Family Practice, 2018, 1–15

doi:10.1093/fampra/cmy122

Systematic Review

Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018

Management of comorbid mental and somatic
disorders in stepped care approaches in primary
care: a systematic review
Kerstin Maehdera,*, Bernd Löwea, Martin Härterb, Daniela Heddaeusb,
Martin Schererc and Angelika Weigela
a
Department of Psychosomatic Medicine and Psychotherapy, University Medical Center Hamburg-Eppendorf,
Germany, bDepartment of Medical Psychology, University Medical Center Hamburg-Eppendorf, Germany and
c
Department of General Practice/Primary Care, University Medical Center Hamburg-Eppendorf, Germany.

*Correspondence to Kerstin Maehder, Department of Psychosomatic Medicine and Psychotherapy, University Medical
Center Hamburg-Eppendorf, Martinistraße 52, 20246 Hamburg, Germany; E-mail: k.maehder@uke.de

Abstract
Background.  Stepped care models comprise a graded treatment intensity and a systematic
monitoring. For an effective implementation, stepped care models have to account for the high
rates of mental and somatic comorbidity in primary care.
Objectives.  The aim of the systematic review was to take stock of whether present stepped care
models take comorbidities into consideration. A further aim was to give an overview on treatment
components and involved health care professionals.
Methods.  A systematic literature search was performed using the databases PubMed, PsycINFO,
Cochrane Library and Web of Science. Selection criteria were a randomized controlled trial of a
primary-care-based stepped care intervention, adult samples, publication between 2000 and 2017
and English or German language.
Results.  Of 1009 search results, 39 studies were eligible. One-third of the trials were conceived for
depressive disorders only, one-third for depression and further somatic and/or mental comorbidity
and one-third for conditions other than depression. In 39% of the studies comorbidities were explicitly
integrated in treatment, mainly via transdiagnostic self-management support, interprofessional
collaboration and digital approaches for treatment, monitoring and communication. Most care teams
were composed of a primary care physician, a care manager and a psychiatrist and/or psychologist.
Due to the heterogeneity of the addressed disorders, no meta-analysis was performed.
Conclusions.  Several stepped care models in primary care already account for comorbidities,
with depression being the predominant target disorder. To determine their efficacy, the identified
strategies to account for comorbidities should be investigated within stepped care models for a
broader range of disorders.

Key words: Collaborative care, comorbidity, mental health care, primary health care, stepped care, telemedicine.

Introduction crucial to provide care that accounts for comorbidity as it may have a
negative impact on illness course, treatment effects, adherence and qual-
Patients with comorbid disorders and undifferentiated symptoms are
ity of life (3,4). One approach to facilitate and structure care pathways
a major challenge in primary care (1). Comorbidity here is defined as
in primary care is ‘stepped care’. To date, it remains unclear whether
the presence of at least two distinct disorders in one individual (2). It is

© The Author(s) 2018. Published by Oxford University Press. All rights reserved.
1
For permissions, please e-mail: journals.permissions@oup.com.
2
current stepped care approaches account for comorbidity. Yet, this ques-
tion is especially important in primary care because for most patients,
primary care is the pivotal setting for diagnostics and the initiation of
evidence-based treatments for single and comorbid disorders (5).
Stepped care models involve (i) a spectrum of interventions, with a
hierarchy of care progressing from the least to the most intensive care,
matched to a person’s needs, and (ii) a systematic monitoring to ensure
the adequateness of a chosen treatment step (6). Due to their gatekeeper
function, primary care physicians often serve as a starting point for the
initiation of such stepped care models. While stepped care models may
include care from primary care physicians and further health care pro-
fessionals, cooperation is not a defining characteristic. This is in contrast
to ‘collaborative care’ approaches where interprofessional cooperation
is a central aspect (7). Problematically, current stepped care models
predominantly design care pathways for single diagnoses, leaving the
frequent comorbidity between both mental and somatic disorders and
the clinical reality of primary care unrecognized (8). The same applies to
most evidence-based guidelines for mental and somatic disorders. These
have a disorder-specific focus and provide little information on the man-
agement of comorbid disorders (9,10). This gap between research on
single conditions and primary care physicians’ everyday practice is one
of the major barriers to guideline implementation in primary care (11).
Stepped care can structure care and improve access to psychologi-
cal and psychiatric treatments. Therefore, evidence-based guidelines for
mental disorders increasingly describe stepped care as the preferred treat-
ment model (12–14). Such care algorithms are particularly useful for
disorders with differing degrees of severity, of longer duration and with
a course susceptible to treatment. Stepped care has been implemented in
a growing number of trials with promising results for different mental
disorders such as depression (15), anxiety disorders (16,17), somatoform
disorders (18) or musculoskeletal pain (19). Systematic reviews on the
efficacy of stepped care in depression treatment indicate a non-inferior-
ity, in some studies a superiority of stepped care as compared with usual
care (20,21). However, with regard to disorders other than depression,
current evidence for stepped care models in primary care is mixed. Thus,
an enlarged evidence base of stepped care models that considers mental
and somatic disorders and their comorbidities is needed.
The present systematic review captured the present state of the
integration of comorbidities in stepped care approaches. Thus, the
aims of this review were to provide an overview on if and how mental
and somatic comorbidities are taken into account in primary-care-
based stepped care models for adults, as tested in randomized con-
trolled trials with an active, usual care or waiting list control group.
Methods
Search strategy
A systematic search was performed using PubMed, PsycINFO,
Cochrane Library and Web of Science, complemented by additional
hand search and author requests to include recently completed stud-
ies. Search syntaxes were developed combining different terms used
for both primary care (e.g. general practice) and stepped care (e.g.
adaptive care) as found in the literature and MeSH terms, filtered by
additional terms or database filter for randomized controlled trials
(see Supplementary 1 for full electronic search strategy for PubMed).
Studies were eligible if they reported on a randomized controlled trial
of a primary-care-based stepped care intervention for adult patient
samples compared with an active control group, waiting list or usual
care. These inclusion criteria were chosen to evaluate the efficacy of
stepped care compared with a control condition. Studies in which the
term ‘stepped’ only referred to a stepped medication algorithm and
Family Practice, 2018, Vol. XX, No. XX
that focused on collaborative care without a structured care algo-
rithm were excluded. Studies were also considered eligible if they
implemented an adaptive treatment algorithm without explicitly
labelling it as ‘stepped care’ if the two core features of stepped care as
described above were fulfilled. Publications could either be in English
or German language and had to be published between 1/2000 and
10/2017 in peer-reviewed journals. The study protocol has not been
Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018
published in advance, and the review has not been registered.
Study selection
After the removal of duplicates, all titles and abstracts were screened
for eligibility. Then, two researchers (AW, KM) independently rated
the remaining full texts and discussed studies with discrepant judge-
ments until consensus was reached.
Data extraction included intervention details, participant charac-
teristics, results and information on if and how comorbidities were
taken into consideration with regard to the treatment algorithm.
A treatment algorithm was considered as accounting for comorbid-
ity if it contained at least one specific treatment component for each
of at least two different disorders.
Quality assessment
The quality of selected studies was assessed according to the Cochrane
Risk of Bias Tool (22) by one researcher (AW). This included the risk
of bias with regard to random sequence generation, allocation conceal-
ment, blinding of participants and personnel, blinding of outcome assess-
ment, incomplete outcome data, selective reporting and other biases.
Analysis
At a content level, eligible studies of primary-care-based stepped care
models were first investigated with regard to addressed disorders
and the percentage of models that accounted for comorbidity within
the applied treatment algorithms. Then, treatment steps and health
care professionals as described in the included studies were analysed
to gain a better understanding of the interventions and the strate-
gies used to account for comorbidities. Finally, an overview of the
efficacy of the stepped care models with regard to their primary out-
comes and their effects on quality of life and disability is provided.
Results
Study selection
Of 1009 eligible studies identified by the search strategy (n  =  558
duplicates removed), 190 full texts were screened, and 39 trials were
included (see Fig. 1 for flowchart). Of those, all were published in
English language, and no study in German language could be iden-
tified. Reasons for exclusion were (i) study protocol (n  =  41), (ii)
no clear stepped care structure of the intervention (n = 26), (iii) no
reported results pertaining to the efficacy of the stepped care model
(n = 39 studies), (iv) no randomized controlled trial (n = 4), (v) the
stepped care model not being based in primary care (n = 8) or (vi)
secondary analyses (n = 33). Differences in study inclusion between
raters were rare and arose predominantly if structured treatment
algorithms were not explicitly labelled as a stepped care approach.
Study characteristics
Setting characteristics
Studies were conducted in 10 countries. The majority of studies was
based in the USA (n = 18, 46%) or the Netherlands (n = 10, 26%).
Participating primary care physicians worked in small practices in
Comorbidity in stepped care
Figure 1.  Flowchart for study inclusion in the systematic review on the management of comorbidities in stepped care
13 studies and in primary care clinics or centres in 26 studies, with
different structures, sizes and patient collectives of different social
and economic statuses.
Participant characteristics
Overall, the studies comprised 16 850 participants (range: 46–2796)
of whom 57% were female (range: 0–100%). The mean age of par-
ticipants was 54  years (mean range: 41.2–81.4  years). The overall
dropout rate varied between 0% and 54% (mean: 18%). Exclusion
criteria that were explicitly mentioned included one or more of the
following conditions: psychotic symptoms, psychosis or schizophre-
nia (n = 31 studies), present substance abuse (n = 22), bipolar dis-
order (n = 24), acute suicidality (n = 18), terminal, life-threatening
illness or physical illness requiring immediate treatment (n = 19) and
significant cognitive impairment or dementia (n = 27).
Disorders and comorbidities addressed in the
stepped care models
Approximately, one-third of the studies implemented a stepped care
model that addressed depression and one or more comorbid mental or
somatic illness(es) (n = 14; 36%) (19,23–35). In case of a combination
of depression and a somatic disorder, the latter usually was the precondi-
tion for study entry, i.e. the presence of the somatic disorders determined
who was approached to participate in the study. While these stepped
care models included patients with two or more disorders, the treatment
steps were designed for more than one disorder in 11 studies and focused
on depression only in 3 studies. The remaining studies realized either a
stepped care model for depression only (n = 13, 33%) (15,36–47), anx-
iety disorders (specific or in general; n = 5; 13%) (16,17,48–50), alcohol-
related disorders (n = 2; 5%) (51,52), medically unexplained symptoms
(n = 1; 3%) (53), heart failure (n = 1; 3%) (54), musculoskeletal pain
(n = 2; 5%) (55,56) or coronary artery disease (n = 1; 3%) (57).
Treatment structure in the primary-care-based
stepped care models
The stepped care interventions differed in length, intensity, involved
professions and the modalities used for diagnostics, monitoring and
treatment. Due to the predominant focus on mental health, most
3
Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018
treatments included psychoeducation, case or care management, a
systematic monitoring, short-term training in problem-solving or
similar self-management strategies, the offer of pharmacotherapy and
the option of referral to mental health specialists. In line with the con-
cept of stepped care, basic care or case management and low-inten-
sity self-management support were usually offered to every patient
in the intervention group, whereas referral to (mental health) spe-
cialist services within or outside the study team was only offered to
patients suffering from greater symptom severity or after insufficient
benefit from the first treatment step(s). An exemplary stepped care
treatment algorithm comprised step 1) watchful waiting and psych-
oeducation, step 2) guided self-help supported by telephone contacts
with a nurse care manager in order to support self-management with
regard to psychiatric symptoms (e.g. behavioural activation) and
somatic symptoms (e.g. medication adherence), step 3)  face-to-face
problem-solving-treatment (seven sessions, delivered by the nurse
care manager) and step 4) referral back to the primary care physician
or further mental health services for pharmacotherapy and psycho-
therapy treatment options (stepped care model in (27)). Treatment
length in the studies varied between 2 and up to 12 months.
Studies made divergent use of telehealth options and digital com-
munication, for example, in the form of psychiatric consultation
via telephone (37), telephone monitoring (17) or shared electronic
health records (15). Basic study and intervention characteristics can
be seen in Table 1.
In the majority of cases (37 of 39 studies), the abovementioned
interventions were compared with a usual care control condition.
Strategies used to account for comorbidity in
primary-care-based stepped care models
Based on the study descriptions in the published papers, 39% of
studies (n  =  15) accounted for comorbidity. The strategies used to
account for comorbid illnesses could be classified as (i) assessment
and treatment components that addressed at least two disorders at
the same time, (ii) interprofessional cooperation and (iii) the use
of telehealth and digitally shared information between health care
professionals (see Fig.  2). The assessment and treatment compo-
nents addressing at least two disorders (i) included diagnostic and
 4

