Curth et al.

Trials (2017) 18:382

DOI 10.1186/s13063-017-2120-3

STUDYPROTOCOL Open Access

Collaborative care for panic disorder,

generalised anxiety disorder and
social phobia in general practice:
study protocol for three cluster-
randomised, superiority trials
1 1* 2
Nadja Kehler Curth , Ursula Ødum Brinck-Claussen , Annette Sofie Davidsen , Marianne
3 7 4 5
Engelbrecht Lau , Merete Lundsteen , John Hagel Mikkelsen , Claudio Csillag , Carsten
1 6 1
Hjorthøj , Merete Nordentoft and Lene Falgaard Eplov

Abstract
Background: People with anxiety disorders represent a significant part of a general practitioner’s patient population.
However, there are organisational obstacles for optimal treatment, such as a lack of coordination of illness management and
limited access to evidence-based treatment such as cognitive behavioral therapy. A limited number of studies suggest that
collaborative care has a positive effect on symptoms for people with anxiety disorders. However, most studies are carried
out in the USA and none have reported results for social phobia or generalised anxiety disorder separately. Thus, there is a
need for studies carried out in different settings for specific anxiety populations.
A Danish model for collaborative care (the Collabri model) has been developed for people diagnosed
with depression or anxiety disorders. The model is evaluated through four trials, of which three will be
outlined in this protocol and focus on panic disorder, generalised anxiety disorder and social phobia. The
aim is to investigate whether treatment according to the Collabri model has a better effect than usual
treatment on symptoms when provided to people with anxiety disorders.
Methods: Three cluster-randomised, clinical superiority trials are set up to investigate treatment according to the
Collabri model for collaborative care compared to treatment-as-usual for 364 patients diagnosed with panic disorder,
generalised anxiety disorder and social phobia, respectively (total n = 1092). Patients are recruited from general
practices located in the Capital Region of Denmark. For all trials, the primary outcome is anxiety symptoms (Beck
Anxiety Inventory (BAI)) 6 months after baseline. Secondary outcomes include BAI after 15 months, depression
symptoms (Beck Depression Inventory) after 6 months, level of psychosocial functioning (Global Assessment of
Functioning) and general psychological symptoms (Symptom Checklist-90-R) after 6 and 15 months.
Discussion: Results will add to the limited pool of information about collaborative care for patients with
anxiety disorders. To our knowledge, these will be the first carried out in a Danish context and the first to
report results for generalised anxiety and social phobia separately. If the trials show positive results, they
could contribute to the improvement of future treatment of anxiety disorders.
(Continued on next page)

* Correspondence: [email protected]
1
Mental Health Center Copenhagen, Mental Health Services, Capital Region
of Denmark, Kildegårdsvej 28, 2900 Hellerup, Denmark
Full list of author information is available at the end of the article

© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Curth et al. Trials (2017) 18:382 Page 2 of 13

(Continued from previous page)

Trial registration: ClinicalTrials.gov, ID: NCT02678624. Retrospectively registered 7 February 2016; last
updated 15 August 2016,
Keywords: Collaborative care, Shared care, Anxiety disorders, Panic disorder, Generalised anxiety
disorder, Social phobia, Cluster-randomised trial, General practice

Background This treatment gap calls for more accessible evidence-

Across surveys, anxiety disorders are found to be the most based treatment opportunities in general practice.
prevalent group of mental disorders in the gen-eral Danish guidelines for anxiety disorders and depression
population, with an estimated lifetime prevalence of 16% suggest that improving the treatment in general practice
on average for any anxiety disorder [1]. Among patients should be done by introducing shared-care interventions such
with anxiety disorders, there is a high level of as collaborative care programs [19, 29]. Collaborative care
comorbidity with other psychiatric disorders such as other models for anxiety disorders and depression are com-plex
anxiety disorders, depression and alco-hol abuse [2–4], interventions; however, there are four criteria which should
but also physical illnesses [5, 6]. The human costs related be fulfilled: (1) a multi-professional approach to treatment,
to anxiety disorders are high and place a significant (2) scheduled follow-up, (3) enhanced interpro-fessional
burden on the economy due to high levels of health care communication and (4) treatment according to a structured
use, lost income and sick leave. In Denmark, the total cost treatment plan [30].
related to anxiety disorders is estimated to be six billion Studies on collaborative care for anxiety disorders are
DKK per year [7], approximately US$900 million. limited. According to a Cochrane review from 2012,
collaborative care between the primary and secondary
The majority of patients with anxiety disorders in Denmark care system provides a useful addition to clinical path-
are treated in general practice [8]. For general practitioners ways for the treatment of anxiety disorders and depres-
(GPs), people with anxiety disorders repre-sent a significant sion and shows significant improvements in treatment
part of their patient population as they use health care outcomes for up to 2 years compared with treatment-as-
services more frequently compared to patients without usual in general practice [30]. However, the review only
anxiety disorders [9–11]. It is estimated that 1–7% of the included four studies reporting results for panic disorder,
patients in general practice have panic disorder [12–14], 4– generalised anxiety disorder or social phobia and they
12% have generalised anxiety disorder [3, 12, 13, 15] and 4– were all conducted in the USA [31–34]. A recent review
6% have social phobia [3, 13]. updated the evidence base for anxiety disorders [35] and
Patients with non-psychotic disorders, such as anxiety included an additional three studies from Germany and
disorders, are poorly recognised and often not treated suf- The Netherlands [36–38] and also found collaborative
ficiently [3, 15–22]. The main obstacles for adequate treat- care superior to treatment-as-usual, with a small effect
ment are poor coordination between sectors and a lack of size for all anxiety disorders combined and a moderate
competent treatment availability in general practice [23]. effect size in a subgroup of studies on patients with panic
Cognitive behavioural therapy (CBT) is an evidence-based disorder. Additional analysis in the review [35] also
treatment for anxiety disorders and recommended in treat- showed a greater effect size in studies performed in the
ment guidelines [19, 24]. In Denmark, GPs can refer patients USA compared to studies performed in Europe in studies
either to an independent psychologist through an existing that included a care manager and in studies using stepped
public scheme, which includes up to 24 sessions with partly collaborative care, although this is to be interpreted with
subsidised fees for patients between 18 and 38 years, or a caution due to the small number of studies and the
public insurance-paid independent psychiatrist, where they heterogeneity between the European studies. Thus,
often have to wait for months due to long wait-ing lists [23]. current evidence builds on few studies all conducted in
GPs can offer a limited number of sessions of psychotherapy, organisational settings not comparable with the Danish or
but only if they receive supervision [25]. It is estimated that Scandinavian ones. Furthermore, no studies report results
only around one third of GPs fulfil these requirements [23, for social phobia or generalised anxiety disorder
26–28]. Furthermore, there is no train-ing in psychotherapy separately. Therefore, there is a need for further research
in GPs’ specialist training and no requirements for the GP to into collaborative care models facilitating a more effective
participate in postgraduate training to improve their skills implementation of the intervention as well as for specific
and ensure that they are continuously updated and provide a anxiety populations in different organisational contexts.
high quality of care.
Curth et al. Trials (2017) 18:382 Page 3 of 13

