Causal associations between schizophrenia and other psychiatric disorders: A Mendelian randomization study

Yin Zhou; Yuxiao Chen; Pengli Wang; Kejing Zhang; Yili Zhang

doi:10.1177/03000605251369855

. 2025 Aug 25;53(8):03000605251369855. doi: 10.1177/03000605251369855

Causal associations between schizophrenia and other psychiatric disorders: A Mendelian randomization study

Yin Zhou ^1,^*, Yuxiao Chen ^2,^*, Pengli Wang ³, Kejing Zhang ⁴, Yili Zhang ^5,^✉

PMCID: PMC12378533 PMID: 40852779

Abstract

Objective

Schizophrenia is a globally prevalent complex neuropsychiatric disorder that is frequently comorbid with various psychiatric disorders, leading to poor prognoses for affected patients. However, the causal relationships between schizophrenia and these comorbid disorders remain unclear.

Methods

We utilized Mendelian randomization to investigate the causal effects of schizophrenia on eight psychiatric disorders, including alcohol use disorder, anorexia nervosa, anxiety disorders, attention-deficit hyperactivity disorder, autism spectrum disorders, bipolar disorder, depression, and obsessive–compulsive disorder, using data from the Psychiatric Genomics Consortium and other extensive Genome-Wide Association Studies. We employed the inverse variance-weighted method as the primary analysis, complemented by Mendelian randomization-Egger, weighted median, Mendelian randomization-Presso, Steiger filtering, leave-one-out sensitivity analysis, and reverse Mendelian randomization to address potential biases and validate the directionality of the causal relationships.

Results

Our analysis revealed that a genetically predicted one-log unit increase in schizophrenia risk was associated with a 70.7% increase in the odds of bipolar disorder (odds ratio: 1.707, 95% confidence interval: 1.58–1.84). We also found strong evidence regarding a causal relationship of schizophrenia with autism spectrum disorders, showing a 17.4% higher odds (odds ratio: 1.174, 95% confidence interval: 1.11–1.24). Additionally, schizophrenia conferred a 14.5% elevated risk of alcohol use disorder (odds ratio: 1.145, 95% confidence interval: 1.09–1.21), while a statistically significant yet clinically marginal association was observed with depression (odds ratio: 1.004, 95% confidence interval: 1.003–1.006). No causal relationships were detected between schizophrenia and attention-deficit hyperactivity disorder, anorexia nervosa, anxiety disorders, or obsessive–compulsive disorder. Sensitivity analyses reinforced these findings, and reverse Mendelian randomization analyses provided no evidence of reverse causal impacts on schizophrenia from the disorders examined.

Conclusion

These findings confirm schizophrenia as a significant genetic risk factor for bipolar disorder, autism spectrum disorders, and alcohol use disorder. Our findings enhance understanding of the interrelationships among psychiatric disorders and offer novel insights into the clinical diagnosis and management of psychiatric comorbidities.

Keywords: Schizophrenia, autism spectrum disorders, bipolar disorder, alcohol use disorder, Mendelian randomization, psychiatric comorbidities

Introduction

Schizophrenia is a severe, complex neuropsychiatric disorder characterized by disorganized thought, hallucinations, delusions, and cognitive impairment, affecting approximately 20 million individuals worldwide.^1,2 The disorder not only exerts profound impacts on patients but also imposes significant burdens on families and societal structures.²

Epidemiologically, schizophrenia is frequently comorbid with several other psychiatric disorders, including attention-deficit hyperactivity disorder (ADHD), anorexia nervosa, autism spectrum disorders (ASD), bipolar disorder, and depressive disorder.^3–8 These comorbidities complicate the clinical trajectory and management of schizophrenia, increasing the likelihood of suicide, harm to others, and psychiatric readmission, thereby deteriorating overall life quality.⁹

Historically, the overlap between schizophrenia and these disorders has been attributed to shared genetic underpinnings, environmental risk factors, and neurobiological alterations.^10–12 The hypothesis of the “increased vulnerability model,” as proposed by Krueger et al.,¹¹ suggests that the presence of one psychiatric condition could increase susceptibility to others, pointing toward interdisorder causal effects. Nonetheless, traditional epidemiological approaches are often limited in establishing clear causality due to inherent confounders and the absence of robust longitudinal data.

Utilizing data from large-scale Genome-Wide Association Studies (GWAS), Mendelian randomization (MR) offers a refined methodological approach to dissect these potential causal relationships.¹³ By employing genetic variants linked to schizophrenia as instrumental variables, MR allows for the assessment of whether these variants also influence outcomes related to comorbid conditions. This approach helps circumvent the usual confounders and reverse causality seen in observational studies.

