challenges of adopting the role of care manager when

REVIEW ARTICLE

Challenges of adopting the role of care managerwhen implementing the collaborative care modelfor people with common mental illnesses:A scoping review

Ariane Girard,1 �Edith Ellefsen,1 Pasquale Roberge,2,3 Jean-Daniel Carrier4 andCatherine Hudon2,31School of Nursing, 2Department of Family Medicine and Emergency Medicine, 3CHUS Research Centre, and4Department of Psychiatry, Faculty of Medicine and Health Sciences, University of Sherbrooke, Sherbrooke,Quebec, Canada

ABSTRACT: This review aimed to identify the main factors influencing the adoption of the roleof care manager (CM) by nurses when implementing the collaborative care model (CCM) forcommon mental illnesses in primary care settings. A total of 19 studies met the inclusion criteria,reporting on 14 distinct interventions implemented between 2000 and 2017 in five countries. Twocategories of factors were identified and described as follows: (i) strategies for the CCMimplementation (e.g. initial care management training and supervision by a mental healthspecialist) and (ii) context-specific factors (e.g. organizational factors, collaboration with teammembers, nurses’ care management competency). Identified implementation strategies were mainlyaimed towards improving the nurse’s care management competency, but their efficacy indeveloping the set of competencies needed to fulfil a CM role was not well demonstrated. There isa need to better understand the relationship between the nurses’ competencies, the caremanagement activities, the strategies used to implement the CCM and the context-specific factors.Strategies to optimize the adoption of the CM role should not be solely oriented towards theindividual’s competency in care management, but also consider other context-specific factors. TheCM also needs a favourable context in order to perform his or her activities with competency.

KEY WORDS: care manager, collaborative care model, implementation, nurses, primary care.

INTRODUCTION

The collaborative care model (CCM) is a well-knownand effective model of care for the treatment of peoplewith common mental illnesses, such as anxiety anddepression (Un€utzer & Ratzliff 2015). More than 70randomized controlled trials (RCTs) have demonstratedthe effectiveness of the CCM in improving anxiety anddepressive symptoms compared to usual care (Archeret al. 2012). Beyond its positive effects on patients’health condition, implementing the CCM also has thepotential to improve access to mental health care inprimary care settings and has proven to be cost

Correspondence: Ariane Girard, �Ecole des Sciences Infirmi�eres,Facult�e de M�edecine et de Sciences de la Sant�e, Universit�e deSherbrooke, 3001, 12e Avenue Nord, Sherbrooke, Qu�ebec, J1H5N4, Canada. Email: [email protected] statement: All authors listed meet the authorship cri-teria according to the latest guidelines of the International Com-mittee of Medical Journal Editors, and all are in agreement withthe Manuscript.Declaration of conflict of interest: Authors declare no conflictof interest.

Ariane Girard, RN, MSc.�Edith Ellefsen, RN, PhD.Pasquale Roberge, PhD.Jean-Daniel Carrier, MD, FRCPC.Catherine Hudon, MD, PhD.Accepted February 03 2019.

© 2019 Australian College of Mental Health Nurses Inc.

International Journal of Mental Health Nursing (2019) ��, ��–�� doi: 10.1111/inm.12584

bs_bs_banner

mailto:

effective (Gilbody et al. 2006; Un€utzer & Ratzliff2015). The CCM can be defined as a multiprofessionalapproach where a team of providers, such as a generalphysician (GP), a care manager (CM), and a mentalhealth specialist (often a psychiatrist), work togetherusing a structured management plan based on evi-dence-based practice and the specific needs of individ-ual patients (Coventry et al. 2014; Gunn et al. 2006;Ratzliff et al. 2016). The CCM is derived from thechronic care model and showcases four essential char-acteristics: it is (i) Team driven; (ii) Population focused,that is the team is responsible for a defined populationof patients; (iii) Measurement guided, that is the teamuses outcome measures to drive clinical decision-mak-ing; and (iv) Evidence based, that is the team adaptsscientifically proven treatments to an individual’s clini-cal situation in order to achieve improved health out-comes (American Psychiatric Association & Academy ofPsychosomatic Medicine 2016).

In the past 20 years, the CCM has been widely imple-mented and tested throughout numerous initiativesacross the USA and in some European countries(Un€utzer & Ratzliff 2015). In Canada, the CCM hasgained the attention of researchers and stakeholders, butfew scientific data on its implementation are available(Sunderji et al. 2016). Even if the CCM is associated withpositive outcomes, its implementation in real-world set-tings outside of clinical trials remains a challenge.Indeed, many factors can influence the implementationof an intervention in healthcare organizations, includingthe characteristics of the individuals involved, the innerand outer settings, the characteristics of the interventionitself, and the implementation process (Damschroderet al. 2009).

Two recent systematic reviews have described themain barriers and enablers to the implementation of theCCM in primary care (Overbeck et al. 2016; Wood et al.2017). Both reviews concluded that three of the main bar-riers encountered were the reticence of primary care pro-viders to adopt the CCM, not knowing the differencebetween what is already done in practice for the manage-ment of depression and the CCM, and providers’ remu-neration (Overbeck et al. 2016; Wood et al. 2017). Onthe other hand, the presence of a leader in the team, hav-ing a GP with an interest in mental health and involvingstakeholders in the implementation process were enablersof CCM implementation (Overbeck et al. 2016; Woodet al. 2017). Among identified barriers and enablers,many were related to the adoption of the CM role.

Care management is a core CCM component(Blasinsky et al. 2006). This role is often played by a

healthcare professional in collaboration with a team ofprimary care providers and mental health specialists.The CM ensures a link exists between patients andproviders. More specifically, the CM role includes vari-ous activities, such as screening and assessing patientsfor anxiety and depression; monitoring patient symp-toms and medication side effects and planning for fol-low-up; providing brief behavioural interventions usingevidence-based techniques (e.g. behavioural activation,motivational interviewing, problem-solving training);providing education to the patient and his family oncommon mental illnesses and a healthy lifestyle; ensur-ing care coordination and facilitating referrals for clini-cally indicated services outside the organization;communicating any significant information regardingthe patient’s health condition to the team of providers;and participating in regularly scheduled caseload con-sultations with the team’s psychiatrist (AIMS Center-University of Washington 2018).

According to the experience of some patients andclinicians with care management, there are advantagesto having a competent and dedicated professional pro-viding a proactive follow-up. First, the patient receivesemotional support and has someone to talk to, hasincreased motivation to achieve personal goals and hasbetter information on his specific condition (Bennettet al. 2013; Levine et al. 2005; Lipschitz et al. 2017).Second, the CM can also improve continuity and effec-tiveness of care and contributes to promote collabora-tion among providers (Blasinsky et al. 2006; Nuttinget al. 2008).

The CM can be a mental health nurse but is alsooften a primary care nurse (PCN) because of theirexperience with the management of people with long-term diseases and the holistic nature of their role(Webster et al. 2016). Patients seem to appreciatewhen the CM is a nurse who works closely with theirGP (Bennett et al. 2013). In their day-to-day practice,PCNs usually have activities similar to those of the CMwhen caring for people with physical long-term diseaseand common mental illnesses (Girard et al. 2017; Poi-tras et al. 2018). A meta-analysis has also demonstratedthe positive impact on mental health conditions forpatients with both depression and physical long-termdiseases when nurses play the CM role (Ekers et al.2013). However, even if PCNs seem to be good candi-dates to fulfil the CM role, there are still many chal-lenges to the adoption of the role when implementingthe CCM in primary care. Given that the CM role isan essential CCM component, it is important to knowwhat might facilitate or challenge its adoption in order


2 A. GIRARD ET AL.

to improve the model’s implementation and adaptationin primary care settings.

AIM

This study aimed to identify the main factors influenc-ing the adoption of the role of CM by PCNs whenimplementing the CCM in primary care settings.

METHODOLOGY

The use of a scoping review method was deemed themost relevant approach to reach this study’s aim. Ascoping review can be defined as ‘[. . .] a process ofsummarising a range of evidence in order to conveythe breadth and depth of a field’ (Levac et al. 2010, p.1). Since the development of the CCM in the 1990sby Katon and colleagues (Katon et al. 1995), a widerange of publications on CCM implementation hasbeen issued clinical trials and quasi-experiments, quali-tative or mixed methods studies, systematic reviews,opinion papers, descriptions of specific initiatives vary-ing from local to national scopes, book chapters, etc.The role of CM is not always the focus of these publi-cations, but they generally offer relevant insights intothe challenges of adopting this role when implement-ing the CCM in primary care. Therefore, any attemptto visualize the main factors influencing the adoptionof the CM role would benefit from taking into accounta variety of publication types. The scoping reviewmethodology of Arksey and O’Malley (2005) was usedto guide the elaboration of this review. This method isdivided into five stages: (i) Stating the research ques-tion; (ii) Identifying relevant studies; (iii) Study selec-tion; (iv) Charting the data; and (v) Collating,summarizing, and reporting results (Arksey & O’Malley2005). Additionally, Levac et al. (2010) published apaper clarifying and expanding on these five stages.Moreover, a sixth stage aiming to consult stakeholdersabout the results was proposed as optional by Arkseyand O’Malley, although Levac et al. (2010) describe itas an essential component of a scoping review. Thestakeholder’s consultation stage pertaining to this scop-ing review is not reported here, as it will be done inanother project currently in progress by the researchteam.

Research question

The research question for a scoping review is usuallybroadly stated, but it minimally needs to be

informative on the concept of interest and the targetpopulation for the study in order to allow for a clearunderstanding of the focus of the review and toestablish an effective search strategy (Levac et al.2010). The research question for this review is Whatare the main factors influencing the adoption of theCM role by nurses when implementing the CCM inprimary care settings for the treatment of adults withanxiety and/or depression?

Identifying relevant studies

The search strategy to identify relevant studies wasdeveloped in collaboration with an information spe-cialist and included three databases: CINAHL,PsycINFO, and MEDLINE. Keyword selection wasbased on the systematic reviews of Wood et al.(2017) and Coventry et al. (2014) and included threeconcepts: Common Mental Disorders (anxiety ordepression), Implementation, and the CCM (seeTable 1 for the complete search strategy). The key-words ‘Nursing’ and ‘Primary Care’ were tested inthe search strategy, but the number of publicationsdropped significantly when adding either of them.The research team therefore decided to remain asbroad as possible in order to include important stud-ies. Moreover, given that the CCM stems fromresearch and was known to be widely documented inpeer-reviewed journals, the team decided to notinclude grey literature in the search strategy and onlyused published articles. The search strategy includedpublications until June 2018.

