- Study Protocol
- Open access
- Published:
Medical comorbidities in bipolar disorder (BIPCOM): clinical validation of risk factors and biomarkers to improve prevention and treatment. Study protocol
International Journal of Bipolar Disorders volume 12, Article number: 15 (2024)
Abstract
Background
BIPCOM aims to (1) identify medical comorbidities in people with bipolar disorder (BD); (2) examine risk factors and clinical profiles of Medical Comorbidities (MC) in this clinical group, with a special focus on Metabolic Syndrome (MetS); (3) develop a Clinical Support Tool (CST) for the personalized management of BD and medical comorbidities.
Methods
The BIPCOM project aims to investigate MC, specifically MetS, in individuals with BD using various approaches. Initially, prevalence rates, characteristics, genetic and non-genetic risk factors, and the natural progression of MetS among individuals with BD will be assessed by analysing Nordic registers, biobanks, and existing patient datasets from 11 European recruiting centres across 5 countries. Subsequently, a clinical study involving 400 participants from these sites will be conducted to examine the clinical profiles and incidence of specific MetS risk factors over 1 year. Baseline assessments, 1-year follow-ups, biomarker analyses, and physical activity measurements with wearable biosensors, and focus groups will be performed. Using this comprehensive data, a CST will be developed to enhance the prevention, early detection, and personalized treatment of MC in BD, by incorporating clinical, biological, sex and genetic information. This protocol will highlight the study's methodology.
Discussion
BIPCOM's data collection will pave the way for tailored treatment and prevention approaches for individuals with BD. This approach has the potential to generate significant healthcare savings by preventing complications, hospitalizations, and emergency visits related to comorbidities and cardiovascular risks in BD. BIPCOM's data collection will enhance BD patient care through personalized strategies, resulting in improved quality of life and reduced costly interventions. The findings of the study will contribute to a better understanding of the relationship between medical comorbidities and BD, enabling accurate prediction and effective management of MetS and cardiovascular diseases.
Trial registration: ISRCTN68010602 at https://www.isrctn.com/ISRCTN68010602. Registration date: 18/04/2023.
Background
Bipolar Disorder (BD) is a severe, complex genetic disorder with a wide range of clinical presentations and a multifactorial etiology (Grande et al. 2016). It ranks as the 6th highest disease burden worldwide and contributes significantly to the global economic burden. BD is a critical public health concern due to its high prevalence (more than 1% of the world's population), chronic nature, high disability rate, and association with both psychiatric and Medical Comorbidities (MC) (Grande et al. 2016). Patients with BD have an elevated risk of cardiovascular disease (CVD) mortality, resulting in a reduced life expectancy of 10–20 years (Miller and Bauer 2014; Laursen 2011; Roshanaei-Moghaddam and Katon 2009), primarily due to a high level of MC in this population (Roshanaei-Moghaddam and Katon 2009), such as type 2 diabetes, CVD, musculoskeletal disorders, renal disorders, neurological disorders, and metabolic syndrome (MetS) (McIntyre et al. 2007). The heightened cardiovascular mortality in individuals with BD is possibly due to multiple factors. Medications like lithium and valproic acid may lead to weight gain and disrupt glucose metabolism, while second-generation antipsychotics are linked to hyperlipidaemia, insulin resistance, and weight gain (Weiner et al. 2011), nonetheless, despite these associations, long-term treatment with antipsychotic drugs has been correlated with lower mortality rates when compared to individuals who do not use antipsychotics (Tiihonen et al. 2009). Lifestyle factors such as poor diet, inadequate exercise, and higher rates of smoking further compound these risks. As a result, common cardiovascular risk factors like obesity, hypertension, diabetes, and hyperlipidemia are more prevalent among those with BD, amplifying their susceptibility to cardiovascular mortality (Weiner et al. 2011; Swartz and Fagiolini 2012). In addition, MC in BD are associated with unfavourable medical outcomes of this mental disorder such as treatment resistance, mood relapse or cognitive disorders (Bauer et al. 2018). Factors contributing to these higher MC rates in BD include pathogenic links between a variety of risk factors such as pathogenic and biological factors (e.g., metabolic, or immune system disorders, genetic susceptibility), medication exposure, unhealthy lifestyles, underutilization of health services, and co-occurring mental disorders (Keck and McElroy 2003; Carney and Jones 2006; Chengappa et al. 2000; Liese et al. 1998). Furthermore, the interplay of genetic susceptibility and underlying physiological processes, including heightened activation of immunometabolic or endocrine systems in both central and peripheral contexts, contributes to the development of both MetS and psychiatric disorders (Penninx and Lange 2018). MetS is a complex and multifaceted medical condition characterized by a cluster of interconnected risk factors that significantly increase the risk of heart disease, stroke, and type 2 diabetes. These risk factors often occur together and share common underlying mechanisms (Eckel et al. 2005). While there is no single universally accepted definition of MetS, several major health organizations and expert groups have agreed on three out of the five criteria listed in Table 1 required to fulfill a diagnosis of MetS (Alberti et al. 2009). The MetS is significantly more prevalent (by 1.98 times) in patients with BD than in the general population (Vancampfort et al. 2013; Godin et al. 2024; Godin et al. 2014). This high prevalence rate is plausibly explained by a range of multiple factors, including medication side effects, unhealthy lifestyles, and pathogenic and biological factors (Firth et al. 2020). For example, certain antipsychotic medications are associated with weight gain and insulin resistance, which in turn increase the risk of developing MetS. In addition, people with BD experience periodic mood fluctuations that affect their ability to maintain a healthy lifestyle (Penninx and Lange 2018). Physical activity (PA) and sleep patterns play important roles in MetS development. MetS is positively associated with physical inactivity and sedentary behaviour, while healthy sleep patterns benefit cardiometabolic health (Amirfaiz and Shahril 2019; Huang and Redline 2019). It is proven that patients with BD are typically less active, more sedentary, and experience poorer sleep quality (Janney et al. 2014; Brochard et al. 2018). Despite the high MC prevalence in BD, patients often receive inadequate diagnoses and treatment, and comprehensive studies on this topic remain limited, hindering personalized plans and preventive treatments. Neglecting comorbidities in BD management creates relevant obstacles to early and effective interventions for patients. For all these reasons, collaborative efforts between mental health and medical services are essential to enhance diagnosis and treatment, improve quality of life, and reduce comorbidity-related mortality in patients with BD.
