Circadian reinforcement therapy in combination with electronic self-monitoring to facilitate a safe post-discharge period of patients with depression by stabilizing sleep: protocol of a randomized controlled trial

The MDB system has been specifically developed for patients suffering from depression in a joint collaboration between software engineers, psychiatrists, and patients [48]. The system can be accessed from all devices using an internet browser. Participants are provided with an iPad and a Fitbit bracelet (version Charge HR). Participants will enter the Fitbit count into the MDB system where it will be displayed along with the other registrations. Each participant is allocated a unique username and password. Participants get access to registrations through a patient interface and investigators through a separate clinician interface. Both interfaces display a similar graphical representation of entered data (see Fig. 1). Patients will be asked to self-assess their sleep and mood twice daily: 0–30 min after waking up, and 0–30 min before going to sleep – and enter it into the MDS.

Morning self-assessments: sleep onset, wake-up time, sleep quality score, number of nighttime awakenings, unlimited qualitative description of their sleep episode, and morning depression severity.

The Fitbit bracelet enables the collection of activity and heart rate. These will be investigated as a predictor for antidepressant response and as markers of the circadian rhythm.

After an introduction to the MDB system and the Fitbit bracelet, participants will enter data into the system daily in a four-week period. Participants will receive a weekly telephone call from the investigator evaluating (a) experience with the MDB system, (b) suicidal ideations, (c) side effects of medications, (d) worsening of depression, and (e) compliance with data entry. Patients will be reminded to contact either the IOS or the psychiatric emergency unit if they experience aggravation in depressive symptoms.

The CRT intervention also includes an individual CRT psychoeducational session at baseline and in one of the next two subsequent days, and with booster sessions at the weekly telephone calls. The initial two CRT sessions are provided by the investigator as an individual 1-h session including an introduction to the CRT principles using a full written CRT guide, and a leaflet version, focusing on how participants can incorporate adequate and timed zeitgeber stimuli into their daytime activities regarding daylight exposure, exercise, meals, and social contact. The investigator then helps the participant to fill in a schedule called “basic strategy for daily structure” and provides the participant with a paper and pen diary “The Chrono schedule” where the daily timing of daylight exposure, exercise, meals, and social contact must be noted for the morning, afternoon and evening time-period (see Fig. 2). Finally, the investigator helps the participant to formulate and write down three problems, within the CRT area, they want to work with, in the following 4 weeks, for example: how to achieve an earlier bedtime, how to achieve more daylight, and how to achieve more regular meals.

The Fitbit will be set to a goal of 4000 steps, alerting with a vibration when this is reached, to nudge participants to be more active.

Participants’ data registrations, in the MDB system, will be evaluated daily (Monday to Friday) by the investigator. Patients will be contacted if data is missing or if the registrations show predefined trigger signs of aggravation of depression or unstable sleep patterns (described in Table 1).

All participants will be following the standard IOS treatment program as described above.

The Mini International Neuropsychiatric Interview (M.I.N.I.) 4.0 (Danish version) is used for diagnosis. This is a short, structured diagnostic interview based on the DSM-IV diagnostic system, and will be conducted to confirm depression diagnosis, assess comorbidity, and any contraindications for participation in the trial [49]. Investigators are certified in the use of the M.I.N.I. instrument.