The development of a Danish model for collaborative with the inclusion and exclusion criteria. These criteria
care (the Collabri model) for panic disorder, generalised are assessed by the GP at recruitment and/or a research
anxiety disorder and social phobia was completed in 2014 assistant at a baseline eligibility interview.
in collaboration between Danish GPs, psychiatrists and
researchers. In this paper the protocol (version 2) for three Inclusion criteria Participating patients should be at least
cluster-randomised controlled, superiority trials 18 years old, diagnosed in general practice according to
investigating collaborative care, according to the Collabri the International Classification of Diseases 10th edition
model, vs. treatment-as-usual for panic disorder, general- (ICD-10) criteria for panic disorder (F41.0), generalised
ised anxiety disorder and social phobia will be outlined. anxiety disorder (F41.1) or social phobia (F40.1) [39].
As we developed specific collaborative care treatment They must speak Danish and give their written consent to
models for each of the three anxiety disorders, and wanted participation.
to examine the effect of each of these three collaborative
treatment models, we designed three cluster-randomised Exclusion criteria Patients cannot participate in the trials
controlled, superiority trials; but as the trials for each of if they are pregnant, have a pending disability pension
three anxiety disorders are similar in terms of aim, design application, are at a high risk of suicide, have a current
and methods, they are jointly presented in this paper. The psychotic condition, obsessive compulsive disorder, or
hypothesis for each of the trials is that treatments bipolar affective disorder, misuse alcohol or other sub-
according to the Collabri model are more effective than stances to an extent that will hinder the patient’s participa-
treatment-as-usual. tion in treatment, if they receive current psychological or
psychiatric treatment for depression or anxiety or have
Methods been receiving such treatment within the last 6 months.
Aim and design Also, referral to the secondary care system before or at
The aim of the three trials is to test if treatment accord-ing the first GP consultation after baseline eligibility
to the Collabri model is more effective than treatment-as- interview, because of a need for more specialised care, is
usual. The null hypothesis for each trial to be rejected is a reason for exclusion. Furthermore, patients cannot
that collaborative care according to the Collabri model participate if it is clinically assessed by the GP that they
(intervention group) and treatment-as-usual (control have dementia or an unstable medical condition that
group) have the same effect on anxiety symptoms for hinders the patient from adhering to treatment. Patients in
patients with panic disorder, generalised anxiety disorder the intervention group are excluded if they want treatment
and social phobia in general practice. The trials are according to the existing public psychologist scheme and
designed as three independent, two-armed, cluster- do not want referral to the psychologist to be preceded by
randomised, clinical superiority trials (see the flow chart treatment according to the Collabri model.
in Fig. 1). All aspects of the study design (including
assessment time points, outcome measures, etc.) are iden- Exclusion criteriaRecruitment and randomisation
tical in the three trials. A total n = 1092 is required based Cluster level
on sample size calculations; 364 patients in each of the The GP recruitment period lasted from May 2014 until July
three trials (diagnoses) on panic disorder, generalised 2015. GP practices were invited to participate via postal
anxiety disorder and social phobia, respectively. letters. This included easy-to-read information about the
project and an invitation to attend information meetings.
Eligibility criteria for participants Eligible GPs were randomised to either provide treatment
Cluster level according to the Collabri model or treatment-as-usual. A
GPs with a registered provider number, practicing in the cluster corresponds to a GP provider number which may
Capital Region of Denmark, are eligible for the trials. For include one or several GPs. Cluster-randomisation was
practical reasons, GPs on the island of Bornholm are not chosen, as a significant risk of bias was identified in the form
eligible. The local branch of the Organisation of General of contamination if randomisation was performed on patient
Practitioners in Denmark and the Capital Region of level. Randomisation was per-formed externally by The
Health Care have negotiated and signed an agreement that Research Centre for Prevention and Health (RCPH) in the
allows the GPs to participate in the study and sets out the Capital Region of Denmark, which used a centralised
terms and conditions for it, including financial computer-generated allocation sequence. The randomisation
reimbursement. was conducted in two phases. The first included 49 GPs and
was stratified by three geographically defined uptake areas.
Individual level The second randomisation included four GPs, and here a
Patients are eligible for the trials if they consult a GP who geographical stratification was not used.
is participating in the trials and additionally comply
Curth et al. Trials (2017) 18:382 Page 4 of 13

Fig. 1 Flow chart for participants. Each trial has a flow chart similar to this

Individual level conducting a diagnostic interview. If the diagnosis accord-

As the trial is randomised on general practice level, the ing to the diagnostic interview differs from the diagnosis
participating patients are allocated to either the inter- given by the GP, the research assistant consults a psych-
vention or control group according to their GP. The GPs iatrist in the Collabri team who contacts the GP in order to
identify and recruit the patients with anxiety disor-ders. In discuss and agree on the diagnosis. Patient recruitment
Danish medical guidelines GPs are recom-mended to use was initiated in November 2014 and ended on 31 Decem-
the Anxiety Symptom Scale (ASS) in conjunction with ber 2016.
ICD-10 criteria for diagnostic investi-gation [24]. The GP
provides written and verbal infor-mation about the Blinding
project, and hands out the Consent Form. Following The research team (who assess baseline eligibility criteria,
recruitment, a research assistant contacts the patient to collect interviewer-based baseline and follow-up assess-
schedule a telephone baseline eligibility interview. The ments, perform statistical analyses and draw conclusions)
aim is to assess inclusion and exclusion criteria and verify are blinded to allocation throughout the entire trial period
the GP’s diagnosis by (from baseline eligibility interview to drawing
Curth et al. Trials (2017) 18:382 Page 5 of 13

conclusions). It is not possible to blind GPs, the Collabri All patients commence their treatment by developing an
team who deliver the intervention, the administrative pro- individually structured treatment plan which is based on
ject team and the participating patients towards interven- disorder-specific manuals (unpublished, available through
tion allocation. It will be highlighted to them that they the corresponding author) following the stepped-care
cannot reveal their allocation group to the research team. principle and complying with national guidelines [19, 24],
Should the allocation be revealed to the research assistant see Fig. 1. The core treatment elements of the model are:
before assessments (baseline eligibility interview or disorder-specific written material and a self-management
interviewer-based assessments), another research assistant book (bibliotherapy), CBT, group-based or individual
is allocated. Furthermore, during statistical analyses, the manualised psychoeducation based on the Chronic
two groups will be coded and anonymised (e.g., X and Y) Disease Management Program (CDSMP)) or as a part of
until final analyses and conclusions have been made. CBT, individual CBT (10–12 sessions, depending on
The referral diagnosis is not revealed to the researchers disease) and/or medication. The treatment period is 3–4
conducting the baseline eligibility interview until the end months. The number of sessions in the Collabri group will
of the interview. This makes the assessment of baseline be obtained and reported.
eligibility criteria more objective. Depending on the disorder and severity, the patients will
be offered treatment elements according to a stepped-care
Interventions plan, offering the least invasive and least resource-
The experimental intervention – the Collabri model demanding treatment first (stepwise scaling up of
for collaborative care treatment efforts) (see Fig. 1).
The Collabri model is developed based on the previously The care manager ensures active and scheduled follow-
mentioned four criteria for collaborative care [30] and has up including regular monitoring and review of
been adapted to Danish conditions by including progression. Monitoring takes place as a minimum every
collaboration with relevant professionals in the local 2 weeks or more often depending of the severity of the
authorities and integrating the existing public psycholo- disorder or when medication is initiated. Regular reviews
gist scheme. In addition, the model integrates elements of the treatment plan will take place at least once a month,
which are found to be essential for collaborative care and must precede stepping up as well as ending treatment.
interventions in an academic literature review [40] or
recommended in current guidelines [19, 24, 41] recom-
mending the use of screening instruments for detection The multi-professional Collabri team In the Collabri
and follow-up, and a stepped-care approach to treatment. treatment, the GP works in combination with the Collabri
In addition, the model incorporates principles of involve- team (the care manager and the psychiatrist/psychologist)
ment of relatives, patient involvement and influence on around the treatment of the patient). The GP has the
treatment and support in self-management. Below, how overall treatment responsibility, including the diagnostic
the Collabri model meets the four criteria for process, initiates the treatment and collaborates with the
collaborative care is outlined. care manager to provide seamless treatment and care.
The model includes a multi-professional approach to The Collabri team is a health care team consisting of
treatment involving a GP, a care manager and a psychi- care managers and psychiatric specialists. The psychiatric
atric specialist. Interprofessional communication consist specialists can either be psychiatrists or psychiatrists in
of weekly meetings between the GP and care manager in combination with a psychologist with specialist training
order to discuss clinical cases and of the psychiatric in psychiatry. The care managers have a medium-long
specialist supervising the care managers in groups twice a health professional education; for example, as a nurse
month and the GPs in groups once a month. When with men-tal health care experience as well as being
needed, the GP, care manager and the psychiatric certified in providing CBT (the minimum requirement is a
specialist have joint consultations and the GPs and care 1-year recognised course). Care managers, GPs and the
managers receive individual supervision from the psychi- psychia-trists ensure adherence to the Collabri model.
atric specialist if required. The weekly meetings between
care manager and GP take place in person, but other Training prior to intervention All care managers attend a
communication can take place via video conferences if it 1-week course introducing them to the Collabri model and
is not possible to meet in person. the CBT methods. In addition they attend a 4-day training
It has not been possible to establish a joint recording course leading to certification in a manua-lised group-
system between primary and secondary care systems; based psychoeducation program (based on the Chronic
however, written communication between the GP and care Disease Management Program (CDSMP)), and an
manager or psychiatric specialist occurs through already additional 2-day course of introduction in an
existing electronic communication systems. individualised psychoeducation manual based on the
Curth et al. Trials (2017) 18:382 Page 6 of 13