Methods
Aims
The BIPCOM project aims to enhance personalized treatment for comorbid medical conditions, including MetS, in individuals with BD. The project's objectives include: (1) identifying MetS prevalence and risk factors in patients with BD using Nordic registries, biobanks, and medical records; (2) conducting an exploratory clinical study (ECS) with 400 subjects recruited from 11 centres to study the clinical profile and 1-year incidence of MetS; and (3) developing a Clinical Support Tool (CST) that utilizes clinical, biological, and genetic data to enhance personalized treatment for comorbidities, including MetS, with consideration for sex differences. MetS in patients with BD serves as a 'pilot case' due to its implications for prevention and precision medicine (PM). The project aims to pioneer PM approaches in BD, leading to innovative, individualized care models for implementation in clinical practice (Kupfer 2005): these efforts seek to improve healthcare delivery, patient stratification, disease prediction, and prevention, ultimately enhancing patient care and outcomes.
The general architecture of the project
The BIPCOM project will combine various methods and approaches. These include a register study, a survey of medical records, and the ECS incorporating the use of actigraphy and Experience Sampling Method (ESM). ESM, an ecologically valid method, provides a comprehensive view of daily life, assessing various constructs like quality of life and psychopathology and psychological mechanisms such as stress-sensitivity and coping. These constructs are challenging to obtain through traditional cross-sectional questionnaires. ESM finds applications in treatment monitoring, ecological momentary interventions, clinical trials, and single-case clinical assessments. Data collection can be simplified using a smartphone app (Verhagen et al. 2016). The project will also include a qualitative study with focus groups to assess the multiple dimensions of MC, seen from the ‘eyes’ of patients, clinicians, and family members.
This comprehensive approach will culminate in the development of a CST that will prove valuable to a diverse range of professionals. The BIPCOM project will last 3 years, but the clinical study will take place over a period of approximately 1 year. In the following sections, we will describe all the components of the project.
Study design
Study 1: the registry study
The consortium, drawing on extensive genetic epidemiology expertise, will leverage Swedish and Norwegian national registers and biobanks with (single-nucleotide polymorphism) SNP-array genotyped samples. It will begin by identifying adult Swedish residents from the Swedish National Patient Register and the Swedish Prescribed Drug Register, with registered treatment contact for BD. A similar process will be implemented with Norwegian registers. Subsequently, the consortium will investigate diagnosed MC, comparing individuals with and without BD. The primary focus of the Swedish and Norwegian teams is to explore the associations between BD onset and MC risk, with a specific emphasis on MetS. Additionally, it will include type 2 diabetes and CVD as relevant outcome variables due to their strong association with MetS. Research outcomes will be determined using a combination of diagnostic data from patient registers and pharmacological treatment details from the prescribed drug register, as shown in Table 2. Furthermore, risk prediction models for cardiometabolic conditions in individuals with BD will be developed, with CVD and type 2 diabetes as primary outcomes, and MetS, and all-cause and cause-specific mortality, as secondary outcomes. Given the absence of complete MetS diagnostic criteria in Swedish population registries, the consortium will use several related medical conditions as proxies (e.g., obesity, hypertension, hyper/dyslipidaemia). Access to genotyping data from approximately 1.3 million participants, derived from Illumina arrays or similar platforms in existing genotyped cohorts, will further enrich the study.
While leveraging national registers offers significant advantages, several limitations will have to be considered when interpreting findings from this study. First, the reliance on proxy conditions to represent MetS may introduce potential misclassification bias. The absence of complete diagnostic data in the registries limits the study's ability to accurately capture the true prevalence of MetS among individuals with BD. Additionally, individual proxies might not fully capture the complex interplay of factors contributing to MetS, potentially underestimating the true association between MetS and these outcomes in BD. Second, the study is inherently limited to information available in the registers, which may lack granularity around certain lifestyle factors, clinical metrics, and disease progression markers over time. Lastly, the utilization of specific-cause mortality outcomes should also be interpreted with caution due to potential discrepancies between recorded and actual causes of death (Brooke et al. 2017).
Study 2: the study of medical records
BIPCOM offers a unique opportunity to gather and analyse extensive datasets from patients with BD utilizing local medical services across five countries. All centres participating to this project use electronic patient records: an ad hoc Medical Record Abstraction Tool (MRAT) will be employed to extract clinical data and information related to 32 predefined variables (Additional file 1: Table S1), including past and current comorbidities and treatments, and a set of biomarkers related to MetS. Patients with BD will be randomly selected from those with inpatient or outpatient visits, aiming for a total sample size of at least 1,500 patients (300 per site). Rigorous measures will be taken to minimize selection bias and ensure data quality, following Jansen's guidelines (Jansen et al. 2005), which involve cross-verification with source documents, close collaboration with local physicians, written guidelines for data collectors, interobserver studies for training, regular data collector meetings, and random checks of completed MRATs against original medical records. Consent is not required as the data used are fully anonymized clinical routine data. MRAT will capture information on various medical comorbidities, including MetS, diabetes mellitus, dyslipidaemias, obesity, endocrine imbalances, CVD, other significant comorbidities, and concurrent prescribed drug use.
Study 3: the Exploratory Clinical Study (ECS)
Participants
Study 3 consists of the ECS conducted in five countries (France, Italy, Germany, Norway, and Spain), with a total of 11 recruiting sites (Additional file 1: Table S2). A total of 400 participants, aged 18–65 years, diagnosed with either BD I or BD II or BD not otherwise specified (NOS) according to the DSM-5, will be enrolled: these participants will be selected from patients with BD who have had at least one contact with the mental health service in the past year. Table 3 shows the inclusion and exclusion criteria. To ensure a representative sample of patients with BD, a specific stratification will be employed, according to the following schema: age will be split into two strata (18–45 and 46–65 years), sex will be split into two strata (male and female) and age of onset (before 21 years and after 21 years) will be used as a proxy of disorder severity.
The study foresees 2 evaluation times: the first at the entry into the study (T0) and the second after 1 year (T1).
Recruitment strategy
A random selection of eligible patients will be conducted within each stratum. Considering an anticipated refusal rate of approximately 30%, we will screen 130 eligible patients per country; the Spanish site will screen about 50 patients and will recruit not more than 30 patients. Trained researchers will conduct assessments using the Structured Clinical Interview for DSM-5 (SCID-5)—Clinical Version, specifically focusing on the BD section, to confirm the BD diagnoses.
