A study protocol for the feasibility and acceptability of a personalized early intervention combining light therapy, lifestyle psychoeducation, and imagery-focused cognitive therapy in individuals at risk for bipolar disorder

Abstract

Background

Bipolar disorder (BD) is a severe mental illness associated with marked functional impairment and reduced life expectancy. Early indicators such as mood instability, circadian rhythm disturbance, and anxiety symptoms often precede the first manic or depressive episode, providing a potential window for preventive intervention. Currently, no structured early intervention program exists for individuals at risk for BD who do not yet meet diagnostic criteria. This study aims to evaluate the feasibility and acceptability of a novel, personalized early intervention program combining light therapy, lifestyle psychoeducation, and imagery-focused cognitive therapy (ImCT) for individuals at risk for BD.

Methods

The study employs a single-case experimental A-B-A design with staggered baseline and multiple daily assessments. Fifty participants aged 16–35 years identified as being at risk for BD by a specialized early detection team will be included. The intervention consists of three core components: (1) a chronotherapeutic intervention (bright light therapy or blue-light blocking glasses) tailored to individual symptom profiles; (2) one session of lifestyle-focused psychoeducation targeting sleep, nutrition, and physical activity; and (3) six sessions of ImCT to address mood instability and maladaptive mental imagery. Feasibility and acceptability will be assessed through drop-out rates, adherence, and participant feedback. Secondary outcomes include changes in depressive, hyperactive, anxiety, and imagery-related symptoms, as well as sleep quality and activity levels, measured through validated questionnaires and actigraphy.

Discussion

By combining chronotherapeutic, psychological, and lifestyle components, this intervention targets multiple mechanisms implicated in BD risk. Findings will inform the development of preventive strategies for individuals in an at-risk mental state for BD. The study will also provide data on the feasibility of integrating early interventions within routine mental health services and guide the design of future randomized controlled trials.

Trial registration

Medical Ethical Committee Brabant (METC Brabant; identifier P2314); ClinicalTrials.gov Identifier: NCT06282250. Registered 20 February 2024.

Background

Bipolar Disorder (BD) is a severe mental illness (SMI) affecting approximately 2–3% of the general population and is associated with a significantly reduced life expectancy, often by as much as 20 years (Faedda et al. 2015). Despite its substantial impact, a clearly defined at-risk mental state for BD remains elusive, in contrast to the established concept of Ultra High Risk (UHR) for psychosis. Emerging research has highlighted key risk factors that may precede the onset of BD by several years. These early warning signs include mood instability, circadian rhythm disturbances, and anxiety symptoms (Kapczinski et al. 2009). Screening tools like the Semi-structured Interview for Bipolar at Risk States (SIBARS) can assist in identifying individuals at heightened risk for BD (Fusar-Poli et al. 2018).

For other SMI’s, the phase before manifestation has been more clearly defined. For example, in the field of psychosis, an ‘at-risk mental state’ has been identified. Recent meta-analyses have confirmed once more that 25% of patients with an at-risk mental state for psychosis, develop this SMI within three years (De Pablo et al. 2021). Studies mention that the transition risk increases over the years, meaning that clinical monitoring and preventive interventions might be beneficial and even crucial. Consequently, an early intervention named the UHR program has been developed specifically for individuals with an at-risk mental state for psychosis. The UHR program has been found to reduce the risk of transitioning into psychosis by 50%, and is successfully administered as care as usual in most SMI healthcare institutions in the Netherlands (Van Der Gaag et al. 2013). Another example is the recent suggestion in literature for an early detection and intervention model for SMI in general, named the Clinical High At-Risk Mental State (CHARMS) categories (McGorry et al. 2021). Given the success of early intervention programs for other SMI’s, the next step is a similar approach to BD risk.

The current study is a step towards the implementation of a transdiagnostic early intervention program in a population of individuals at risk for developing bipolar disorder. This program integrates the following evidence-based interventions to address depressive or attenuated mania symptoms and anxiety: (1) chronotherapeutic intervention in the form of light therapy (Bright Light Therapy [BLT]) or blue-light blocking glasses), (2) psycho-education on lifestyle factors, and (3) Imagery-focused Cognitive Therapy (ImCT). All three intervention elements are explained below.