CDSMP principles. The psychiatric specialists also attend Interventional Trials (SPIRIT) figure (Fig. 2) for an over-
the training except the group psychoeducation training. view of time schedule, assessment instruments and source
All GPs in the intervention group participate in a 1-day of data collection.
training course in the principles of collaborative care and
the Collabri model. Primary outcome measure The primary outcome is anx-
iety symptoms measured with the Beck Anxiety Inventory
Fidelity Fidelity assessments ensure that the Collabri (BAI) 6 months after baseline. The BAI is a 21-item
intervention is carried out according to the described Collabri general questionnaire for anxiety, measuring symptoms
model. Assessors will monitor fidelity to the model 6 months during the last week rated on a four-point Likert-scale
after study initiation and at least once more during the from 0 (never) to 3 (almost all the time), where the
inclusion period. Fidelity will be moni-tored using a Collabri maximum score is 63 [45]. The BAI has shown excel-lent
fidelity scale (unpublished, available through the psychometric properties, with internal consistency:
corresponding author). In order to improve implementation, α = 0.92 and 1-week test-retest reliability: r = 0.75 in a
an action plan will be developed where needed based on the community sample [46].
outcome of the fidelity assessments.
Secondary outcome measures The secondary outcome
The control intervention – treatment-as-usual measures are the BAI at 15 months, depression symptoms
Participants whose GP is randomised to the control group (Beck Depression Inventory (BDI-II)) [47] at 6 months,
will receive treatment-as-usual as offered by their GP. level of psychosocial functioning (Global Assessment of
Guidelines from the Danish Health Authority and The Functioning (GAF-F)) [48] and general psychological
Danish College of General Practitioners are available for symptoms (Symptom Checklist (SCL-90-R)) [49] at 6 and
guidance [19, 24]. According to the guidelines a physical 15 months after baseline.
health evaluation is performed prior to diagnosis and the The BDI-II is a 21-item general questionnaire for
ASS instrument in conjunction with ICD-10 criteria is measuring depression symptoms during the last 14 days
recommended for detection, diagnostic investigation and rated on a four-point Likert-scale from 0 to 3, where the
monitoring of the disorder [42]. If relevant, the GP can maximum score is 63. A meta-analysis of the BDI II’s
prescribe medication and/or provide psychotherapy if they internal consistency estimates yielded a mean coefficient
attend supervision. The GP can also refer patients to an alpha of 0.86 for psychiatric patients and 0.81 for non-
independent psychologist (partly publicly subsidised) or psychiatric subjects [50]. The test-retest reliability for the
to a public psychiatrist or to treatment in the secondary BDI II for psychiatric patients has been reported to range
health care system (fully publicly subsidised). Accessible from 0.48 to 0.86, and for non-psychiatric subjects from
treatment can vary between general practices as these 0.60 to 0.83. The SCL-90-R is the revised version of
guidelines only provide recommendations for treatment or SCL-90, a multi-dimensional questionnaire consisting of
management of the disorders, as opposed to required 90 questions about general psychological symptoms. The
minimum standards. questionnaire consists of nine subscales from which a
joint measure (Global Severity Index (GSI)) can be
Assessments and outcome measures calculated. The internal consistencies of the SCL-90-R are
Baseline eligibility interview satisfactory with range from α = 0.74 (‘aggression’) to α =
At baseline, patients attend a telephone eligibility interview 0.97 (GSI) [51].
where the inclusion and exclusion criteria are assessed by a The split version of the GAF on psychosocial function-
research assistant. To confirm the diagnosis the Diagnos-tic ing ranges from 0 to 100 points, with a higher number
and Statistical Manual of Mental Disorders, 4th edition indicating a higher level of functioning. Studies suggest a
(DSM-IV) MINI International Neuropsychiatric Interview good validity of the GAF-F scale as well as usefulness of
[43] is used including ICD-10-specific questions, for the measuring the level of psychosocial functioning among
inclusion diagnoses. The GP assesses comorbidity at incident patients with schizophrenia. A high inter-rater
recruitment. Information on personality disorder traits are reliability can be achieved with little training of the raters
obtained through the Standardised Assessment of Person- [52].
ality: Abbreviated Scale (SAPAS) [44].
Explorative outcome measures The explorative out-come
Outcome measures measures are the BDI-II at 15 months, personal and social
Participants are interviewed and asked to fill out a ques- performance (PSP) [52], health-related qual-ity of life
tionnaire at baseline, 6 and 15 months’ follow-up. In the (EQ-5D-3 L) [53], functional impairment (Sheehan
following each assessment instrument will be described. Disability Scale (SDS)) [54], quality of life (WHO-five
See the Standard Protocol Items: Recommendations for Well-being Index (WHO-5)) [55] and self-
Curth et al. Trials (2017) 18:382 Page 7 of 13

