Multiple Indicators Multiple Causes Research Articles

Conducting cross-cultural, multi-lingual or multi-country scale development and validation in health care research: A 10-step framework based on a scoping review.

Valid, reliable and cross-cultural equivalent scales and measurement instruments that enable comparisons across diverse populations in different countries are important for global health research and practice. We developed a 10-step framework through a scoping review of the common strategies and techniques used for scale development and validation in a cross-cultural, multi-lingual, or multi-country setting, especially in health care research. We searched MEDLINE, Embase, and PsycINFO for peer-reviewed studies that collected data from two or more countries or in two or more languages at any stages of scale development or validation and published between 2010-22. We categorised the techniques into three commonly used scale development and validation stages (item generation, scale development, and scale evaluation) as well as during the translation stage. We described the most commonly used techniques at each stage. We identified 141 studies that were included in the analysis. We summarised 14 common techniques and strategies, including focus groups or interviews with diverse target populations, and involvement of measurement experts and linguists for item content validity expert panel at the item generation stage; back-and-forth translation, collaborative team approach for the translation stage; cognitive interviews and different recruitment strategies and incentives in different settings for scale development stage; and three approaches for measurement invariance (multigroup confirmatory factor analysis, differential item functioning and multiple indicator multiple causes) for scale evaluation stage. We provided a 10-step framework for cross-cultural, multi-lingual or multi-country scale development and validation based on these techniques and strategies. More research and synthesis are needed to make scale development more culturally competent and enable scale application to better meet local health and development needs.

Identifying neurofunctional domains across substance use disorders

ABSTRACT Background: Substance use disorders (SUDs) are heterogeneous across multiple functional domains. Various frameworks posit that domains (e.g., executive function) contribute to the persistence of SUDs; however, the domains identified in different studies vary. Objectives: We used factor analysis to identify the underlying latent domains present in a large sample (N = 5,244, 55.8% male) with a variety of SUDs to yield findings more generalizable than studies with a narrower focus. Method: Participants (1,384 controls and 3,860 participants with one or more SUDs including alcohol, cocaine, cannabis, and/or opioid use disorders) completed the Semi-Structured Assessment for Drug Dependence and Alcoholism, the NEO Personality Inventory, and the Wisconsin Card Sorting Test. Exploratory factor analysis (EFA) and fit indices (root mean-squared error of approximation (RMSEA), Comparative Fit Index (CFI), and Tucker-Lewis Index (TLI)) were used to examine different latent variable models. A multiple indicators, multiple causes (MIMIC) approach-tested associations of the latent variables with sociodemographics, substance use, and a history of abuse/neglect. Results: A six-factor model (predominant alcohol, predominant cocaine, predominant opioid, externalizing, personality, and executive function) provided the best fit [RMSEA = 0.063 (90% CI 0.060, 0.066), CFI = 0.98, TLI = 0.96]. All factors were moderately correlated (coefficient = 0.25–0.55, p < .05) with the exception of executive function. MIMIC analysis revealed different patterns of associations (all p < .0001) with sociodemographics, substance use, and a history of abuse/neglect among the factors. Conclusions: The domains identified, particularly executive function, were parallel to those observed previously. These factors underscore the heterogeneous nature of SUDs and may be useful in developing more targeted clinical interventions.

A Bayesian Moderated Nonlinear Factor Analysis Approach for DIF Detection under Violation of the Equal Variance Assumption

AbstractResearch has shown that multiple‐indicator multiple‐cause (MIMIC) models can result in inflated Type I error rates in detecting differential item functioning (DIF) when the assumption of equal latent variance is violated. This study explains how the violation of the equal variance assumption adversely impacts the detection of nonuniform DIF and how it can be addressed through moderated nonlinear factor analysis (MNLFA) model via Bayesian estimation approach to overcome limitations from the restrictive assumption. The Bayesian MNLFA approach suggested in this study better control Type I errors by freely estimating latent factor variances across different groups. Our experimentation with simulated data demonstrates that the BMNFA models outperform the existing MIMIC models, in terms of Type I error control as well as parameter recovery. The results suggest that the MNLFA models have the potential to be a superior choice to the existing MIMIC models, especially in situations where the assumption of equal latent variance assumption is not likely to hold.

