*3.2. Instruments*

Clinical evaluation. Sociodemographic data, family and personal psychiatric history, presence and characteristics of current diagnoses will be collected, along with age at onset, pharmacological treatments, tobacco consumption (if so, the Fagerström dependence questionnaire will be applied), caffeine consumption, assistance framework and treatment program, withdrawal time, relapses and previous suicide attempts, number of hospital admissions and duration and presence/absence of psychosocial problems. The healthrelated quality of life assessment will also be incorporated (SF-36 questionnaire) [66].

To rule out that the comorbid mental disorder in DD is secondary, the Spanish version of the Psychiatric Research Interview for Substance and Mental Disorders for DSM-5 (PRISM-5) [67] will be used. The severity of the SUD will be assessed using the Drug Abuse Screening Test (DAST-20) [68], the intensity of the depressive symptomatology using the Hamilton depression scale (HDRS) [69], the psychotic symptomatology with the PANSS scale [70] and suicidal risk using the Plutchik scale [71], all of them the Spanish version.

The evolution will be followed at 3, 6 and 12 months of inclusion in the study (application in pdf with closed fields that can be filled in on the computer), preferably through the contact professional of the center of origin or by the team's researchers in a hetero-assessed manner, with the patient and consulting their medical history.

Circadian rhythmicity pattern. A Kronowise KW6® ambulatory device (Kronohealth SL, University of Murcia, Spain) will be used to objectively record circadian rhythmicity. This integrates the measurement of peripheral/distal skin temperature, activity level by means of a 3-axis accelerometer and body position, as well as four light channels (average and peak of visible light, blue or circadian light that mimics the spectrum of melanopsin of retinal ganglion cells and infrared radiation). This integrated register allows for collection of variables that influence circadian rhythmicity (body position, physical activity, exposure to light, sleep and food schedules, etc.) in an objective and validated way [20,72]. An easy-touse programming and reading-analysis software (CircadianwareTM, Kronohealth, Murcia, Spain) has been developed and provides a complete circadian report. The device consists of a clock system placed around the non-dominant wrist, with a micro-USB connection for programming, data download and battery charging. Patients in study 1 should wear the device for 2 days, and those in study 2 will be monitored continuously during the first two weeks and at weeks 4 and 8, except in moments of personal hygiene, during which they can be removed (indicated in the "event" option of the Kronowise KW6®). In addition to the usual sleep schedule, on the days of registration, bedtime and waking up times will be considered by the event marker available in the device. The use of ambulatory instruments in mental health represents a significant progress both in assessment and, as recently evidenced, in therapeutic managemen<sup>t</sup> [73,74].

Self-assessed information on circadian typology will also be collected using the Composite Morningness Scale (CSM) [75], mood rhythmicity pattern (MrhI) [76] and seasonal variations (SPAQ) [77], in all cases with validated instruments in the Spanish population. For study 2, patients will fill in eight unipolar visual analogue scales [78] daily after natural light exposition and during the two months of the intervention in order to assess subjective activation and mood status.

Genetic determinations. A 2 mL sample of saliva (Salivette Oragene-DNA (OG-500)) will be taken after 30 min of not having had any liquid or solid intake, including chewing gums or smoking. The saliva samples will be stored at room temperature in a light-preserved place until processed. For the extraction of genomic DNA, an aliquot of saliva will be used, following the manufacturer's protocol (Oragene). A Qubit fluorometer (Invitrogen, Carlsbad, CA) will be used to determine DNA concentrations. The DNA samples will be adjusted to a final concentration of 10 ng/ul in TE-4 (10 mM Tris-HCl, 0.1 mM EDTA, pH 7.5) and stored at 4 ◦C until use. Molecular genetics analyses will be performed using previously validated real-time PCR-based methodologies, using Taqman probes for the PER2 polymorphisms (rs934945), PER3 (rs2640909), BDNF rs6265 (Val66Met), APOE rs429358 and rs7412 (E2, E3 and E4). The VNTR in MAOA will be genotyped by analyzing the amplified PCR fragments and run on agarose gel electrophoresis.

