Blended versus face-to-face Cognitive Behavioral Therapy for major depression in specialized mental health care * a pilot randomized controlled trial examining health care efficiency

The health-economic analyses combine clinical outcomes with cost estimates.

Measures of these primary variables are described in this section.

Clinical outcomes

Primary clinical outcomes are 1) recovery from Depression as assessed by the

Mini-International Neuropsychiatric Interview (MINI), 2) changes in Depressive

symptom severity as measured by the Inventory of Depressive Symptomatology *

Self Report (IDS-SR), 3) quality-adjusted life years (QALY*s), derived from the

Euro Quality of Life questionnaire (EQ-5D-3L) and 4) health-related quality of

life, tapped by the SF-36 Health Survey.

The Mini-International Neuropsychiatric Interview (MINI; Sheehan et al., 1992;

Van Vliet et al., 2007) is a brief clinician-administered structured diagnostic

interview for assessing the presence of DSM-IV and ICD-10 psychiatric

disorders. The interview takes 45 minutes to conduct at the most, depending on

the amount of disorders that are present. In this study, independent, trained

research assistants, who are blind to the random allocation, will administer

the MINI either over the telephone or face-to-face at the clinic (based on

patient preference). At T0 and T3, the MINI will be administered in full. At T1

and T2, current state will be assessed through a focused telephone or face to

face interview, in which only MINI section A & B are administered (assessing

depressive disorders and dysthymic disorder). At T0, the lifetime-version is

administered (which also yields current state). At T1-T3, the MINI is only

focused on *current* symptoms of the participant.

The IDS-SR (Rush et al., 1986; Rush et al., 1996; Rush et al., 2000) is a

self-report measure of the severity of depressive symptoms in the past week. It

consists of 28 items which each have four response categories, ranging from 0

(low severity) to 3 (high severity). The total score varies between 0 and 84,

with higher scores being indicative of a higher severity of depressive

symptoms. For this study, treatment response is defined as a symptom reduction

of the baseline IDS-SR symptom severity score of at least 50%.

The EQ-5D-3L (EuroQol Group, 1990; Lamers et al., 2006) will be administered at

every assessment moment (T0-T3) to assess general well-being. The questionnaire

consists of five questions that tap mobility, self-care, daily activities, pain

and mood. Each item has three response categories, ranging from 0 (no problems)

to 3 (severe problems). In addition to this, participants use a VAS scale to

rate their health on a scale ranging from 0 (worst possible health) to 100

(best possible health). The answers on the five questions are combined in a

number sequence that corresponds with the five answers, for example 03210. The

total number of possible sequences is 35=243. Each sequence stands for a

certain health state. On these health states, a value (utility) has been placed

(Lamers et al., 2006), which in turn is used to determine the quality-adjusted

life years (QALYs). This is done by calculating the QALYs gained between the

follow-up periods by weighing the length of time spent in a particular health

condition by the utility score (Drummond & Sculpher, 2005).

The Short Form Health Survey (SF-36) will be used to assess health-related

quality of life (Aaronson et al., 1998; Bech et al., 2003; Ware et al., 2000).

This questionnaire consists of 36 questions that are scored on a 8 multi-item

scales, which assess physical functioning, role limitations caused by physical

health problems, bodily pain, general health perceptions, vitality, social

functioning, role limitations caused by emotional problems and general mental

health. The raw scores are converted to a scale ranging from 0 to 100 with

higher scores being indicative of better levels of functioning. SF-36 scores

will provide a second source to determine QALYS, through the application of the

Brazier algorithm (Brazier, Roberts & Deverill, 2002).



Cost outcomes

Cost estimates include 1) the costs of offering the treatments and 2) Patients*

out-of-pocket costs (non-medical), such as the costs of travelling to the

health services and the patients* time costs of travelling, waiting and

receiving treatment, which are determined with the standard cost prices as

listed in the pertinent Dutch guideline for economic evaluation (Tan et al.,

2012). Furthermore, 3) Costs stemming from health care uptake, including costs

of medication, and 4) Costs stemming from productivity losses due to

absenteeism and lesser efficiency while at work (presenteeism) are assessed

with The Trimbos/iMTA questionnaire for Costs associated with Psychiatric

Illness (TiC-P; Hakkaart * van Roijen et al., 2002).

TiC-P is the most widely used health service receipt interview for economic

evaluations in the Netherlands. The TiC-P consists of 46 questions, divided

over two parts. Part 1 entails the health care uptake at relevant health care

providers in the past 4 weeks, such as medication intake, and the number of

contacts within the mental health care setting, with the GP and with other

medical specialists. To determine the costs associated with these contacts, the

care consumption is multiplied by the cost price described in the before

mentioned guideline (Tan et al., 2012). Part 2 of the TiC-P entails loss of

productivity in the past 4 weeks. This is measured by enquiring about the

number of days absent from work and the number of days with reduced efficiency

due to feeling ill. The costs of productivity losses will be based on the

gender and age specific friction costs, as outlined in the Dutch guideline for

costing (Tan et al., 2012).