TABLE 1: Search strategy

Common mental

illnesses [AND] Implementation [AND] Collaborative care

AB ((‘common

mental disorder*’OR ‘common

mental illness*’OR ‘panic disorder*’OR ‘panic attack*’OR agoraphobi*OR ‘social anxiety

disorder*’ OR ‘social

phobia’ OR depress*OR dysthymi*)OR (anxi*N3 (sympt* or

ill* or disease*or condition*or disorder*)))

AB (implement*OR ‘quality

improvement’

‘process*evaluation’

OR feasibilit*OR barrier*OR difficult*OR enabler*OR facilitat*OR adopt*)

AB (‘integrat* care’

OR ‘collaborati*care’ OR ‘case

manage*’ OR

‘care manage*’)


NURSE CARE MANAGER AND IMPLEMENTATION OF THE COLLABORATIVE CARE MODEL 3

Study selection

Two authors (AG and JDC) independently screenedtitles and abstracts to include articles reporting studiesabout CCM implementation in primary care clinics forthe treatment of adults with depression and/or anxiety.Studies were excluded during abstract screening if they(i) reported solely the authors’ opinion or were an exper-imental study reporting only clinical results or (ii) wereconducted in a specialized setting (e.g. HIV clinics) ortargeted a specific clientele (e.g. children or adolescents,post-traumatic stress, or severe mental illness patients)or (iii) if the publication language was neither Englishnor French. The authors met after screening theabstracts of the first 150 and 300 studies to reach consen-sus for all studies and to refine selection criteria whererelevant. For instance, the criterion ‘primary care set-ting’ was updated to ‘medical clinic in primary care’. Asindicated in Fig. 1, after removing duplicates, a total of975 titles and abstracts were screened and the full textsof 104 studies were obtained for analysis. The same twoauthors independently read the 104 studies and metonce to discuss discrepancies until consensus for inclu-sion was reached. A total of 19 studies met inclusion cri-teria at this stage: (i) at least three CCM components;

(ii) at least some of the CMs were nurses; (iii) the settingwas medical clinic in primary care; (iv) studies reportingon CCM implementation and adoption of the CM role;and (v) the CCM targeted adults with anxiety and/ordepression.

RESULTS

Charting the data

The research team first discussed important variablesto extract from individual studies. The process of chart-ing the data was iterative and conducted throughoutthe analysis process. All studies were linked to a speci-fic CCM intervention, which was implemented in aspecific context. Table 2 summarizes studies’ maincharacteristics and the CCM interventions. Qualityassessment of the included studies is not a commonstage of a scoping review and was not undertaken inthis review. The aim of a scoping review was to presentan overview of existent and relevant literature on aresearch topic regardless of methodological quality orrisk of bias and allow a more complete examination ofall types of research activity (Pham et al. 2014; Triccoet al. 2018).

Articles identified from: MEDLINE, PsychINFO and CINAHL (n = 1481)

Duplicates removed(n = 506)

Title and abstract screened (n = 975)

Did not meet inclusion criteriain neither English nor French

(n = 871)

Full text screened (n = 104)

Exclusion from full text (n =86). Reasons: No nurse among care managers, not compatiblewith collaborative care model, experimental design withoutdata on implementation, doesnot include medical clinic in primary care, does not target

adults with anxiety and/ordepression.

Included for analysis(n = 19)

Articleincluded fromreference listscreen (n = 1)

FIG. 1: PRISMA flow chart.


4 A. GIRARD ET AL.

TABLE

2:

Description

ofselected

stud

ies

Authors,year/cou

ntry

ofim

plemen

tation

Nam

eof

theoriginal

interven

tion

/number

of

sites/totalnumber

ofCMstrained

Healthconditions

targeted

bythe

interven

tion

Aim

ofthestudy

Studysample

Design/datacollection

Mainfactorsrelatedto

the

adop

tion

oftherole

ofCM

Ben

nettet

al.(2013)/

UK

Pro-ActiveCareand

itsevaluationforEnduring

Dep

ressiontrial(ProCEED)1

42general

practices

42CMs(nurses)

Dep

ression(twoor

moredocumen

ted

episod

esof

dep

ression

within

thepreviou

s3years)

[...]explore

both

patients’and

practicenurses’

perspectiveson

theirexperience

ofreceivingand

providing

proactive

care

(p.2)

15CMs26

patients

Qualitativeapproach/

semi-structuredinterview

(faceto

face)afterthe

trialof

theProCEED

interven

tion

Com

peten

cyof

theCM

incare

managem

ent(m

otivation,

clinical

experience

inmen

tal

health/level

ofcomfort,

empathy,

capacityto

support

thepatienten

gagingin

achange

ofbeh

aviourbyusingan

eviden

ce-based

approach,

patiented

ucation

)

Collaborationwithteam

mem

bers(relationship

with

patients

andGPs)

Blasinskyet

al.(2006)/

USA

ImprovingMood:

PromotingAccessto

CollaborativeTreatmen

t

(IMPACT)2

18primarycare

clinics

(7sitesrepresentedby

diverse

healthcare

organizations)

17CMs(nurses

or

psychologists)

Late-lifedep

ression

[...]discuss

the

issues

involved

in

sustainingthemod

el

inaprimarycare

practice(p.719)

15participants:mem

bersof

theresearch

team

,GPs,CMs,

supervisingpsychiatrist,

program

mecoordinator,

recruiter


review

ofgrantproposals

that

described

the

interven

tion

-as-planned

toestablishthecompon

ents

oftheprogram

me,

site

visits

attwopoints

during

implemen

tation

,semi-structured

interview

(telep

hon

e)1year

aftertheinterven

tion

IMPACTen

ded

Organization-related

factors

(workloadin

primarycare,

prioritiesof

theleaders,

remunerationof

theCMs)

Colem

anet

al.(2017a)/

USA

Careof

Men

tal,Physical,

andSubstance-use

Syndromes

(COMPASS)3

172clinics(18healthcare

system

sacross

8states)N

umber

ofCMstrained

not

specified

(registered

nurses,licen

sedpractical

nurses,social

workers,

med

ical

assistants,physician

assistants,nursepractitioners,

men

talhealthpractitioners)

Dep

ressionwithpoorly

controlleddiabetes

and/orcardiovascular

disease

[...]describes

the

COMPASScare

managers’perception

s

oftheprogram

,their

perceived

role

inthe

program

,andthe

relation

ship

between

care

manager

characteristics,their

care

practices

and

control

ofthreekey

patienthealthou

tcom

es

fortheprogram

:

dep

ression,diabetes

andhypertension

(p.87)

93CMs(55%

werenurses)

Quantitative

approach/O

nlin

e

survey

9–18

mon

thsafter

beginningtheim

plemen

tation

ofCOMPASS


factors

(com

plexity

oftheclientele)


mem

bers(com

munication,

relation

ship

withpatients)

Com

peten

cyof

theCM

(clin

ical

experience

in

men

talhealthandcare

managem

ent,capacityto

supportthepatient

engagingin

achange

of

beh

aviourbyusingan

eviden

ce-based

approach,

know

ledge

ofpharmacological

treatm

entandchangingbeh

aviours)

Strategies(training)

(Con

tinu

ed)



TABLE

2:

(Con

tinu

ed)

Authors,year/cou

ntry

ofim

plemen

tation

Nam

eof

theoriginal

interven

tion

/number

of

sites/totalnumber

ofCMstrained

Healthconditions

targeted

bythe

interven

tion

Aim

ofthestudy

Studysample



the

adop

tion

oftherole

ofCM

Colem

anet

al.(2017b

)/

USA

COMPASS

Idem

Idem

Dep

ressionwithpoorly

controlleddiabetes

and/orcardiovascular

disease

Todescribeanational

effort

to

disseminateandim

plemen

t

aneviden

ce-based

collaborative

care

managem

entmod

el[...]

across

multiple,real-w

orld

diverse

clinical

practice

sites(p.69)

N/A

‘[...]description

ofthe

COMPASSinitiative

brough

t

together

bydiverse

healthcare

organisationsacross

theUnited

Statesin

order

toim

plemen

tan

efficaciou

smod

elof

care

for

patients

withdep

ressionand

uncontrolleddiabetes

and/or

cardiovasculardisease’.(p.70)

Strategies(supervision

,training)


mem

bers

(roleclarification

)

Curran

etal.(2012)/

USA

Coordinated

AnxietyLearning

andManagem

ent(C

ALM)4

17clinics(w

ithin

four

diverse

healthcare

organizations

includingprimarycare

clinics)

14CMs(five

nurses,sixsocial

workers,threepsychologists)

Gen

eralized

anxiety

disorder,panic

disorder,

post-traumatic

stress

disorder,and/orsocial

anxietydisorder

Iden

tify

facilitators/

barriersto

implemen

tingand

sustainingtheCALM

interven

tion

14CMs

18GPs

16administrators

13PCNs


semi-structuredinterviews

(faceto

face)duringthefinal

year

oftheRCT(excep

tfor

someCMswhowere

interviewed

atthemid-point

andat

theconclusion

oftheinterven

tion

)

Com

peten

cyof

theCM

(warm,

engaging,

visible,motivation)


factors

(workloadin

primarycare,resources,

complexity

oftheclientele)


mem

bers

(logistics,location

oftheCM,

relation

ship

withpatients)

Gasket

al.(2006)/

USA

ThePathwaysStudy5

nineGroupHealthCooperative

primarycare

clinicsin

western

Washington

threeCMs(nurses)

People

withdiabetes

whoalreadyreceiving

antidep

ressantmed

ication

orpsychotherapyfrom

non

psychiatristclinicians,

butwhostill

had

high

dep

ressionscores

(PHQ-9

≤10).