Fig. 2 Time schedule, assessment instruments and source of data collection

efficacy (Personal Control subscale from the revised self-care and disturbing and aggressive behaviours on a
version of the Illness Perception Questionnaire (IPQ-R) scale from 1 to 100 with a higher number indicating
[56] and two subscales from the Chronic Disease Self- higher level of function.
Efficacy Scales (SECD-32); and the Obtain Help from The EQ-5D-3 L is a measure of health status in five do-
Community, Family, Friends Scale and Control/Manage mains: mobility, self-care, usual activities,
Depression Scale) [57] at 6 and 15 months. pain/discomfort and anxiety/depression and also includes
Here, a measure of apathy will also be collected through a Visual Analogue Scale from 0 (worst imaginable health
the Diagnostic Apathia Scale, which consists of six items status) to 100 (best imaginable health status).
[58] and side effects from treatment will be measured The SDS is a composite of three items designed to
with the Patient-Rated Inventory of Side Effects (PRISE) measure the extent to which three major domains in the
which identifies and evaluates the toler-ability of patient’s life are impaired by symptoms, and can be
symptoms/side effects in nine domains [59]. summed into a joint measure of global functional
The PSP measures personal and social functional level impairment that ranges from 0 (unimpaired) to 30 (highly
in the domains of: socially useful activities (for example, impaired). The WHO-5 consists of five items that
work and education), personal and social relationships, measure the subjective experience of quality of
Curth et al. Trials (2017) 18:382 Page 8 of 13

life or psychological wellbeing. Each item is rated on questionnaire after 8 days, they will be reminded and, if
a six-point Likert scale from 0 (not present) to 5 necessary, they will be contacted in order to collect the
(constantly present). missing data. The BAI and BDI-II data can be collected
The IPQ-R has 12 subscales of which the Personal via telephone if it is not possible for the participants to fill
Control subscale consists of six items about one’s own them out by themselves. Information about intervention-
beliefs about the ability to affect the disorder. Each item is specific services and treatment will be registered by the
rated on a five-point Likert-scale from 1 (disagree very care managers and psychiatrists throughout the inter-
much) to 5 (agree very much). vention period.
The two subscales of SECD-32 consist of four and six
items about how confident one is in doing certain activ- Safety measures
ities. Each item is rated on a 10-point Likert scale from The following safety measures are collected:
one (not at all confident) to 10 (very confident).
Information about life-threatening conditions, the use of Self-reported anxiety and depression symptoms
inpatient and outpatient physical and mental health measured with the BAI [45] and BDI-II [47]
services and number of days of admission and former Suicidal ideation obtained from the questions
treatment is collected through the National Patient concerning suicidality in the MINI baseline
Registry which holds information about all patient con- eligibility interview [43]
tacts in the secondary health care system [60]. Death (natural, accident, suicide, homicide/violence
Medication use is retrieved from the Danish National or unknown) obtained from the Danish Register of
Prescription Registry which holds information about all Causes of Death [62]
sales of medication [61]. Life-threatening conditions for reasons other than
Information about deaths is collected through The suicide attempts obtained from the National Patient
Danish Register of Causes of Death, which is based on Registry [60]
information of all deaths in Denmark since 1943 [62]. Number of physical and mental health outpatient
Sickness leave, employment and use of other social services, admissions and inpatient days obtained
services will be obtained from the Danish Register for from the National Patient Registry [60]
Evaluation of Marginalisation (DREAM), which contains Number of sickness leave days obtained from the
information about contact with the labour market for the DREAM database [63]
entire Danish population [63]. These data are collected at
baseline, 6 and 15 months. Training and inter-rater reliability
At 6 months the patient’s feeling of being supported in Trained research assistants will perform the baseline
their recovery by their primary health care provider (care eligibility interviews as well as the objective assessments.
manager and GP in the intervention group and GP in the Based on sound records, inter-rater reliability is assessed
control group) is assessed through the INSPIRE question- throughout the training and assessment period and
naire, which has two sections – one about support (20 discrepancies are discussed with the other raters and at least
items) and one about relationship (seven items) [64]. one of the intervention developers. In the first 3 months, bi-
Participants also rate their general satisfaction with monthly or monthly meetings are held to dis-cuss potential
treatment through the CSQ-8 questionnaire [65] together difficulties in rating the objective measures of the PSP scale
with project-specific questions. The CSQ-8 consists of and the GAF-F scale. Subsequently, reliability ratings related
eight items which are rated on a scale from 0 to 4. to these measures are performed approxi-mately every 3
Baseline information, such as former treatment, educa- months throughout the assessment period.
tion and marital status, will be collected through Statistics
Denmark, which is the central authority on Danish statis- Power and sample size calculation
tics [66], and the National Patient Registry. Differences in clinically relevant treatment response for
the primary outcome measure BAI is set at 4 points based
Data collection on international academic literature for the BDI [67, 68],
Interview-based assessments such as the GAF-F and the as academic literature on the BAI could not be found.
PSP will be conducted through telephone inter-views by International academic literature suggests that a standard
research assistants who are thoroughly trained in using the deviation (SD) of 12 for the BAI can be used in the
instruments. Other data are collected either through self- sample size calculation [36, 68, 69]. The intraclass
reported questionnaires or registers (see Fig. 2 for details correlation coefficient (ICC) is set at 0.04 based on a
of sources of data collection). Self-reported review on ICC for anxiety and depression and other
questionnaires are completed online or in a paper version. mental disorders in primary care [70]. Sample size calcu-
If participants have not completed the lations based on these numbers show that 364 persons
Curth et al. Trials (2017) 18:382 Page 9 of 13

for each trial on panic disorder, generalised anxiety countries [74], indicate that a GP with 1600 registered
disorder and social phobia should be included in order to patients will on average see 37 patients with panic dis-
reject the null hypothesis that the intervention and control order, 24 patients with generalised anxiety disorder and
groups have improved their symptom level equally when 32 patients with social phobia per year. By including a
the power is set at 0.8 and the significance level at 0.05. minimum of 48 GPs in each trial we find it realistic to
include 364 patients with panic disorder, 364 patients with
Power calculations for the secondary outcomes have generalised anxiety disorder and 364 patients with social
been estimated to be more than 0.8 based on calcula-tions phobia, adding up to a total of 1092 patients. Thus, each
with 182 participants in each group (see Table 1). As we GP should recruit approximately seven to eight patients
have not been able to find estimates for the BDI- with each anxiety disorder over a 12-month period. With a
II and the GAF-F for patients with anxiety disorders in conservative caseload of 100 pa-tients per care manager
general practice, estimates for patients with depression in per year (25 at a time) and eight care managers, it should
general practice have been used as reference for power be possible in terms of capacity to include up to 1600
calculations for the BDI-II [36], whereas patients with patients in the study during 12 months of which 800
social phobia recruited from psychiatric clinics have been would be in the intervention group.
used as reference for power calculations for the GAF-F
[71]. Estimates for the SCL-90-R are based on a mixed Because there is a risk of GP dropout it might be ne-
anxiety and depression population in an out-patient cessary to reduce the number of GPs to 44. This affects
setting [72]. the sample size calculations for the trials and thereby the
number of patients needed to include. Thus, 374
Statistical analyses participants are required in each of the three trials and
Data analyses will be carried out according to the statis- each GP should recruit eight to nine patients with each
tical principle ‘intention-to-treat’ [73]; thus, once a per- anxiety disorder.
son meeting the eligibility criteria is included in the study
the person stays in the study population, including follow- Project organisation
up, regardless of whether the person later meets the The project is led by a steering group which ensures the
exclusion criteria. All continuous outcome measures will progress of the research project. A lead project manager
be analysed using generalised linear models. Multi-level, will ensure that the general management of the project
linear mixed models with repeated measures will be used together with a project manager in charge of implemen-
to handle the cluster-randomisation and un-equal loss to tation of the intervention and a project manager in charge
follow-up. The models will be estimated using an of the research project. Administrative staff members
unstructured covariance matrix if possible. If not possible, support the project managers. Two PhD stu-dents and
other covariance matrices, such as inde-pendent, research assistants perform data collection and analyses.
interchangeable, autoregressive, and Toeplitz, will be
estimated, and the best fitting structure selected based on
Bayes’ information criterion. The analysis levels are: GP, Discussion
patient and time. This model is based on the as-sumption The design of the trials presents several strengths. The
that data are missing at random or missing completely at randomisation is carried out externally and is computer
random. Explorative subgroup analyses will be made for based which ensures an adequate allocation sequence and
patients with somatic comorbidity and personality concealment, thus reducing the risk of (cluster) selection
disorders. bias. In order to further limit the risk of selec-tion bias,
data will be analysed according to the intention-to-treat
Feasibility principle and attempts to increase follow-up rates will be
Twelve months’ prevalence rates, based on a meta- made. Also, the trials are rando-mised at cluster level in
analysis including studies conducted in European an attempt to eliminate the