Our current study employs two-sample MR to explore how schizophrenia may causally impact the incidence of alcohol use disorder (AUD), anorexia nervosa, anxiety disorders, ADHD, ASD, bipolar disorder, depression, and obsessive–compulsive disorder (OCD). Our findings indicate that schizophrenia significantly elevates the risk of developing bipolar disorder, ASD, and AUD, highlighting the critical need for comprehensive psychiatric assessment and early intervention strategies in managing schizophrenia and its associated comorbidities.

Methods

Study design

In this investigation, we leveraged publicly available GWAS datasets to perform bidirectional two-sample MR. This approach was utilized to explore the potential causal relationships between schizophrenia and eight psychiatric disorders, namely, AUD, anorexia nervosa, anxiety disorders, ADHD, ASD, bipolar disorder, depression, and OCD.

Ethical compliance for each GWAS dataset was ensured by the respective institutional ethics committees that originally approved the studies. Our analysis was conducted using publicly accessible summary-level data; therefore, no additional ethical approval was required for our MR investigations.

We followed three fundamental instrumental variable (IV) assumptions in our MR study (Figure 1): 1. The selected genetic variants are robustly associated with the exposure. 2. The selected variants are independent of any confounders of the exposure–outcome relationship. 3. The selected variants affect the outcome only through the exposure, not via alternative pathways.

Selection of genetic instruments for schizophrenia

To identify appropriate genetic instruments for schizophrenia, we extracted summary statistics from the GWAS dataset provided by the Psychiatric Genomics Consortium (PGC), specifically the analysis conducted by Trubetskoy et al.¹⁴ This dataset comprised 127,906 European participants, including 52,017 diagnosed with schizophrenia and 75,889 control participants. Detailed information about the data sources is included in Table S1, with a more extensive description of participant characteristics available in the abovementioned study.¹⁴

Single-nucleotide polymorphisms (SNPs) associated with schizophrenia were selected as genetic instruments based on their achievement of genome-wide significance (p < 5 × 10⁻⁸). A clumping procedure was subsequently performed to ensure the independence of SNPs, with SNPs pruned at a stringent linkage disequilibrium threshold of R² < 0.01 within a 500-kb window. Instrumental strength was quantified using the F statistic, with an F statistic >10 considered sufficiently informative. Finally, 149 SNPs were selected with an average F statistic of 44.5 (range: 29.4–175.3; Table S2), thus representing sufficient instrumental strength.

Data sources for other psychiatric disorders

To rigorously assess the association between schizophrenia and the risk of comorbid psychiatric disorders, we conducted a comprehensive search of the public IEU GWAS database (https://gwas.mrcieu.ac.uk/). This search aimed to include the largest and most robust GWAS available for various psychiatric disorders, such as AUD, anorexia nervosa, anxiety disorders, ADHD, ASD, bipolar disorder, depression, and OCD. We prioritized GWAS with the largest sample sizes and the highest number of cases to ensure robustness in our results. Detailed information regarding these GWAS can be found in Table S1.

Prior to conducting MR analyses, we harmonized the SNPs identified from the exposure GWAS with those available in the outcome GWAS, aligning alleles on the same strand to ensure accuracy and consistency in our genetic comparisons.

Statistical analyses

Our primary analytical method was the inverse variance-weighted (IVW) approach. We then performed a suite of sensitivity analyses to evaluate the robustness of the IVW results against potential biases, including MR-Egger, weighted median, MR-Presso, and Steiger filtering. We further visually assessed SNP heterogeneity through scatter plots, forest plots, and leave-one-out analysis. These methods offer distinct theoretical advantages for controlling various biases and are designed to be robust against specific assumption violations, albeit with some trade-offs in statistical efficiency.

The IVW method,¹⁵ which assumes that all genetic instruments are valid, provides the highest statistical power. It functions as a weighted linear regression of SNP-outcome effects on SNP-exposure effects, constrained to a zero intercept. We first quantified SNP heterogeneity using Cochran’s Q statistic (Table S3). For associations with Cochran’s Q p-value <0.05, we employed a random-effects model, while for those with Cochran’s Q p-value ≥0.05, we applied a fixed-effects model as recommended.

MR-Egger regression,¹⁶ unlike IVW, does not fix the intercept to zero. This allows the slope coefficient of the MR-Egger to provide a corrected, albeit less precise, causal estimate even in the presence of pleiotropy. The intercept in the MR-Egger regression acts as an indicator of the average pleiotropic effect across the genetic variants. An intercept with a p-value greater than 0.05 suggests the absence of pleiotropic bias.