[...]explore

what

happen

edin

the

interactionsbetween

nurses

andpatients

with

bothdep

ressionand

diabetes

duringthe

Pathways-Study(p.232)

ThreeCMs

25patients


contentanalysisof

records

of30

sessionsbetweena

CM

andapatient:recording

ofsessionswas

routinely

madebyCMsforpurposes

ofsupervision

and

fidelityassessmen

t


factors

(com

plexity

oftheclientele)

Com

peten

cyof

theCM

(clin

ical

experience

inmen

talhealth,capacity

tosupportthepatientin

engaging

achange

ofbeh

aviourbyusing

aneviden

ce-based

approach)

Gasket

al.(2010)/

UK

Exploratory

RCTof

collaborativecare

for

dep

ressionin

theUK6

fourprimarycare

organizationsin

the

northernUK

EightCMs(twomen

tal

healthnurses,on

e

counsello

r,five

graduate

men

talhealthworkers)

Dep

ression

Apply

thenormalization

process

mod

el(N

PM)to

inform

theprocess

of

implemen

tation

of

collaborativecare

inboth

future

full-scaletrials,

andthewider

healthcare

settings

follo

wingthetrial

EightCMs

11patients

12GPs

fourPCNs

fourpsychiatrists

fourpsychologists

six

men

talhealthworkers



(faceto

face)andfocus

grou

pbeforeandaftertheRCT


mem

bers(relationship

withGPs,

role

clarification

,interprofessional

respect)


factors(type

ofclientele

inprimarycare)


)

deJonget

al.(2009)/

NL

TheDep

ression

initiative

(NL)

78GPsworkingin

20

healthcare

centres

participated

inthetrial

nineCMs(twopractice

nurses,fourcommunity

psychiatric

nurses,three

social

workers)

Major

dep

ressive

disorder

Description

ofa

collaborativecare

mod

el

formajor

dep

ressive

disorder

andof

the

factorsinfluen

cingits

implemen

tation

in

primarycare

settings

intheNetherlands(p.1)

N/A

TheIM

PACTcollaborative

care

mod

elwas

adapted

fortheNetherlands

Com

peten

cyof

theCM

(clin

ical

experience

inprimarycare,capacity

toapply

abrief

psychological

interven

tion

,

specifickn

owledge

ontreatm

ents,motivation)


mem

bers

(com

munication,location

oftheCM)

(Con

tinu

ed)


6 A. GIRARD ET AL.

TABLE

2:

(Con

tinu

ed)

Authors,year/cou

ntry

ofim

plemen

tation

Nam

eof

theoriginal

interven

tion

/number

of

sites/totalnumber

ofCMstrained

Healthconditions

targeted

bythe

interven

tion

Aim

ofthestudy

Studysample



the

adop

tion

oftherole

ofCM

Levineet

al.(2005)/

USA

IMPACT

18primarycare

clinics

(seven

sitesrepresented

bydiverse

healthcare

organizations)

17CMs(nurses

or

psychologist)and490GPs

Late-life

dep

ression

[...]describephysicians’

satisfaction

withcare

for

patients

withdep

ression

beforeandafterthe

implemen

tation

ofa

primarycare-based

collaborativecare

program

(p.383)

450GPs

Quantitative

approach/

auto-rep

ortedsurvey

before

and12

mon

thsafterthe

interven

tion


mem

bers

(com

municationbetweenGPsandtheCM)

Lipschitzet

al.(2017)/

USA

TheDep

artm

entof

VeteransAffairs

(VA),

alon

gwithother

healthcare

system

s,has

adop

teda

general

‘collaborativeapproach’

tohealthcare.Som

esites

havechosen

toim

plemen

ta

mod

elthat

addsded

icated

telephon

e-based

dep

ression

care

managersto

theteam

ofem

bed

ded

men

talhealth

practitioners

VAmed

ical

centres

(number

ofded

icated

CM

sitesvs

embed

ded

sitesare

notegiven)

Number

ofCMstrained

not

specified

(CMsare

oftenanurse)

Dep

ression

Com

pared

asite

that

implemen

tedded

icated

CM

toasite

that

had

implemen

tedthe

embed

ded

mod

el(w

ithou

t

ded

icated

CM):(i)What

doesaded

icated

care

manager

offerin

addition

toan

embed

ded

mod

el?’

(p.2)

(ii)What

arethe

barriersto

implemen

ting

aded

icated

care

manager?(p.3)

CM

site:

twoCMs

threeprimarycare

clinicians

twoleaders,two

men

talhealthstaff

EMBED

site:twocare

managem

entstafffrom

atelephon

e-based

program

meoneleader

twoprimary

care

clinicians

onemen

talhealthstaff

Qualitativeapproach/sem

i-

structuredinterviews(telep

hon

e)


mem

bers

(logistics,location

oftheCM,

communicationandrelation

ship,

role

clarification

)


factors(priorities

oftheleaders)

Com

peten

cyof

theCM

(warm,

communicationskills,patiented

ucation

)

Mølleret

al.(2018)/

DNK

TheCollabriMod

el7-8

24GPclinicsin

the

capital

region

ofDen

mark

eigh

tCMs

(sixpsychiatric

nurses,on

e

occupational

therapist,and1health

visitorin

theUK,both

withpsychiatric

experience)

Anxietyand

dep

ression

[...]explore

theexperiences

withcurren

ttreatm

ent

practices

amon

gGPs,clinic

staffandCMsandto

exam

inetheirview

son

and

perception

sof

future

collaborativecare.Iden

tify

enablers

andbarriersfor

successfulim

plemen

tation

ofaspecificDanish

collaborativecare

mod

el[...](p.2)

Singlecase

studyof

theCollabriMod

el:

twoCMs

oneGP

Multipracticestudy:

10GPs

threePCNs

onereception

ist

Researcherscombined

a

multiple

case

studyof

GPs’practiceandtheir

staffwithcurren

tmen

tal

healthtreatm

entanda

singlecase

studyof

the

CollabriMod

el

Datawerecollected

through

outtheim

plemen

tation

:

directob

servations,

semi-structuredinterviews,

ethnographic

conversation,

fieldnotes


mem

bers(location

oftheCM,logistic,interprofessional

respect)

Com

peten

cyof

theCM

(clin

ical

experience

inmen

talhealth,capacityto

adapthim

/herself)


factors(w

orkloadin

primarycare,complexity

oftheclientele)

Strategies(training,

supervision

) (Con

tinu

ed)



TABLE

2:

(Con

tinu

ed)

Authors,year/cou

ntry

ofim

plemen

tation

Nam

eof

theoriginal

interven

tion

/number

of

sites/totalnumber

ofCMstrained

Healthconditions

targeted

bythe

interven

tion

Aim

ofthestudy

Studysample



the

adop

tion

oftherole

ofCM

Murphyet

al.(2014)/

UK

Service

evaluationof

aPCN-led

collaborative

care

initiative

inrural

North

EastEngland

that

initially

reported

outcom

esin

2008

9

eigh

tGPpractices

13CMs(nurses)

Mod

erate-to-severe

dep

ression

[...]investigateto

what

degreetheservicewas

maintained

over

time

andto

what

degree

dep

ressionsymptom

levelsof

patients

being

follo

wed

up

improved(p.828)

Six

CMs

Clin

ical

data

from

218patients

Pre–p

oststudy/statistical

analysisof

dep

ression

symptom

levelscores

(PHQ-9

outcom

es)to

explore

within-groupchange

and


(telep

hon

e)4yearsafter

theserviceim

plemen

tation

Com

peten

cyof

theCM

(perception

of

limited

skillsbynurses)


)

Nuttinget

al.(2008)/

USA

Re-Engineering

Systemsforprimary

CareTreatmen

tof

Dep

ression

(RESPECT-D

epression)10

60clinics(5

diverse

healthcare

organizations)

Total

number

of

CMstrained

not

specified

(nurses,

psychologists,

social

workers)

Dep

ression

Toexam

inethebarriers

toadop

tingdep

ression

care

managem

entam

ong

primarycare

cliniciansandCMs

18CMs

42primarycare

providers(G

Ps,PCNs,

nursepractitioners,

med

ical

assistant),

sevenmen

talhealth

specialists

(psychiatrist,

psychologist)

Qualitativeapproach/sem

i-

structuredinterviews

(telep

hon

e)duringthe

disseminationphase,

6mon

thsafterthetrial

oftheRESPECT-D

epression


factors(rem

uneration

oftheCM,workloadin

primarycare,

complexity

oftheclientele)


mem

bers(location

oftheCM,communication)


)

Nuttinget

al.(2007)/

USA

RESPECT-D

epression

Idem

Idem

Dep

ression

Tounderstandthe

characteristicsof

organizationsandthe

interven

tion

compon

ents

that

wereassociated

with

implemen

tation

and

disseminationof

the

RESPECT-D

epression

compon

ents

Idem

Idem


mem

bers

(locationof

theCM,relation

ship

withtheteam

)


factors(w

orkload

inprimarycare,remunerationof

theCM,

prioritiesof

theleaders,complexity

oftheclientele)


)

Overbecket

al.(2018a)/

DNK

TheCollabrimod

el

24GPclinicsin

the

capital

region

of

Den

mark

eigh

tCMs(six

psychiatric

nurses,

oneoccupational

therapist,andon

e

healthvisitorin

the

UK,bothwith

psychiatric

experience)

Anxietyand

dep

ression

Toexplore

CMs’

experience

oftheir

workandthechallenges

they

face

when

implemen

tingtheirrole

inacollaborativecare

interven

tion

intheCapital

Regionof

Den

mark(p.167)

EightCMs



duringthetrialof

the

Collabrimod

el


mem

bers

(locationof

theCM,role

clarification

,

relation

ship

andcommunicationwith

theGPsandother

primarycare

providers)

Com

peten

cyof

theCM

(motivation,

capacityto

adapthim

/herself,leadership,

clinical

experience

inmen

talhealth/

psychiatryandprimarycare,1year

trainingin

cogn

itivebeh

aviouraltherapy)


)

(Con

tinu

ed)


8 A. GIRARD ET AL.

TABLE

2:

(Con

tinu

ed)

Authors,year/cou

ntry

ofim

plemen

tation

Nam

eof

theoriginal

interven

tion

/number

of

sites/totalnumber

ofCMstrained

Healthconditions

targeted

bythe

interven

tion

Aim

ofthestudy

Studysample



the

adop

tion

oftherole

ofCM

Web

ster

etal.(2016)/

UK

Aservice

develop

men

twith

theaim

oftraining

practicenurses

to

deliver

brief

beh

avioural

activation

interven

tion

s

within

acollaborative

care

fram

ework

Oneprimarycare

clinic

intheNorth

ofEngland

threeCMs(nurses)

Patients

with

dep

ressionandon

e

ormorelong-term

healthconditions

[...]to

exam

ine

poten

tial

barriersand

facilitatorsto

engaging

withtheinterven

tion

from

thepatientand

clinicianperspective

inorder

togu

ide

future

service

develop

men

tand

research

inthisarea

(p.3)

ThreeCMs

fourpatients

five

GPs

onehealthcare

assistant

onemen

tal

healthspecialist

Pilo

tstudyusingaqualitative

approach/sem

i-structured

interviews(faceto

face)more

than

2mon

thsafterthe

implemen

tation

oftheservice


factors(priorities

oftheleaders,workloadin

primarycare,

complexity

oftheclientele)


mem

bers

(relationship

withpatients)