Table 1 Power calculations for the secondary outcomes

Secondary outcomes Mean difference (MD) Standard deviation (SD) Type 1 error Calculated power
of the pooled mean
BDI-II 4 [77] 11 [36, 77] 5% 93%
GAF-F 11 [71] 5% 99%
5a
SCL-90-R 23 [72] 50 [72] 5% 99%
a
For the GAF-F the expected mean difference has been conservatively estimated to be 5 points as this is considered clinically
relevant BDI Beck Depression Inventory, GAF Global Assessment of Functioning, SCL Symptom Checklist
Curth et al. Trials (2017) 18:382 Page 10 of 13

potential risk of contamination otherwise introduced by Split Scale for Functional Level; IPQ-R: Illness Perception Questionnaire
Revised; MINI: Mini International Neuropsychiatric Interview; PRISE: Patient-
individual randomisation. This risk could be present
Rated Inventory of Side Effects; PSP: Personal and Social Performance;
where GPs trained in collaborative care principles, and SAPAS: Standardised Assessment of Personality: Abbreviated Scale; SCL-90-
receiving supervision from a project psychiatrist, would R:Symptom Checklist-90-Revised; SDS: Sheehan Disability Scale; WHO-
not be able to differentiate between intervention and 5: WHO-five Well-being Index
control patients when using their acquired skills and ad-
Acknowledgements
vice from the psychiatrist, which is likely to affect the re-
Rie Poulsen, for excellent assistance in the design development,
sults, and thereby making it more difficult to detect and Lone Tonsgaard for English revision of the manuscript.
differences between the intervention and the control
group. To ensure that the intervention is provided as Funding
intended, we monitor fidelity to the Collabri model at The trials are financed by a grant from the Danish Ministry of Health to enhance
the collaboration between the mental health care system and general practice
least twice during the intervention period. (in Danish ‘Pulje til styrket samarbejde mellem behandlingspsykiatrien og almen
The primary outcome is self-reported and patients are praksis (shared care)’). No current or future sponsors of the trials will have any
not blinded towards allocation, which might result in role in the design, collection of data, analysis of data, data interpretation, or in
publication of data from the trials.
information bias and possibly overestimation of effects
[75]. However, the BAI has been found usable for asses-
Availability of data and materials
sing the severity of anxiety symptoms in patients with Not applicable.
anxiety disorders in a general practice setting [76], and
other outcome measures will be assessor blinded for Authors’ contributions
allocation; for example, the secondary outcome measure LFE developed the trial, participated in the planning and design development,
and read and critically revised the manuscript for important intellectual content.
of psychosocial functioning (GAF-F). A limitation, as a NC drafted the manuscript and participated in the planning and design
result of the cluster-randomisation, may be that patients development. UBC, ASD, MEL, JHM, ML, CC, CH and MN participated in the
within a cluster have some similarity and may differ from planning and design development, and read and critically revised the manuscript
for important intellectual content. Managerial responsibility and supervision is
patients from other clusters. However, sample size provided by LFE and MN. The authors comply with the Vancouver guidelines
calculations including an ICC have been carried out, and have read and approved the final version of the manuscript and have
aiming to reduce this problem. Furthermore, we cannot be agreed to be accountable for all aspects of the work in ensuring that questions
related to any part of the work are appropriately investigated and resolved. All
sure that all eligible patients are asked to participate by
authors read and approved the final manuscript.
their GP, as patients are not systematically screened for
eligibility. Ethics approval and consent to participate
The results of these trials will add to the limited pool of The trials will be conducted in compliance with this protocol, the
knowledge about collaborative care for patients with Helsinki Declaration, Good Clinical Practice guidelines and
following the rules for informed consent.
anxiety disorders. To our knowledge, they will be the first Electronically stored data will be kept in entry-restricted files and other
carried out in a Danish context and the first report-ing data will be stored in sealed cupboards. Personally identifiable data
results for generalised anxiety and social phobia will be treated according to the Danish law on the handling of personal
information (in Danish ‘Lov om behandling af personoplysninger’).
separately. The GP obtains the patients’ consent to participation, which is based
If the trials show positive results, they could contribute on verbal information and written material. Written declarations of
to the improvement of future treatment for patients with consent must be signed, and the participants will receive a copy.
Participants will be informed of their rights to exit the study at any point
panic disorder, generalised anxiety disorder or social and without consequences for their future treatment. If they withdraw
phobia in general practice. from the study they can specify how their data are used.
The authors of this protocol are not aware of previous trials reporting any risks or
adverse events of collaborative care. In order to minimise the possible risks of
Trial status discomfort during interviews these will be conducted in a flexible manner with
breaks if necessary. If any of the participants present acute suicidal ideation
Recruitment of participants within the clusters is ongoing
during interviews, the assessor will consult a clinician or contact the emergency
and continued until 31 December 2016. services if necessary. There are currently no known circumstances that can lead
to exclusion from participation once a patient is included. If such circumstances
arise, the participant will be notified.
Additional file The trial protocol has been evaluated and approved by the Regional Ethics
Committees of the Capital Region (identification no: H-3-2013-203). It is also
Additional file 1: A word file with A Standard Protocol Items: approved by the Danish College of General Practitioners’ Multi Practices
Recommendations for Interventional Trials (SPIRIT) Checklist is Committee, the Danish Data Protection Agency (journal no: 2007-58-0015, local
included as ‘Additional file 1’. (DOC 123 kb) journal no: RHP-2014-012) and has been registered at http://www.clinicaltrials.
gov (identification no: NCT02678624). No substantial deviation from the protocol
will be implemented without the prior review and approval of relevant authorities.
Abbreviations Results will be published in international journals.
BAI: Beck Anxiety Inventory; BDI-II: Beck Depression Inventory II; A Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT)
CBT: Cognitive behavioural therapy; CDSMP: Chronic Disease Management Checklist has been completed and is available in Additional file 1. The (SPIRIT)
Program; CSQ-8: Client Satisfaction Questionnaire; DREAM: Danish Register figure for an overview of time schedule, assessment instruments and source of
for Evaluation of Marginalisation; GAF-F: Global Assessment of Functioning data collection has been completed and is available in Fig. 2.
Curth et al. Trials (2017) 18:382 Page 11 of 13

Consent for publication 10. Kujanpää T, Jokelainen J, Auvinen J, Timonen M. Generalised anxiety disorder
Not applicable. symptoms and utilisation of health care services. A cross-sectional study from the
‘Northern Finland 1966 Birth Cohort’. Scand J Prim Health Care. 2016;34:151–8.

Competing interests 11. Bélanger L, Ladouceur R, Morin CM. Generalized anxiety

The authors declare that they have no competing interests. disorder and health care use. Can Fam Physician. 2005;51:1363.
12. Üstün TB, Sartorius N. Mental illness in general health care. An international
study. Chichester, New York, Brisbane, Toronto, Singapore: Wiley; 1995.