1. Chronotherapeutic intervention: Bright Light Therapy (BLT) and blue-light blocking glasses

Light Therapy (LT) is a chronotherapeutic intervention, also known as ‘bright light therapy’ (BLT) and affects the suprachiasmatic nucleus by stimulating light-sensitive ganglion cells in the human brain (Tao et al. 2020). Bright light stabilizes (US/UK) circadian rhythm-related melatonin secretion in the brain and can be delivered through a light box with light tubes and a screen, amongst other options (Pail et al. 2011; Terman and Terman 2005). BLT is a transdiagnostic intervention that has been found effective in multiple SMI and non-SMI populations and is part of the clinical guidelines for the treatment of bipolar disorder (Tao et al. 2020; Videnovic et al. 2017; Rosenthal et al. 1984). In the current program, a BLT intervention will be offered to at-risk individuals presenting with depressive symptoms. On the other hand, blue-light blocking glasses will be offered to at-risk patients with hyperactive symptoms. Glasses blocking blue light reduce the activity of photosensitive retinal ganglion cells (ipRGCs), which are important in circadian rhythm regulation. The glasses have been found to positively affect sleep in mood symptoms (Hester et al. 2021). The inclusion criterion for either BLT or blue-light blocking glasses is based on symptom scores of depressive symptoms or hyperactive symptoms, as determined by the cut-off scores (> 5) of the Self-Rated Inventory of Depressive Symptoms (IDS-SR) (Rush et al. 2000) and the Self Rating Mania Scale (ASRM) (Altman et al. 1997) questionnaires, respectively. Both questionnaires are further explained in the Methods section.

2. Lifestyle-oriented psycho-education (PE)

Psycho-education (PE) is one of the most commonly applied psychosocial interventions for SMI’s and recommended in international clinical guidelines (Regeer et al. 2015). PE is comparable to cognitive behavioral therapy when it comes to the chances of relapse and symptom burden (Parikh et al. 2012). PE enhances patients’ understanding of BD, promotes adherence and contributes to relapse prevention. The content of the PE is generally focused on topics such as mood stability, self-management, lifestyle and personal experiences (Stoornissen 2021). In the current program, PE will be focused on the ‘lifestyle wheel’ (Arts 2024). Specifically, the factors of sleep, food and movement will be addressed. The main goal of the PE session will be insight in individual mood patterns, the take-home being to use sleep, food and movement to maintain rhythm and stabilize mood symptoms.

3. Imagery-focused Cognitive Therapy (ImCT)

Growing evidence indicates that mental imagery drives mood fluctuations in bipolar disorder. Experimental studies by Holmes and colleagues (Holmes et al. 2008) show that mental imagery can amplify emotional states: vivid positive imagery increases (hypo)manic affect, whereas intrusive negative imagery intensifies depressive symptoms. Such imagery-driven mood shifts may contribute to the subsyndromal inter-episodic mood instability that is strongly linked to functional impairment and poorer prognosis in bipolar disorder (Altshuler et al. 2006; Henry et al. 2008; Marangell et al. 2009). Negative, future-oriented imagery is common in BD, and individuals at risk for BD display particularly potent mood responses to positive imagery (O’Donnell et al. 2018). suggesting heightened imagery-emotion coupling. To address these mechanisms, Holmes and colleagues developed imagery-based cognitive therapy (ImCT) (Hales et al. 2011; Holmes et al. 2016), which reduces mood instability by modifying unhelpful imagery and strengthening adaptive imagery strategies. In bipolar disorders, ImCT has been effective in the treatment of BD (Van Den Berg et al. 2023). In our early-intervention program, we assess the feasibility, acceptability, and effectiveness of the imagery techniques included in the imagery-focused CBT manual of Holmes and colleagues. The imagery techniques included in the present study are (1) imagery rescripting, (2), metacognitive imagery techniques (3), promoting positive imagery de novo and (4) a visuospatial task, which additionally have been described and found effective in the work of Janssen and colleagues (Julious 2005). These techniques were selected because they are accessible CBT techniques that have been found effective in BD populations (Van Den Berg et al. 2023) and populations at risk for psychosis (Julious 2005) The ImCT intervention specifically targets mood instability, anxiety, aimed at reducing destabilizing imagery and supporting mood regulation.