Participant retention strategies in study 3
Due to the characteristics of the study population group (people with BD, possibly of working age), we have allowed for a 20% dropout rate. To maximise response and participation in the ECS, we propose a reimbursement of transport costs for everyone who decides to participate to the study, based on the availability of funds and the permissions of each ethics committee. Participants may be contacted using several methods of communication (post/phone/email); contact details of all participants, including general practitioner (GP) and clinician details, will be recorded in a ‘keeping in touch’ form and each contact, or contact attempt, made with participants will be recorded in a bespoke contact log. All participants will remain in the study and follow-up data will be sought unless consent for participation in data collection is explicitly withdrawn.
Clinical measures
A "Patient Schedule" (PS) will be created to capture socio-demographic, clinical, and treatment-related data of each participant at baseline (T0) and during the 1-year follow-up (T1). Table 4 provides detailed information about all assessment instruments to be administered to participants at T0.
Assessments will be conducted through participants' illness history and BD treatments. A structured physical examination, based on the Physical Examination Essential Checklist (Additional file 1), will be administered, with any signs of ill health promptly reported to clinicians.
During the physical examination, patients will complete the Knee extension test (Bohannon 1986) and the handgrip squeeze test (Sayer et al. 2006) measured through a dynamometer. Additionally, clinical electrocardiograms (ECGs) will be performed to identify QTc prolongation or the presence of arrhythmias. The primary aim of the ECS is to identify common associations among the five MetS diagnostic criteria in patients with BD at both T0 and T1 using clustering techniques and multiple correspondence analyses. Each stratum within each site will include approximately 14 subjects, totaling approximately 57 patients from the 400-sample pool, ensuring data representativeness and statistical power. The 1-year follow-up phase will begin immediately after the baseline assessment. At the end of this period, participants will undergo reassessment using a set of tools (Additional file 1: Table S3) to evaluate the clinical progression or improvement in MetS criteria, focusing on changes among patients meeting these criteria.
Biological measures
MetS is defined based on clinical and laboratory measurements (Alberti et al. 2009) and is strongly associated with metabolic liver disease, considered by some authors as an additional component of the syndrome (Kotronen et al. 2009; Kotronen and Yki-Järvinen 2008). There is a myriad of biomarkers associated with MetS that are either included in the pathway to its development and can help in its prediction or are potential mediators of its outcomes. They have been categorized into cardiometabolic, dyslipidemias, markers of oxidative stress, and inflammatory markers (Robberecht and Hermans 2016). Cardiometabolic biomarkers include insulin resistance, thyroid function, and cortisol levels, with the latter two being particularly relevant for patients with BD (Chakrabarti 2011; Dziurkowska and Wesolowski 2021). Adipokines, including adiponectin and leptin, are significantly associated with MetS, with its ratio being proposed as a biomarker of adipose tissue dysfunction (Lee and Shin 2020). Dyslipidemias involve elements additional to those required for MetS definition (Robberecht and Hermans 2016). Furthermore, oxidative stress markers like oxidized low-density lipoprotein (OxLDL) are linked to MetS and its consequences (Holvoet et al. 2008). Inflammation plays a crucial role in MetS, with markers such as high-sensitivity C-reactive protein (hsCRP), serum amyloid A, and cystatin C being associated (Robberecht and Hermans 2016; Magnusson et al. 2013; Ridker 2007). A systematic literature review (Srikanthan et al. 2016) has informed the selection of various biomarkers for MetS, including additional analytes to those mentioned so far: pro-inflammatory cytokines (interleukin-6—IL-6, Tumour Necrosis Factor-alpha—TNF-α), markers of pro-oxidant status (OxLDL, uric acid), and leptin, which are increased in MetS. There is limited available information on most of these biomarkers in this population, as well as their relationship with psychometric variables both cross-sectionally and longitudinally. During the ECS, patients will also be asked to undergo two blood and two saliva samples, one at T0 and one at 1-year follow-up (T1): 24 selected measurements and biomarkers, shown in Table 5, will be collected, shipped to a centralized laboratory, and assessed.
Digital measures
During the ECS, we will employ a digital tool to measure PA and sleep patterns. We will compare about 100 patients who meet (MetS +) or do not meet (MetS−) MetS diagnostic criteria at baseline (about 20 per country) with 100 psychologically healthy control subjects (about 20 per country), matched for both age (by 10-year age group) and sex. Healthy controls will be interviewed to obtain sociodemographic data and information on health status, specifically regarding psychological, cardiovascular, or metabolic diseases. Patients with BD and healthy controls will wear ActiGraph GT9X Link or ActiGraph GT3X devices three times during a year, each time lasting 1 week totalling 3 weeks over a year to assess total PA, intensity-specific PA, sedentary time, and circadian rhythms. In both groups, we will explore associations between PA and different variables, (such as sedentary time, sociodemographic characteristics, and clinical markers) using generalized linear mixed models (GLMM). These models will consider wear time, various activity levels, heart rate, inter-daily stability, intra-daily variability, and rhythm amplitude as factors. Additionally, PA and circadian rhythm variables will predict changes in MetS components between baseline and follow-up assessments. In the same week during the measurement via ActiGraph, the same subgroup of patients undergoing PA monitoring will also complete a dietary assessment using the MyFood24 web portal. Furthermore, the ECS includes an ESM study, prompting patients eight times daily via smartphone to report mood, stressors, eating behaviour, and other variables. This signal-contingent reporting will be complemented by event-contingent reports of eating episodes. This approach will ensure comprehensive data collection in natural settings, will avoid lengthy retrospective self-reports, and will capture participants' environmental, social, and psychological states.
Data analyses
Data quality management
Data quality management and storage are critical aspects of any biomedical study. These processes ensure that the data collected during the study are accurate, reliable, and securely stored for analysis and future reference. In order to guarantee appropriate data quality management, we will adopt the following steps:
-
(i)
Data collection protocols: we will develop well-defined data collection protocols to ensure consistency in data gathering. These protocols may include standardized procedures, guidelines, and training for data collectors to minimize errors and biases.
-
(ii)
Data validation: we will use validation procedures to check the integrity and accuracy of the data. This involves cross-verifying data entries, identifying outliers, and resolving any inconsistencies or missing values.