Com

peten

cyof

theCM

(motivation,

feelingconfiden

twiththeirskillsin

deliveringapsychosocialinterven

tion

)

Strategies(training,

supervision

)

Whiteb

irdet

al.(2014)/

USA

Dep

ression

Improvemen

tacross

Minnesota-Offeringa

New

Direction

(DIA

MOND)11

99clinics(21

differenthealthcare

organizations)

99CMs(registered

nurses,licen

sedpractical

nurses,certified

med

ical

assistants,clinical

social

workers)—

theclinics

hired

theCM

they

wanted

Dep

ression

Toiden

tify

thecare

mod

el

factorsthat

werekeyfor

successfulim

plemen

tation

of

collaborativedep

ressioncare

in

astatew

ideMinnesotaprimary

care

initiative

(p.699)

42clinics

Mixed

method

design

incorporatingboth

qualitativedatafrom

clinic

site

visits

(DIA

MOND

datareportstrategy

grou

p

usedin

implemen

tation

;

adiscussiongu

idefocused

onbarriersandfacilitators;

narrative

meeting,

field

notes)andquantitative

measuresof

patient

activation

(PHQ-9

≥10)

and6-mon

thremission

(PHQ-9

<5)

Caremod

elfactors

iden

tified

from

thesite

visits

weretested

for

associationwithratesof

activation


mem

bers

(locationof

theCM,role

clarification

)

Com

peten

cyof

theCM

(leadership,available)


)

Williamset

al.(2011)/

USA

DIA

MOND

Idem

Idem

Dep

ression

Examined

andcompared

the

firsttwoclinicsim

plemen

ting

DIA

MOND

atMayo

TwoMayofamily

clinics

Enrolledpatients

(n=247,

n=219)

Retrospective

study/

Outcom

esmeasures

betweenthetwoclinics

werecompared

asfollo

ws:

Percentage

ofpatients

‘activated

’into

CCM;

6-mon

thdep

ression

remission

rates;PHQ-9,Mood

disordersquestion

naire,Alcoh

ol

UsedDisorders

Iden

tification

test,

andGAD-7

scores;

andthedropou

trates

Theanalysisalso

includes

number

ofCMs(fulltime)


mem

bers(location

oftheCM,relation

ship

betweentheCM,

thepatient,andtheprimarycare

providers)

(Con

tinu

ed)



TABLE

2:

(Con

tinu

ed)

Authors,year/cou

ntry

ofim

plemen

tation

Nam

eof

theoriginal

interven

tion

/number

of

sites/totalnumber

ofCMstrained

Healthconditions

targeted

bythe

interven

tion

Aim

ofthestudy

Studysample



the

adop

tion

oftherole

ofCM

Wozniaket

al.(2015)/

CAN

TeamCARE12

fournon

metropolitan

primarycare

networks

inAlberta

fourCMs(fou

rPCNs

trained

asCM)

Type2diabetes

anddep

ression

[...]describethedegreeof

implemen

tation

fidelityof

the

TeamCareinterven

tion

atthe

organizational

level,including

thedeliveryof

interven

tion

compon

ents,to

determine

whether

itsexecution

affected

itseffectiven

ess(p.84)

36participants:

administrators,specialists,

CMs,other

staff

Mixed

method

design/D

atawere

collected

through

out

theim

plemen

tation

:

Sem

i-structuredinterview

(faceto

face),grou

p

meeting,

documen

treview

,fieldnotes

Com

peten

cyof

theCM

(ability

tolearn

quickly,

effectivecommunicationskills,

beingmotivated

,capable

orconfiden

t,

beingadaptable

andwellorganized

,

clinical

experience

inmen

talhealth,

andcollaborativecare)


mem

bers

(locationof

theCM,communication

andrelation

ship,interprofessional

respect)


,training)

CM,care

manager;CCM,collaborativecare

mod

el;GP,general

physician;PCN,primarycare

nurse;

PHQ-9,PatientHealthQuestion

naire-9

item

;GAD-7,generalized

anxietydisor-

ders-7item

.

Inadditionto

thestudiesincluded

inthisreview

,thefollo

wingpublicationswereconsulted

inrelation

tothecontext

ofim

plemen

tation

ofCCM

interven

tion

s:

1.Buszew

iczet

al.(201

0).

2.Un€utzer

etal.(200

1).

3.Rossom

etal.(201

7).

4.Roy-Byrneet

al.(201

0).

5.Katon

etal.(200

3).

6.Richardset

al.(200

8).

7.Brinck-C

laussen

etal.(201

7).

8.Keh

lerCurthet

al.(201

7).

9.Ekers

andWilson

(200

8).

10.Dietrichet

al.(200

4).

11.Institute

forClin

ical

SystemsIm

provemen

t(201

4).

12.Johnsonet

al.(201

2).


10 A. GIRARD ET AL.

Characteristics of the CCM interventions

Across the 19 selected studies, a total of 14 individualCCM interventions implemented in five countriesbetween 2000 and 2017 were identified. The imple-mentation contexts of these 14 CCM interventionswere not always described, and some data were diffi-cult to find (e.g. number of CMs trained, number ofclinics where the CCM intervention was implemented,initial training of the CM, location of the CM). Almostall CCM interventions included in this scoping reviewtargeted all adults (18 years and older), except for theIMPACT intervention which focused on older people(60 years and older). The health conditions targeted bythe CCM interventions varied as follows: depression(n = 8), depression with long-term conditions, such asdiabetes and cardiovascular disease (n = 4), anxiety(n = 1), and anxiety and depression (n = 1).

The CM was always central to the intervention. TheCM activities were generally similar between studies,except for variations in the specific psychosocial or psy-chotherapeutic interventions delivered to support beha-viour change.

Among the CCM interventions, five hired onlyPCNs (Bennett et al. 2013; Gask et al. 2006; Murphyet al. 2014; Webster et al. 2016; Wozniak et al. 2015).Half of the CCM interventions (n = 8) included PCNsand CMs trained from a variety of mental health disci-plines (e.g. psychologist, social worker, other mentalhealth worker, community psychiatric nurse, mentalhealth nurse) (Blasinsky et al. 2006; Coleman et al.2017a,b; Curran et al. 2012; de Jong et al. 2009; Gasket al. 2010; Levine et al. 2005; Møller et al. 2018; Nut-ting et al. 2007, 2008; Overbeck et al. 2018a; Whitebirdet al. 2014; Williams et al. 2011). In one CCM inter-vention, the initial training of the CM was not clearlyspecified (Lipschitz et al. 2017).

Generally, the process of hiring or choosing theright CM was described neither in the selected studiesnor in the other related studies cited below Table 2. Inthree of the CCM interventions, leaders of each orga-nization or GPs involved in the intervention wereappointed to hire the right CM (de Jong et al. 2009;Webster et al. 2016; Williams et al. 2011). Some stud-ies mentioned eligibility criteria for CMs, such as mini-mal experience in mental health, primary care nursing(Williams et al. 2011) or in care management of long-term conditions (Gask et al. 2006). In the Collabrimodel, eligibility criteria were more specific as theresearch team was looking for nurses with psychiatricor mental health experience and at least 1 year of

training in cognitive behavioural therapy (Møller et al.2018; Overbeck et al. 2018a). Information on the expe-rience and competency of the CMs beyond their initialtraining were seldom specified in the selected studies(Blasinsky et al. 2006; Levine et al. 2005; Lipschitzet al. 2017; Wozniak et al. 2015).

Collating, summarizing, and reporting results

Data were summarized with a thematic qualitativeanalysis using NVivo 11 software (Miles et al. 2014).All the selected studies were attentively read by thefirst author to identify relevant themes according to theresearch question. This step was followed by the con-struction of a thematic tree by regrouping emergentthemes into more general themes (i.e. main factors).One author (EE) independently analysed half of theselected studies to validate the themes identified bythe first author. Two authors (AG and EE) schematizedthe data (Fig. 2) to allow for visualization of thebreadth of the data and identification of relationshipsbetween emerging themes. At the end of the process,all authors agreed on the main concepts emerging fromanalysis. Two categories of factors which may influencethe adoption of the CM role were identified as follows:(i) strategies to implement CCM interventions, whichwere specifically geared towards adoption of the CMrole and (ii) factors specific to the implementation con-text. These two factors appear to be an important con-sideration based on the themes identified in thethematic qualitative analysis process.

Essential strategies

Two important strategies related to care managementwere mentioned in nearly all selected studies: theCM’s initial care management training and supervisionby a mental health specialist. These implementationstrategies of CCM interventions seemed to improve theadoption of the CM role.

The initial care management training was mentionedin all but one of the included studies (Lipschitz et al.2017). Training duration varied between 1.5 days and1 week. The content generally included a psychosocialor psychotherapeutic approach for the management ofanxiety or depression, such as behavioural activation(Coleman et al. 2017a,b; Webster et al. 2016; White-bird et al. 2014; Williams et al. 2011; Wozniak et al.2015), problem-solving therapy or techniques (Murphyet al. 2014; Webster et al. 2016), motivational inter-viewing (Bennett et al. 2013; Coleman et al. 2017a,b;



Curran et al. 2012; Whitebird et al. 2014; Williamset al. 2011; Wozniak et al. 2015) or cognitive beha-vioural strategies (Curran et al. 2012; Møller et al.2018; Overbeck et al. 2018a). Two CCM interventions(IMPACT and the Pathways Study) opted for an effec-tive and well-known training programme to teach clini-cians to deliver problem-solving therapy in primarycare (Blasinsky et al. 2006; Gask et al. 2006). The con-tent of the training also included an introduction to theCCM components specific to the intervention (e.g.tracking systems, treatments algorithms, relevant guide-lines), as well as an overview of pharmacological treat-ments for depression and/or anxiety, and occasionallysuicidal risk assessment (Murphy et al. 2014; Wozniaket al. 2015). In two studies, CMs received additionaltraining specific to their needs before and during theimplementation process (Coleman et al. 2017a,b; Woz-niak et al. 2015). For instance, Wozniak et al. (2015)specified that CMs asked for further training duringthe implementation process, and they were referred toexisting online training programs on specific topics:Diabetes Boot Camp, ASIST suicide Training, andChoices and Changes.

Various learning methods were used during training:role-playing (Blasinsky et al. 2006; Curran et al. 2012;Gask et al. 2006; Levine et al. 2005), didactic materials(Curran et al. 2012; Gask et al. 2006), observation of

videotape and feedback (Blasinsky et al. 2006; Gasket al. 2006; Levine et al. 2005; Nutting et al. 2007,2008), and case discussions (Bennett et al. 2013; Nut-ting et al. 2007, 2008; Wozniak et al. 2015).