Primary research question and hypotheses

The primary aim of this study will be to investigate whether this personalized early intervention program is feasible and acceptable for participants.

Secondary research questions and hypotheses

The specific secondary objective of the study will be to investigate the preliminary effectiveness of the intervention by assessing whether the intervention is associated with a significant decrease in mood symptoms and anxiety symptoms. Also, we aim to assess whether the intervention is associated with more perceived control over mental imagery, significant improvement in sleep quality, and activity levels.

By assessing both clinical outcomes and participant feedback, the study seeks to establish the potential for such interventions in preventing the onset of bipolar disorder in individuals at risk. The research will explore the mechanisms through which these interventions may exert their effects. Ultimately, the findings from this study could inform future randomized controlled trials and contribute to the development of a standardized early intervention protocol for at-risk individuals.

Method

Design

This study employs a single-case experimental design (SCED) with an A-B-A phase structure, a staggered baseline across participants, and multiple daily measurements (three times per day) to capture within-day fluctuations in mood, imagery, sleep and activity levels. Staggering refers to varying duration of the initial baseline phase (A₁) across participants, such that the intervention is introduced at different time points. The varying baseline period reduces the likelihood that observed changes are attributable to time-related confounds, and is consistent with methodological recommendations for multiple-baseline SCEDs (Lancaster et al. 2004). In addition, the varying baseline design will serve as each participant’s control phase, allowing for comparisons of pre- and post-intervention data within the same individual.

Although the design includes planned A–B phase alternations, the second A phase represents a monitoring-only phase. This within-subjects design enables the assessment of treatment effects while accounting for individual variability. The SCEDs have 3 phases: (1) Baseline (Phase A1), lasting up to 2 weeks (2), Treatment (Phase B) lasting 4 to 7 weeks, and (3) Follow-up (Phase A2), lasting up to 2 weeks (Sim and Lewis 2012; Zyto et al. 2020; AKWA 2025).

Sample size and power analysis

The National Institute for Health Research recommends sample sizes for a feasibility study in a range of 24 to 50 in total (Sim and Lewis 2012; Zyto et al. 2020; AKWA 2025). This study meets those requirements. This calculation is based on the drop-out rates of previous studies on this topic in this healthcare institution. In addition, we performed a power and sample size analysis for randomization test in a single case experimental design (see Table 1). This analysis confirmed that we need at least N = 10 patients per intervention to demonstrate the effect of therapy. N= 10 patients per intervention will thus provide sufficient power to demonstrate effects. Sample sizes used in this analysis are based on previous meta-analyses studying the effect of light therapy interventions on mood and sleep, wherein small to medium effect sizes are found (d = 0.3 to d = 0.6) (Tao et al. 2020; Beck 1993). Since there will be three possible combinations of interventions within the early intervention program, power and sample size calculation is based on the number of participants needed per intervention combination to reach reliable conclusions.

Table 1.

Power for n = 30 and required sample size for Power = 0.8 for the effect baseline-treatment

Effect-size (d)	Power (n=)	N (power = 0.8)
0.60	0.94	4
0.50	0.85	4
0.40	0.84	7
0.30	0.87	10
0.20	. 80	15

When there are 30 clients, the power to find even a small effect is 0.87. Since sleep quality has only 2 measurements per day, a separate sample size calculation was made for multiple scenarios. In conclusion, a sample of n = 45 is needed to find an effect size of d = 0.3 for sleep-related data. With 45 clients, the power to find even a small effect is 0.86 (see Table 2).

Table 2.