Table 1.  Characteristics of the included studies in the systematic review on the management of comorbidities in stepped care (published between 2000 and 2017)

First author, Country/ Condition Sample size Intervention/treatment options Control condition Professionals involved Primary outcome Comorbidity
year setting addressed in treatment

Araya et al., Chile, 3 Depression n = 240, Structured psychoeducational Usual care Primary care Significantly (sign.) greater Not specified as part
2003 (36) primary care only groups, monitoring, structured physicians (PCP), reduction in depression of the treatment
clinics women pharmacotherapy program for psychoeducational symptoms [Hamilton Rating algorithm
patients with severe or persistent group leaders (social Scale for Depression (HDRS)]
depression workers and nurses) at 3 and 6 months in the
intervention group (IG)
Aragonès Spain, 20 Depression n = 338 Case management, antidepressant Usual care PCP, case managers/PC Severity of depressive Not specified as part
et al., 2012 primary care treatment algorithm, self-help nurses, psychiatrists symptoms (PHQ-9) sign. better of the treatment
(38) centres activities, psychoeducation, in IG at 3 and 6 months, no algorithm
systematic evaluation, referral longer sign. at 12 months;
options to specialized care response rates and remission
rates sign. higher in the IG at
3,6 and 12 months
Bekelman USA, four Heart failure n = 392 Case reviews; use of electronic Usual care PCP, nurse No sign. difference in Integrated treatment
et al., 2015 Veterans medical record for communication; coordinators, improvement on the Kansas for heart-related
(54) Affairs depression care: psychoeducation, cardiologist, City Cardiomyopathy health and depression
centres problem-solving treatment, self- psychiatrist Questionnaire (KCCQ)
management, antidepressant between IG and control group
medication management; telehealth (CG) at 12 months
monitoring of heart and depressive
measures, self-care support
Coulton UK, 53 Hazardous n = 529 Twenty-minutes behaviour Usual care and Practice and research No sign. differences in the Not specified as part
et al., 2017 general alcohol use in change counselling by a practice 5-minute structured nurses, alcohol reduction of average drinks per of the treatment
(51) practices elderly patients or research nurse, Motivational advice session by therapists, specialist day (ADD) between IG and algorithm
Enhancement Therapy (MET) practice or research alcohol treatment CG (Alcohol Use Disorders
by alcohol therapist, referral to a nurse personnel Identification Test (AUDIT)
specialist alcohol treatment service
Coventry UK, 36 Depression n = 387 Eight sessions of brief psychological Usual care PCP, psychological Sign. greater reduction in Focus on depression
et al., 2015 general (and anxiety) therapy by case manager; wellbeing practitioners depressive symptoms [Self- (and anxiety) but links
(23) practices & diabetes collaborative meetings between as case managers, reported Symptom Checklist— to diabetes and/or
and/or patient, case manager and practice nurse, Depression Section (SCL-D13)] cardiovascular disease
cardiovascular practice nurse; case management; psychological therapists in the IG at 4 month (e.g. in case review
disease supervision as supervisors, study and management)
psychiatrist
Dobscha USA, five musculosceletal n = 401 Telephone/face-to-face assessments, Usual care Psychologist care Greater improvement in pain- Integrated treatment
et al., 2009 primary care pain psychoeducation workshop, case manager, internist, related disability [Roland- for pain and
(55) clinics reviews, self-management support, physical therapist Morris Disability Questionnaire depression
assistance with pain treatment, (RMDQ)] and in pain intensity
mental health consultation [Chronic Pain Grade (CPG)] in
the IG at 12 months; in patients
with depressive symptoms
greater improvement in
depression severity in the IG at
2 months (PHQ-9)
Family Practice, 2018, Vol. XX, No. XX

Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018

Table 1.  Continued

First author, Country/ Condition Sample size Intervention/treatment options Control condition Professionals involved Primary outcome Comorbidity
year setting addressed in treatment