Neuropsychological evaluation. The battery of standardized neuropsychological tests is based on the evaluation of functions that may be affected in our sample. A measurement is avoided based on aspects such as reaction time or fine motor skills, since these may be influenced by the effects of the various psychopharmacological treatments used by patients, especially those with DD. The functions assessed and the tests applied are the following:


(d) Executive functions. Trail Making Test (Part A and B) and block of reverse digits from the WAIS-III Digits test for working memory; the Tower of Hanoi in its computerized version of 4 disks for planning and problem resolution; and the Wisconsin Card Sorting Task (WCST, Computerized Wisconsin Card Sort Task Version 4, 2003, Estevez-Gonzalez, Barcelona, Spain) for cognitive flexibility, concept creation, problem solving, inhibition and learning.

Finally, information will be collected with the Prefrontal Symptom Inventory (ISP) [80], consisting of 46 items that evaluate three factors: executive control problems (motivational, control and attention problems), social behavior problems and emotional control problems. This inventory has been validated in an adult clinical population with SUD, and convergence data with neuropsychological measures are also available [81].

Personality characteristics. Two personality questionnaires will be used that have been sensitive to numerous psychopathological disorders, including SUDs, and with positive data also in DD. Both are available in validated Spanish versions with population norms:


#### *3.3. General Procedure*

Study 1. Candidates to participate in the study will be added gradually, as they are derived by the psychiatrist or psychologist in the therapeutic team from the clinical centers, who will have briefly explained what the study consists of. The study objective will be explained to all patients at the first visit (inclusion), along with the measurements to be carried out and number of sessions involving their participation. Those who accept will sign an informed consent (tutors in the case of incapacitation) and will be evaluated in the patient's usual care center, preferably in morning sessions (9:00 a.m.–2:00 p.m.).

The collection of epidemiological and clinical information, as well as the application of the instruments, will be carried out by means of an individualized data collection booklet that will be completed either by the experimenter (hetero-application) or by the patient (self-application), depending on the case. Those tests that have a computerized version will be presented and answered on a computer. These are mostly performance tests that the subjects must answer individually. For the completion of the information, 3 individual sessions with the patient will be required, with a variable duration depending on the time it takes to respond but that will not exceed 2.5 h. The breakdown of sessions is presented in Figure 2.

A report of the results will be prepared for the clinical centers, and the return to the patient will be carried out by us or by a professional of the therapeutic team, according to the center's decision.

Study 2. Outpatients of both sexes with a diagnosis of SUD and dual depression who have provided a circadian registry of peripheral temperature with a partial restoration of circadian rhythmicity will be proposed to participate in a chronobiological intervention complementary to the usual treatment. Those patients of both groups who agree to participate (given the link with the centers, a very high affirmative response is expected) will be assigned alternately to the treatment condition (habits and exposure to light) or to the control condition, with only routine treatment and ambulatory evaluations of rhythmic and

subjective measures, as well as scheduled visits. The instructions of habits and exposure to natural light that will be stipulated for patients are presented in Box 1.

**Figure 2.** Breakdown of measurements for the sessions in study 1.

**Box 1.** Instructions of habits and exposure to natural light stipulated for patients assigned to treatment condition in study 2. The text will be printed on a plasticized card that will be delivered on the day of inclusion to the patient so that they have it on hand and can check whenever they want.

## **Hourly habits:**


#### **Daily exposure to natural light:**

Go out every day between 9:00 and 11:00 a.m. for a one-hour walk. You can take short rest breaks, but it is better to keep moving and walking while exposed to daylight. Do not wear sunglasses or glasses that darken in contact with light. If weather conditions make it impossible to go outside, place yourself at the time of the walk next to a window with outside light for an hour while doing some activity (for example reading).

The intervention will last two months with the following visits and planned evaluations:

• Visit 1 (inclusion). Explanation of the study differentiated according to the assigned condition and collection of informed consent. Placement of the Kronowise KW6® device that will be worn continuously until the next visit, emphasizing that the patient should register daily the time of getting up, having meals and going to bed. Specification of the rules to be followed in relation to hourly habits and exposure to natural light and delivery of the reminder card for treatment condition. Delivery and collection of the subjective assessment in 8 visual analogue scales (4 of activation and 4 of affective state) to be filled out by the patient each day until the next visit.


For the follow-ups at 3 (12 weeks), 6 and 12 months of inclusion, the same information will be collected from the patients in study 1 with a personal visit whenever possible.

