sackett, D. L., et al., Evidence based medicine: What it is and whatit isn’t. BMJ 312(7023):71–72, 1996.

18. Walshe, K., and Rundall, T. G., Evidence-based management:From theory to practice in health care. Milbank Q. 79(3):429–457,IV–V, 2001.

19. Niedzwiedzka, B. M., Barriers to evidence-based decision makingamong Polish healthcare managers. Health Serv. Manage. Res.16(2):106–115, 2003.

20. Lindstrom, R. R., Evidence-based decision-making in healthcare:Exploring the issues though the lens of complex, adaptive systemstheory. Healthc. Pap. 3(3):29–35, 2003; discussion 66–71.

21. Mitton, C., and Donaldson, C., Health care priority setting: Prin-ciples, practice and challenges. Cost. Eff. Resour. Alloc. 2(1):3,2004.

22. Dobrow, M. J., et al., The impact of context on evidence utilization:A framework for expert groups developing health policy recom-mendations. Soc. Sci. Med., 2006.

23. Landry, R., Amara, N., and Lamari, M., Utilization of social scienceresearch knowledge in Canada. Res. Policy 30:333–349, 2001.

24. Amara, N., Ouimet, M., and Landry, R., New evidence on instru-mental, conceptual and symbolic utilization of university researchin government agencies. Sci. Commun. 26(1):75–106, 2003.

25. Holmberg, H., et al., Economic evaluation of screening for prostatecancer: A randomized population based programme during a 10-year period in Sweden. Health Policy 45(2):133–147, 1998.

26. Fishbein, M., and Ajzen, I., Belief, Attitude, Intention, and Be-havior: An Introduction to Theory and Research, Addison-Wesley,Reading, MA, 1975.

27. Triandis, H., The self and social behavior in differing culturalcontexts. Psychol. Rev. 96:506–520, 1989.

28. Gagnon, M. P., and Godin, G., An adaptation of the theory ofinterpersonal behaviour to the study of telemedicine adoption byphysicians. Int. J. Med. Inform. 71(2–3):103–115, 2003.

29. Emin, S., Les facteurs determinants la creation d’entreprise par leschercheurs publics: Application des modeles d’intension. Revue del’Entrepreunariat 3(1):1–19, 2004.

30. Jeffery, R. W., How can Health Behavior Theory be made moreuseful for intervention research? Int. J. Behav. Nutr. Phys. Act.1(1):10, 2004.

31. Bandura, A., Self-Efficacy: The Exercise of Control, Freeman, NewYork, WH, 1997.

32. Bandura, A., and A., L. E., Negative self-efficacy and goal effectsrevisited. J. Appl. Psychol. 88(1):87–99, 2003.

33. Beauchamp, M. R., Bray, S. R., et al. Pre-competition imagery,self-efficacy and performance in collegiate golfers. J. Sports Sci.20(9):697–705, 2002.

34. Benight, C. C., and Bandura, A., Social cognitive theory of post-traumatic recovery: The role of perceived self-efficacy. Behav. Res.Ther. 42(10):1129–1148, 2004.

35. Casey, L. M., Oei, T. P., et al., An integrated cognitive model ofpanic disorder: The role of positive and negative cognitions. Clin.Psychol. Rev. 24(5):529–555, 2004.

36. De Souza, G. A., Da Silva, A. M., et al., Self-efficacy as a mediatorfor improvement in oral health clinical indices. Pesqui. Odontol.Bras 16(1):57–62, 2002.

37. Lavelle, E., Smith, J., et al., The writing approaches of secondarystudents. Br. J. Educ. Psychol. 72(Pt 3):399–418, 2002.

38. Oetker-Black, S. L., et al., Preoperative teaching and hysterectomyoutcomes. Aorn J. 77(6):1215–1218, 1221–1231, 2003.

39. Shon, K. H., and Park, S. S., Medication and symptom managementeducation program for the rehabilitation of psychiatric patients inKorea: The effects of promoting schedule on self-efficacy theory.Yonsei. Med. J. 43(5):579–589, 2002.

40. 33(8), 2003.41. Caron, F., Godin, G., et al., Evaluation of a theoretically based

AIDS/STD peer education program on postponing sexual inter-course and on condom use among adolescents attending highschool. Health Educ. Res. 19(2):185–197, 2004.

42. Helfand, M., Using evidence reports: Progress and challengesin evidence-based decision making. Health Aff. (Millwood)24(1):123–127, 2005.

43. Nicklin, W., and Stipich, N., Enhancing skills for evidence-basedhealthcare leadership: The Executive Training for Research Ap-plication (EXTRA) program. Can. J. Nurs. Leadersh 18(3):35–44,2005.

44. Mitton, C., and Patten, S., Evidence-based priority-setting: whatdo the decision-makers think? J. Health Serv. Res. Policy 9(3):146–152, 2004.

45. Browman, G. P., Snider, A., and Ellis, P., Negotiating for change.The healthcare manager as catalyst for evidence-based practice:Changing the healthcare environment and sharing experience.Healthc. Pap. 3(3):10–22, 2003.

46. Scott’s Government, Index Scott’s Government Index, SouthamInformation Products Group, Don Mills, Ont., 2001.

47. Ahire, S., and Devaray, S., An empirical comparison of Statisti-cal Construct Validation Approaches. IEEE Trans. Eng. Manage.48(3):319–329, 2001.

Springer

Research-Based-Decision-Making in Canadian Health Organizations: A Behavioural Approach

Documents