Even though the importance of training CMs washighlighted, little data on the impact of such trainingwere reported. One study suggested that training inbehavioural activation has the potential to improve theconfidence of PCNs regarding the care of clients withmental health needs (Webster et al. 2016). Morespecifically, in the IMPACT study, CMs became morecompetent in delivering problem-solving therapy as thenumber of completed training sessions increased(Un€utzer et al. 2001).

In addition to initial training in care management,CMs had access to supervision by a mental health spe-cialist (e.g. psychologist, psychiatrist, or other mentalhealth professionals) or by a GP with an interest inmental health. The aim of the supervision was gener-ally to provide an opportunity for CMs to discuss theircases and concerns, or to receive recommendations onmedication management or possible changes in thepsychological approach. In de Jong’s study (2009),supervision was based on problem-solving therapy ses-sions recorded with patient permission.

Supervision frequency ranged from weekly (Curranet al. 2012; Gask et al. 2010; Nutting et al. 2007,

Competency of the CM in care managementMotivation to work with people with mental health issues and desire to help themClinical experiences in mental health, collaborative care, care managementProfessional skills:

Capacity to support the patient engaging in a change of behaviour by using evidence-based techniques (motivational interviewing, behaviouralactivation, problem-solving technique, cognitive behavioural strategies)Capacity to collaborate (work in team)

Interpersonal and personal skills: EmpathyCommunication skillsLeadershipBeing available and attentive to patients needsCapacity to adapt him/herself to different situations

Specific knowledge:Psychological and pharmacological treatments, psychosocialinterventions, community resources

Organization-related factorsWorkload and the diversity of the primary careclienteleComplex psychosocialproblems of the clienteleRemuneration of the CMLeaders' priorities

Collaboration with team membersLocation of the CMLogistics (time and space to meetwith the team)Communication between CMs and the teamRelationship between GPs, patients,and CMsInterprofessional respectRole clarification

Context-specific factors

Initial care management training Supervision by a mental health specialist

Implementat ion strategies of CCMinterventions ap pear ing essential to the

ad option of the CM role

Adoption of the CM role

FIG. 2: Main factors influencing the adoption of the CM role when implementing the CCM. CM, care manager; CCM, collaborative caremodel; GP, general physician.


12 A. GIRARD ET AL.

2008), twice a month (Gask et al. 2006), every 6 weeks(de Jong et al. 2009), to three or four times a year(Bennett et al. 2013). Generally, the specialists wereavailable when the CMs needed them for the durationof the CCM intervention implementation.

The importance of supervision was outlined in mostof the selected studies (Coleman et al. 2017a; Gasket al. 2010; Møller et al. 2018; Murphy et al. 2014;Nutting et al. 2007, 2008; Overbeck et al. 2018a; Web-ster et al. 2016; Whitebird et al. 2014; Wozniak et al.2015). Supervision by a mental health specialist wasdeemed to have many benefits, such as increasingCMS’ confidence in their capacity to take care of peo-ple with mental health problems (Nutting et al. 2008;Whitebird et al. 2014). The CM role can be emotion-ally stressful, and supervision seems to help the CMdeal with the negative impact of the role on their psy-chological and mental well-being (Murphy et al. 2014;Webster et al. 2016). For instance, in Murphy et al.’study (2014), one nurse stopped delivering the inter-vention because she found it too emotionally stressful,as she did not have access to supervision due to timerestraints and lack of funding.

Context-specific factors

Even though these strategies had the potential toimprove CMs’ performance, several factors specific tothe context of implementation may also influence theadoption of the role. Three main factors related to thecontext of implementation were identified as follows:organization-related factors, collaboration with teammembers, and competency of the CM in care manage-ment. Although interdependent, these three context-specific factors are presented separately.

Organization-related factorsAspects specific to the organization where the imple-mentation took place seem to influence the adoption ofthe CM role and the sustainability of the entire CCMintervention. The workload in primary care settingsappears to be an obstacle to the adoption of the rolebecause it is sometimes difficult for primary care provi-ders, specifically for GPs, to have additional time forthe team meetings or consultations (Blasinsky et al.2006; Møller et al. 2018; Nutting et al. 2008; Websteret al. 2016). Managing the demands of a programmetargeted towards a specific clientele while answering tothe wide range of needs in primary care patients mayalso be challenging (Blasinsky et al. 2006; Overbecket al. 2018a). However, in one study, GPs and nurses

reported little to no increase in their workload, andsome clinicians even mentioned their workloaddecreased with the CCM implementation (Curranet al. 2012).

Furthermore, the clientele often has complex psy-chosocial problems making it sometimes difficult toengage patients in their care (Curran et al. 2012; Nut-ting et al. 2007, 2008). CMs interviewed in theincluded studies frequently mentioned having to dealwith many other related problems during consultations(e.g. health complications, housing issues, work prob-lems), which made it harder to assist patients inmanaging or solving mental health-related problems(Gask et al. 2006). The time needed to complete theCM tasks for these patients with complex needs couldbecome an implementation problem (Coleman et al.2017a). For instance, in one study, nurses stopped theirCM activities because of time/staffing constraints (Mur-phy et al. 2014). There was a significant need to benefitfrom support of other professionals, such as socialworkers and community resources (Coleman et al.2017a; Curran et al. 2012). The necessity to adjust theintervention to account for comorbidities was also oftenmentioned (Gask et al. 2010; Møller et al. 2018; Nut-ting et al. 2008).

In some cases, remuneration of the CM was tied tothe implementation of the CCM intervention, whichturned into a barrier to the sustainability of the entireCCM when research funding ceased. The issue oftenrevolved around who is going to pay for that additionalprofessional (Blasinsky et al. 2006; Curran et al. 2012;Nutting et al. 2008). Moreover, leaders’ priorities influ-enced the degree of adoption of the CM role and ofthe other components of the CCM as well (Lipschitzet al. 2017; Nutting et al. 2007; Overbeck et al. 2018a;Webster et al. 2016). The lack of a shared visionamong clinicians and leaders regarding the process ofchanging practices was a serious barrier to the dissemi-nation of one CCM intervention (Nutting et al. 2007).The two most recent studies of this review concludedby stressing the importance of addressing the needs ofspecific populations and engaging relevant stakeholdersin the intervention design and implementation process(Coleman et al. 2017b; Møller et al. 2018).

Collaboration with team membersThis category of context-specific factors refers to thecapacity of the CM and the team of providers to col-laborate and work as a team. The role of CM relies onthe collaboration between primary care providers,patients, and mental health specialists. However, the



capacity of the CM to collaborate efficiently with theteam of providers and patients relies on many charac-teristics of the organization and of the individualsinvolved in the care.

First, the physical location of the CM seems toinfluence the degree of collaboration that can bereached between them and the GPs and other provi-ders (Curran et al. 2012; de Jong et al. 2009; Wozniaket al. 2015). Indeed, the CM was not always locatedwithin the primary care clinic. At times, the CMworked in a centralized clinic or a hospital department,but some clinicians preferred when the CM waslocated on site (Møller et al. 2018; Nutting et al. 2008;Whitebird et al. 2014). The colocation of CMs with pri-mary care providers, especially GPs, could enhance thecommunication and the relationship between membersof the whole team. In fact, colocation generates morefrequent ‘nonformal’ meeting opportunities such asduring lunchtime or group meetings (Curran et al.2012).

Moreover, clinicians appeared to appreciate face-to-face communication because it helped rapid informa-tion sharing, thus optimizing interactions with the CM(Curran et al. 2012; Lipschitz et al. 2017; Nutting et al.2008; Overbeck et al. 2018a). For instance, Whitebirdet al. (2014) found better 6 months postinterventionremission rates when the clinicians and CM sharedface-to-face information when referring a patient. Incontrast, in one study where the CM was located out-side of the primary care clinic, the CMs and the GPsfelt that they did not really collaborate but rather pro-ceeded to an exchange of information or a ‘transfer’ ofpatients (Møller et al. 2018).

In order to improve the interaction between theCM and the team of providers, some authors proposedto include a logistical component when implementingthe CCM (Lipschitz et al. 2017; Overbeck et al.2018a), such as dedicated moments and meeting spaces(Lipschitz et al. 2017). Collaboration between the CMand the team is an essential component of the CCM,and it was deemed important that the CM, the GPs,and other team members have the appropriate timeand space to collaborate (Curran et al. 2012; Lipschitzet al. 2017; Møller et al. 2018; Overbeck et al. 2018a).

Beyond the colocation of the CM, the use of stan-dardized tools such as the Patient Health Question-naire (PHQ-9) was a valuable strategy to enhancecommunication among clinicians as it acted as a refer-ence point to rapidly monitor the evolution of thepatient’s health condition (Blasinsky et al. 2006; Lips-chitz et al. 2017; Nutting et al. 2007). Nevertheless,

communication difficulties were sometimes associatedwith the inability of computer systems to signal prob-lems with specific patient results or the limited accessto the electronic patient record by some members ofthe team (Coleman et al. 2017a; de Jong et al. 2009;Lipschitz et al. 2017).

The relationship between CMs and GPs was also animportant ingredient in the adoption of the CM role(Lipschitz et al. 2017; Nutting et al. 2008; Overbecket al. 2018a; Williams et al. 2011). CMs who had previ-ously worked with the GPs and the primary care provi-ders seemed to have an advantage over those newlyintroduced to the team. The discontinuity of the indi-vidual fulfilling the role through implementationseemed to influence the team’s trust in the CM (Wil-liams et al. 2011; Wozniak et al. 2015).

Furthermore, in some studies, GPs’ attitude regard-ing the CM may have influenced the adoption of therole (Møller et al. 2018; Wozniak et al. 2015). Depend-ing on the culture of the organization, some GPs wereaccustomed to referring patients to a psychiatrist orother mental health specialist and exchange informa-tion directly with that person without an intermediarylike the CM (Gask et al. 2010). In addition, when thereis a hierarchy between the GP and the CM, it can beharder for the CM to use his or her leadership in orderto optimize collaboration (Wozniak et al. 2015). Inter-professional respect and trust truly need to be instilledin the relationship, but this was not always the caseaccording to some selected studies (Gask et al. 2010;Møller et al. 2018; Wozniak et al. 2015).

Given that the patient is the core member of thecollaborative care team, the quality of the relationshipbetween them and the CM also influenced the capacityto build an efficient collaboration. Again, a CM whowas already acquainted with a patient appeared to havean advantage, simply because less efforts were requiredcompared to building a new relationship (Websteret al. 2016). Patients’ level of motivation to engage intreatment seemed to influence their relationship withthe CM (Bennett et al. 2013), and lack of motivationquickly proved challenging to the CM in fulfilling hisor her role (Coleman et al. 2017a).