Examining individuals’ adoption of flex route transit

ABSTRACT This study aims to comprehensively identify the contributing factors affecting the adoption of flex route transit. A survey was conducted in 2019, and a multiple indicators multiple causes (MIMIC) model was utilized to capture individuals’ preferences toward car, fixed route, and flex route transit. Subsequently, the multinomial logit model with attitudinal features is employed to identify the key factors influencing the shift from car or fixed route to flex route transit. Results indicate that several factors contribute to flex route transit adoption, with the top three being car restriction, travel cost, and age groups. Car restriction is positively associated with flex route transit adoption (marginal effects 11.67%). Travel costs (≥1.49 US$) and elderly groups also contribute to an increase in flex route transit adoption, with marginal effects of 9.18% and 8.01%, respectively. These findings provide insightful implications for decision-makers in identifying potential adopters and improving flex route transit adoption.

Psychometric characteristics of the career adapt-abilities scale in Thai undergraduate students: a multiple indicators multiple causes model.

This study aimed to evaluate the psychometric properties of the Career Adapt-abilities Scale (CAAS) and analyze the relationships between sociodemographic variables and career adaptability using the MIMIC model with DIF. In this cross-sectional study, the CAAS, consisting of four sub-scales-concern, control, curiosity, and confidence-each comprising six items, was administered to 620 undergraduate students from 34 universities in Thailand. Among the participants, women constituted 66.77%, while men accounted for 32.58% (Mean Age = 20.33). To explore the influence of sociodemographic characteristics on specific CAAS item responses, a Multiple-Indicators, Multiple-Causes model with Differential Item Functioning (MIMIC-model with DIF) was employed. The findings revealed robust internal consistency and reliability in the CAAS. Both the four-factor and second-order factor models exhibited excellent fit indices, emphasizing the significance of each item within the measure. Sociodemographic characteristics did not demonstrate a significant direct effect on the four CAAS subscales, it is worth noting the exception regarding paternal education. Paternal education was found to have a statistical significance impact on two specific CAAS items: "looking for opportunities to grow as a person" and "becoming curious about new opportunities." These findings underscore the intricate influence of paternal education on specific aspects of career adaptability, suggesting that certain dimensions of career adaptability may be shaped by factors extending beyond the conventional sociodemographic variables examined in this study. Future research could delve deeper into the complex interplay of sociodemographic factors and individual attributes to provide a more holistic perspective on career adaptability in Thailand.

Confirmatory factor analysis of the Hyperacusis Impact Questionnaire, Sound Sensitivity Symptoms Questionnaire, and Screening for Anxiety and Depression in Tinnitus, including preliminary analyses of the parent versions for use with children.

We previously reported the results of exploratory factor analysis (EFA) of the Hyperacusis Impact Questionnaire (HIQ), the Sound Sensitivity Symptoms Questionnaire (SSSD) and the Screening for Anxiety and Depression in Tinnitus (SAD-T). Confirmatory Factor Analysis (CFA) is necessary to confirm the latent constructs determined using EFA. CFA should use different samples but with similar characteristics to those used for EFA. The aim was to use CFA to confirm latent constructs derived using EFA of the HIQ, SSSQ and SAD-T. We further evaluated the psychometric properties of parent versions of these questionnaires (indicated by -P), which are intended for use with children. This was a retrospective cross-sectional study. Data for 323 consecutive adults and 49 children who attended a Tinnitus and Hyperacusis Therapy Clinic in the UK within a six-month period were included. Data were collected retrospectively from the records of patients held at the audiology department. CFA with the weighted least-squares mean and variance-adjusted estimator was applied to assess the previously proposed factor structures of the HIQ, SSSQ and SAD-T. The internal consistency of the scales was assessed via Cronbach's alpha (α). The items of the HIQ, SSSQ and SAD-T were tested for measurement invariance regarding age and gender using the multiple indicator multiple cause (MIMIC) model. All questionnaires showed good to excellent internal consistency, with α = 0.93 for the HIQ, 0.87 for the SSSQ, and 0.91 for the SAD-T. The parent versions showed acceptable to good internal consistency, with α = 0.88 for the HIQ-P, 0.71 for the SSSQ-P, and 0.86 for the SAD-T-P. CFA showed that the HIQ, SSSQ, and SAD-T were all one-factor questionnaires and the factors generally were similar to those obtained for the EFA. The MIMIC model showed that all three questionnaires can be considered as measurement invariant, with scores similar across genders and ages. The HIQ, SSSQ and SAD-T are internally consistent one-factor questionnaires that can be used in clinical and research settings to assess the impact of hyperacusis, the severity of sound sensitivity symptoms, and to screen for anxiety and depression symptoms. Future studies should further explore the psychometric properties of the parent versions of the HIQ and SSSQ and SAD-T.