To further evaluate bCBT compared to CBTAU, a number of explorative measures

are administered.

The Quick Inventory of Depressive Symptomatology (QIDS-SR; Rush et al., 2003),

a shortened version of the IDS-SR, will be administered monthly during the

study period to more closely monitor the depressive symptom change trajectory

(at weeks 4, 8, 12, 16, 24, 28). This questionnaire consists of 16 questions.

The total score ranges from 0 to 27, with higher scores being indicative of a

higher severity of depressive symptoms.



The five item version of The Mastery Scale (Pearlin, & Schooler, 1987) is

administered at every assessment moment (T0-T3) to assess changes in locus of

control. Locus of control could potentially mediate treatment effect and

facilitate relapse prevention (Badamgarav et al., 2003; Neumeyer-Gromen et al.,

2004) The questionnaire consists of five questions, which are scored on a

five-point Likert-scale, ranging from 1 (totally disagree) to 5 (totally

agree). The total score ranges from 5 to 30, with higher scores being

indicative of a higher level of experienced control.



The 12-item version of the Work Alliance Inventory (WAV-12, Reynolds et al.,

1995; Andrusyna et al., 2001; Stinckens et al., 2009) is used to let patients

rate the quality of the work alliance between patient and therapist at T1 (week

10). The questionnaire is administered to investigate whether the blended

treatment for has an effect on the quality of the work alliance.

The questionnaire consists of 12 items, which are scored on a five-point

Likert-scale, ranging from 1 (seldom or never) to 5 (always). The raw scores

range from 12 to 60, with higher scores being indicative of a better alliance

between therapist and patient.



The depression scale of the Cognition Checklist (Beck et al., 1987) will be

used to assess the frequency of automatic thoughts relevant to depression, at

T0-T3. The depression scale consists of 14 questions on which a patient rates

the frequency with which a thought typically occurs on a five-point

Likert-scale ranging from 0 (never) to 4 (always). Scores range from 0-56 with

higher scores being indicative of more frequent negative depressive cognitions.



Demographic characteristics such as age, sex, education, employment and marital

status will be collected with a general demographic questionnaire. Additional

questions are asked concerning clinical depression characteristics such as age

of onset, number of months depressed in past 4 years, duration of current

episode, medical illnesses and treatment status. In addition, patients indicate

their treatment preference (BCBT / CBTAU). Finally, participants are asked

about their computer use: number of hours spent behind a computer and reasons

for use.

Patient evaluations

The Client Satisfaction Questionnaire-8 (CSQ-8; Larsen et al., 1979) will be

administered at week 30 (T3). The CSQ consists of 8 questions with

item-specific response categories. The total score ranges from 8 to 32, with

higher scores being indicative of higher levels of client satisfaction.

The System Usability Scale (SUS, Brooke, 1996) will be administered at week 20

amongst the participants randomized to the bCBT group. The SUS consists of 10

questions with 5 response options, ranging from 0 (strongly disagree) to 4

(strongly agree).The total scores are converted to a scale ranging from 0 to

100. Higher scores are indicative of higher usability of the online platform

that is used for the Internet sessions in the blended therapy.



Process data

Data for process analyses are obtained from the administration of the

participating mental health care institutions and through usage statistics of

the online platform. We will consider the aspects that will feed the study flow

chart according to the CONSORT guidelines (Moher et al., 2010; Schulz et al.,

2010), such as the losses and exclusions for each group, with reasons. In

addition to this, we will pay specific attention to:

- The extent to which treatment was provided parallel to bCBT or CBTAU, such as

pharmacotherapy, and the nature of this treatment.

- Therapy drop-out (number of completed sessions);

- The total number of face-to-face contacts and cancellations (amount of

therapy received);

- Treatment fidelity of both patients and therapists. In order to measure

treatment fidelity of therapists face-to-face sessions are audio-taped in both

treatment groups (when a patient has provided consent for this). A randomly

selected sample of these session-recordings will be checked for therapist*s

treatment integrity by independent raters. In the bCBT condition the raters

also include a random selection of the written feedback therapists have

provided on the online sessions. The sessions recordings also provide insight

in the extent to which patients complied with making their homework exercises.

- Time investment by both the patient and the therapist. In the bCBT group time

investment can be calculated based on the number of face-to-face contacts and

the amount of time spent working with the online platform. In the CBTAU group

time investment will be estimated based on the number of face-to-face contacts

and the treatment fidelity rating.

- Time investment will be discussed in more detail in qualitative interviews

for which we will invite a random selection of 10 patients per treatment group

after week 30. The interviews focus on the feasibility and usability of the CBT

treatment provided.