-
(iii)
Data cleaning: data cleaning involves the process of correcting errors, removing duplicates, and addressing missing or erroneous data points. This step is essential for ensuring the reliability of the final dataset.
-
(iv)
Quality control: regular quality control checks will be performed throughout the study to identify and rectify any potential issues with data collection or recording. These checks will help maintain data accuracy and completeness.
-
(v)
Data monitoring: continuous monitoring of the data collection process will allow study researchers to identify any deviations from the established protocols and take corrective actions promptly.
-
(vi)
Data auditing: an independent review or audit of the data can be performed to validate the accuracy and compliance of the data with study protocols and ethical guidelines.
Study 1: the registry study
For prediction modelling studies in Sweden, we will employ multivariate Cox proportional hazards regression analysis to develop 1-, 2- and 5-year risk prediction models separately for: CVD, and type 2 diabetes, as primary outcomes, as well as for proxy measures of MetS—hypertension, obesity and hyper/dyslipidaemia, and all-cause and cause-specific mortality as secondary outcomes, in individuals with BD. This will be done by combining traditional risk factors of cardiometabolic disorders (i.e., age, sex, a diagnosis/medication prescription for hypertension, diabetes mellitus (type 1 and type 2), hyperlipidaemia, obesity, and smoking; and family history), with potentially relevant novel predictors associated with both BD and an increased risk of cardiometabolic conditions (i.e., psychiatric comorbidity, use of psychotropic medication, and socio-demographic factors). We will use a limited backward stepwise procedure to determine whether to retain novel candidate predictors based on their p-values. Traditional risk factors will be kept in the model. With this approach we aim to retain predictors considered as traditional cardiometabolic risk factors based on clinical practice and previous studies, and to consider potentially relevant novel risk factors, with the final model being easily applicable by clinician, with satisfactory face validity. A bootstrapping method will be used for internal validation of the models with 200 bootstrap samples. To assess the discrimination of the models, the ROC curve and C-index will be used. To assess the calibration of the models we will use the Integrated Brier Score and calibration plots. We will also compare the performance of the models that includes only traditional risk factors with the models that include additional novel risk factors by calculating the Net Reclassification Index (NRI) and the Integrated Discrimination Improvement (IDI) index. We will also test the performance of the models across the following subgroups: males/females, and aged younger than 50/50 and older. Machine Learning (ML) algorithms for survival analysis will be tested on Swedish and Norwegian register data, considering a large number of available potential risk indicators such as demographics, electronic medical records, and genotypes. The overarching research question is to explore if ML algorithms based on many predictors may improve the predictive accuracy. For the ML algorithms, the data will be split into a 60% training set, 20% validation set, and 20% test set. Four models will be trained: penalized Cox regression, random survival forest, gradient boosting machine, and survival support vector machine. Hyperparameters will be tuned on the validation set. Model discrimination will be evaluated with C-index and cumulative/dynamic AUC. Calibration will be assessed with the Integrated Brier Score. Performance metrics on the test set will determine the final model. Sensitivity analyses will explore model accuracy when considering predictors related to patient history and dispensed medications only. We will also assess model performance separately in Sweden and Norway. Furthermore, we will merge register data from Norway and Sweden with biobank data (N ~ 1.3 million) to evaluate whether adding genotypes improves predictive power.
Study 2: the study of medical records
First, the anonymized data will be collected using a list of 32 predefined items (see Additional file 1: Table S1), some of which are similar to the variables used in the registry study. These data will then be aggregated and presented descriptively. In the second step, the data will be compared with analogous data from all centres included in the study to identify regional differences. Finally, the analysis will be made by comparing this data with corresponding samples gathered from the general population as well as the ongoing H2020 projects REALMENT (grant agreement No. 964874) and TIMESPAN (grant agreement No. 96538). The aim is to identify differences between the study population and the general population.
Study 3: the exploratory clinical study
Following the sign of informed consent, we will assess at baseline, the distribution of all possible combinations of the 5 MetS criteria (32 dispositions with repetition of two elements) among BD patients. A subject scoring positive at 3 criteria out of 5 will be defined as MetS + . At baseline, MetS + subjects will be compared to MetS—(less than 3 out of 5 criteria) for 15 predefined items, using either t-test for independent samples (continuous items) or Fisher Exact test (for binary items). Due to test multiplicity, individual tests p-value will be adjusted using Benjamini–Hochberg procedure.
The evolution in MetS status (MetS + /MetS−) from baseline to 1 year follow-up will be evaluated using multistate models, in order to identify what items might affect the transition probabilities. The 5 MetS criteria combinations will be explored using multiple correspondence analyses in order to identify possible clusters based on the predefined items. All the analyses will be performed using R (R Foundation, version 4.3.1 or newer). All tests will be two-sided.
Power and sample size
Study 1: the registry study
To determine the required sample size for fitting a Cox proportional hazards model, recent guidelines were followed (Riley et al. 2019). With an estimated cohort of 40,000 incident patients with BD and an expected 12,000 outcome events based on a 31% prevalence rate (Engin 2017) the minimum required sample size was estimated to be 4,199 individuals for a model with 24 candidate predictors. This assumed an average 2-year follow-up, a shrinking factor of 0.90, and an anticipated R2 of 0.05. For the ML models, sample size calculations based on statistical power are not directly applicable, since these techniques focus on optimizing predictive performance (i.e., reducing a loss function) rather than hypothesis testing. Instead, sample size and model complexity will be considered when training (e.g., learning curves plots by sample size and several parameters).
With a large cohort of 40,000 patients, our study is adequately powered to develop and validate robust predictive models for MetS in patients with BD using both standard survival analysis and ML techniques. This will allow the identification of modest predictive effects across a range of candidate variables and the construction of accurate predictive models for MetS incidence in this population (Engin 2017; Peduzzi et al. 1995).
Study 2: the study of medical records
Considering a confidence level of 95% and a standard error of 5%, the study would require the inclusion of at least 385 patients to provide a valid representation of the population under investigation. However, our study takes on an additional layer of complexity as we aim to make a cross-country comparison, analysing differences within the study population. To accomplish this and maintain robust statistical power, we need to consider a power level of 0.9 and an alpha level of 0.05. These choices are essential to ensure that our study is adequately powered to detect any meaningful differences between the countries being examined. Given these criteria and assuming an effect strength (η2) of 0.015, we should aim to include a minimum of 346 patient records from each site, thus totalling 1,730 patient records for the four countries comparative analysis (France, Germany, Italy, Norway, and Spain). This larger sample size is necessary to account for the population differences and to draw meaningful conclusions about the healthcare conditions and outcomes in each country.