Drummond UK, six Alcohol use n = 112 Forty-minute session of behaviour Usual care and Practice nurse, local Greater but not sign. better Not specified as part
et al., 2009 general disorder change counselling, a maximum of 5-minute directive alcohol specialist reduction in the alcohol of the treatment
(52) practices four 50-minute sessions of MET, advice session by services consumption measured over algorithm
referral to the local community practice nurse 180 days (Time Line Follow
alcoholTeam Back method) in the IG
Ell et al., USA, two Depression and n = 387 Problem-solving therapy, Enhanced usual PCP, social work Sign. higher chance of Focus on depression
Comorbidity in stepped care

2010 (24) public diabetes antidepressant medication, monthly care diabetes depression response in the IG [Symptom but self-management
safety-net telephone symptom monitoring and clinical specialists, Checklist Depression Scale strategies and service
clinics relapse prevention, care and service assistant patient (SCL-20)], sign. higher chance system navigation for
system navigation navigator, psychiatrist of remission in the IG after diabetes as well
6 and 18 months, not after
12 months (SCL-20), no sign.
differences in remission as
measured with PHQ-9
Fihn et al., USA, four Stable n = 703 case reviews, psychoeducation, Usual care PCP, cardiologist, No sign. improvement on the Not specified as part
2011 (57) affiliated ischaemic heart care management, referrals or internist, clinical nurse Seattle Angina Questionnaire of the treatment
Departments disease assessments specialist, pharmacist (SAQ) or in self-perceived algorithm
of Veterans health in the IG after
Affairs 12 months, higher guideline
health care concordance in the IG
system
Fortney USA, five Depression n = 364 ‘practice-based collaborative care Active comparison, PBCC: on-site PC Remission and response (SCL- Not specified as part
et al., 2013 Federally (PBCC)’: on-site nurse depression see left providers, on-site 20) sign. better in the TBCC of the treatment
(37) Qualified care managers, systematic nurse depression care after 6,12 and 18 months; algorithm
Health monitoring, psychoeducation, managers (registered sign. group × time interaction
Centres self-management, antidepressant nurses or licensed for depression severity in
treatment, referral to specialized practical nurses), referral both groups with greater
care possible‘telemedicine-based to off-site mental health improvement in the TBCC
collaborative care (TBCC)’: providers possible
off-site: nurse depression TBCC: on-site PC
care managers, psychiatrist, providers, off-site
pharmacist, psychologist; telephone depression care
monitoring and case management; manager, off-site
psychoeducation; self-management; pharmacist, off-site
antidepressant treatment psychologist, off-site
psychiatrist
Gilbody UK, 32 Depression n = 705 Collaborative care program; case Usual care PCP, case manager Sign. lower depression severity Not specified as part
et al., 2017 primary care management: telephone support, (background in (PHQ-9) in the IG at 4 month; of the treatment
(39) centres symptom monitoring, behavioural mental health nursing sign. smaller number of algorithm
activation; antidepressant treatment or psychology), depression diagnoses in the IG
psychiatrist after 12, not at 4 months
5

Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018

 6

Table 1.  Continued

First author, Country/ Condition Sample size Intervention/treatment options Control condition Professionals involved Primary outcome Comorbidity
year setting addressed in treatment

Hedrick USA, one Depression n = 354 Patient education, antidepressant Consult-liaison CC: PC providers, Sign. greater improvement in Not specified as part
et al., 2004 Veteran medication, adjuvant medication, (CL) care (access psychologist, depression symptoms (SCL- of the treatment
(41) Affairs cognitive-behavioural therapy to on-site psychiatrist, social 20) in the IG at 3 months, no algorithm
primary care (CBT) group, psychologist/ services, including workers, psychology longer sign. at 12 months
clinic psychiatrist appointment, referral psychiatrist) technician
to MH specialty care, telephone CL: PC provider,
support possible referral to
on-site psychiatrist,
psychologist or social
worker or specialty
mental health clinics
Huijbregts NL, 18 Depression n = 150 Care management, self-help Usual care PCP, depression care Response in depression Not specified as part
et al., 2013 primary care manual, problem-solving treatment, manager, consultant sign. greater in IG at 3 and of the treatment
(47) centres antidepressant medication psychiatrist 9 months, not sign. at 6 and algorithm
12 months
Katon et al., USA, nine Depression and n = 329 Antidepressant medication, Usual care PCP, depression Sign. less depression severity Not specified as part
2004 (26) primary diabetes problem-solving treatment, clinical specialist nurse, in the IG (SCL-90), higher of the treatment
clinics telephone and in-person psychiatrist supervisor, rating of patient-rated algorithm
appointments, psychiatric team psychologist global improvement at 6 and
consultation, referral to specialty 12 months, higher satisfaction
care, monthly telephone contacts with care at 6 and 12 months,
during maintenance phase, case no differences in HbA1c
reviews
Katon et al., USA, 14 Depression and n = 214 Support for self-management and Usual care PCP; study nurses; Sign. greater overall 12-month Integrated treatment
2010 (25) clinics diabetes and/or medication adherence; patient supervising psychiatrist, improvement in the IG in the for depression and
coronary heart education; pharmacotherapy PCP and psychologist combined outcome (glycated comorbid illness
disease for depression, hyperglycaemia, haemoglobin levels, low-
hypertension and hyperlipidaemia; density lipoprotein cholesterol
monitoring; case reviews; levels, systolic blood pressure
monitoring telephone calls in and depression scores (SCL-20)
maintenance phase
Kroenke USA, 6 depression and n = 250 Antidepressant medication, Usual care PCP, nurse care Sign. greater reduction in Integrated treatment
et al., 2009 community musculosceletal pain self-management program, manager, supervising depression severity in the IG for depression and
(19) clinics & pain telephone and in-person contacts physician depression at 12 months (SCL-20); sign. pain
5 Veteran during continuation phase, referral specialist greater reduction in pain in
Affairs to psychiatrist the IG [Brief Pain Inventory
general (BPI)]; sign. greater global
medicine improvement in pain in the IG
clinics (7-point-scale)
Family Practice, 2018, Vol. XX, No. XX

Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018

Table 1.  Continued

First author, Country/ Condition Sample size Intervention/treatment options Control condition Professionals involved Primary outcome Comorbidity
year setting addressed in treatment

Kroenke USA, five Chronic n = 250 Case reviews, automated symptom Usual care PCP, physician pain Sign. better improvement in Not specified as part
et al., 2014 primary care musculosceletal monitoring (interactive voice- specialist, nurse care pain in the IG after 12 months of the treatment
(56) clinics in pain recorded telephone calls or by manager (BPI) algorithm
one Veterans internet), care management through
Affairs in-person and telephone contacts,
medical analgesic optimization algorithm
Comorbidity in stepped care

centre
Lin et al., USA, four depression n = 228 Psychoeducation, behavioural usual care PCP, on-site psychiatrist Sign. less interference in Not specified as part
2000 (42) primary care activation, on-site psychiatric family, work + social activities of the treatment
clinics consultation, antidepressant in IG; trend towards more algorithm
medication, monitoring improvement in social
functioning in IG, no diff. in
role performance [Sheehan
Disability Scale (SDS)]
Muntingh NL, 43 Anxiety n = 180 Guided self-help and support Usual care PCP, psychiatric nurses Sign. greater reduction of Not specified as part
et al., 2014 general disorders by care manager, brief CBT as care managers, anxiety symptoms in the IG of the treatment
(16) practices interventions for generalized consulting psychiatrists, over 12 months [Beck Anxiety algorithm
anxiety disorder and panic disorder specialized mental Inventory (BAI)]
by care manager, antidepressant health care personnel
medication, referral to specialized
mental health care
Menchetti Italy, eight depression n = 227 Brief psychological interventions, Usual care PCP, consultant Higher share of remission (not Not specified as part
et al., 2013 primary pharmacotherapy, possible mental psychiatrist sign.) in IG after 3 months of the treatment
(43) care sites in health referral (PHQ-9) algorithm
three Italian
regions
Oladeji Nigeria, six depression n = 234 Psychoeducation, activity Usual care PCP, PC workers Focus on feasibility, better Not specified as part
et al., 2015 primary care scheduling, adapted version (nurses, community symptom improvement of the treatment
(40) centres of problem-solving treatment, health officers, (PHQ-9) algorithm
antidepressant medication community health
extension workers),
supervising
psychiatrists
Oosterbaan NL, 20 ‘Common n = 158 Guided self-help, antidepressant Usual care PCP, psychiatric Sign. greater improvement in Adaptation of
et al., 2013 general mental medication, CBT and nurses, psychologists, the IG mid-term at 4 months treatment components
(35) practices disorders’ pharmacotherapy in specialized psychiatrist [Clinical Global Impression of to the different
mental health care centre, further Improvement Scale (CGI-I)], disorders
specialized mental health care no sign. differences at 8 and
12 months
7

Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018

 8

Table 1.  Continued

First author, Country/ Condition Sample size Intervention/treatment options Control condition Professionals involved Primary outcome Comorbidity
year setting addressed in treatment

Patel et al., India, 12 depression and n = 2796 Case management by lay health Enhanced usual PCP, lay health Sign. greater chance of recovery Adaptation of
2010 (33) primary care anxiety counsellors, psychoeducation, care counsellors, visiting in the IG after 6 months treatment components
centres and antidepressant medication, psychiatrist (=clinical [Clinical Interview Schedule— to depression and/or
12 general interpersonal psychotherapy, specialist) Revised (CIS-R)] anxiety
practice adherence management, referral to
facilities clinical specialist
Pols et al., NL, 27 subthreshold n = 236 Watchful waiting with introductory Usual care PCP, practice nurses, No sign. difference in Not specified as part
2017 (27) primary care depression and meeting with practice nurse, written specialized mental cumulative incidence of of the treatment
centres diabetes and/or guided self-help and monitoring health care personnel depression [Mini-International algorithm
coronary heart calls by practice nurse, problem- Neuro-psychiatric Interview
disease solving treatment, referral to GP, (MINI)]
referral to specialized mental
health, antidepressant medication
Rollman USA, four generalized n = 191 Case reviews, psychoeducation, Usual care PCP, care managers, Sign. greater reduction in Not specified as part
et al., 2005 primary care anxiety workbook for self-management study team for case anxiety symptoms in the IG of the treatment
(50) practices disorder and with feedback, anxiolytic reviews after 12 months [Structured algorithm
panic disorder pharmacotherapy, referral to a Interview Guide for the
community mental health specialist Hamilton Anxiety Rating Scale
(SIGH-A)]
Rollman USA, seven Depression n = 302 Telephone-based case management Usual care PCP, nurse care Sign. greater improvement in Not specified as part
et al., 2009 university- after coronary (+151 non- by nurse care manager: manager, local mental health-related quality of the treatment
(32) based and artery bypass depressed psychoeducation, workbook, psychiatrists and of life in the IG at 8 months algorithm
community graft (CABG) CABG- antidepressant medication, watchful psychologists follow-up [Short Form 36
hospitals patients as waiting; referral to local mental Mental Component Score (SF-
comparison health specialist 36 MCS)]
group)
Rollman USA, six Anxiety n = 329 Case reviews, psychoeducation, Usual care PCP, care managers, Sign. higher improved mental Not specified as part
et al., 2017 affiliated disorders workbook for self-management psychiatrist, health-related quality of life of the treatment
(17) primary care with telephone care management, psychologist in the IG at 12 months (SF-36 algorithm
practices anxiolytic pharmacotherapy; MCS)
referral to a community mental
health specialist, telephone contacts
during continuation phase, use
of electronic medical records for
communication
Roy-Byrne USA, six Panic disorder n = 232 Psychoeducation; workbook on Usual care PCP, behavioural health Sign. higher proportion of Focus on panic
et al., 2005 primary care panic disorder; brief CBT by specialists, consulting responders and remitters, disorder but options
(49) clinics behavioural health specialists; care psychiatrist, greater sign. improvement in to account for
coordination; telephone contacts; WHO Disability Scale and social anxiety and
medication algorithm Short Form 12 (SF-12) mental depression)
health in the IG
Family Practice, 2018, Vol. XX, No. XX

Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018

Table 1.  Continued

First author, Country/ Condition Sample size Intervention/treatment options Control condition Professionals involved Primary outcome Comorbidity
year setting addressed in treatment

Roy-Byrne USA, 17 Anxiety n = 1004 Psychopharmacotherapy, CBT, Usual care PCP, anxiety clinical Sign. greater improvement Not specified as part
et al. 2010 primary care disorders care management and self- specialists, supervising in global anxiety symptoms of the treatment
(48) clinics management support, web-based psychiatrist and in the IG at all time-points algorithm
monitoring system, telephone calls psychologist [Brief Symptom Inventory-12
in continuation phase (BSI-12)]
Seekles et al., NL, 32 Depression and n = 120 Watchful waiting; guided self-help Usual care PCP, psychiatric No sign. difference in Adaptation of
Comorbidity in stepped care

2011 (34) general anxiety by care manager: either problem- nurses, psychologists, improvement between IG and treatment components
practices solving-based treatment (book specialized mental CG [Inventory of Depressive to depression and/or
or internet) with feedback or health care personnel Symptomatology (IDS) anxiety
exposure-based therapy (book) & Hospital Anxiety and
with feedback; problem-solving Depression Scale (HADS)]
treatment; pharmacotherapy and/
or referral to specialized mental
health care
Silverstone Canada, two Depression n = 1489 ‘Group 3’: screening + usual care + Group 1: screening Not specified Sign. reduction in depression Not specified as part
et al., 2017 primary care online CBT program as option Group 2: screening symptoms in all groups (PHQ- of the treatment
(44) clinics ‘Group 4 (in one clinic)’: + usual care 9), no sign. group differences algorithm
stepped care (watchful waiting,
self-management information,
additional visits, medication,
outside referral options)
Stoop et al., NL, 24 Depression n = 46 Disease Management program Usual care PCP, practice nurses Sign. lower level of anxiety in Adaptation of
2015 (28) general and anxiety for Co-morbid Depression and the IG at post-intervention and treatment components
practices in patients Anxiety (DiMaCoDeA): meetings 6 months-follow-up [General to depression and/or
allied to the with diabetes, for psychoeducation, ‘Coping with Anxiety Disorder-7 (GAD-7)]; anxiety
primary care asthma or depression course’/‘Coping with sign. lower level of depression
organization chronic anxiety course’, referral to GP, at post-intervention (PHQ-9),
PoZoB obstructive booster sessions but not in the fully adjusted
pulmonary model, not sign. at 6-month
disease (COPD) follow-up
Unützer USA, 18 depression n = 1801 psychoeducation, depression usual care PCP, depression care Sign. lower depression severity Not specified as part
et al., 2002 primary care care management, antidepressant managers (nurses in the IG (SCL-20), sign. higher of the treatment
(15) clinics medication, problem-solving or psychologists), response and remission rates algorithm
treatment supervising team in the IG
psychiatrist, liaison
PCP
van Beljouw NL, 18 Depression n = 263 Watchful waiting, guided self-help UUsual care PCP, nurses (mental Sign. short-term reduction in Not specified as part
et al., 2015 general or exercise program, problem- (stepped-wedge- health or home depression symptoms (PHQ-9) of the treatment
(45) practices and solving treatment or life-review, trial) care nurses), after 3 months, sign. dropout algorithm
1 home care referral to GP physiotherapists
organization
9

Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018

Table 1.  Continued
10

First author, Country/ Condition Sample size Intervention/treatment options Control condition Professionals involved Primary outcome Comorbidity
year setting addressed in treatment

van der NL, 36 Medically n = 81 Psychiatric consultation Usual care PCP, psychiatrist, Sign. greater improvement in Treatment options for
Feltz- general unexplained for PCPs, psychoeducation, CBT therapist, further severity of main MUS in the different psychological
Cornelis practices symptoms psychopharmacotherapy, CBT, referral options IG: no sign. difference in the comorbidities and
et al., 2006 (MUS) referral to psychiatrist, self- improvement in psychological pain
(53) management, further psychotherapy symptoms [Symptom
options, further referral options Checklist-90-Revised
(physical therapist, social worker) (SCL-90-R)]
van der NL, 67 Depression n = 239 Individual counselling in home Usual care PCP, community Sign. better improvement Not specified as part
Weele et al., general visits by a community psychiatric psychiatric nurse, in depression severity of the treatment
2012 (46) practices nurse, ‘Coping with depression mental health [Montgomery-Åsberg algorithm
course’, referral back to PCP professionals Depression Rating Scale
(MADRS)] in the CG at
6 months, no longer sign. at
12 months
van’t Veer- NL, 33 Prevention of n = 170 Watchful waiting, psychoeducation Usual care PCP, home care nurses, Sign. reduced cumulative Adaptation of
Tazelaar general depression and by home care nurse, adapted course specially trained incidence of depressive and treatment components
et al., 2009 practices anxiety in late ‘Coping With Depression and community psychiatric anxiety disorders in the IG to depression and/or
(29) life Anxiety’ and visits/phone calls by nurse after 12 months (MINI) anxiety
home care nurse, brief problem-
solving treatment, referral to PC
Vera et al., USA (Puerto depression and n = 179 Psychoeducation, care management, Usual care PCP, care manager, Sign. greater reduction of Focus on depression
2010 (30) Rico), 14 chronic medical antidepressant medication, CBT consulting psychiatrist, depressive symptoms in the IG
internal conditions psychologists, at 6 months (SCL-20)
medicine or
primary care
clinics
Zhang et al., China subthreshold n = 240 Watchful waiting, telephone Usual care PCP, trained social No sign. difference in the Adaptation of
2014 (31) (Hongkong), depression and/ counselling by trained socials workers, psychiatrist cumulative probability of onset treatment components
primary care or anxiety workers for self-management, face- of major depression or anxiety to depression and/or
clinics to-face problem-solving treatment disorder between IG and CG in anxiety
by trained social workers, referral 15 months [Structured Clinical
to PCP, referral to a psychiatrist Interview for DSM-IV (SCID)]
Family Practice, 2018, Vol. XX, No. XX

Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018

Comorbidity in stepped care
Figure 2.  Strategies used to account for comorbid disorders in diagnostics and treatment in stepped care
monitoring appointments (via phone or face-to-face) addressing the
symptoms of, for example, anxiety and depression, self-management
strategies like goal setting and activity scheduling for the benefit
of, for example, diabetes and depression and additional treatment
options for patients suffering from, for example, further mental ill-
ness like anxiety disorders (e.g. the ‘Coping with anxiety course’).
Interprofessional cooperation (ii) comprised case management, case
reviews and improved referral processes.
Telehealth strategies (iii) that allowed the management of more
than one disorder included the use of telephone or internet-based
treatment and monitoring components (e.g. case management, auto-
mated symptom monitoring). Additionally, shared electronic health
records were employed for better coordination between the health
care professionals.
Involved health care professionals
With regard to involved health care professionals in stepped care,
primary care physicians’ main tasks were the responsibility for
treatment decisions, case reviews and the prescription of medica-
tion. Primary care physicians were supported in 33 studies by case
managers, with different backgrounds predominantly in nursing,
and to a lesser extent in psychology or social work. Case manag-
ers were responsible for care coordination, monitoring and the
training in self-management strategies. The stepped care team was
complemented in 24 trials by a study psychiatrist as supervisor and/
or consultant. Psychologists or other mental health specialists were
frequently integrated either as part of the treatment team or as a
referral option if indicated. In a small number of studies additional
professionals, such as pharmacists, community or lay health care
workers, cardiologists, social workers or physiotherapists were part
of the stepped care team.
Efficacy of the primary-care-based stepped care
models with regard to evaluated primary outcomes
The heterogeneity of outcomes and the complexity of interventions
prevented us from applying meta-analytic analyses. However, the
following description of the primary outcomes allows for an overall
impression of the efficacy of the stepped care models. It is struc-
tured along the main disorders addressed. The first group comprises
comorbidity studies that addressed depression and one or more
11
Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018
mental and/or somatic disorders (labelled ‘comorbidity group’).
Studies of the second group addressed depression only (labelled
‘depression group’), and the last group comprises studies on condi-
tions other than depression (labelled ‘mixed group’).
The primary outcome was of psychological nature (e.g. depres-
sive symptoms) in 30 studies and of somatic nature (e.g. HbA1c as
diabetes marker) in 4 studies. Five studies combined both.
Comorbidity group
Of the 14 studies, addressing depression and one or more somatic
and/or mental disorder, 6 studies reported results in favour of stepped
care compared with the control condition (19,23,25,29,30,32). Four
studies could not state significant differences between both groups
(27,31,34,35), while four studies revealed mixed results with regard
to different outcome variables and the stability of treatment effects
(24,26,28,33). Taking a specific look at depression outcomes only
(e.g. reduction in depressive symptoms; response, remission, or inci-
dence rates; irrespective of whether depression was part of the pri-
mary or the secondary outcome), study results showed a significant
beneficial effect compared with the control condition in 7 out of 14
stepped care studies (19,23,25,26,29,30,32). No such positive effect
on depression symptoms (reduction or incidence) was found in 6 out
of 14 studies (27,28,31,33–35), while Ell et al. (24) reported a supe-
riority of stepped care with regard to response but not with regard
to remission as assessed with the Patient Health Questionnaire 9
(PHQ-9) at all points of measurement.
Depression group
In the 13 stepped care models addressing depression, the primary
outcomes predominantly focused on change in depression symp-
toms or remission and/or response rates with the exception of Lin
et al. (42) who focused on disability. In two studies, the stepped care
model reduced depression symptoms significantly greater than the
control condition (15,36). In the trial of Fortney et al. (37), the tele-
medicine-based intervention was superior to the practice-based trial
arm. In six studies, stepped care proved to be significantly superior
than the control condition with significant differences on the pri-
mary outcomes at some but not at all points of measurements (as
predefined as primary outcome) or not in every component of com-
posite primary outcomes (38,39,41,42,45,47). In four studies, no
12
significant results in favour of stepped care compared with the con-
trol condition with regard to the primary depression outcome were
found (40,43,44,46).
Mixed group
In the 12 studies addressing one or more disorders other than
depression, the primary outcomes were mainly disorder-specific
(except for Rollman et  al. (17) with mental health-related quality
of life as primary outcome). Seven studies revealed an overall supe-
riority of the stepped care model compared with the control condi-
tion (16,48–50,53,55,56), four did not find a benefit of stepped care
regarding the primary outcome (51,52,54,57). Rollman et  al. (17)
found significant improvement in quality of life after 12 but not after
24 months.
3) b) Results for quality of life and disability of the primary-care-
based stepped care models
The results for quality of life and disability are reported for all stud-
ies combined. The heterogeneity of outcomes and interventions pre-
vented us from applying meta-analytic analyses.
With regard to quality of life, five studies reported a significant
superiority of stepped care compared with the control condition
(15,19,25,36,39), whereas no significant differences were found in
11 studies (23,31,35,40,41,45,51,52,55–57). Mixed results emerged
in further nine studies, mainly with a significant improvement in the
mental component summary of the SF-36 in the intervention group
but no such advantage in the physical component summary of the
SF-36 (16,17,24,32,37,38,48–50). Quality of life was not measured
or reported in 14 studies (26–30,33,34,42–44,46,47,53,54).
The majority of studies (n = 24) did not assess or report compari-
sons of changes in disability. Yet, seven studies showed a significant
better outcome for the intervention group (15, 19,30,32,48,49,55).
In five studies, no superiority of the intervention group was found
(23,24,40,43,56). Mixed results with regard to disability were found
in three studies with significant improvements in the intervention
group only in some subdomains of functioning or at some time-
points of measurement (41,42,53).
Risk of bias
Overall, the risk of bias of the primary analyses with regard to ran-
dom sequence generation, allocation concealment and blinding of
outcome assessment was rather low. A  slightly higher risk due to
unclear or high risk was detected for incomplete outcome data, selec-
tive reporting and other biases (e.g. selective dropout or recruitment
in only one clinic). Due to the inclusion of psychological interven-
tions, a blinding of participants and personnel was usually not pos-
sible with the exception of additional care-team members in some
studies (such as consulting pharmacists or psychiatrists). A lack of
blinding resulted in a high risk of bias with regard to possible effects
of expectations of both caregivers and patients.
Discussion
The present systematic review investigated the consideration of
mental and somatic comorbidity in primary-care-based stepped care
models. Fifteen out of 39 identified studies considered comorbidity
in that they encompassed two or more disorders in their treatment
algorithms. Depression was the most frequently targeted condition,
both alone and comorbid to further mental or somatic disorders.
The care algorithms were mainly composed of self-management
Family Practice, 2018, Vol. XX, No. XX
support, psychotherapeutic approaches in varying intensity, (psy-
cho-)pharmacotherapy and optional referral to further specialist ser-
vices. Most health care teams included primary care physicians, case
or care managers and a psychiatrist and/or psychologist. Strategies
to account for comorbidities comprised (i) modular treatment, (ii)
collaboration between different health care professionals and (iii)
digital approaches for both treatment and communication.
Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018
In the majority of the 39 identified studies, stepped care models
in general showed superior (n = 16) or mixed (n = 10) results with
regard to their primary outcomes.
Results were mixed with regard to the outcome quality of life
(only reported in n  =  25 studies), with significant improvements
mainly in mental but not in physical parameters. Disability, as
another patient reported outcome, was not reported in the majority
of studies; the overall picture in those studies reporting disability
was positive to mixed. Due to the complexity of the interventions,
the heterogeneity of the disorders and the lack of transdiagnostic
outcomes, no meta-analysis was performed, and results did not
allow to judge whether the consideration of comorbidity in stepped
care increased the efficacy of these treatment models. Thus, future
research is needed to assess the efficacy of managing comorbidities
within stepped care models in terms of quality of life and disability.
The applied strategies to account for comorbidity via modu-
lar treatment options echo interventions used in collaborative care
and integrative treatment models (e.g. depression and diabetes (58);
depression and coronary disease (59)). There is evidence that such
integrative care models improved psychological (mainly depressive)
symptoms, whereas smaller effects were observed in chronic somatic
conditions and depending on the condition (58–62).
All examined stepped care models applied interprofessional
cooperation with case/care managers and health care professionals
from medical or psychological specialties as a major care compo-
nent to account for comorbidities. This result highlights again the
conceptual overlap between stepped and collaborative care. Lessons
learned from collaborative care research indicate that interprofes-
sional cooperation and task delegation are particularly helpful for
an integrated treatment of both mental and somatic complaints
(63). Yet, establishing interprofessional cooperation in routine care
requires clear care pathways, role definitions and good communica-
tion between the professionals involved, as otherwise these factors
might impede on its implementation (64). Within distinct stepped
care algorithms, interprofessional collaboration can be established
in such a structured way that clearly defines responsibilities along
the different treatment steps. With regard to the implementation of