Publisher’s Note 13. Grandes G, Montoya I, Arietaleanizbeaskoa MS, Arce V, Sanchez A. The burden
of mental disorders in primary care. Eur Psychiatry. 2011;26:428–35.
Springer Nature remains neutral with regard to jurisdictional
14. Toft T, Fink P, Oernboel E, Christensen K, Frostholm L, Olesen F. Mental disorders in
claims in published maps and institutional affiliations.
primary care: prevalence and co-morbidity among disorders. Results from the
Functional Illness in Primary care (FIP) study. Psychol Med. 2005;35:1175–84.
Author details
1
Mental Health Center Copenhagen, Mental Health Services, Capital Region of 15. Munk-Jørgensen P, Allgulander C, Dahl A, Foldager L, Holm M,
2 Rasmussen I, Virta A, Huuhtanen M, Wittchen H. Prevalence of
Denmark, Kildegårdsvej 28, 2900 Hellerup, Denmark. Research Unit for
General Practice and Section of General Practice, University of Copenhagen, generalized anxiety disorder in general practice in Denmark, Finland,
Øster Farimagsgade 5, PO Box 2099, 1014 Copenhagen K, Denmark. Norway, and Sweden. Psychiatr Serv. 2006;57:1738–44.
3
Stolpegård Psychotherapy Centre, Mental Health Services, Capital Region of 16. Preventive outpatient treatment for severe affective disorder (depression and
4 mania)—a health technology assessment [Forebyggende ambulant behandling
Denmark, Stolpegårdsvej 20, 2820 Gentofte, Denmark. Mental Health Center
ved svær affektiv lidelse (depression og mani)—en medicinsk
Frederiksberg, Mental Health Services, Capital Region of Denmark, Nordre
5 teknologivurdering]. Danish Health Authority [Sundhedsstyrelsen]. 2006.
Fasanvej 57-59, 2000 Frederiksberg, Denmark. Mental Health Center North
http://www.sst.dk/~/media/9E293BABE8CC4823B3A92CCAAA10564B.ashx.
Zealand, Mental Health Services, Capital Region of Denmark, Dyrehavevej 48,
6 Accessed 30 Oct 2016.
3400 Hillerød, Denmark. Institute for Clinical Medicine, University of
17. Treatment of psychiatric non-psychotic disorders [Behandling af psykiske
Copenhagen, Mental Health Center Copenhagen, Mental Health Services,
Capital Region of Denmark, Kildegårdsvej 28, 2900 Hellerup, Denmark. lidelser af ikke- psykotisk karakter]. Danish College of Psychiatrists [Dansk
7 Psykiatrisk Selskab]. 2001. http://dpsnet.dk/wp-content/uploads/2015/01/
Independent General Practitioner, Copenhagen, Denmark.
ikkepsykose_010102.pdf. Accessed 30 Oct 2016.
Received: 26 December 2016 Accepted: 26 July 2017 18. Improved patient courses for patients with non-psychotic psychiatric disorders
[Bedre patientforløb for patienter med psykiske lidelser af ikke-psykotisk
karakter]. The Danish College of Psychiatrists and The Danish College of
General Practitioners [Dansk Psykiatrisk Selskab og Dansk Selskab for Almen
References
Medicin]. 2004. http://dpsnet.dk/wp-content/uploads/2015/01/ 2004-11-10-
1. Kessler RC, Aguilar-gaxiola S, Alonso J, Chatterji S, Lee S, Ormel J, Üstün TB,
rapport_bedreptforlob.pdf. Accessed 30 Oct 2016.
Wang PS. The global burden of mental disorders: an update from the WHO World
19. Reference program for anxiety disorders for adults [Referenceprogram for
Mental Health (WMH) Surveys. Epidemiol Psichiatr Soc. 2009;18:23–33.
angstlidelser hos voksne]. Danish Health Authority [Sundhedsstyrelsen].
2. Alonso J, Angermeyer MC, Bernert S, Bruffaerts R, Brugha TS,
2007. http://www.sst.dk/~/media/63928C3E9D2845D19A0A0783AAB92CD8.
Bryson H, et al. Prevalence of mental disorders in Europe: results
ashx. Accessed 30 Oct 2016.
from the European Study of the Epidemiology of Mental Disorders
20. Francis JL, Ph D, Weisberg RB, Dyck IR, Culpepper L, Smith K,
(ESEMeD) project. Acta Psychiatr Scand. 2004;109 Suppl 420:21–7.
Edelen MO, Keller MB. Characteristics and Course of Panic
3. Kroenke K, Spitzer R, Williams J, Monahan P, Lowe B. Anxiety
Disorder and Panic Disorder With Agoraphobia in Primary Care
disorders in primary care: prevalence, impairment,
Patients. Prim Care Companion J Clin Psychiatry. 2007;9(3):173–9.
comorbidity, and detection. Ann Intern Med. 2013;146:317–26.
4. Brown TA, Campbell LA, Lehman CL, Grisham JR, Mancill RB. Current 21. Goodwin RD, Faravelli C, Rosi S, Cosci F, Truglia E, de Graaf R,
and lifetime comorbidity of the DSM-IV Anxiety and Mood Disorders in a Wittchen HU. The epidemiology of panic disorder and agoraphobia in
large clinical sample. J Abnorm Psychol. 2001;110:585–99. Europe. Eur Neuropsychopharmacol. 2005;15:435–43.
5. Muller JE, Koen L, Stein DJ. Anxiety and medical disorders. 22. Christensen KS, Fink P, Toft T, Frostholm L, Ørnbøl E, Olesen F.
Curr Psychiatry Rep. 2005;7:245–51. A brief case-finding questionnaire for common mental disorders:
6. Meier SM, Mattheisen M, Mors O, Mortensen PB, Laursen TM, The CMDQ. Fam Pract. 2005;22:448–57.
23. Hauge-Helgestad A, Schlepern Johansen K, Hansen J. Treatment of people
Penninx BW. Increased mortality among people with anxiety
with anxiety and depression. Mapping of the treatment and discussion about
disorders: total population study. Br J Psychiatry. 2016;209:216–21.
7. Psychiatric illness and the labour market. Discussion paper [Psykisk sygdom og perspectives of implementing collaborative care [Behandling af mennesker med
arbejdsmarkedet. Debatoplæg]. Danish College of Psychiatrists, Danish Union angst og depression. Kortlægning af behandlingsfelt og diskussion af
for Psychologists, Union for Danish Psychiatrists and Children-Youth perspektiverne]. Danish Health Institute [Dansk Sundhedsinstitut]. 2012.
Psychiatrists, Danish Regions [Dansk Psykiatrisk Selskab, Dansk Psykolog http://www.kora.dk/media/763524/behandling-af-mennesker-med-angst-og-
Forening, Danske Psykiateres og Børne- Ungdomspsykiateres organisation, depression.pdf. Accessed 30 Oct 2016.
Danske regioner]. 2009. http://www.regioner.dk/media/1238/psykisk- sygdom-og- 24. Clinical guide for general practice. Anxiety disorders, diagnostics and treatment
arbejdsmarkedet.pdf. Accessed 30 Oct 2016. [Klinisk vejledning for almen praksis. Angsttilstande. Diagnostik og behandling].
8. Interventions for people with psychiatric illness—development in diagnoses Danish College of General Practitioners. 2010. http://vejledninger.dsam.dk/
and treatment [Indsatsen for mennesker med psykiske lidelser—udvikling i media/files/7/angsttilstande-med-links.pdf. Accessed 30 Oct 2016.
diagnoser og behandling. Bilagsrapport 2]. The Government’s Committee 25. Agreement for general practice, changed 1 Mar 2014 [Overenskomst for almen
on Psychiatry [Regeringens udvalg om psykiatri]. 2013. http://www.sum.dk/ praksis. Af 3 Jun 1991, senest ændret ved aftale af 1 Mar 2014]. Danish College
Aktuelt/Nyheder/Psykiatri/2013/Oktober/~/media/Filer%20- for General Practitioners [Praktiserende Lægers Organisation]. 2014. https://
%20Publikationer_i_pdf/2013/Rapport-psykiatriudvalg-okt-2013/21-10- www.laeger.dk/sites/default/files/ok14.pdf. Accessed 30 Oct 2016.
2013/ 21102013Indsatsenformenneskermedpsyliskelidelserbilag2web.ashx 26. Fjeldsted R, Christensen KS.Evaluation of the psychologist scheme for persons
- Publikationer_i_pdf/2013/Rapport-psykiatriudvalg-okt-2013/21-10-2013/ with mild to moderate depression [Evaluering af ordning med
21102013Indsatsenformenneskermedpsyliskelidelserbilag2web.ashx. psykologbehandling af personer med let til moderat depression]. Research Unit
Accessed 30 Oct 2016. for General Practice, Aarhus University [Forskningsenheden for Almen Praksis,
9. Flachs E, Eriksen L, Koch M, Ryd J, Dibba E, Skov-Ettrup L, Juel K. The Århus Universitet]. 2011. http://www.dp.dk/wp-content/uploads/
disease burden in Denmark—diseases [Sygdomsbyrden I Danmark— Evaluering_af_ordning_med_psykologbehandling_til_trykning_220611.pdf.
Sygdomme]. National Institute of Public Health, University of Southern Accessed 30 Oct 2016.
Denmark [Statens Institut for Folkesundhed, Syddansk Universitet]. 2015. 27. Davidsen A. Talking therapy as part of the general practitioner’s
https:// sundhedsstyrelsen.dk/da/sygdom-og-behandling/~/media/ normal working day [Samtaleterapi ind i den praktiserende
00C6825B11BD46F9B064536C6E7DFBA0.ashx. Accessed 30 Oct 2016. læges hverdag]. Ugeskr Laeger. 2010;5:2025–9.
Curth et al. Trials (2017) 18:382 Page 12 of 13