Finally, another important area in which collabora-tion could be optimized was the clarification of therespective roles of the CM and of each team member(Coleman et al. 2017b; Lipschitz et al. 2017; Overbecket al. 2018a; Whitebird et al. 2014). Some activities ofthe CM may overlap with those of other clinicians.This made it especially important to clarify the CMs’activities, referral modalities, and how the CM would


14 A. GIRARD ET AL.

be interacting with both patients and providers (Lips-chitz et al. 2017; Whitebird et al. 2014).

Competency of the CM in care managementCompetency in care management refers to the specificknowledge and the professional, interpersonal, and per-sonal skills needed to provide quality care managementfor people with common mental illnesses and/or physi-cal long-term diseases in primary care settings. In otherwords, competency in care management refers to thecapacity of the CM to perform the activities related tothe role of CM.

One of the most important ingredients that the CMneeded was the motivation to work with people withmental health problems and the desire to help them.The CM had to be interested in this clientele in orderto fully engage in their care (Bennett et al. 2013; Cur-ran et al. 2012; de Jong et al. 2009; Webster et al.2016; Wozniak et al. 2015). In three studies where theintervention targeted people with both physical andmental health problems, the nurse CMs were oftenmore comfortable in their skills to address the physicalconditions (Coleman et al. 2017a; Murphy et al. 2014;Webster et al. 2016). Interestingly, sometimes theinterest or motivation regarding mental health wouldgrow throughout the implementation process (Websteret al. 2016). Coleman et al. (2017a) also demonstratedthat greater familiarity and feeling comfortable withcaring for people with depression could lead to betterhealth outcomes for patients with both depression anddiabetes and/or cardiovascular disease.

Previous clinical experience of the CM seemed toinfluence their level of comfort with mental healthissues (Bennett et al. 2013) and their capacity to deli-ver psychosocial interventions (Gask et al. 2006). Forinstance, in Bennett et al.’s study (2013), the nurseCMs who were more confident with mental healthissues at the beginning of the trial had a more thera-peutic, counselling-oriented approach and were lessdirective.

Opinions diverged regarding the specific clinicalexperience needed to play the CM role in the selectedstudies. In one study, participants held various opin-ions, with preference ranging from clinical experiencein collaborative care in both long-term disease andmental health, to experience in care management oflong-term disease or mental health (Wozniak et al.2015). In the study of Møller et al. (2018), a nurse withexperience in mental health reported that if CMs wereonly nurses with psychiatric experience, it would be amajor barrier to the sustainability of the model due to

the limited number of these experienced nurses com-pared to the needs of the population. However, findingthe person with the right qualities and who can workwell in the clinic’s setting seemed to be a challenge tothe implementation of the CCM in other studies (Lips-chitz et al. 2017; Whitebird et al. 2014).

Generally, the CM should have practical and theo-retical knowledge on psychotherapeutic and pharmaco-logical treatments for depression and anxiety (Colemanet al. 2017a; de Jong et al. 2009). The CM should alsohave the skills to support patients engaging in a changeof behaviour by using an evidence-based psychologicalapproach (Bennett et al. 2013; Coleman et al. 2017a;Gask et al. 2006; de Jong et al. 2009; Overbeck et al.2018a; Webster et al. 2016), including the capacity toinform and educate patients on their physical and men-tal health conditions (Bennett et al. 2013; Lipschitzet al. 2017). In addition, a set of CM interpersonalskills has been identified, including the capacity toshow empathy, attentiveness to the clientele’s needs,and personal availability (Bennett et al. 2013; Curranet al. 2012; de Jong et al. 2009; Wozniak et al. 2015).The CM’s capacity to adapt to different situations,related to factors such as clientele diversity and chal-lenges with working in collaboration with a variety ofactors, is likewise essential (Møller et al. 2018; Over-beck et al. 2018a). The importance of having strongcommunication skills to enhance collaboration betweenteam members was also documented (Bennett et al.2013; Wozniak et al. 2015). Two studies also pointedout the importance of leadership by CMs when imple-menting such a new role in a primary care clinic (Over-beck et al. 2018a; Whitebird et al. 2014).

DISCUSSION

Two categories of factors seeming to influence theadoption of the CM role when implementing the CCMwere identified and described as follows: strategies toimplement CCM interventions and context-specific fac-tors. Identified strategies, that is initial care manage-ment training and supervision by a mental healthspecialist, appeared essential in order to allow nursesto perform the CM role adequately when caring forpeople with anxiety or depression and comorbidities.These implementation strategies mainly aim to improvethe CM’s competencies.

The implementation strategies are therefore relatedto only one of the context-specific factors, that is theindividual’s competency in care management. Accord-ing to our results, a competent CM should be



motivated by the role; have professional, interpersonal,and personal skills, such as empathy and the ability toprovide psychosocial interventions to help modify beha-viour with efficiency; possess specific knowledge on thetreatment of common mental illnesses; and have someexperience in care management, mental health care orcollaborative care. Other important aspects include thedrive to help the patient, leadership, and clinical expe-rience, which have also been identified as context-spe-cific factors influencing the activities of nurses inprimary care settings (Poitras et al. 2018). For all thesereasons, recruiting a competent CM can be challeng-ing, as is the case in other types of interventions ori-ented towards the care management of a complexclientele (Hudon et al. 2017).

Apart from previous experience of the CM, therewas limited information regarding the initial competen-cies of the CM in the selected studies. Attention wasmore focused on describing the main tasks or activitiesof the CM rather than the individual’s competency incare management. Indeed, description of the compe-tencies needed to fulfil the CM role was often absentin the 14 CCM interventions. Therefore, it is difficultto assess whether the CCM implementation strategies(training and supervision) were effective in improvingCM competencies.

In the IMPACT study, Un€utzer et al. (2001) founda positive relationship between the number of trainingsessions and the CM’s competence in delivering prob-lem-solving therapy in primary care. However, the rat-ing of competency was based on the CM’s skill todeliver with fidelity the seven stages of problem-solvingtherapy in primary care (Hegel et al. 2000). Imple-menting such a specific training programme in real-world settings could be challenging, especially giventime and cost constraints associated with the manage-ment of primary health care (Hegel et al. 2000; Web-ster et al. 2016).

Specific learning methods used to train the CM dur-ing CCM implementation interventions (e.g. case dis-cussion, role-playing, observation, and feedbackthrough video recording, use of didactic materials)were similar to those found in an integrative review onmental health education programmes for generalisthealth professionals (Brunero et al. 2012). According tothis review of 25 studies, the most frequent pedagogi-cal approach used to train generalists in mental healthwas experiential learning, with some studies combiningboth didactic and experiential styles of learning (Bru-nero et al. 2012). Experiential learning refers to con-structing knowledge and meaning through real-life

experience (Yardley et al. 2012). However, the efficacyof experiential learning in changing behaviour orimproving the quality of professional activities is notwell documented (Brunero et al. 2012; Thistlethwaiteet al. 2012). This is not surprising, as instruments usedto measure competencies often do not consider theprofessional’s real-world activities, even though activi-ties are the main medium through which competenciesare demonstrated in clinical practice. As highlighted inthe study of Brunero et al. (2012), the instrumentsused to assess competency often measure knowledge,skills or attitudes. Scientific data on the relationshipsbetween competencies, role-specific activities, and thelearning methods used to train CMs remain sparse,especially considering the impact of context-specificfactors.

In addition to the competency of the CM in caremanagement, the adoption of the CM role is affectedby other context-specific factors. Consequently, whenimplementing the CCM, strategies to optimize theadoption of the CM role should not be oriented solelytowards the competencies of the CM, who also needs afavourable context in order to perform their activitieswith competency.

One of the main activities of the CM is to collabo-rate within a team. However, the results of this reviewshow that the capacity to collaborate efficiently is oftenhard to achieve for a variety of reasons, including orga-nizational and individual factors. For instance, variousfactors were identified, such as the location of the CM,the difficulty to find the time and space to meet withthe team, the lack of clarity regarding the respectiveroles of the CM and other team members, GPs’ atti-tudes towards the CM role, and the lack of efficientmechanisms of communication. These factors alsoemerged from numerous studies on enablers and barri-ers affecting interprofessional collaboration in primarycare and in the caring for people with long-term condi-tions (Chong et al. 2013; Gucciardi et al. 2016; San-galeti et al. 2017). Studies also mention the importanceof having a clinical information system that allows theteam (primary care providers and specialists) to sharedata, using standardized tools and electronic dischargesummaries, dedicating time and space for meetings,clarifying team members’ roles and responsibilities, anddeveloping relationships between providers andpatients (Chong et al. 2013; Gucciardi et al. 2016; San-galeti et al. 2017). These results highlight the impor-tance of including specific strategies aiming to improvecollaborative practices when implementing the CCM inprimary care settings (Lipschitz et al. 2017; Wozniak


16 A. GIRARD ET AL.

et al. 2015). In fact, even though CCM interventionsusually include strategies to enhance communicationbetween the CM, GPs, and the mental health special-ist, such as meetings or electronic records, these strate-gies are often not sufficiently adapted to the setting,and there is sometimes no guidance provided on howtheir effects should be enhanced or who needs to beinvolved in the process (Overbeck et al. 2018b).

Furthermore, collaborative and competency factorsrelated to the CM role are also influenced by organiza-tion-related factors. Some factors related to the organi-zation’s context, such as the workload in primary careand the psychosocial complexity of the clientele, arenot surprising results given they were both highlightedin the two recent systematic reviews on CCM imple-mentation enablers and barriers (Overbeck et al. 2016;Wood et al. 2017). Indeed, by its underlying mission ofpromoting and managing the health of the population,primary care has a wide range of clientele for which itis responsible for addressing any health-related prob-lem (Fiscella & McDaniel 2018). Therefore, whendeveloping an intervention in primary care, it appearsimportant to be mindful of the needs of the range ofpeople consulting in primary care.

One of the first CCM interventions (i.e. IMPACT)targeted people suffering from a specific mental healthcondition (i.e. depression). However, a study on thesustainability of the IMPACT intervention has shownthat it was adapted matter-of-factly by organizations forother conditions and even for generic disease manage-ment (Blasinsky et al. 2006). Since 2000, there hasbeen an evolution regarding the choice of health condi-tions targeted by CCM interventions.