Study 3: the exploratory clinical study
We estimated the sample size based on planned comparisons for 15 variables of interest (Table 6) for a two independent groups design (MetS + vs MetS−) assuming an expected proportion of Mets (at least three criteria out of five) of approximately 31%. With the estimated cohort of 400 subjects (180 Mets + , 220 Mets−), assuming a power of at least 80% and a significance level of 0.002 (to account for test multiplicity), we are going to be able to identify, at baseline, a significant effect size (ES) of 0.4 for independent t-tests (continuous variables) or an ES of 0.395 for a comparison of two independent proportions (assuming normal asymptotic test).
Discussion
The final output of the project: the clinical support tool
The project's primary outcome is the CST, a risk calculator designed to enhance prevention, diagnosis, and treatment of comorbidities in Patients with BD. The CST evaluates key factors such as BMI, PA, job type, diet, lifestyle choices, medical family history, stress events, and biomarkers. It will categorize patients into low, medium, high, or diagnosed comorbidity risk levels. The CST includes three components: an electronic risk assessment questionnaire for physicians and patients, a best practice paper-based recommendation prompt with comorbidity management guidance a risk graph, and a customized educational sheet outlining the patient's risk level. The beta version of the CST will undergo refinement with input from BIPCOM stakeholders. We will test it with 5–10 patients per country from the ECS who were not used in prediction model development. Initially, we will perform retrospective validation using ECS baseline data to assess MetS prediction. Subsequently, the beta CST will be evaluated by 2 clinicians per country for user interface, data access, and clinical applicability. Importantly, the CST will not suggest therapeutic decisions, which will remain the responsibility of treating clinicians. This novel tool addresses a crucial need for evaluating MetS risk in patients with BD and provides valuable insights for future clinical translation, addressing practical, ethical, and user-related considerations in PM in Psychiatry implementation.
Patient and public involvement: the qualitative study and focus groups
In the BIPCOM project our aim is to acquire current insights regarding the identification and treatment of medical co-occurring conditions in individuals with BD. This includes understanding the impact of these conditions, both addressed and unaddressed needs, and how healthcare services can assist in their management. We will conduct a qualitative study using Focus Groups (FGs) to achieve this. FGs will be conducted simultaneously in six European sites (Additional file 1: Table S4), with three groups per location. Each recruiting site will run three groups, each consisting of 6–7 participants. Of the three groups, one will include patients with BD, the second one BD clinicians, and the third one caregiver, resulting in a total of 18 sessions. Before starting the group sessions, research questions applicable to all groups will be defined. In each group there will be a moderator and an observer. Data analysis will follow the Framework Method (Gale et al. 2013), involving coding and the development of an analytical framework, according to recommended best practice standards in qualitative research using the Consolidated criteria for reporting qualitative research (COREQ) (Tong et al. 2007) We will use nVIVO software for analysis. The insights gathered from the FGs will be crucial to finalize the CST that offers suggestions to aid clinical decision-making in managing comorbidities in BD. The study results will have significant implications for enhancing prevention, early detection, and effective treatment of comorbidities in individuals with BD. A specific and more detailed study protocol regarding the qualitative research will be submitted separately.
Data transfer
All transfers of study data will be informed by and comply with the European Parliament and the Council of Europe’s Directive 95/46/EC (GDPR) on the protection of individuals concerning the handling of personal data and on the free flow of such information between EU countries. To ensure the security and integrity of data during such transfer, an appropriate documented standard procedure has been established, spelled out in a Data Management Plan, and will be followed without exception. Any study data that are to be transferred between research sites will be anonymized prior to transfer.
Data storage
All essential trial records will be securely stored by Istituto Centro San Giovanni di Dio, Fatebenefratelli (IRCCS FBF) (Brescia, Italy), and local sites, adhering to regulatory requirements and ensuring privacy compliance. Access will be limited to authorized personnel with role-based access permissions. Regular data backups and a disaster recovery plan will safeguard data integrity. Long-term archiving will ensure data accessibility for future research and replication.
Data access and quality assurance
At all recruiting sites, we adhere to local research ethics committee requirements and national/EU laws for collecting, recording, sharing, and securely storing person-identifiable data. Personal information about participants is kept confidential throughout the study and afterward. Data access is limited to authorized personnel and provided to relevant bodies for audit purposes only. After the conclusion of the BIPCOM project, analysis data sets will be made available in accordance with the IRCCS FBF Data Management Policy, ensuring compliance with legal, ethical, and funder requirements for data registration, storage, accessibility, and disposal.
Strength and limitations of the proposal
This project is novel and innovative in several key ways. BIPCOM adopts a comprehensive approach to studying MetS and other MC in BD, filling a gap in research that typically focuses on specific comorbidities. The BIPCOM project employs a multifaceted approach, combining register studies, ecological momentary assessments, medical record analysis, qualitative studies with FGs, digital measures, as well as extensive genetic epidemiology expertise, large-scale genetic data, and a broad array of other clinical and biological assessments. Case-register studies will uncover the prevalence, course characteristics, and mortality rates among patients with BD, enabling the development of prediction models. ML algorithms and predictive modeling will be employed to analyze the vast dataset, enhancing the understanding of the relationship between BD and MetS. Unlike many existing studies reliant on self-reporting questionnaires, BIPCOM will conduct direct, standardized medical assessments and passive activity monitoring using an actigraphy measurement. This approach will enhance the identification of emerging medical conditions, will clarify their progression, and will aid in identifying potential risk factors. The project includes screening specific biomarkers from biological samples, offering valuable insights into the underlying mechanisms of MC, coupled with data collected during the ECS. This comprehensive approach is designed to provide a holistic understanding of comorbidities in BD patients and to facilitate the development of effective prevention and treatment strategies.