28. Nielsen HG, Söderström M. Group supervision in general practice as 48. Pedersen G, Hagtvet KA, Karterud S. Generalizability studies of the Global
part of continuing professional development. Dan Med J. 2012;59:2. Assessment of Functioning-Split version. Compr Psychiatry. 2007;48:88–94.
29. Reference program for unipolar depression for adults [Referenceprogram for Unipolar 49. Derogatis L. SCL-90-R. Symptom Checklist-90-R. Administration, scoring and
Depression Hos Voksne]. Danish Health Authority [Sundhedsstyrelsen]. 2007. procedures manual. 3rd ed. Minneapolis: National Computer Systems, Inc; 1994.
http://www.sst.dk/~/media/6F9CE14B6FF245AABCD222575787FEB7.ashx. Accessed 50. Beck AT, Steer RA, Garbin MG. Psychometric properties of
30 Oct 2016. the Beck Depression Inventory: twenty five years of
30. Archer J, Bower P, Gilbody S, Lovell K, Richards D, Gask L, evaluation. Clin Psychol Rev. 1988;8:77–100.
Dickens C, Coventry P. Collaborative care for depression and 51. Prinz, et al. Comparative psychometric analyses of the SCL-90-R and its short
anxiety problems (Review). Cochrane Database Syst Rev. versions in patients with affective disorders. BMC Psychiatry. 2013;13:104.
2012;10:CD006525. doi:10.1002/ 14651858.CD006525.pub2. 52. Morosini PL, Magliano L, Brambilla L, Ugolini S, Pioli R. Development,
31. Roy-Byrne PP, Craske MG, Stein MB, Sullivan G, Bystritsky A, reliability and acceptability of a new version of the DSM-IV Social and
Katon W, Golinelli D, Sherbourne CD. A randomized effectiveness Occupational Functioning Assessment Scale (SOFAS) to assess
trial of cognitive-behavioral therapy and medication for primary care routine social functioning. Acta Psychiatr Scand. 2000;101:323–9.
panic disorder. Arch Gen Psychiatry. 2005;62:290–8. 53. Rabin R, De Charro F. EQ-SD: a measure of health status from
32. Roy-Byrne PP, Katon W, Cowley DS, Russo J. A randomized the EuroQol Group. Ann Med. 2001;33:337–43.
effectiveness trial of collaborative care for patients with panic 54. Sheehan D, Harnett-Sheehan K, Raj B. The measurement of
disorder in primary care. Arch Gen Psychiatry. 2001;58:869–76. disability. Int Clin Psy-chopharmacology. 1996;11 Suppl 3:89–95.
33. Rollman BL, Belnap BH, Mazumdar S, Houck PR, Zhu F, Gardner W, 55. Bech P. Measuring the dimensions of psychological general
Reynolds CF, Schulberg HC, Shear MK. A randomized trial to improve well-being by the WHO-5. QoL Newsl. 2004;32:15–6.
the quality of treatment for panic and generalized anxiety disorders in 56. Moss-Morris R, Weinman J, Petrie K, Horne R, Cameron L,
primary care. Arch Gen Psychiatry. 2005;62:1332–41. Buick D. The Revised Illness Perception Questionnaire (IPQ-
34. Roy-byrne P, Craske MG, Sullivan G, Rose RD, Edlund MJ, Lang AJ, R). Psychol Health. 2002; 17:1–16.
Bystritsky A, Welch SS, Chavira DA, Golinelli D, Campbell-sills L, 57. Lorig K, Stewart A, Ritter P, González V, Laurent D, Lynch J.
Sherbourne CD, Stein MB. Delivery of evidence-based treatment for Outcome measures for health education and other health care
multiple anxiety disorders in primary care. JAMA. 2010;303:1921–8. interventions. Thousand Oaks: Sage Publications; 1996.
35. Muntingh ADT, Van Der Feltz-cornelis CM, Van Marwijk HWJ, Spinhoven P, 58. Hellström L, Eplov LF, Nordentoft M, Ostergaard SD, Bech P. The
Van Balkom AJLM. Collaborative care for anxiety disorders in primary care: Diagnostic Apathia Scale predicts the ability to return to work following
a systematic review and meta-analysis. BMC Fam Pract. 2016;17:62. depression or anxiety. Acta Neuropsychiatr. 2014;26:364–71.
36. König H-H, Born A, Heider D, Matschinger H, Heinrich S, Riedel-Heller 59. Katz AJ, Dusetzina SB, Farley JF, Ellis AR, Gaynes BN, Castillo WC,
SG, Surall D, Angermeyer MC, Roick C. Cost-effectiveness of a primary Stürmer T, Hansen R. Distressing adverse events after antidepressant
care model for anxiety disorders. Br J Psychiatry. 2009;195:308–17. switch in the Sequenced Treatment Alternatives to Relieve Depression
37. Oosterbaan DB, Verbraak MJPM, Terluin B, Hoogendoorn AW, Peyrot (STAR*D) trial: influence of adverse events during initial treatment with
WJ, Muntingh A, Van Balkom AJLM. Collaborative stepped care v. citalopram on development of subsequent adverse events with an
Care as usual for common mental disorders: 8-month, cluster alternative antidepressant. Pharmacotherapy. 2012;32:234–43.
randomised controlled trial. Br J Psychiatry. 2013;203:132–9. 60. Lynge E, Sandegaard JL, Rebolj M. The Danish National Patient
38. Muntingh A, van der Feltz-Cornelis C, van Marwijk H, Spinhoven P, Assendelft W, Register. Scand J Public Health. 2011;39 Suppl 7:30–3.
de Waal M, Adèr H, van Balkom A. Effectiveness of collaborative stepped care for 61. Kildemoes HW, Sorensen HT, Hallas J. The Danish National
anxiety disorders in primary care: a pragmatic cluster randomised controlled trial. Prescription Registry. Scand J Public Heal. 2011;39 Suppl 7:38–41.
Psychother Psychosom. 2014;83:37–44. 62. Helweg-Larsen K. The Danish Register of Causes of Death.
39. Bertelsen A. WHO ICD-10—Mental and Behavioural Disorders— Scand J Public Health. 2011;39 Suppl 7:26–9.
classification and diagnostic criteria [WHO ICD-10—Psykiske 63. The DREAM database, Statistics Denmark [DREAM databasen, Danmarks
Lidelser Og Adfærdsmæssige Forstyrrelser—Klassifikation Og Statistik]. The National Labour Market Authority in Denmark. http://
Diagnostiske Kriterier]. 1st edition. Gyldendal Akademisk; 2000. www.dst.dk/da/TilSalg/Forskningsservice/Data/Andre_Styrelser.aspx. Accessed
40. Eplov L, Lundsteen M, Birket-Smith M. Shared care for non-psychotic 30 Oct 2016.
diseases. recommendations based on a systematic literature review 64. Williams J, Leamy M, Bird V, Le C, Sam B, Francesca N. Development and
[Shared care for ikke-psykotiske sygdomme. Anbefalinger på baggrund af evaluation of the INSPIRE measure of staff support for personal recovery. Soc
en systematisk litteraturundersøgelse]. Danish Regions. 2009. Psychiatry Psychiatr Epidemiol. 2014. doi: 10.1007/s00127-014-0983-0.
41. Generalised anxiety disorder and panic disorder (with or without 65. Larsen DL, Attkisson CC, Hargreaves WA, Nguyen TD.
agoraphobia) in adults. management in primary, secondary and Assessment of client/ patient satisfaction: development of a
community care. National Institute for Health and Clinical Excellence. general scale. Eval Program Plann. 1979;2:197–207.
2011. guidance. nice.org.uk/cg113NHS. Accessed 30 Oct 2016. 66. Statistics Denmark [Danmarks statistik]. https://www.dst.dk/en.
42. Instructions in the use of services in the agreement about general practice Accessed 30 Oct 2016.
[Vejledning i anvendelse af overenskomstens ydelser. Overenskomst om almen 67. Swindle RW, Rao JK, Helmy A, Plue L, Zhou XH, Eckert GJ, Weinberger M.
praksis 1 Sep 2014]. The Regions’ Salary Board, The Danish College for General Integrating clinical nurse specialists into the treatment of primary care patients
Practitioners [Regionernes Lønnings- og Takstnævn (RLTN), Praktiserende with depression. Int J Psychiatry Med. 2003;33:17–37.
lægers Organisation (PLO)]. 2014. https://www.laeger.dk/sites/default/files/ 68. Proudfoot J, Ryden C, Everitt B, Shapiro DA, Goldberg D, Mann A,
vejledning_i_anvendelse_af_ydelser_i_ok14.pdf. Accessed 30 Oct 2016. Tylee A, Marks I, Gray JA. Clinical efficacy of computerised
43. Sheehan D, Lecrubier Y, Harnett Sheehan K, Janavs J, Weiller E, cognitive-behavioural therapy for anxiety and depression in primary
Keskiner A, Schinka J, Knapp E, Sheehan M, Dunbar G. The validity care: randomised controlled trial. Br J Psychiatry. 2004;185:46–54.
of the Mini International Neuropsychiatric Interview (MINI) according 69. Proudfoot J, Goldberg D, Mann A, Everitt B, Marks I, Gray J. Computerized,
to the SCID-P and its reliability. Eur Psychiatry. 1997;12:232–41. interactive, multimedia cognitive-behavioural program for anxiety and
44. Hesse M, Moran P. Screening for personality disorder with the depression in general practice. Psychol Med. 2003;33:217–27.
Standardised Assessment of Personality: Abbreviated Scale (SAPAS): 70. Adams G, Gulliford M, Ukoumunne O, Eldridge S, Chinn S, Campbell
further evidence of concurrent validity. BMC Psychiatry. 2010;10:10. M. Patterns of intra-cluster correlation from primary care research to
45. Osman A, Hoffman J, Barrios FX, Kopper BA, Breitenstein JL, Hahn SK. inform study design and analysis. J Clin Epidemiol. 2004;57:785–94.
Factor structure, reliability, and validity of the Beck Anxiety Inventory in 71. Yonkers K, Dyck IR, Keller MB. An eight-year longitudinal
adolescent psychiatric inpatients. J Clin Psychol. 2002;58:443–56. comparison of clinical course and characteristics of social phobia
46. Beck AT, Epstein N, Brown G, Steer RA. An inventory for measuring clinical among men and women. Psychiatric Services. 2001;52:637–43.
anxiety: psychometric properties. J Consult Clin Psychol. 1988;56:893–7. 72. Stant AD, Ten Vergert EM, den Boer PCAM, Wiersma D. Cost-
47. Beck A, Ward C, Mendelson M, Mock J, Erbaugh J. An inventory effectiveness of cognitive self-therapy in patients with depression
for measuring depression. Arch Gen Psychiatry. 1961;4:561–71. and anxiety disorders. Acta Psychiatr Scand. 2008;117:57–66.
Curth et al. Trials (2017) 18:382 Page 13 of 13