Power for n = 45 and required sample size for Power = 0.8 for the effect baseline-treatment (sleep data)

Effect-size (d)	Power (n=)	N (power = 0.8)
0.60	0.85	6
0.50	0.84	8
0.40	0.83	10
0.30	0.86	15
0.20	. 80	15

The results show that with an effect size of d = 0.4, 45 clients is sufficient to have 0.8 probability to find a significant effect (see Table 3). When the effect-size is 0.6 or larger, 30 or 18 clients are sufficient to have high power to find a significant effect.

Table 3.

Power for n = 45 and required sample size for Power = 0.8 for the effect baseline-follow-up

Effect-size (d)	Power (n=10)	N (power=0.8)
1	0.91	5
0.8	0.85	6
0.6	0.84	10
0.4	0.81	15
0.3	0.80	21

Conclusion

This design has sufficient power to find even a small effect (d = 0.3). This is due to the large number of measurements within the phases. Thus, when it is indeed realistic to measure the dependent variable three times a day for seven weeks, a minimum of 45 clients are expected to be sufficient. Considering potential drop-out, the current study will aim to include 50 participants. This calculation is based on drop-out rates of previous studies on this topic in this healthcare institution.

Participants

The study will recruit participants from the Early Detection and Intervention Team (EDI-Team) at the mental health care institute Eindhoven (GGzE, The Netherlands) who have been identified as being at risk for BD. The recruitment is taking place from September 2024 to September 2027.

Inclusion criteria are:

• Age 16 to 35 years.

• Identified as at risk for developing BD by the EDI-Team at GGzE.

• Proficient in Dutch or English.

• Able to provide informed consent.

• Willing to complete daily monitoring throughout the study period.

Exclusion criteria are:

• Moderate/severe learning difficulties (IQ < 70), organic brain disease, or severe neurological impairment estimated by a clinician and based on neuro(psycho)logical test results, imaging such as MRI/CT and/or information provided by the participant.

• Recent receipt of BLT (within the last three weeks).

• Current severe substance or alcohol misuse estimated by a clinician.

• Eye conditions contraindicating the use of BLT or inability to attend study sessions.

• Any health concerns that could compromise the participant’s safety, as assessed by the clinician or investigator.

Early intervention

The early intervention program consists of three main components with a duration of 4–7 weeks; light therapy (BLT or blue-light blocking glasses), psycho-education (PE), and Imagery-focused Cognitive Therapy (ImCT). Based on their symptoms, participants will start with one of three options:

Week 1–3: chronotherapeutic intervention

1a. Bright Light Therapy: in case of depressive symptoms (IDS-SR score > 5), BLT will be administered for a minimum of one week and a maximum of three weeks. A week consists of five consecutive weekdays (Monday to Friday) of one 30-minute session per day, in the morning between 7:30 AM and 10:00 AM. At the end of week one, the effect of the light therapy is evaluated using the Self-Rated Inventory of Depressive Symptoms (IDS-SR) (Rush et al. 2000). If the patient has achieved remission (IDS-SR < 6), BLT has been successful and the patient can discontinue BLT. If there is insufficient or no response (IDS-SR > 5), BLT is extended by another week. If necessary, a third week is optional. On the last day of BLT, the effects and possible side effects will be evaluated. Patients receive BLT in the Light Café setting of GGzE. During their visit, participants are, like all Light Café visitors, offered a cup of coffee or tea and some fruit by trained healthcare professionals (specialized nurse). While in front of the light tube, patients will have a conversation with the health care professional present, to discuss topics related to lifestyle. Advice can be given on sleep, exercise, nutrition, relaxation, social interaction etc. This conversation will be structured.

1b. Blue-light blocking glasses: in case of attenuated mania symptoms (ASRM score > 5), blue-light blocking glasses will be used for a minimum of one week and a maximum of three weeks. A week consists of five consecutive weekdays (Monday to Friday), of wearing the glasses two hours before bedtime. Treatment can be done at home, and the patient can resume his or her normal activities during this time. At the end of week one, the effect of the blue-light blocking glasses is evaluated using the Self Rating Mania Scale (ASRM). If the patient has achieved remission (ASRM < 6), the glasses have been successful and the patient can discontinue wearing the glasses. If there is insufficient or no response (ASRM > 5), use of the glasses will be extended for another week. If necessary, a third week is optional. On the last day of use, the effects and possible side effects will be evaluated.