Determinants of residents’ support for urban tourism in times of uncertainty: Exploring the case of the city of Madrid

Residents are key actors in the management of tourist destinations, especially in times of uncertainty such as those that characterize the current context of multidimensional crisis (ecological, political, social or economic). The role of tourism in these crises, its contribution to their acceleration or its work in recovery, makes it necessary to deepen the support of residents for tourism in complex times such as the present. The context generated by the COVID-19 health crisis is a good framework for this reflection, although it has not been sufficiently explored. This article draws on Social Exchange Theory to explain residents' support for tourism in urban destinations during periods of uncertainty such as those generated by the health crisis, focusing on factors that have so far been little explored: risk perception, community involvement and the management of tourism constraints. A structural multiple indicator multiple cause (MIMIC) model is proposed to test the theoretical model. Using data derived from a survey of 342 residents in the city of Madrid (Spain), the results show that, in addition to benefits as a predictor of residents' support for tourism in contexts of uncertainty, risk perception and the management of tourism constraints are also relevant determinants of this support. The findings have important implications for decision-makers and public managers of urban destinations in the current period of multiple crises.

Cultural adaptation to Bolivian Quechua and psychometric analysis of the Patient Health Questionnaire PHQ-9

ObjectiveCultural adaptation of the Patient Health Questionnaire-PHQ-9 to Bolivian Quechua and analysis of the internal structure validity, reliability, and measurement invariance by sociodemographic variables.MethodsThe PHQ-9 was translated and back-translated (English-Quechua-English) to optimise translation. For the cultural adaptation, experts, and people from the target population (e.g., in focus groups) verified the suitability of the translated PHQ-9. For the psychometric analysis, we performed a Confirmatory Factor Analysis (CFA) to evaluate internal validity, calculated α and ω indices to assess reliability, and performed a Multiple Indicator, Multiple Cause (MIMIC) model for evaluating measurement invariance by sex, age, marital status, educational level and residence. We used standard goodness-of-fit indices to interpret both CFA results.ResultsThe experts and focus groups improved the translated PHQ-9, making it clear and culturally equivalent. For the psychometric analysis, we included data from 397 participants, from which 73.3% were female, 33.0% were 18–30 years old, 56.7% reported primary school studies, 63.2% were single, and 62.0% resided in urban areas. In the CFA, the single-factor model showed adequate fit (Comparative Fit Index = 0.983; Tucker-Lewis Index = 0.977; Standardized Root Mean Squared Residual = 0.046; Root Mean Squared Error of Approximation = 0.069), while the reliability was optimal (α = 0.869—0.877; ω = 0.874—0.885). The invariance was confirmed across all sociodemographic variables (Change in Comparative Fit Index (delta) or Root Mean Square Error of Approximation (delta) < 0.01).ConclusionsThe PHQ-9 adapted to Bolivian Quechua offers a valid, reliable and invariant unidimensional measurement across groups by sex, age, marital status, educational level and residence.

Measurement equivalence of the English and French versions of the self-efficacy to manage chronic disease scale: a Scleroderma Patient-Centered Intervention Network (SPIN) study.

The Self-Efficacy to Manage Chronic Disease (SEMCD) scale is widely used, including in systemic sclerosis (SSc). The SEMCD has been validated in SSc, but the metric equivalence of the English and French versions has not been assessed (i.e., whether psychometric properties are equivalent across English and French). Participants were adults from the Scleroderma Patient-Centered Intervention Network (SPIN) Cohort (N = 2159) who completed baseline measures in English (n = 1473) or French (n = 686) between May 2014 to July 2020. Analyses assessed internal consistency reliability via Cronbach's alpha and McDonald's omega, convergent validity via Pearson's correlations, structural validity via confirmatory factor analysis (CFA), and differential item functioning via the Multiple-Indicator Multiple-Cause (MIMIC) model. Internal consistency reliability was high in English (α = .93, ω = .93) and French (α = .92, ω = .93). All correlations between the SEMCD and measures of health outcomes were moderate to large, statistically significant, and in the hypothesized direction in both languages. The CFA demonstrated that the one-factor model of self-efficacy, overall, fit reasonably well (CFI = .96, TLI = .93, SRMR = .03, RMSEA = .14). Standardized factor loadings were large (.76 to .88). Three items displayed statistically significant uniform DIF and all six displayed nonuniform DIF; all DIF was of minimal magnitude. Comparison of unadjusted and DIF-adjusted models indicated that DIF did not meaningfully impact total score (ICC = 0.999, r = 0.999). Scores from English- and French-speaking adults with SSc can be combined for analysis or compared.