Patients who agree to participate will sign a second informed consent that includes, for the intervention condition, the commitment to adhere to the recommendations of daily habits and daily exposure to natural light and to wear the Kronowise KW6® device during the established periods, with specification of the control about their activity both night and day. Patients assigned to the control group will sign another simplified informed consent. Thanks to continuous ambulatory registry, adherence to both the schedules and the daily session of exposure to natural light will be controlled, along with the quantification of the intensity and type of light exposed and duration/intensity of physical exercise throughout the record. This prevents us from requiring the patient to register daily, even for shorter time periods of essential information, such as that collected with sleep diaries.

#### *3.4. Statistical Analysis*

The time series obtained from the Kronowise KW6® records, using the Circadianware™ software, will be subjected to a classic rhythmometric analysis with the cosinor method (maximum, minimum, mesor, amplitude, acrophase and % of the rhythm), as well as to an analysis non-parametric (interdaily stability, intraday variability, relative amplitude, etc.). The circadian function index (CFI) is calculated with the average of IS, IV and RA, ranging from 0 (absence of rhythm) to 1 (robust circadian rhythm).

The SNPStats program [85] will be used to calculate allelic and genotypic frequencies, as well as Hardy–Weinberg equilibrium analysis (with an *χ*2 test) to explore the association of SNP genotypes, with quantitative measures of the evaluated variables. A corrected linear regression model will be used for age and sex (when applicable). An approach based on the false discovery rate (FDR, q value) will be applied for the correction of multiple statistical tests, using the QVALUE program. The exploration of gene–gene and environment–gene interactions will be carried out with the MDR program and the multilocus genetic profile (MGP) method.

All independent and dependent variables collected in the study will be incorporated through a double entry system in a computer file after conversion of the data involving hours to the centesimal system. In the dependent variables for which population scales are available (i.e., percentiles), these will be considered or Z or T scores will be calculated as appropriate. Descriptive statistics and a complete correlational analysis will be calcu-

lated for the different variables considered. Subsequently, different covariance analyses (ANCOVA/MANCOVA) will be carried out for the rhythmic pattern, neuropsychological performance and personality factors, as dependent variables, introducing group (DD, SUD, SMI) as an independent variable and the age as covariate in all cases. The same analyses will be carried out for the SUD and dual depressive patients, incorporating sex as a factor. If the conditions for the analyses are not met, equivalent non-parametric tests will be used. The need to apply Bonferroni correction for multiple comparisons will also be assessed. The consideration of other independent variables to be included as covariates will be determined from the descriptive and correlational analyses (residential/outpatient treatment, age at onset of the disorder, months of abstinence, etc.). All analyses will also be performed comparing only the DD and SMI groups according to the type of mental disorder. For study 2, repeated measures analyses will be carried out with the different temporal measurements and with intervention (treatment and control), diagnosis (SUD and dual depression) and sex as factors, as well as contrasts between groups at each time considered. In all cases, the eta squared partial statistic (ηp 2) will be estimated to measure the effect sizes. Linear or logistic regression models will be carried out, if applicable, that include as predictive variables those present in the clinical history in relation to the measurements made in both studies and, of these, in relation to the information of the follow-ups. The analyses will be carried out with the SPSS/PC+ statistical package (SPSS Inc., Chicago, IL, USA), and the statistical tests will be considered bilaterally, with a type I error established at 5%.

#### *3.5. Management and Collection of Research Data*

The protocols for the two studies are in accordance with Spanish legislation (Biomedical Research Law, BOE 4 July 2007, Research on data collection in humans). Our research adheres to the ethical standards of the Declaration of Helsinki [86] and of research in chronobiology [87]. Furthermore, the procedures will be carried out in accordance with international recommendations in the field of ethics of human genetic studies [88]. Participation in the studies does not imply risks for patients, as there are no invasive registries or interventions with known risky side effects.

Data collected from the research group for the Project will be digitized and stored on the University's Microsoft OneDrive for Business. The Microsoft Agreement includes Terms and Conditions that are compliant with EU Data Protection Law and the National Bioethics Committee rules and regulations. Only researchers working for the Project will have access to the data, using their username and passwords to access the files.