Among the study limitations, we acknowledge the absence of more detailed assessments of target organs with the evaluation of echocardiography and ECG for the heart and ultrasound and elastography for the liver. However, properly conducting these assessments in a multicenter study presents several challenges, especially when faced with financial constraints. The primary concerns revolve around the need for standardization, quality control, and the associated costs; echocardiography is highly operator-dependent, and the accuracy and reliability of its results depend significantly on the skill and experience of the operator. Standardizing the exact protocols for performing and interpreting echocardiography across multiple centers (e.g. positioning of patients, measurement procedures, and criteria for abnormal findings) is challenging. Similarly, ultrasound/elastography assessment of the liver can be equipment and operator-dependent (Fang and Sidhu 2020). The costs of standardizing equipment, protocols, and training across multiple centers can be prohibitive, especially for studies with limited budgets, as BIPCOM, and this has prevented us from the adoption of these assessments.
Conclusions
BIPCOM aims to guide for enhancing healthcare for patients with BD, influencing clinical practices, decision-making, and regulatory frameworks for various stakeholders, including physicians, healthcare providers, and policymakers. The study will investigate system-based clinical and biological markers to monitor patients with BD conditions and identify potential risks and protective factors for comorbidities. Identifying clear risk factors enables early detection and timely interventions, leading to more effective and personalized care, improving overall patient health, and reducing healthcare costs. BIPCOM will also explore shared pathways for predicting adverse events and advancing targeted diagnostics and treatments. It will contribute to policy-making by informing health service administrators, politicians, professional groups, Non-governmental Organizations (NGOs), and business leaders, providing evidence for future funding and policy changes in line with emerging clinical insights. Additionally, BIPCOM will address social challenges, optimize clinical choices, enhance quality of life, reduce healthcare costs, and training young scientists and clinicians in personalized BD treatment approaches. In conclusion, the BIPCOM project is a commendable effort to address the complex issue of comorbidities in individuals with BD. By combining multidisciplinary research methods, cutting-edge technology, and the involvement of various stakeholders, it aims to pave the way for personalized, effective, and comprehensive healthcare for BD patients. The outcomes of this project will have the potential to not only improve the quality of life of those with BD, but also to inform future healthcare policies and practices in the field of psychiatry and broader healthcare in general.
Availability of data and materials
The datasets generated and/or analysed during the current study will be available in the Zenodo repository (see https://zenodo.org/).
Abbreviations
- ALT:
-
Alanine transaminase
- AST:
-
Aspartate transaminase
- BD:
-
Bipolar disorder
- BMI:
-
Body mass index
- CST:
-
Clinical support tool
- CVD:
-
Cardiovascular disease
- ECGs:
-
Electrocardiograms
- ECS:
-
Exploratory clinical study
- ECs:
-
Etichal committees
- EF:
-
Effect size
- ESM:
-
Experience sampling method
- FG:
-
Focus group
- GGT:
-
Gamma-glutamyl transferase
- HDL:
-
High-density lipoprotein
- HsCRP:
-
High-sensitivity C-reactive protein
- IDI:
-
Discrimination improvement index
- LDL:
-
Low-density lipoprotein
- LMM:
-
Linear mixed models
- MC:
-
Medical comorbidities
- MetS:
-
Metabolic syndrome
- ML:
-
Machine learning
- MRAT:
-
Medical record abstraction tool
- NRI:
-
Net reclassification index
- NGOs:
-
Non-governmental Organizations
- PA:
-
Physical activity
- PM:
-
Precision medicine
- PS:
-
Patient schedule
- QOL:
-
Quality of life
- SCID-5:
-
Structured clinical interview for DSM-5
- TGAAb/TPOAb:
-
Thyroid antibodies
- TNF-α:
-
Tumour necrosis factor alpha
- TSH:
-
Thyroid-stimulating hormone
References
Alberti KGMM, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, et al. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. 2009;120(16):1640–5.
Amirfaiz S, Shahril MR. Objectively measured physical activity, sedentary behavior, and metabolic syndrome in adults: systematic review of observational evidence. Metab Syndr Relat Disord. 2019;17(1):1–21.
Babor TF, Higgins-Biddle J, Saunders JB, Monteiro M. AUDIT-the alcohol use disorders identification test: guidelines for use in primary heath care. 2nd ed. Geneva: WHO; 2001.
Bauer M, Andreassen OA, Geddes JR, Vedel Kessing L, Lewitzka U, Schulze TG, et al. Areas of uncertainties and unmet needs in bipolar disorders: clinical and research perspectives. Lancet Psychiatry. 2018;5(11):930–9.
Bernstein DP, Fink L, Handelsman L, Foote J, Lovejoy M, Wenzel K, et al. Initial reliability and validity of a new retrospective measure of child abuse and neglect. Am J Psychiatry. 1994;151(8):1132–6.
Bohannon RW. Manual muscle test scores and dynamometer test scores of knee extension strength. Arch Phys Med Rehabil. 1986;67(6):390–2.
Brazier JE, Harper R, Jones NM, O’Cathain A, Thomas KJ, Usherwood T, et al. Validating the SF-36 health survey questionnaire: new outcome measure for primary care. BMJ. 1992;305(6846):160–4.
Brochard H, Godin O, Geoffroy PA, Yeim S, Boudebesse C, Benizri C, et al. Metabolic syndrome and actigraphy measures of sleep and circadian rhythms in bipolar disorders during remission. Acta Psychiatr Scand. 2018;138(2):155–62.
Brooke HL, Talbäck M, Hörnblad J, Johansson LA, Ludvigsson JF, Druid H, et al. The Swedish cause of death register. Eur J Epidemiol. 2017;32(9):765–73.
Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2):193–213.
Carney CP, Jones LE. Medical comorbidity in women and men with bipolar disorders: a population-based controlled study. Psychosom Med. 2006;68(5):684–91.
Chakrabarti S. Thyroid functions and bipolar affective disorder. J Thyroid Res. 2011;2011:306367.
Chengappa KN, Levine J, Gershon S, Kupfer DJ. Lifetime prevalence of substance or alcohol abuse and dependence among subjects with bipolar I and II disorders in a voluntary registry. Bipolar Disord. 2000;2(3 Pt 1):191–5.
Dziurkowska E, Wesolowski M. Cortisol as a biomarker of mental disorder severity. J Clin Med. 2021;10(21):5204. https://doi.org/10.3390/jcm10215204.
Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet. 2005;365(9468):1415–28.
Engin A. The definition and prevalence of obesity and metabolic syndrome. Adv Exp Med Biol. 2017;960:1–17.
Fabbian F, De Giorgi A, Maietti E, Gallerani M, Pala M, Cappadona R, et al. A modified Elixhauser score for predicting in-hospital mortality in internal medicine admissions. Eur J Intern Med. 2017;40:37–42.