Driving Stress Partial Model Considering Time Domain Applying for Two Wheelers Drivers: A Case of Primary Arterial Roads

The physiological response of motorcyclists, which is substantially correlated with driving stress, can be utilised to assess traffic safety. Heart rate and respiratory rate can be used to measure driver tension using the heart rate variability method. This investigation was conducted in Aceh province along the urban arterial roads Jl. T. Nyak Arief and Jl. T. Muhammad Daud Bereueh. This study seeks to provide the anticipated empirical findings for determining the stress threshold of motorcyclists on Banda Aceh's primary arterial roads. The Polar Vantage V2 Smartwatch Multisport measures the pulse rate and respiratory rate of voluntarily participating motorcycle riders. This research was conducted using direct observation and direct observation related to volunteer travel activities as a sample, with volunteers selected at random and questioned about their willingness to participate. Direct interview surveys with volunteers were used to collect socioeconomic information about volunteers. The Multiple Indicators Multiple Causes (MIMIC) model empirically examined which road conditions and land use variables influence driving tension among motorists. The results revealed that the highest heart rate (HR) was recorded in the commercial land use area, at 94.029 beats per minute, while the highest HR was recorded at road bends, at 94.244 beats per minute. The time domain stress model is influenced by land use variables, with the most significant types of land use being offices and education, with the forming variables consisting of SDNN, RMSSD, and TINN.

Direct effect of personality traits and work engagement on job crafting: A structural model

Research in job crafting is relevant to creating healthier organizations and is attracting increasing attention. This study contributes to identifying the predictors of job crafting from a new conceptualization that considers three dimensions, i.e., job crafting towards strength, job crafting towards interests, and job crafting towards development. On this basis, the present study aims to explore two predictors of job crafting in the personal domain, the Big Five personality traits and Work Engagement. The study sample consists of 327 Spanish employees who completed standardized questionnaires. A Multiple Indicator Multiple Cause (MIMIC) like Structural Equational Model (SEM) was applied to the data. Results showed that job crafting towards strength was significantly and positively predicted by Work Engagement (dedication and absorption) as well as extraversion, agreeableness, conscientiousness, and intellect/imagination. Furthermore, job crafting towards interest was predicted by Work Engagement (absorption), extraversion, and intellect/imagination. Last, job crafting towards development was predicted by Work Engagement (dedication and absorption), agreeableness, conscientiousness, and intellect/imagination. Vigor and neuroticism were unrelated to job-crafting dimensions. These findings can be useful for managers and organizations looking to facilitate job crafting as a means of promoting employees' performance as well as satisfaction and well-being.

Validation of the Modified Weight Bias Internalization Scale (WBIS-M) among First-Generation Asian Immigrants.

Internalized weight bias (i.e., the application of negative weight-based stereotypes to one's self-evaluation) is associated with adverse physical and mental health outcomes. The 10-item version of the Modified Weight Bias Internalization Scale (WBIS-M) is one of the most commonly used measures to assess internalized weight bias. However, studies examining the psychometric properties of the WBIS-M are limited in racially minoritized U.S.-based populations. The current study sought to examine the factor structure of the 10-item version of the WBIS-M in a sample of first-generation Asian immigrants, as both native and foreign-born Asian populations have consistently demonstrated higher rates of weight-related concerns relative to other racially minoritized groups. Confirmatory factor analysis was used to assess the factor structure, and a multiple indicator multiple cause modeling approach was used to assess associations between internalized weight bias and sociodemographic predictors. Results confirm the unidimensionality of the WBIS-M and provide preliminary support for a nine-item version. Identifying as a woman (p = .027) and perceiving oneself as overweight (p < .001) were significantly associated with greater internalized weight bias. When working with Asian and Asian American clients, mental health service providers should be aware of potential weight-related concerns and may consider using the WBIS-M to assess for internalized weight bias.

Customers’ behavioral intentions when visiting upscale restaurants: Enjoying the experience or posturing?