In case of both depressive and hyperactive symptoms, as indicated with a score > 5 on both the IDS-SR and the ASRM, treatment will be based on the highest symptom score (BLT in case of a higher IDS-SR score; blue-light blocking glasses in case of a higher ASRM score).

1c. In case of stable mood, with no heightened scores on either depressive or attenuated mania symptoms, participants will not receive a chronotherapeutic intervention and will therefore continue immediately with psycho-education after baseline.

Week 4: psycho-education

One week after finalizing the chronotherapeutic intervention, participants will attend one individual 60-minute PE session with a trained psychologist and a person with lived experience with BD. PE will entail information about the at-risk mental state for BD, the importance of lifestyle factors, and guidance on improving physical activity, sleep hygiene, and managing stress. The content of the PE session is based on the PE guideline in the Netherlands, as described in the manual of the Dutch Expertise Centre for Bipolar Disorders (KenBis) and proven effective by Zyto and colleagues (KenBis) (Hales et al. 2018). Specifically, the session is focused on the meaning of mood swings and their relevant symptoms, and the relevant stress factors for developing these symptoms (in line with session 1–2 and 8 of the KenBis protocol). Next, relevant self-management and lifestyle advice, such as circadian rhythm and diet (in line with session 7 of the KenBis protocol), are discussed. Finally, the role of the session leader with lived experience with mood swings is important (in line with session 12 of the KenBis protocol). In addition, information is added about the at risk mental state for BD, in line with the PE Ultra High Risk (UHR) protocol for individuals at risk for psychosis (Leon et al. 1999).

Week 5–7: Imagery-focused cognitive therapy

One week after finalizing the PE session participants will be provided with six sessions of 60 min of ImCT by a trained psychologist. The ImCT interventions in the current study are adapted from the protocol of Holmes et al. (2008). There will be two sessions per week, of which one will be online. Participants will practice techniques at home between sessions. To maximize adherence, individualized attention such as telephone calls following missed daily questionnaires, will be implemented. ImCT treatment will start with 2 sessions focused on an in-depth identification of relevant images the participant has. This microformulation identifies triggers, quality aspects and appraisals of mental imagery leading to depressive episodes or hyperactive episodes. Next, 2 sessions will focus on practicing imagery techniques. Finally, the final 2 sessions will focus on the consolidation. All sessions are based on the imagery-focused CBT manual of Holmes and colleagues (Hirschfeld et al. 2000), and in line with the protocol used in the work of Janssen and colleagues (Julious 2005).

Adapted from Holmes and colleagues (Hirschfeld et al. 2000), participants will receive one or more of the following imagery techniques:

• Imagery Rescripting (ImRs) is focused on changing the negative mental imagery into more functional or neutral imagery.

• Metacognitive imagery techniques are focused on creating emotional distance from the negative mental image, which reduces the power of this image. The goal of this technique is to clarify the negative mental image is “just an image” and in that way controllable by the participant.

• Promoting positive imagery is focused on creating a new, positive mental image to promote compassion and reduce anxiety related to the mental image.

• Tetris is a visuospatial working memory task used to compete with the negative metal image. Tetris is used when participants experience frequent, changing images, making it difficult to identify one as most negative or impactful.

Sociodemographic variables

At baseline, participants will provide sociodemographic information, including age, sex, education level and employment status, history of psychiatric care and current medication use. This information is obtained from the digital patient file combined with oral information from the participant.

Clinical variables

At baseline, participants will provide Psychiatric History (i.e. total duration of psychiatric or psychological treatments (in years), current medication use and comorbidities. This information is obtained from the digital patient file combined with oral information from the participant.

Measurements and outcome variables:

Various self-report measures, daily measures and questionnaires will be administered during this study. Below is a timeframe provided to present all assessments administered in this study (See Table 4).

Table 4.