interprofessional cooperation, the different primary care systems and
professional roles have to be taken into account. Despite the pre-
dominance in this systematic review of studies conducted in the USA,
the diversity of settings, ranging from multi-professional primary care
clinics to single practices and to the inclusion of lay health care work-
ers, illustrates the adaptability of stepped care models to the respect-
ive settings.
The third strategy to account for comorbidities was the use of
digital approaches for treatment, monitoring and interprofessional
cooperation, for example, as optional psychotherapy via inter-
active video (37) or shared electronic health records (55). Telehealth
in treatment and monitoring is on the rise, and there is evidence
for their effectiveness in primary care, such as for depression (65).
Technology may enhance the array of treatment options in primary
care, especially for mild disorders, for patients in rural areas (66)
and for patients with complex treatment needs, such as comorbidity.
With regard to digital strategies for the cooperation between health
Comorbidity in stepped care
care professionals, there is hope that shared electronic health care
records may improve cooperation and continuity of care, reduce
redundant testing, treatment and costs (67,68), yet the overall scarce
evidence-base calls for further research (67,69). Country-specific
data-management regulations, health care systems and treatment
infrastructures require an adaptation of technology-based strategies
within stepped care approaches.
Taken together, evaluations of stepped care models that account
for comorbidity have shown first promising results in randomized
controlled trials with regard to different outcome domains. Also,
they comprise management strategies that have already been proven
useful in other care models.
Strengths and limitations
The present review was, to the best of our knowledge, the first to shed
light on the integration of comorbidity treatment in current stepped
care model studies. The broad search strategy adopted in this review
allowed the inclusion of a wide range of studies even when a stepped
intervention was not labelled as such but implemented an adap-
tive treatment algorithm and systematic monitoring. Yet, the term
‘stepped care’ is difficult to delineate since adaptive treatment models
have been studied for a much longer time than the label ‘stepped care’
is used, and there is significant overlap between stepped and collabo-
rative care. While the overall delineation between stepped care and
collaborative care is difficult to draw and the models have been com-
bined in some studies (18), the main characteristic of stepped care is
the structured care pathway and its systematic monitoring.
The heterogenic spectrum of disorders and varying degrees of
treatment duration, treatment intensity and illness severity ham-
pered a meta-analytic approach regarding the efficacy of integrat-
ing comorbidity treatment in stepped care. Thus, whether stepped
care models integrating comorbidity treatment are more successful
than monomorbid trials cannot be answered based on the current
evidence and requires further research. In addition, while effect-
ive stepped care models for two or three frequently co-occurring
disorders would improve primary care, the essential challenge is to
develop care for multimorbidity. While the definitions of both con-
cepts, comorbidity and multimorbidity, remain controversial, the
concept of multimorbidity, in its comprehensive meaning, encom-
passes a broader perspective that takes into account aspects like
biopsychosocial factors, health care consumption or the burden
of disease (70). The stepped care models of the included studies in
this review overwhelmingly addressed a small number of clearly
defined disorders, not taking a multimorbid perspective which led
to the decision to focus on comorbidity. Due to the lack of pri-
mary-care-based stepped care models concerned with multimor-
bidity, this review cannot add specific knowledge on how to care
for multimorbid patients. Yet, the identification of useful strategies
to account for more than one disorder is a first step on the way
to a much larger toolbox that may help primary care to tackle the
challenge of multimorbidity.
The exclusion of particular mental disorders such as schizo-
phrenia, bipolar disorders and/or present substance abuse limited
the range of possible comorbidities to be considered in stepped care
models. This aspect illustrates the difference between scientific effi-
cacy studies and clinical reality and highlights the urgent need for
effectiveness studies.
With regard to the risk-of-bias-rating, the overall study quality
was found to be acceptable, and the exclusion of non-randomized
and/or non-controlled trials strengthened the validity of outcome
comparisons with usual care.
13
Conclusion and future research
This systemic review provides a first overview of primary care-based
stepped care models with regard to the management of mental and
somatic comorbidities. Fifteen out of 39 identified studies have imple-
mented strategies that enable the integrated treatment of two or more
disorders. These strategies could be classified in modular assessment
Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018
treatment components, interprofessional cooperation and the use of tel-
ehealth options for both treatment and interprofessional communica-
tion. While these strategies have proven useful in other care models, the
complex stepped care models and their current level of evaluation did
not allow to estimate both the efficacy and the effectiveness of stepped
care models accounting for comorbidity. In view of the challenges posed
by different comorbidity patterns and by multimorbidity, future studies
should adopt a broader perspective than addressing a limited number
of fixed comorbid disorders. Primary care, with its holistic approach, is
the pivotal setting for developing stepped care models that allow for an
adaptable care in case of different co-occurring disorders and different
patient priorities with regard to health outcomes.
Further research in stepped care needs to address the questions
of effectiveness both with regard to the stepped care algorithms and
the required collaborating providers while acknowledging patient
preferences, especially in case of comorbidity. In order to allow for
a comparative evaluation of effectiveness, transdiagnostic outcome
parameter such as quality of life and disability should be assessed
and reported. Since the investigated stepped care models predomi-
nantly addressed depression, there is an urgent need to expand
research to further and particularly to severe mental disorders such
as schizophrenia and psychosis.
The identified primary-care-based stepped care models with their
different strategies may serve as a blueprint for the development of
such future care models.
Supplementary material
Supplementary material is available at Family Practice online.
Acknowledgements
We would like to thank Olaf von dem Knesebeck, head of the Institute of
Medical Sociology at University Medical Center Hamburg-Eppendorf, for his
valuable and constructive feedback on the final manuscript.
Declaration
Funding: the PhD position of KM and the position of DH are funded by the
German Federal Ministry of Education and Research (BMBF) as part of the
trial COMET (grant no. 01GY1602) (Collaborative and Stepped Care in
Mental Health by Overcoming Treatment Sector Barriers; NCT03226743).
Ethical approval: ethical approvals were not required for this work.
Conflict of interest: none.
References
1. World Organization of National Colleges, Academies and Academic
Associations of General Practitioners/Family Physicians, Region Europe.
The European Definition of General Practice/Family Medicine. http://
www.woncaeurope.org/sites/default/files/documents/Definition%20
3rd%20ed%202011%20with%20revised%20wonca%20tree.pdf
(accessed on 2 April 2018).
2. Valderas JM, Starfield B, Sibbald B, Salisbury C, Roland M. Defining
comorbidity: implications for understanding health and health services.
Ann Fam Med 2009; 7: 357–63.
14
3. Bekhuis E, Boschloo L, Rosmalen JG, de Boer MK, Schoevers RA. The
impact of somatic symptoms on the course of major depressive disorder. J
Affect Disord 2016; 205: 112–8.
4. Katon WJ. Epidemiology and treatment of depression in patients with
chronic medical illness. Dialogues Clin Neurosci 2011; 13: 7–23.
5. Starfield B, Shi L, Macinko J. Contribution of primary care to health sys-
tems and health. Milbank Q 2005; 83: 457–502.
6. Richards DA, Bower P, Pagel C et al. Delivering stepped care: an analysis
of implementation in routine practice. Implement Sci 2012; 7: 3.
7. Gunn J, Diggens J, Hegarty K, Blashki G. A systematic review of complex
system interventions designed to increase recovery from depression in pri-
mary care. BMC Health Serv Res 2006; 6: 88.
8. Cross SPM, Hickie I. Transdiagnostic stepped care in mental health. Public
Health Res Pract 2017; 27: e2721712.
9. Lugtenberg M, Burgers JS, Clancy C, Westert GP, Schneider EC. Current
guidelines have limited applicability to patients with comorbid conditions:
a systematic analysis of evidence-based guidelines. PLoS One 2011; 6:
e25987.
10. Mutasingwa DR, Ge H, Upshur RE. How applicable are clinical prac-
tice guidelines to elderly patients with comorbidities? Can Fam Physician
2011; 57: e253–62.
11. Carlsen B, Glenton C, Pope C. Thou shalt versus thou shalt not: a meta-
synthesis of GPs’ attitudes to clinical practice guidelines. Br J Gen Pract
2007; 57: 971–8.
12. National Institute for Health and Care Excellence. Common Mental
Health Problems: Identification and Pathways to Care (CG123). https://
www.nice.org.uk/guidance/cg123/evidence/full-guideline-pdf-181771741
(accessed on 14 August 2017).
13. Deutsche Gesellschaft für Psychiatrie und Psychotherapie, Psychosomatik und
Nervenheilkunde; Bundesärztekammer; Kassenärztliche Bundesvereinigung;
Arbeitsgemeinschaft der Wissenschaftlichen Medizinischen
Fachgesellschaften. S3-Leitlinie/Nationale VersorgungsLeitlinie Unipolare
Depression. Langfassung, 2. Auflage, Version 5 (S3-guideline/national
guideline Unipolar Depression — long version, 2nd ed., version 5). 2015.
http://www.awmf.org/uploads/tx_szleitlinien/nvl-005l_S3_Unipolare_
Depression_2017-05.pdf. (accessed on 3 August 2017).
14. olde Hartman TC, Rosendal M, Aamland A, et  al. What do guidelines
and systematic reviews tell us about the management of medically unex-
plained symptoms in primary care? BJGP Open 2017; 1. BJGP-2016-
0868. doi:10.3399/bjgpopen17X101061.
15. Unützer J, Katon W, Callahan CM et al.; IMPACT Investigators. Improving
Mood-Promoting Access to Collaborative Treatment. Collaborative care
management of late-life depression in the primary care setting: a rand-
omized controlled trial. JAMA 2002; 288: 2836–45.
16. Muntingh A, van der Feltz-Cornelis C, van Marwijk H et al. Effectiveness
of collaborative stepped care for anxiety disorders in primary care: a prag-
matic cluster randomised controlled trial. Psychother Psychosom 2014;
83: 37–44.
17. Rollman BL, Belnap BH, Mazumdar S et al. Telephone-delivered stepped
collaborative care for treating anxiety in primary care: a randomized con-