High-level gastronomy is booming and, in many cases, has become the main purpose for travel. This research aims to investigate the relationships between the performance of high-level restaurants, the affective components they generate, and how they can affect customer behavior both in regard to the restaurant and the destination in which it is located. The empirical work was analyzed by means of a multiple indicators multiple causes model (MIMIC) and structural equation modeling (SEM). The survey was administered to customers who visited Michelin-star restaurants. The main results indicate that enjoyment of the experience and the status it provides are important factors motivating consumers to visit these types of restaurants and that their behavioral intentions can influence their family, friends, acquaintances, and destination. Therefore, it is suggested that restaurant owners and destination managers develop joint strategies beneficial for the gastronomic tourism sector.

Uncovering Differential Item Functioning effects using MIMIC and mediated MIMIC models.

The aim of this study was twofold: first, to examine the presence of bias across gender in a scholastic achievement test named the Academic Achievement Test (AAT) for the Science Track. Second, to understand the underlying mechanism that causes these bias effects by examining the effect of general cognitive ability as a mediator. The sample consisted of 1,300 Saudi high school students randomly selected from a larger pool of 173,133 participants to reduce the effects of excessive power. To examine both goals, the Multiple Indicators Multiple Causes (MIMIC) approach for detecting Differential Item Functioning (DIF) items was used. The results showed that 13 AAT items exhibited DIF effects for different gender groups. In most of these items, male participants were more likely to answer them correctly than their female counterparts. Next, the mediated MIMIC approach was applied to explore possible underlying mechanisms that explain these DIF effects. The results from this study showed that general cognitive ability (i.e., General Aptitude Test - GAT) seems to be a factor that could explain why an AAT item exhibits DIF across gender. It was found that GAT scores fully explain the DIF effect in two AAT items (full mediation). In most other cases, GAT helps account for only a proportion of the DIF effect (partial mediation). The results from this study will help experts improve the quality of their instruments by identifying DIF items and deciding how to revise them, considering the mediator's effect on participants' responses.

Assessing the validity and reliability of the International Anxiety Questionnaire and the International Depression Questionnaire in two bereaved national samples.

The International Anxiety Questionnaire (IAQ) and International Depression Questionnaire (IDQ) are self-report measures of ICD-11 Generalized Anxiety Disorder (ICD-11 GAD) and ICD-11 Single Episode Depressive Disorder (ICD-11 DD). This study tested the psychometric properties of these scales in two samples of bereaved adults from the United Kingdom and the Republic of Ireland. Confirmatory factor analysis (CFA) was used to test the combined dimensionality and measurement invariance of the IAQ and IDQ across the United Kingdom (n = 1012) and Irish (n = 1011) samples. Differential item functioning (DIF) was tested using multiple indicator multiple cause (MIMIC) modelling while convergent validity was also assessed. CFA results supported a correlated two-factor model in both samples. The MIMIC model showed that the IDQ item "Had recurrent thoughts of death or suicide" showed DIF and the effect was small. Internal reliability of the scales were high and convergent validity was supported. The prevalence of ICD-11 GAD was 18.6% and 16.1% and ICD-11 DD was 13.8% and 10.5% in the United Kingdom and Irish samples, respectively. Findings of the study provide support for the validity, measurement invariance, and reliability of the IAQ and IDQ among two bereaved national samples.

Comparing MIMIC and MIMIC-interaction to Alignment Methods for Investigating Measurement Invariance concerning a Continuous Violator

Measurement invariance holds when a latent construct is measured in the same way across different levels of background variables (continuous or categorical) while controlling for the true value of that construct. Using Monte Carlo simulation, this paper compares the multiple indicators, multiple causes (MIMIC) model and MIMIC-interaction to a novel use of alignment optimization (AO) for detecting measurement noninvariance when the violator is a continuous variable. Results showed that MIMIC and MIMIC-interaction in sequential likelihood ratio tests and Wald tests with a Bonferroni correction provided a good balance between identifying invariant and noninvariant (linear violations) items when n ≥ 500 in terms of classification accuracy (CA). AO (CA ≥ .86) was as competitive as MIMIC and MIMIC-interaction to linear invariance violations but was far better under nonlinear quadratic violations when n ≥ 1,000 (i.e., 100 per group for 10 groups).