Timeframe of assessments

Questionnaire	Assessment	Psychometric scale	Items	Duration	Administered
					T0: Baseline	T1: Post treatment	T2: Three month follow-up	Daily T0-T1 T2
IDS-SR	Depressive symptoms	Likert scale	12	5 min	X	X	X
ASRM	Attenuated mania symptoms	Likert scale	12	10 min	X	X	X
BAI	Anxiety	Likert scale	12	5 min	X	X	X
MICQ-BD	Control over mental imagery	Likert scale	12	5 min	X	X	X
LIFE-RIFT	Lifestyle characteristics	7-point scale	7	5 min	X	X	X
Move Monitor	Lifestyle characteristics Sleep data Activity data	Continuous measurement	n.a.	5–7 days	X	X	X
Retrospective self-report assessment for qualitative review of therapy	Treatment quality	5-point Likert scale (1–5)	32	10 min		X
Diary item on mood, anxiety, sleep, activity	Level of item	Likert scale (0–10)	10	< 1 min per item				X

1. Feasibility and Acceptability: Assessed through participant dropout rates, adverse events, and a retrospective self-report evaluation of treatment quality.

   We hypothesize that no serious adverse side effects will occur, drop-out rates will remain low and participants will provide positive qualitative feedback on their treatment experience.

Secondary hypotheses

1. Depressive symptom scoresare measured with the Inventory of Depressive Symptomatology (IDS-SR), which has demonstrated internal consistency in clinical and community samples (Cronbach’s α = 0.92–0.94; (Rush et al. 2000)).

   We hypothesize that there will be a significant reduction in depressive symptoms, as measured by the Inventory of Depressive Symptomatology - Self Report (IDS-SR) between T0 and T1.

• 2.Attenuated Mania symptom scores are measured with the Altman Self-Rating Mania Scale (ASRM), which shows concurrent validity with clinician-rated mania scales (r ≈ 0.72–0.77.72.77) and good diagnostic accuracy (sensitivity = 85.5%, specificity = 87.3%; (Altman et al. 1997))

  We hypothesize that there will be a significant reduction in attenuated mania symptoms, as measured by the ASRM between T0 and T1.

• 3.Anxiety symptom scores are measured with the Beck Anxiety Inventory (BAI), which shows internal consistency (α ≈ 0.92) and adequate test–retest reliability (r≈ 0.75; (Bouwmeester and Jongerling 2020))

  We hypothesize that there will be a significant reduction in anxiety symptoms, as measured by the Beck Anxiety Inventory (BAI), between T0 and T1.

• 4.Imagery symptom scores are measured with the Mental Imagery and Coping with BD Questionnaire (MICQ-BD), a 14 item self-report instrument with a five-point likert scale from “not at all” to “a lot” (Cronbach’s α = 0.70; (25,37). Higher scores are associated with more perceived control over problematic mental imagery. We hypothesize that there will be a significant improvement in control over imagery-related symptoms, as measured by the Mental Imagery in Bipolar Disorder Scale (MIC-BD) between T0 and T1.

• 5.Sleep Quality is measured with a daily assessments by Experience Sampling Method (ESM) and objective data collection using a movement sensor (Move Monitor © ….) which objectively measures sleep based on heart rate, movement and 24-hour activity levels.

  We hypothesize that Sleep quality will mediate the relationship between the intervention and the improvement in depressive symptoms or attenuated mania symptoms. We hypothesize that Sleep quality will mediate the relationship between the intervention and changes in physical activity levels.

• 6.Activity Levels are measured through Move Monitor sensor and three daily assessments (ESM). Additionally, functional impairment on daily activities such as work and school are measures with the Longitudinal Interval Follow-up Evaluation - Range of Impaired Functioning Tool (LIFE-RIFT), which has shown internal consistency (α = 0.78–0.84) and interrater reliability (ICC = 0.94; (38)).

  We hypothesize that physical activity levels will significantly increase following the intervention.

Daily measurements (ESM)

Thrice-daily measurements are administered daily to capture short-term, within-day fluctuations in mood, affective reactivity, and circadian-related processes. These dynamic processes are not adequately captured by retrospective questionnaires administered at baseline or follow-up. ESM therefore complements standard assessments by providing granular, ecologically valid information on daily symptom variability.