trolled trial. J Gen Intern Med 2017; 32: 245–55.
18. Löwe B, Piontek K, Daubmann A et al. Effectiveness of a stepped, collab-
orative, and coordinated health care network for somatoform disorders
(Sofu-Net): a controlled cluster cohort study. Psychosom Med 2017; 79:
1016–24.
19. Kroenke K, Bair MJ, Damush TM et al. Optimized antidepressant therapy
and pain self-management in primary care patients with depression and
musculoskeletal pain: a randomized controlled trial. JAMA 2009; 301:
2099–110.
20. Firth N, Barkham M, Kellett S. The clinical effectiveness of stepped care
systems for depression in working age adults: a systematic review. J Affect
Disord 2015; 170: 119–30.
21. van Straten A, Seekles W, van ‘t Veer-Tazelaar NJ, Beekman AT, Cuijpers P.
Stepped care for depression in primary care: what should be offered and
how? Med J Aust 2010; 192(11 suppl): S36–9.
22. Higgins JPT, Altman DG, Sterne JAC. Chapter 8: Assessing risk of bias in
included studies. In Higgins JPT, Green S (eds). Cochrane Handbook for
Family Practice, 2018, Vol. XX, No. XX
Systematic Reviews of Interventions Version 5.1.0 (updated March 2011).
The Cochrane Collaboration, 2011. www.handbook.cochrane.org.
23. Coventry P, Lovell K, Dickens C et al. Integrated primary care for patients
with mental and physical multimorbidity: cluster randomised controlled
trial of collaborative care for patients with depression comorbid with dia-
betes or cardiovascular disease. BMJ 2015; 350: h638.
24. Ell K, Katon W, Xie B et  al. Collaborative care management of major
Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018
depression among low-income, predominantly Hispanic subjects with dia-
betes: a randomized controlled trial. Diabetes Care 2010; 33: 706–13.
25. Katon WJ, Lin EH, Von Korff M et al. Collaborative care for patients with
depression and chronic illnesses. N Engl J Med 2010; 363: 2611–20.
26. Katon WJ, Von Korff M, Lin EH et al. The Pathways Study: a randomized
trial of collaborative care in patients with diabetes and depression. Arch
Gen Psychiatry 2004; 61: 1042–9.
27. Pols AD, van Dijk SE, Bosmans JE et al. Effectiveness of a stepped-care inter-
vention to prevent major depression in patients with type 2 diabetes melli-
tus and/or coronary heart disease and subthreshold depression: a pragmatic
cluster randomized controlled trial. PLoS One 2017; 12: e0181023.
28. Stoop CH, Nefs G, Pommer AM, Pop VJ, Pouwer F. Effectiveness of a
stepped care intervention for anxiety and depression in people with dia-
betes, asthma or COPD in primary care: a randomized controlled trial. J
Affect Disord 2015; 184: 269–76.
29. van’t Veer-Tazelaar PJ, van Marwijk HW, van Oppen P et al. Stepped-care
prevention of anxiety and depression in late life: a randomized controlled
trial. Arch Gen Psychiatry 2009; 66: 297–304.
30. Vera M, Perez-Pedrogo C, Huertas SE et  al. Collaborative care for
depressed patients with chronic medical conditions: a randomized trial in
Puerto Rico. Psychiatr Serv 2010; 61: 144–50.
31. Zhang DX, Lewis G, Araya R et al. Prevention of anxiety and depres-
sion in Chinese: a randomized clinical trial testing the effectiveness of
a stepped care program in primary care. J Affect Disord 2014; 169:
212–20.
32. Rollman BL, Belnap BH, LeMenager MS et al. Telephone-delivered collab-
orative care for treating post-CABG depression: a randomized controlled
trial. JAMA 2009; 302: 2095–103.
33. Patel V, Weiss HA, Chowdhary N et al. Effectiveness of an intervention led
by lay health counsellors for depressive and anxiety disorders in primary
care in Goa, India (MANAS): a cluster randomised controlled trial. Lancet
2010; 376: 2086–95.
34. Seekles W, van Straten A, Beekman A, van Marwijk H, Cuijpers P. Stepped
care treatment for depression and anxiety in primary care. a randomized
controlled trial. Trials 2011; 12: 171.
35. Oosterbaan DB, Verbraak MJ, Terluin B et  al. Collaborative stepped
care v. care as usual for common mental disorders: 8-month, cluster ran-
domised controlled trial. Br J Psychiatry 2013; 203: 132–9.
36. Araya R, Rojas G, Fritsch R et  al. Treating depression in primary care
in low-income women in Santiago, Chile: a randomised controlled trial.
Lancet 2003; 361: 995–1000.
37. Fortney JC, Pyne JM, Mouden SB et al. Practice-based versus telemedicine-
based collaborative care for depression in rural federally qualified health
centers: a pragmatic randomized comparative effectiveness trial. Am J
Psychiatry 2013; 170: 414–25.
38. Aragonès E, Piñol JL, Caballero A et al. Effectiveness of a multi-compo-
nent programme for managing depression in primary care: a cluster rand-
omized trial. The INDI project. J Affect Disord 2012; 142: 297–305.
39. Gilbody S, Lewis H, Adamson J et al. Effect of collaborative care vs usual
care on depressive symptoms in older adults with subthreshold depression:
the CASPER randomized clinical trial. JAMA 2017; 317: 728–37.
40. Oladeji BD, Kola L, Abiona T et  al. A pilot randomized controlled trial
of a stepped care intervention package for depression in primary care in
Nigeria. BMC Psychiatry 2015; 15: 96.
41. Hedrick SC, Chaney EF, Felker B et al. Effectiveness of collaborative care
depression treatment in Veterans’ Affairs primary care. J Gen Intern Med
2003; 18: 9–16.
42. Lin EH, VonKorff M, Russo J et al. Can depression treatment in primary
care reduce disability? A stepped care approach. Arch Fam Med 2000; 9:
1052–8.
Comorbidity in stepped care
43. Menchetti M, Sighinolfi C, Di Michele V et al. Effectiveness of collabora-
tive care for depression in Italy. A randomized controlled trial. Gen Hosp
Psychiatry 2013; 35: 579–86.
44. Silverstone PH, Rittenbach K, Suen VYM et al. Depression outcomes in
adults attending family practice were not improved by screening, stepped-
care, or online CBT during a 12-week study when compared to controls in
a randomized trial. Front Psychiatry 2017; 8: 32.
45. van Beljouw IM, van Exel E, van de Ven PM et al. Does an outreaching
stepped care program reduce depressive symptoms in community-dwelling
older adults? A randomized implementation trial. Am J Geriatr Psychiatry
2015; 23: 807–17.
46. van der Weele GM, de Waal MW, van den Hout WB et  al. Effects of a
stepped-care intervention programme among older subjects who screened
positive for depressive symptoms in general practice: the PROMODE ran-
domised controlled trial. Age Ageing 2012; 41: 482–8.
47. Huijbregts KM, de Jong FJ, van Marwijk HW et al. A target-driven col-
laborative care model for major depressive disorder is effective in primary
care in the Netherlands. A randomized clinical trial from the depression
initiative. J Affect Disord 2013; 146: 328–37.
48. Roy-Byrne P, Craske MG, Sullivan G et  al. Delivery of evidence-based
treatment for multiple anxiety disorders in primary care: a randomized
controlled trial. JAMA 2010; 303: 1921–8.
49. Roy-Byrne PP, Craske MG, Stein MB et  al. A randomized effectiveness
trial of cognitive-behavioral therapy and medication for primary care
panic disorder. Arch Gen Psychiatry 2005; 62: 290–8.
50. Rollman BL, Belnap BH, Mazumdar S et al. A randomized trial to improve
the quality of treatment for panic and generalized anxiety disorders in pri-
mary care. Arch Gen Psychiatry 2005; 62: 1332–41.
51. Coulton S, Bland M, Crosby H et al. Effectiveness and cost-effectiveness of
opportunistic screening and stepped-care interventions for older alcohol
users in primary care. Alcohol Alcohol 2017; 52: 655–64.
52. Drummond C, Coulton S, James D et al. Effectiveness and cost-effective-
ness of a stepped care intervention for alcohol use disorders in primary
care: pilot study. Br J Psychiatry 2009; 195: 448–56.
53. van der Feltz-Cornelis CM, van Oppen P, Adèr HJ, van Dyck R.
Randomised controlled trial of a collaborative care model with psychi-
atric consultation for persistent medically unexplained symptoms in gen-
eral practice. Psychother Psychosom 2006; 75: 282–9.
54. Bekelman DB, Plomondon ME, Carey EP et  al. Primary Results of the
Patient-Centered Disease Management (PCDM) for Heart Failure Study:
a randomized clinical trial. JAMA Intern Med 2015; 175: 725–32.
55. Dobscha SK, Corson K, Perrin NA et  al. Collaborative care for chronic
pain in primary care: a cluster randomized trial. JAMA 2009; 301:
1242–52.
15
56. Kroenke K, Krebs EE, Wu J et al. Telecare collaborative management of chronic
pain in primary care: a randomized clinical trial. JAMA 2014; 312: 240–8.
57. Fihn SD, Bucher JB, McDonell M et al. Collaborative care intervention for
stable ischemic heart disease. Arch Intern Med 2011; 171: 1471–9.
58. Atlantis E, Fahey P, Foster J. Collaborative care for comorbid depression and
diabetes: a systematic review and meta-analysis. BMJ Open 2014; 4: e004706.
59. Tully PJ, Baumeister H. Collaborative care for comorbid depression and
Downloaded from https://academic.oup.com/fampra/advance-article-abstract/doi/10.1093/fampra/cmy122/5233007 by Insead user on 26 December 2018
coronary heart disease: a systematic review and meta-analysis of ran-
domised controlled trials. BMJ Open 2015; 5: e009128.
60. Lemmens LC, Molema CC, Versnel N, Baan CA, de Bruin SR. Integrated
care programs for patients with psychological comorbidity: a systematic
review and meta-analysis. J Psychosom Res 2015; 79: 580–94.
61. Woltmann E, Grogan-Kaylor A, Perron B et al. Comparative effectiveness
of collaborative chronic care models for mental health conditions across
primary, specialty, and behavioral health care settings: systematic review
and meta-analysis. Am J Psychiatry 2012; 169: 790–804.
62. van Eck van der Sluijs JF, Castelijns H, Eijsbroek V et al. Illness burden
and physical outcomes associated with collaborative care in patients with
comorbid depressive disorder in chronic medical conditions: a systematic
review and meta-analysis. Gen Hosp Psychiatry 2018; 50(suppl C): 1–14.
63. Goodrich DE, Kilbourne AM, Nord KM, Bauer MS. Mental health col-
laborative care and its role in primary care settings. Curr Psychiatry Rep
2013; 15: 383.
64. Wood E, Ohlsen S, Ricketts T. What are the barriers and facilitators to
implementing Collaborative Care for depression? A  systematic review. J
Affect Disord 2017; 214: 26–43.
65. Linde K, Sigterman K, Kriston L et al. Effectiveness of psychological treat-
ments for depressive disorders in primary care: systematic review and
meta-analysis. Ann Fam Med 2015; 13: 56–68.
66. Raney L, Bergman D, Torous J, Hasselberg M. Digitally driven integrated
primary care and behavioral health: how technology can expand access to
effective treatment. Curr Psychiatry Rep 2017; 19: 86.
67. Fontaine P, Ross SE, Zink T, Schilling LM. Systematic review of health
information exchange in primary care practices. J Am Board Fam Med
2010; 23: 655–70.
68. Wu H, LaRue EM. Linking the health data system in the U.S.: challenges
to the benefits. Int J Nurs Sci 2017; 4: 410–417.
69. Hersh WR, Totten AM, Eden KB et al. Outcomes from health information
exchange: systematic review and future research needs. JMIR Med Inform
2015; 3: e39.
70. Le Reste JY, Nabbe P, Manceau B et al. The European General Practice
Research Network presents a comprehensive definition of multimorbidity
in family medicine and long term care, following a systematic review of
relevant literature. J Am Med Dir Assoc 2013; 14: 319–25.