Furthermore, a UK survey has shown an increase inthe workload of GPs and nurses in primary carebetween 2007 and 2014 (Hobbs et al. 2016). GPs alsomentioned challenges in caring for people with com-plex needs in primary care, as they do not always per-ceive that their practice is prepared to manage the careof this clientele (Osborn et al. 2015). There is a needto increase the connections between primary care,community services and social services for both accessto and coordination of care (Osborn et al. 2015). It isalso worth emphasizing the challenges that PCNs facewhen supporting patients in resolving more complexproblems and the need to train CMs in optimizing col-laborative practice and delivering psychosocial inter-ventions to change behaviours. Nevertheless, thedegree of difficulty in coordinating care will varydepending on the functioning of the organization. Ifthe CM needs to invest a considerable amount of time

to coordinate care because of organizational con-straints, it might be pertinent to delegate that functionto other members of the team who will ensure a bridgebetween professionals and services (Hunt et al. 2016).

To our knowledge, this is the first review with anemphasis on the main factors influencing the adoptionof the CM role when implementing the CCM in pri-mary care. This review has contributed to identify agap in the literature regarding the competency of CMsin care management and a lack of knowledge on therelevant strategies to use to overcome specific contex-tual factors when implementing the CCM. The resultspresented come from 14 individual CCM interventionsimplemented in five countries—including the largestinitiative in the USA—and are based on studies imple-mented over an important period of time (17 years).The review was conducted following a well-recognizedmethodology by a team with a variety of research andclinical backgrounds (nursing (AG-EE), family medi-cine (CH), psychology (PR), psychiatry (JDC)).

However, this study did not include grey literaturedocuments, nor did we validate the results with agroup of stakeholders as proposed by Levac et al.(2010). These steps could have contributed to identify-ing other factors or to refining our interpretation. Oneof the limitations of the scoping review design is that itdoes not allow us to assess the effectiveness of an inter-vention, nor to confirm the existence of a relationshipbetween two concepts; rather, a scoping review pro-vides a narrative or descriptive account of availableresearch (Arksey & O’Malley 2005). In that respect,the broad range of publication types and designsamong studies included in this review allow for a morecomplete description of the concept of interest.

Additionally, it was occasionally difficult to locateinformation on the context of implementation for indi-vidual CCM interventions (e.g. number of professionalsincluded, number of sites, the location of the CM, ini-tial training of the CM). Information was often notstandardized from one study to another, making itmore difficult to extract the data. When relevant, welooked at other publications reporting on the sameCCM intervention as the included studies, but theadditional information on the context was often relatedto the patients rather than the professionals or theorganizations participating in the study. This caveat ofthe published literature limits the accuracy of the infor-mation we report in the context of implementation, aswell as our ability to understand the full impact of themain factors and the relationships between them. Asystematic review of reviews including 70 articles on



barriers and enablers to achieving change in primarycare also pointed out the lack of information about thecontext in which barriers and facilitators occur inimplementation studies (Lau et al. 2016). Additionally,the aim of this scoping review was to identify the mainfactors influencing the adoption of the CM role in avariety of primary care settings. Indeed, the diversity ofthe contexts found in included studies allowed for thedescription of a wide range of factors. However, resultsof this review might not be transferable to every con-text, and the impact of the individual factors might bedifferent between countries due to specificity of eachhealthcare systems and of local needs.

CONCLUSION

The competency of the CM in care managementappears to be an important ingredient to facilitate theadoption of the role when implementing the CCM.However, there is a need to better understand the rela-tionships between the individual’s competencies, theCM activities, the strategies used while implementing aCCM intervention, and the other context-specific fac-tors. Considering that PCNs have the capacity andopportunity to fulfil many activities pertaining to therole of CM in their day-to-day practice, it is importantto understand which strategies to use and when to usethem, and how they might improve their competenciesspecific to care management. In addition, the adoptionof the CM role when implementing the CCM in pri-mary care can be optimized by developing or selectingimplementation strategies that can overcome other con-text-specific factors. Given the current lack of scientificdata on the impact of these strategies, researchers, andstakeholders (i.e. patients, administrators, primary careproviders, mental health specialists, etc.) are encour-aged to evaluate them and to expand current knowl-edge on the relationships between implementationstrategies and context-specific factors.

RELEVANCE FOR CLINICAL PRACTICE

This scoping review has contributed to mapping thegap in the literature concerning the adoption ofthe CM role when implementing a CCM intervention.The results can help researchers and stakeholders tobe mindful of factors specific to their context in orderto develop or select appropriate implementation strate-gies and to elaborate an implementation plan adaptedto their local needs. CMs’ activities have a significantoverlap with those of PCNs, but nurses tend to have a

lack of confidence in their ability to provide psychoso-cial interventions, which is partly congruent with a lackof training. Given the holistic nature of PCNs’ practiceand their relationship with patients, they have thepotential to contribute to improving the quality of carefor people with common mental illnesses in primarycare. Thus, the CCM, including the role of CM, is agood framework for primary care clinics to optimizethe care of people with mental health problems byleveraging staff and collaboration mechanisms that arealready in place.

ACKNOWLEDGEMENTS

The correspondent author was financially supportedthrough her doctoral scholarship by Quebec Network onNursing Intervention Research, Ordre des infirmi�eres etinfirmiers du Qu�ebec, and Minist�ere de l’�Education etEnseignement sup�erieur Qu�ebec.

REFERENCES

AIMS Center-University of Washington. (2018). CoCMBehavioral Health Care Manager: Sample JobDescription, Typical Workload & Resource Requirements.January 26, 2018. http://aims.uw.edu/sites/default/files/CareManagerJobDescription_0.pdf. Accessed: 13 July 2018.

American Psychiatric Association & Academy of PsychosomaticMedicine. (2016). Dissemination of integrated care withinadult primary care settings: the Collaborative Care Model.https://www.psychiatry.org/psychiatrists/practice/professional-interests/integrated-care/get-trained/about-collaborative-care. Accessed: 13 July 2018.

Archer, J., Bower, P., Gilbody, S. et al. (2012). Collaborativecare for depression and anxiety problems (Review).Cochrane Database Systematic Review, 10, 1–277.

Arksey, H. & O’Malley, L. (2005). Scoping studies: Toward amethodological framework. International Journal of SocialResearch Methodology, 8, 19–32.

Bennett, M., Walters, K., Drennan, V. & Buszewicz, M.(2013). Structured pro-active care for chronic depressionby practice nurses in primary care: A qualitativeevaluation. PLoS ONE, 8, e75810.

Blasinsky, M., Goldman, H. H. & Un€utzer, J. (2006). ProjectIMPACT: A report on barriers and facilitators tosustainability. Administration and Policy in Mental Healthand Mental Health Services Research, 33, 718–729.

Brinck-Claussen, U. Ø., Kehler Curth, N., Davidsen, A. S.et al. (2017). Collaborative care for depression in generalpractice: Study protocol for a randomised controlled trial.Trials, 18, 1–12.

Brunero, S., Jeon, Y. H. & Foster, K. (2012). Mental healtheducation programmes for generalist health professionals:An integrative review. International Journal of MentalHealth Nursing, 21, 428–444.


18 A. GIRARD ET AL.

http://aims.uw.edu/sites/default/files/CareManagerJobDescription_0.pdf

http://aims.uw.edu/sites/default/files/CareManagerJobDescription_0.pdf

https://www.psychiatry.org/psychiatrists/practice/professional-interests/integrated-care/get-trained/about-collaborative-care



Buszewicz, M., Griffin, M., McMahon, E. M., Beecham, J. &King, M. (2010). Evaluation of a system of structured,pro-active care for chronic depression in primary care: Arandomised controlled trial. BMC Psychiatry, 10, 1–9.

Chong, W. W., Aslani, P. & Chen, F. T. (2013). Shareddecision-making and interprofessional collaboration inmental healthcare: A qualitative study exploringperceptions of barriers and facilitators. Journal ofInterprofessional Care, 27, 373–379.

Coleman, K. J., Hemmila, T., Valenti, M. D. et al. (2017a).Understanding the experience of care managers andrelationship with patient outcomes: The COMPASSinitiative. General Hospital Psychiatry, 44, 86–90.

Coleman, K. J., Magnan, S., Neely, C. et al. (2017b). TheCOMPASS initiative: Description of a nationwidecollaborative approach to the care of patients withdepression and diabetes and/or cardiovascular disease.General Hospital Psychiatry, 44, 69–76.

Coventry, P. A., Hudson, J. L., Kontopantelis, E. et al.(2014). Characteristics of effective collaborative care fortreatment of depression: A systematic review and meta-regression of 74 randomised controlled trials. PLoS ONE,9, e108114.

Curran, G. M., Sullivan, G., Mendel, P. et al. (2012).Implementation of the CALM intervention for anxietydisorders: A qualitative study. Implementation Science, 7,1–11.

Damschroder, L. J., Aron, D. C., Keith, R. E., Kirsh, S. R.,Alexander, J. A. & Lowery, J. C. (2009). Fosteringimplementation of health services research findings intopractice: A consolidated framework for advancingimplementation science. Implementation Science, 4, 1–15.

de Jong, F. J., van Steenbergen-Weijenburg, K. M., Huijbregts,K. M. L. et al. (2009). The Depression Initiative.Description of a collaborative care model for depressionand of the factors influencing its implementation in theprimary care setting in the Netherlands. InternationalJournal of Integrated Care, 9, 1–9.

Dietrich, A. J., Oxman, T. E., Kroenke, K. et al. (2004).Going to scale: Re-engineering systems for primary caretreatment of depression. Annals of Family Medicine, 2,301–304.

Ekers, D. & Wilson, R. (2008). Depression case managementby practice nurses in primary care: An audit. MentalHealth in Family Medicine, 5, 111–119.

Ekers, D., Murphy, R., Archer, J., Ebenezer, C., Kemp, D.& Gilbody, S. (2013). Nurse-delivered collaborative carefor depression and long-term physical conditions: Asystematic review and meta-analysis. Journal of AffectiveDisorders, 149, 14–22.

Fiscella, K. & McDaniel, S. H. (2018). The complexity,diversity, and science of primary care teams. AmericanPsychologist, 73, 451–467.

Gask, L., Ludman, E. & Schaefer, J. (2006). Qualitative studyof an intervention for depression among patients withdiabetes: How can we optimize patient–professionalinteraction? Chronic Illness, 2, 231–242.

Gask, L., Bower, P., Lovell, K. et al. (2010). What work has tobe done to implement collaborative care for depression?Process evaluation of a trial utilizing the NormalizationProcess Model. Implementation Science, 5, 1–11.