The ESM questions are designed to minimize participant burden by using brief assessments over a limited study period. Participants receive no additional financial compensation for their participation but are compensated for their travel costs. The ESM procedures build on previously implemented and evaluated experience sampling methodology (ESM) assessments within this institution (Van Den Berg et al. 2023).

Study procedures

Patients at risk for BD referred to GGzE will be screened for eligibility to participate in this study. BD risk is previously determined by the Early Detection and Intervention team with the following criteria:

One or more of the following criteria:

1. A score of > 6 on the Mood Disorder Questionnaire (39).

2. A known familial risk of BD, as determined by whether individuals have one or more first-degree family members with a BD diagnosis.

3. Symptoms indicating the at-risk mental state of BD as assessed with the Semi-structured Interview for Bipolar Risk States (SIBARS) (Fusar-Poli et al. 2018). The responsible clinicians (psychiatrists, psychologists or specialized nurse practitioners) ask the potential participant for permission to hand over their contact information to the researchers if they consider participation. Subsequently, the researcher makes an appointment with the patient to give more information about the study. After two weeks in which the participant can consider their participation in this study, the researcher makes an appointment to check if the participant still agrees to participate. If so, the informed consent form can be signed by the participant and the researcher. After a staggered baseline phase of up to two weeks, participants will begin the intervention, which lasts six to nine weeks, followed by a 3-month follow-up period where they will complete the same assessments as in the baseline phase. The flow diagram of the study procedure can be seen in Fig.1.

Fig. 1.

Study procedure

Data collection

Data collection will be based on a protocol for gathering sociodemographic and clinical data. Participants will be monitored throughout the study using daily diary assessments, and outcomes will be assessed at baseline, post-intervention, and follow-up.

Data analysis

Statistical methods

First, descriptive analyses will be used to establish mean scores, standard deviations and to identify outliers. Missing data from our primary outcome variable, will be listwise deleted. In our other outcome variables, data will only be included if pre- and post-measures are completed. Second, the data will be checked for violations of assumptions (such as error independence, non-normal distributions) for variance-based models. Non-parametric randomization test procedures will be performed via https://architecta.shinyapps.io/SingleCaseDesignsv02/ (40). All other statistical analyses will be performed using the Statistical Package for the Social Sciences (SPSS Statistics Windows, version 25.0) and R Studio. All analyses will be Bonferroni-adjusted for multiple tests within each domain by dividing the alpha of 0.05 by the number of tests within that domain.

Descriptive statistics

Mean scores and standard deviations or percentages of gender, age at inclusion, education level and work status will be presented to describe demographic and clinical characteristics of the study sample. Also, the mean levels of hyperactive mood symptoms (ASRM, diary items), depressive mood symptoms (IDS-SR, diary items), anxiety (BAI, diary items), imagery characteristics (MIC-BD, diary items), sleep quality (diary items, movement data) and activity levels (LIFE-RIFT, diary items, movement data).

Results

The study will run from September 2024 and data collection is expected to be concluded in September 2027. The first results of the data are expected in December 2027. Current data collection is at 30% completion.

Ethics and dissemination

Ethical considerations

This protocol was approved by the Medical Ethical Committee Brabant (METC Brabant; identifier P2314) and registered at clinicaltrials.gov (identifier NCT06282250). Additionally, this protocol was approved by the Institutional Review Board of GGzE, Institute for Mental Healthcare Eindhoven and De Kempen, Eindhoven, The Netherlands, as well as by the Ethical Review Board of Tilburg University, Tilburg, The Netherlands.

Written informed consent is obtained from all participants, and this study is conducted according to the principles of the Declaration of Helsinki (Association 2013) and in accordance with the Medical Research Involving Human Subjects Act (WMO). Participants have the option to discuss their participation in the study with an independent expert and may discontinue their involvement at any point. All data are stored anonymously and can only be accessed by research team members.

Dissemination

The final dataset resulting from this study will be the property of the research team and will not be shared without the explicit permission of the principal investigator. Once completed, the findings from this study will be disseminated through presentations at national and international conferences. Additionally, the results will be submitted for publication in peer-reviewed scientific journals, ensuring wide accessibility to the research community.