Cross-Country (Brazil and Iran) Invariance of Fractionation of Executive Functions in Early Adolescence

ABSTRACT Cultural background can influence cognition, including executive functions (EFs), abilities that encompass skills responsible for self-regulation of thoughts and behavior. The seminal unity and diversity model of EFs proposes the existence, in adulthood, of at least three correlated but separable EF latent (shared variance in more than one task/indicator) domains: inhibition, updating and shifting. However, evidence of the cross-cultural generality of this framework is lacking, especially in adolescence, an age during which these domains become more clearly separable. We tested whether this EF fractionation could be observed in early adolescents (9- to 15-year-olds) from metropolitan areas in Brazil (São Paulo) and Iran (Tehran) (total sample: 739; 407 Iranians; 358 girls). Participants carried out two open-access tasks that are representative of each EF domain and that were adapted to each cultural context. Seven latent model configurations were tested. The three-correlated latent factor structure had adequate fit, and multiple-group confirmatory factor analysis invariance testing showed invariance for country at the level of the latent factor structure (configural), factor loadings (metric), and partial invariance at the intercept (scalar) level. Iranians had higher scores in all domains. Multiple indicators multiple causes invariance testing showed model invariance across age (except for one task) and parental education. Performance in all domains improved with age and only minimally with parental schooling. We conclude that EF fractionation into three domains is present in the first half of adolescence in two samples from underrepresented populations in the literature, suggesting a potential generality of EF latent unity/diversity development at this age.

Supplemental Material for Bayesian Regularization in Multiple-Indicators Multiple-Causes Models

Supplemental Material for Bayesian Regularization in Multiple-Indicators Multiple-Causes Models

A Multiple Indicator Multiple Cause (MIMIC) model of the Self-Compassion Scale Youth (SCS-Y) and investigation of differential item functioning in China, Hong Kong and UK adolescents

A Multiple Indicator Multiple Cause (MIMIC) model of the Self-Compassion Scale Youth (SCS-Y) and investigation of differential item functioning in China, Hong Kong and UK adolescents

Specificity of polygenic signatures across symptom dimensions in bipolar disorder: an analysis of UK Bipolar Disorder Research Network data

Current definitions and clinical heterogeneity in bipolar disorder are major concerns as they obstruct aetiological research and impede drug development. Therefore, stratification of bipolar disorder is a high priority. To inform stratification, our analysis aimed to examine the patterns and relationships between polygenic liability for bipolar disorder, major depressive disorder (MDD), and schizophrenia with multidimensional symptom representations of bipolar disorder. In this analysis, data from the UK Bipolar Disorder Research Network (BDRN) were assessed with the Operational Checklist for Psychotic Disorders. Individuals with bipolar disorder as defined in DSM-IV, of European ancestry (self-reported), aged 18 years or older at time of interview, living in the UK, and registered with the BDRN were eligible for inclusion. Psychopathological variables obtained via interview by trained research psychologists or psychiatrists and psychiatric case notes were used to identify statistically distinct symptom dimensions, calibrated with exploratory factor analysis and validated with confirmatory factor analysis (CFA). CFA was extended to include three polygenic risk scores (PRSs) indexing liability for bipolar disorder, MDD, and schizophrenia in a multiple indicator multiple cause (MIMIC) structural equation model to estimate PRS relationships with symptom dimensions. Of 4198 individuals potentially eligible for inclusion, 4148 (2804 [67·6%] female individuals and 1344 [32·4%] male individuals) with a mean age at interview of 45 years (SD 12·03) were available for analysis. Three reliable dimensions (mania, depression, and psychosis) were identified. The MIMIC model fitted the data well (root mean square error of approximation 0·021, 90% CI 0·019-0·023; comparative fit index 0·99) and suggests statistically distinct symptom dimensions also have distinct polygenic profiles. The PRS for MDD was strongly associated with the depression dimension (standardised β 0·125, 95% CI 0·080-0·171) and the PRS for schizophrenia was strongly associated with the psychosis dimension (0·108, 0·082-0·175). For the mania dimension, the PRS for bipolar disorder was weakly associated (0·050, 0·002-0·097). Our findings support the hypothesis that genetic heterogeneity underpins clinical heterogeneity, suggesting that different symptom dimensions within bipolar disorder have partly distinct causes. Furthermore, our results suggest that a specific symptom dimension has a similar cause regardless of the primary psychiatric diagnosis, supporting the use of symptom dimensions in precision psychiatry. Wellcome Trust and UK Medical Research Council.