Gilbody, S., Bower, P. & Whitty, P. (2006). Costs andConsequences of enhanced primary care for depression:Systematic review of randomised economic evaluations.British Journal of Psychiatry, 189, 297–308.

Girard, A., Hudon, C., Poitras, M. E., Roberge, P. &Chouinard, M. C. (2017). Primary care nursing activitieswith patients affected by physical chronic disease andcommon mental disorders: A qualitative descriptive study.Journal of Clinical Nursing, 26, 1385–1394.

Gucciardi, E., Espin, S., Morganti, A. & Dorado, L. (2016).Exploring interprofessional collaboration during theintegration of diabetes teams into primary care. BMCFamily Practice, 17, 1–14.

Gunn, J., Diggens, J., Hegarty, K. & Blashki, G. (2006). Asystematic review of complex system interventionsdesigned to increase recovery from depression in primarycare. BMC Health Services Research, 6, 1–11.

Hegel, M. T., Barrett, J. E. & Oxman, T. E. (2000). Trainingtherapists in problem-solving treatment of depressivedisorders in primary care: Lessons learned from the“Treatment Effectiveness Project”. Families, Systems, &Health, 18, 423–435.

Hobbs, F. D. R., Bankhead, C., Mukhtar, T. et al. (2016).Clinical workload in UK primary care: A retrospectiveanalysis of 100 million consultations in England, 2007-14.Lancet, 387, 2323–2330.

Hudon, C., Chouinard, M. C., Lambert, M., Diadiou, F.,Bouliane, D. & Beaudin, J. (2017). Key factors of casemanagement interventions for frequent users of healthcareservices: A thematic analysis review. British MedicalJournal Open, 7, 1–7.

Hunt, C. M., Spence, M. & McBride, A. (2016). The role ofboundary spanners in delivering collaborative care: Aprocess evaluation. BMC Family Practice, 17, 1–10.

Institute for Clinical Systems Improvement. (2014). TheDIAMOND Program: Treatment for patients withdepression in primary care. https://www.icsi.org/_asset/rs2qfi/DIAMONDWP0614.pdf. Accessed: 15 July 2018.

Johnson, J. A., AI Sayah, F., Wozniak, L. et al. (2012).Controlled trial of a collaborative primary care teammodel for patients with diabetes and depression: Rationaleand design for a comprehensive evaluation. BMC HealthServices Research, 12, 1–12.

Katon, W., Von Korff, M. & Lin, E. (1995). Collaborativemanagement to achieve treatment guidelines: Impact ondepression in primary care. Journal of the AmericanMedical Association, 273, 1026–1031.

Katon, W., Von Korff, M., Lin, E. et al. (2003). Improvingprimary care treatment of depression among patients withdiabetes mellitus: The design of the Pathways Study.General Hospital Psychiatry, 25, 158–168.

Kehler Curth, N., Brinck-Claussen, U. Ø., Davidsen, A. S.et al. (2017). Collaborative care for panic disorder,



https://www.icsi.org/_asset/rs2qfi/DIAMONDWP0614.pdf

https://www.icsi.org/_asset/rs2qfi/DIAMONDWP0614.pdf

generalized anxiety disorder and social phobia in generalpractice: Study protocol for three cluster-randomised,superiority trials. Trials, 18, 1–13.

Lau, R., Stevenson, F., Ong, B. N. et al. (2016). Achievingchange in primary care-causes of the evidence to practicegap: Systematic reviews of reviews. ImplementationScience, 11, 1–39.

Levac, D., Colquhoun, H. & O’Brien, K. K. (2010). Scopingstudies: Advancing the methodology. ImplementationScience, 5, 1–9.

Levine, S., Un€utzer, J., Yip, J. Y. et al. (2005). Physicians’satisfaction with a collaborative disease managementprogram for late-life depression in primary care. GeneralHospital Psychiatry, 27, 383–391.

Lipschitz, J. M., Benzer, J. K., Miller, C. et al. (2017).Understanding collaborative care implementation in theDepartment of Veterans Affairs: Core functions andimplementation challenges. BMC Health ServicesResearch, 17, 1–11.

Miles, M. B., Huberman, A. M. & Salda~na, J. (2014).Qualitative Data Analysis: A Methods Sourcebook, 3rdedn. Thousand Oaks, CA: SAGE Publications.

Møller, M. C., Mygind, A. & Bro, F. (2018). Who needscollaborative care treatment? A qualitative study exploringattitudes towards and experiences with mental healthcareamong general practitioners and care managers. BMCFamily Practice, 19, 1–12.

Murphy, R., Ekers, D. & Webster, L. (2014). An update todepression case management by practice nurses inprimary care: A service evaluation. Journal of PsychiatricMental Health Nursing, 21, 827–833.

Nutting, P. A., Gallagher, K. M., Riley, K., White, S.,Dietrich, A. J. & Dickinson, W. P. (2007). Implementing adepression improvement intervention in five health careorganizations: Experience from the RESPECT-depressiontrial. Administration and Policy in Mental Health andMental Health Services Research, 34, 127–137.

Nutting, P. A., Gallagher, K., Riley, K. et al. (2008). Caremanagement for depression in primary care practice:Findings from the RESPECT-depression trial. Annals ofFamily Medicine, 6, 30–37.

Osborn, R., Donald, M., Schneider, E. C. et al. (2015).Primary care physicians in then countries reportchallenges caring for patients with complex health needs.Health Affairs, 34, 1–11.

Overbeck, G., Davidsen, A. S. & Kousgaard, M. B. (2016).Enablers and barriers to implementing collaborative carefor anxiety and depression: A systematic qualitative review.Implementation Science, 11, 1–16.

Overbeck, G., Kousgaard, M. B. & Davidsen, A. S. (2018a).The work and challenges of care managers in theimplementation of collaborative care model: A qualitativestudy. Journal of Psychiatric Mental Health Nursing, 24,167–175.

Overbeck, G., Kousgaard, M. B. & Davidsen, A. S. (2018b).Enactments and experiences of ‘enhancedinterprofessional communication’ in collaborative care—a

qualitative study. Journal of Interprofessional Care, 1–8.https://doi.org/10.1080/13561820.2018.1538109

Pham, M. T., Rajic, A., Greig, D., Sargeant, J. M.,Papadopoulos, A. & McEwen, S. A. (2014). A scopingreview of scoping reviews: Advancing the approach andenhancing the consistency. Research Synthesis Methods, 5,371–385.

Poitras, M. E., Chouinard, M. C., Fortin, M., Girard, A.,Crossman, S. & Gallagher, F. (2018). Nursing activities forpatients with chronic disease in family medicine groups: Amultiple-case study. Nursing Inquiry, e12250, 1–15.

Ratzliff, A., Un€utzer, J., Katon, W. & Stephens, K. A. (2016).Integrated Care: Creating Effective Mental and PrimaryHealth Care Teams. New Jersey: John Wiley & Sons.

Richards, D. A., Lovell, K., Gilbody, S. et al. (2008).Collaborative care for depression in UK primary care: Arandomized controlled trial. Psychological Medicine, 38,279–287.

Rossom, R. C., Solberg, L. I., Magnan, S. et al. (2017).Impact of a national collaborative care initiative forpatients depression and diabetes or cardiovascular disease.General Hospital Psychiatry, 44, 77–85.

Roy-Byrne, P., Craske, M. G., Sullivan, G. et al. (2010).Delivery of evidence-based treatment for multiple anxietydisorders in primary care: A randomized controlledtrial. Journal of the American Medical Association, 303,1–19.

Sangaleti, C., Schveitzer, M. C., Peduzzi, M., Pavone Zoboli,E. L. C. & Soares Baldini, C. (2017). Experiences andshared meaning of teamwork and interprofessionalcollaboration among health care professionals in primaryhealth care settings: A systematic review. JBI Database ofSystematic Reviews and Implementation Reports, 15,2723–2788.

Sunderji, N., Ghavam-Rassoul, A., Ion, A. & Lin, E. (2016).Driving Improvements in the Implementation ofCollaborative Mental Health Care: A Quality Frameworkto Guide Measurement, Improvement and Research.Toronto, Canada. https://www.researchgate.net/publication/312539646

Thistlethwaite, J. E., Davies, D., Ekeocha, S. et al. (2012).The effectiveness of case-based learning in healthprofessional education. A BEME systematic review:BEME guide No.23. Medical Teacher, 34, e421–e444.

Tricco, A. C., Lillie, E., Zarin, W. et al. (2018).PRISMA Extension for scoping reviews (PRISMA-ScR):Checklist and explanation. Annals of Internal Medicine.1–7. https://doi.org/10.7326/m18-0850

Un€utzer, J. & Ratzliff, A. H. (2015). Evidence base and coreprinciples. In: L. E. Raney (Ed). Integrated Care:Working at the Interface of Primary Care and BehavioralHealth (pp. 3–16). Washington DC: American PsychiatricPublishing.

Un€utzer, J., Katon, W., Williams, J. W. et al. (2001).Improving primary care for depression in late life: Thedesign of a multicenter randomized trial. Medical Care,39, 785–799.


20 A. GIRARD ET AL.

https://doi.org/10.1080/13561820.2018.1538109

https://www.researchgate.net/publication/312539646

https://www.researchgate.net/publication/312539646

https://doi.org/10.7326/m18-0850

Webster, L. A. D., Ekers, D. & Chew-Graham, C. A. (2016).Feasibility of training practice nurses to deliver apsychosocial intervention within a collaborative careframework for people with depression and long-termconditions. BMC Nursing, 15, 1–11.

Whitebird, R. R., Solberg, L. I., Jaeckels, N. A. et al. (2014).Effective implementation of collaborative care fordepression: What is needed? The American Journal ofManaged Care, 20, 699–706.

Williams, M., Angstman, K., Johnson, I. & Katzelnick, D.(2011). Implementation of a care management model fordepression at two primary care clinics. Journal ofAmbulatory Care Management, 34, 163–173.

Wood, E., Ohlsen, S. & Ricketts, T. (2017). What are thebarriers and facilitators to implementing CollaborativeCare for depression? A systematic review. Journal ofAffective Disorders, 214, 26–43.

Wozniak, L., Soprovich, A., Rees, S., Al Sayah, F., Majumdar,S. R. & Johnson, J. A. (2015). Contextualizing theeffectiveness of a collaborative care model for primarycare patients with diabetes and depression (Teamcare): Aqualitative assessment using RE-AIM. Canadian Journalof Diabetes, 39, S83–S91.

Yardley, S., Teunissen, P. W. & Dornan, T. (2012).Experiential learning: Transforming theory into practice.Medical Teacher, 34, 161–164.



challenges of adopting the role of care manager when

Documents