Discussion

Investigating the feasibility, acceptability and effectiveness of a personalized early intervention program, combining light therapy (BLT or blue-light blocking glasses), psycho-education (PE), and Imagery-focused Cognitive Therapy (ImCT), could be a step towards preventing BD in the future. Delaying or preventing a severe mental illness is possible in the field of psychosis spectrum research. For BD, however, few studies have systematically tested such interventions in individuals who have not yet fulfilled diagnostic criteria for BD but exhibit relevant risk factors or familial history. The present study builds on evidence-based interventions for BD, by extending the use of chronotherapeutic intervention, psycho-education and psychological intervention to individuals at risk for BD. This date will be useful in both clinical practice and scientific research.

This study was explicitly focused on feasibility and acceptability as primary outcomes. Also, a staggered baseline approach was chosen to serve as each participant’s internal control, allowing for within-subject comparisons over time while minimizing ethical and logistical challenges associated with forming an external comparison group.

Participants might find it challenging to wear the movement sensor or fill in the daily diary items. However, attempts have been made to reduce the number of questions asked and because the overall intervention protocol is relatively brief, up to a maximum of nine weeks, the burden on participants should remain manageable. This design choice aims to reduce dropout attributable to perceived inconvenience.

In terms of statistical power, earlier analyses indicated that a sample size of approximately 30 would be sufficient to detect moderate effects for key outcome measures. Nevertheless, we are opting to include up to 50 participants to further bolster the study’s robustness and account for any unforeseen variability or dropouts. Thus, while this remains a first step, the chosen sample size is expected to provide adequate power for detecting clinically meaningful changes in mood, anxiety, and activity levels.

These proposed methods and interventions integrate well into the expanding literature on transdiagnostic early intervention approaches. Although the at-risk mental state for BD has not been as clearly delineated as in psychosis, there is a growing consensus that circadian rhythm disturbance, mood, and anxiety symptoms are key early markers. By proactively addressing these symptoms, rather than waiting for a formal BD diagnosis, the study aligns with preventive mental health paradigms.

Should initial findings demonstrate feasibility, acceptability, and preliminary efficacy, future research could expand on several fronts. A randomized controlled trial (RCT) with a larger sample would offer stronger evidence for causal effects, while also allowing more robust subgroup analyses (e.g., by intervention type, age cohort, baseline symptom severity, or presence of comorbidities).

Conclusion

Taken together, these considerations underscore the innovation and potential impact of this early intervention protocol. By combining chronotherapeutic strategies, psycho-education, and ImCT, the study addresses multiple domains pertinent to BD risk, offering a uniquely comprehensive framework for preventing or delaying the onset of this severe mental illness. The feasibility data and preliminary efficacy outcomes may ultimately contribute to better outcomes for individuals at risk for bipolar disorder.

Author contributions

ET: conceptualization, methodology, investigation, writing – original draft, project administration, visualizationLS: conceptualization, methodology, writing – review & editing, supervisionKB: conceptualization, methodology, writing – review & editing, supervisionIB: conceptualization, methodology, writing – review & editing, supervision, project administration.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The datasets generated and/or analyzed during the current study are not publicly available at this stage because data collection is ongoing. Upon reasonable request and following completion of the study, anonymized data may be made available by the corresponding author in accordance with institutional and ethical guidelines.

Declarations

Ethics approval

This study was approved by the Medical Ethical Committee Brabant (METC Brabant; identifier P2314) and registered at ClinicalTrials.gov (Identifier: NCT06282250). The protocol was additionally approved by the Institutional Review Board of GGzE, Institute for Mental Healthcare Eindhoven and De Kempen (Eindhoven, the Netherlands) and the Ethical Review Board of Tilburg University (Tilburg, the Netherlands). Written informed consent will be obtained from all participants prior to enrollment. The study will be conducted in accordance with the principles of the Declaration of Helsinki (Association 2013) and the Medical Research Involving Human Subjects Act (WMO).

Competing interests

The authors declare no competing interests.