Sociodemographic data are collected at baseline together with a treatment outcome expectancy rating. Psychometric assessments are performed at baseline and at the endpoint visit. Rating window is set as last 3 days for all instruments except for the Pittsburgh Sleep Quality Index (PSQI) that covers the last month. The severity of depression is assessed by the Hamilton Depression Rating scale, 17 item version, which includes the core symptoms of depression (HAM-D₆), and the Bech-Rafaelsen Melancholia Rating Scale (MES) focusing more on the cognitive symptoms of depression [50, 51]. The Hamilton rater is blinded to treatment allocation and is trained in the use of the HAM-D₁₇ by repeated groups ratings with co-researchers. Self-assessed depression severity is measured by the Major Depression Inventory (MDI), a 10-item questionnaire covering both the ICD-10 and the DSM-IV diagnostic criteria for depression [52]. The response categories of each item are from “at no time” (0) to “all the time” [5], with a score range from 0 (zero) to 50 and a cut-off score of 21 for mild depression, 26 for moderate depression and 31 for severe depression. The WHO-5 Wellbeing index is used as a measure of Quality of life [53]. The WHO-5 consist of 5 items, each rated from “all of the time” (5) to “at no time” (0). The raw score is calculated by summing items multiplicated by four, giving a score range from 0 to 100, where 0 is worst possible and 100 is the best possible quality of life. The Morningness-Eveningness Questionnaire (MEQ) is a 19 items self-assessed questionnaire constructed to assess chronotype. A score below 42 indicates “evening type” and a score above 58 indicates “morning type”; and a score between 42 and 58 indicates “intermediate type” [54]. Sleep quality is assessed by the PSQI containing 11 self-reported items. These items are transformed into 7 “component scores” , and a “global score”. The components scores are “Subjective sleep quality (1)”, “Sleep latency (2)”, “Sleep duration (3)”, “Sleep efficiency (4)”, “Sleep disturbances (5)”, “Sleep medication (6)” and “Daytime dysfunction (7)”. A “Global Score” of 5 or above indicates poor sleep quality [55]. The System Usability Score (SUS) is used to evaluate the usability of the MDB system [56].

The Dim Light Melatonin Onset (DLMO) will be determined by saliva sampling every 45 min in the evening from 5 h before expected bedtime until 1 h after expected bedtime at baseline and endpoint (6 samples at each assessment day). The patient will collect samples at home according to a delivered 1-page guideline. The important rule from this guideline is to wear orange goggles (provided by investigators) for the entire 6-h episode of saliva sampling (to prevent suppression of melatonin secretion by blue light). The samples will be refrigerated and kept frozen until the melatonin assay. All samples obtained from a single patient will be assayed in the same batch to avoid inter-assay variability. The DLMO, a marker of circadian phase, is determined as the time point where melatonin production rises [57] using the hockey-stick method [58]. Patient-reported sleep data is used to calculate sleep midpoint. From sleep midpoint and DLMO, we can calculate the Phase Angle Difference (PAD) as a measure of entrainment [59]. The normal value of PAD is 6 h [56]. We expect patients to have a phase delayed DLMO relative to sleep midpoint with a PAD of approximately 5 h.

Participants are included while still on the inpatient ward or at latest 3 weeks after initiating treatment at the IOS and are followed for 4 weeks (see Spirit flow diagram in Fig. 3).

Sample size calculations were done using the SAS 9.4 software.

The sample size for the primary outcome was calculated from an expected HAM-D₁₇ mean score of 18 at baseline, a decrease to a score of 13 in the Standard group arm and to 10 in the CRF group, an expected pooled standard deviation of 6, an alpha level of 0.05 and a power of 0.80: a total of 128 participants is needed.

The sample size for the secondary outcome was calculated from an expected MDI mean score of 27 at baseline, a decrease to 22 in the Standard group and to 16 in the CRF group, with an expected pooled standard deviation of 12, an alpha level of 0.05 and a power of 0.80: a total of 128 participants will be needed.

The sample size for the tertiary outcome was calculated from an expected PAD of 5 h in the CRT group at baseline (a phase delay) normalizing to 6 h at endpoint, an unchanged PAD of 6 h in the Standard group, a PAD standard deviation of 80 min, an alpha level of 0.05 and power of 0.80: a total of 58 patients will be needed. We will collect saliva samples in 58 consecutive patients from participant randomization number 72.

When the IOS receive a referral, the attending physician will pass on information from the patient’s case file to the investigator, regarding age, dementia, psychotic episodes, and bipolar illness. If no exclusion criteria are present, investigators will contact the referring ward.