Fang C, Sidhu PS. Ultrasound-based liver elastography: current results and future perspectives. Abdom Radiol (NY). 2020;45(11):3463–72.
First MB. Structured clinical interview for the DSM (SCID). In: Cautin RL, Lilienfeld SO, editors. The encyclopedia of clinical psychology. Hoboken: Wiley; 2014.
Firth J, Solmi M, Wootton RE, Vancampfort D, Schuch FB, Hoare E, et al. A meta-review of “lifestyle psychiatry”: the role of exercise, smoking, diet and sleep in the prevention and treatment of mental disorders. World Psychiatry. 2020;19(3):360–80.
Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12(3):189–98.
Gale NK, Heath G, Cameron E, Rashid S, Redwood S. Using the framework method for the analysis of qualitative data in multi-disciplinary health research. BMC Med Res Methodol. 2013;18(13):117–117.
Godin O, Etain B, Henry C, Bougerol T, Courtet P, Mayliss L, et al. Metabolic syndrome in a French cohort of patients with bipolar disorder: results from the FACE-BD cohort. J Clin Psychiatry. 2014;75(10):1078–85.
Godin O, Olié E, Fond G, Aouizerate B, Aubin V, Bellivier F, et al. Incidence and predictors of metabolic syndrome onset in individuals with bipolar disorders: a longitudinal study from the FACE-BD cohort. Acta Psychiatr Scand. 2024;149(3):207–18.
Grande I, Berk M, Birmaher B, Vieta E. Bipolar disorder. Lancet. 2016;387(10027):1561–72.
Heatherton TF, Kozlowski LT, Frecker RC, Fagerström KO. The Fagerström test for nicotine dependence: a revision of the Fagerström tolerance questionnaire. Br J Addict. 1991;86(9):1119–27.
Holvoet P, Lee D, Steffes M, Gross M, Jacobs DRJ. Association between circulating oxidized low-density lipoprotein and incidence of the metabolic syndrome. JAMA. 2008;299(19):2287–93.
Huang T, Redline S. Cross-sectional and prospective associations of actigraphy-assessed sleep regularity with metabolic abnormalities: the multi-ethnic study of atherosclerosis. Diabetes Care. 2019;42(8):1422–9.
Janney CA, Fagiolini A, Swartz HA, Jakicic JM, Holleman RG, Richardson CR. Are adults with bipolar disorder active? Objectively measured physical activity and sedentary behavior using accelerometry. J Affect Disord. 2014;152–154:498–504.
Jansen ACM, van Aalst-Cohen ES, Hutten BA, Büller HR, Kastelein JJP, Prins MH. Guidelines were developed for data collection from medical records for use in retrospective analyses. J Clin Epidemiol. 2005;58(3):269–74.
Keck PE, McElroy SL. Bipolar disorder, obesity, and pharmacotherapy-associated weight gain. J Clin Psychiatry. 2003;64(12):1426–35.
Kotronen A, Yki-Järvinen H. Fatty liver: a novel component of the metabolic syndrome. Arterioscler Thromb Vasc Biol. 2008;28(1):27–38.
Kotronen A, Peltonen M, Hakkarainen A, Sevastianova K, Bergholm R, Johansson LM, et al. Prediction of non-alcoholic fatty liver disease and liver fat using metabolic and genetic factors. Gastroenterology. 2009;137(3):865–72.
Kupfer DJ. The increasing medical burden in bipolar disorder. JAMA. 2005;293(20):2528–30.
Laursen TM. Life expectancy among persons with schizophrenia or bipolar affective disorder. Schizophr Res. 2011;131(1–3):101–4.
Lee KW, Shin D. Prospective Associations of serum adiponectin, leptin, and leptin-adiponectin ratio with incidence of metabolic syndrome: The Korean Genome and Epidemiology Study. Int J Environ Res Public Health. 2020;17(9):3287. https://doi.org/10.3390/ijerph17093287.
Liese AD, Mayer-Davis EJ, Haffner SM. Development of the multiple metabolic syndrome: an epidemiologic perspective. Epidemiol Rev. 1998;20(2):157–72.
Magnusson M, Hedblad B, Engström G, Persson M, Nilsson P, Melander O. High levels of cystatin C predict the metabolic syndrome: the prospective Malmö Diet and Cancer Study. J Intern Med. 2013;274(2):192–9.
McIntyre RS, Soczynska JK, Beyer JL, Woldeyohannes HO, Law CWY, Miranda A, et al. Medical comorbidity in bipolar disorder: re-prioritizing unmet needs. Curr Opin Psychiatry. 2007;20(4):406–16.
Miller C, Bauer MS. Excess mortality in bipolar disorders. Curr Psychiatry Rep. 2014;16(11):499.
Peduzzi P, Concato J, Feinstein AR, Holford TR. Importance of events per independent variable in proportional hazards regression analysis. II. Accuracy and precision of regression estimates. J Clin Epidemiol. 1995;48(12):1503–10.
Penninx BWJH, Lange SMM. Metabolic syndrome in psychiatric patients: overview, mechanisms, and implications. Dialogues Clin Neurosci. 2018;20(1):63–73.
Rabin R, de Charro F. EQ-5D: a measure of health status from the EuroQol Group. Ann Med. 2001;33(5):337–43.
Ridker PM. C-reactive protein and the prediction of cardiovascular events among those at intermediate risk: moving an inflammatory hypothesis toward consensus. J Am Coll Cardiol. 2007;49(21):2129–38.
Riley RD, Snell KI, Ensor J, Burke DL, Harrell FEJ, Moons KG, et al. Minimum sample size for developing a multivariable prediction model: part II - binary and time-to-event outcomes. Stat Med. 2019;38(7):1276–96.
Robberecht H, Hermans N. Biomarkers of metabolic syndrome: biochemical background and clinical significance. Metab Syndr Relat Disord. 2016;14(2):47–93.
Rosa AR, Sánchez-Moreno J, Martínez-Aran A, Salamero M, Torrent C, Reinares M, et al. Validity and reliability of the Functioning Assessment Short Test (FAST) in bipolar disorder. Clin Pract Epidemiol Ment Health. 2007;7(3):5–5.
Roshanaei-Moghaddam B, Katon W. Premature mortality from general medical illnesses among persons with bipolar disorder: a review. Psychiatr Serv. 2009;60(2):147–56.