73. Porta N, Bonet C, Cobo E. Discordance between reported intention-

to-treat and per protocol analyses. J Clin Epidemiol. 2007;60:663–9.
74. Wittchen H-U, Jacobi F. Size and burden of mental disorders in Europe—a critical
review and appraisal of 27 studies. Eur Neuropsychopharmacol. 2005;15:357–76.
75. Wood L, Egger M, Gluud LL, Schulz KF, Jüni P, Altman DG, Gluud C,
Martin RM, Wood AJG, Sterne JAC. Empirical evidence of bias in
treatment effect estimates in controlled trials with different interventions
and outcomes: meta-epidemiological study. Br Med J. 2008;336:601–5.
76. Muntingh ADT, van der Feltz-Cornelis CM, van Marwijk HWJ, Spinhoven P,
Penninx BWJH, van Balkom AJLM. Is the Beck Anxiety Inventory a good tool to
assess the severity of anxiety? A primary care study in the Netherlands Study of
Depression and Anxiety (NESDA). BMC Fam Pract. 2011;12:66.
77. Buszewicz M, Griffin M, McMahon EM, Beecham J, King M. Evaluation of
a system of structured, proactive care for chronic depression in primary
care: a randomised controlled trial. BMC Psychiatry. 2010;10:61.

Submit your next manuscript to BioMed Central

and we will help you at every step:
• We accept pre-submission inquiries
• Our selector tool helps you to find the most relevant journal
• We provide round the clock customer support
• Convenient online submission
• Thorough peer review
• Inclusion in PubMed and all major indexing services
• Maximum visibility for your research

Submit your manuscript at

www.biomedcentral.com/submit