Sayer AA, Syddall HE, Martin HJ, Dennison EM, Roberts HC, Cooper C. Is grip strength associated with health-related quality of life? Findings from the Hertfordshire Cohort Study. Age Ageing. 2006;35(4):409–15.
Skinner HA. The drug abuse screening test. Addict Behav. 1982;7(4):363–71.
Srikanthan K, Feyh A, Visweshwar H, Shapiro JI, Sodhi K. Systematic review of metabolic syndrome biomarkers: a panel for early detection, management, and risk stratification in the West Virginian population. Int J Med Sci. 2016;13(1):25–38.
Swartz HA, Fagiolini A. Cardiovascular disease and bipolar disorder: risk and clinical implications. J Clin Psychiatry. 2012;73(12):1563–5.
Tiihonen J, Lönnqvist J, Wahlbeck K, Klaukka T, Niskanen L, Tanskanen A, et al. 11-year follow-up of mortality in patients with schizophrenia: a population-based cohort study (FIN11 study). Lancet. 2009;374(9690):620–7.
Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. Int J Qual Health Care. 2007;19(6):349–57.
Ustün TB, Chatterji S, Kostanjsek N, Rehm J, Kennedy C, Epping-Jordan J, et al. Developing the World Health Organization Disability Assessment Schedule 2.0. Bull World Health Organ. 2010;88(11):815–23.
Vancampfort D, Vansteelandt K, Correll CU, Mitchell AJ, De Herdt A, Sienaert P, et al. Metabolic syndrome and metabolic abnormalities in bipolar disorder: a meta-analysis of prevalence rates and moderators. Am J Psychiatry. 2013;170(3):265–74.
Verhagen SJW, Hasmi L, Drukker M, van Os J, Delespaul PAEG. Use of the experience sampling method in the context of clinical trials. Evid Based Ment Health. 2016;19(3):86–9.
Weiner M, Warren L, Fiedorowicz JG. Cardiovascular morbidity and mortality in bipolar disorder. Ann Clin Psychiatry. 2011;23(1):40–7.
Acknowledgements
The BIPCOM consortium also includes, in alphabetical order: Maximilian Bayas (Frankfurt, Germany), Frank Bellivier (Paris, France), Narcís Cardoner Álvarez (Barcelona, Spain), Pietro Carmellini (Siena, Italy), Federico Cevoli (Créteil, France), Julia Clemens (Dresden, Germany), Philippe Courtet (Montpellier, France), Lorena Di Consoli (Milan, Italy), Giuseppe Delvecchio (Milan, Italy), Maja Dobrosavljevic (Örebro, Sweden), Bruno Etain (Paris, France), Hendrik Friedrichsen (Frankfurt, Germany), Adrienne Kelemen (Dresden, Germany), Despoina Koukouna (Siena, Italy), Eugenia Mato (Barcelona, Spain), Dídac Mauricio (Barcelona, Spain), Romina Miranda-Olivos (Barcelona, Spain), Lisa Möbius (Dresden, Germany), Chiara Moltrasio (Milano, Italy), Caroline Mohn-Haugen (Oslo, Norway), Isabelle Nuss (Frankfurt, Germany), Emilie Olie (Montpellier, France), Agnes Pelletier (Créteil, France), Zillur Rahman (Oslo, Norway), Davide Rampi (Cremona, Italy), Jonathan Repple (Frankfurt, Germany), Eugenia Resmini (Barcelona, Spain), Julia Schneider (Dresden, Germany), Elena Toffol (Treviso, Italy).
Furthermore, the Consortium acknowledges the invaluable help of Cristina Kullmann, Dr.Scie, Fabio Pareto, Dr.Scie, and Cosimo Ottomano, M.D. (Synlab, Monza-Brescia, Italy) for the planning and realization of all steps related to laboratory analyses. Acknowledgments are due to Renzo Rozzini, M.D. and to Massimo Barbieri, M.D. (Fondazione Poliambulanza, Brescia, Italy) for their active contribution to the medical examination of Italian patients. We also thank Giovanni Aresi, PhD (Catholic University of Sacred Heart, Milan) for his contribution and supervision in the design of the qualitative research.
Funding
Funded by the Joint Transnational Call for Proposals (JTC) 2022 on "Prevention in Personalised Medicine " within the framework of ERA PerMed, ERAPERMED2022-087—BIPCOM. This project was supported by: 1. Fondazione Regionale per la Ricerca Biomedica (Italy); 2. Bundesministerium für Bildung und Forschung (Germany); 3. VINNOVA (Sweden); 4. Norges Forskningsråd (Norway); 5. Agence Nationale de la Recherche (France); 6. Departament de Salut, Generalitat de Catalunya (Spain); 7. Sächsisches Staatsministerium für Wissenschaft und Kunst (Germany); under the frame of ERA PerMed.
Author information
Authors and Affiliations
Consortia
Contributions
Conceptualization: G.d.G., E.C, M.M, M.L, O.A, M.G.A, H. L, M.B, P.R, R.C, F. K., A.R. Statistical Analysis: S.C, N.K, H.L., M.G.A. Writing—Original Draft Preparation: E.C, G.d.G, M.M. Writing—Review & Editing: all co-authors have reviewed several drafts of this study protocol and have approved the current version.
Corresponding author
Ethics declarations
Ethics approval and consent to participate
This study was approved by the Ethics Committees (ECs) of the coordinating center (Comitato Etico IRCCS San Giovanni di Dio—FBF, approval no. 36-2023) and of the Frankfurt site (Ethical Committee of the Department of Medicine at Goethe University Frankfurt am Main, approval no. 2023-1397). Requests of permission to the ECs of the other participating sites have been submitted and are currently assessed.
Consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Additional file 1: Table S1.
Socio-demographic, clinical characteristics, and biomarkers to be used for the study of medical records. Table S2. List of participating and recruiting centres. Table S3. List of assessment tools at follow-up. Table S4. List of participating and recruiting centres in qualitative research.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
de Girolamo, G., Andreassen, O.A., Bauer, M. et al. Medical comorbidities in bipolar disorder (BIPCOM): clinical validation of risk factors and biomarkers to improve prevention and treatment. Study protocol. Int J Bipolar Disord 12, 15 (2024). https://doi.org/10.1186/s40345-024-00337-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s40345-024-00337-8