The weighted median¹⁷ provides a robust consensus causal estimate by taking the median of the ratio estimates distribution. This method is effective when more than 50% of the information is derived from valid variants and is less influenced by variants with pleiotropic effects, serving as an implicit outlier removal strategy.

The MR-Presso,¹⁸ a variation on the IVW method, includes a global test to assess for overall horizontal pleiotropy. When pleiotropy is detected, the MR-Presso outlier test identifies individual pleiotropic outliers by calculating the residual sum of squares. Causal estimates are then recalculated using the IVW method, excluding any detected outliers.

Meanwhile, we utilized the Steiger filter to ascertain the directionality of the causal relationships identified in our MR analyses. The Steiger test essentially assesses the Z-scores derived from the genetic associations with the exposure and outcome. A significant result from this test indicates that the genetic variants are more strongly associated with the exposure than with the outcome, thereby validating the directionality of the MR analysis.

To visually assess the causal relationship and identify heterogeneity or outliers, we employed MR scatter plots and generated forest plots based on individual SNP estimates. Additionally, to assess the robustness of the combined causal estimate and determine if any single SNP disproportionately influenced the results, we performed a leave-one-out sensitivity analysis, iteratively removing each SNP and recalculating the estimate; stability across all iterations indicates that the finding is not reliant on any single instrumental variable.

To explore the potential reverse causal relationships, where the other psychiatric disorders may influence schizophrenia, we conducted reverse-direction MR analyses. For anxiety disorders, AUD, ASD, and anorexia nervosa, due to a shortage of SNPs reaching genome-wide significance (p < 5 × 10⁻⁸), we instead selected SNPs meeting a less stringent significance level (p < 1 × 10⁻⁵). For OCD, reverse MR analysis was not feasible due to the lack of valid instrumental SNPs even at a relaxed significance threshold of p < 1 × 10⁻⁵. Subsequent IVW, MR-Egger, and weighted median analyses were conducted, consistent with the methods applied in our primary analyses. Details regarding the GWAS datasets used for these analyses are provided in Table S4.

All statistical procedures were executed via R (Version 4.0.3), utilizing the TwoSampleMR, MendelianRandomization, and MR-Presso packages. Effects were quantified as odds ratios (ORs) with 95% confidence intervals (CIs) for dichotomous outcomes. For multiple testing, the statistical significance was determined using a Bonferroni-corrected p-value threshold of <0.05.

Results

Schizophrenia and other psychiatric disorders: principal results

Initially, we utilized the IVW method to investigate the causal relationships between schizophrenia and eight other psychiatric disorders, including AUD, anorexia nervosa, anxiety disorders, ADHD, ASD, bipolar disorder, depression, and OCD. The IVW method is preferred as the primary analysis in MR studies for its optimal efficiency when all genetic variants are valid instruments. The results of IVW analysis are presented in Figure 2, with corresponding scatter plots displayed in Figure S1.

Figure 2. — Mendelian randomization (MR) estimates derived from the inverse variance-weighted (IVW) method to assess the causal effect of schizophrenia on other psychiatric disorders. Random- or fixed-effects models were selected based on Cochran’s Q statistic, with statistical significance defined as Bonferroni-corrected p < 0.05. OR: odds ratio; CI: confidence interval.

Our analysis revealed robust evidence for several causal associations. A genetically predicted one-log unit increase in schizophrenia risk was associated with the following: (a) a substantial (70.7%) increase in bipolar disorder risk (OR: 1.707, 95% CI: 1.58–1.84, p_adj = 2.87 × 10⁻⁴²); (b) a 17.4% higher ASD susceptibility (OR: 1.174, 95% CI: 1.11–1.24, p_adj = 4.46 × 10⁻⁸); and (c) a 14.5% increased likelihood of AUD (OR: 1.145, 95% CI: 1.09–1.21, p_adj = 3.64 × 10⁻⁶). Additionally, although a statistically significant association of schizophrenia with depression was observed (OR: 1.004, 95% CI: 1.003–1.006, p_adj = 3.11 × 10⁻⁵), the minimal effect magnitude suggests limited clinical relevance. Following rigorous Bonferroni correction, we found no compelling evidence supporting the causal effects of schizophrenia on ADHD, anorexia nervosa, anxiety disorders, or OCD.

Sensitivity analysis

To further validate the causal relationships between schizophrenia and other psychiatric disorders, we conducted MR-Egger, weighted median, MR-Presso, and leave-one-out analyses. These sensitivity analyses, although less powerful than the IVW method, offer robustness against various forms of bias (see Methods section for details).

Schizophrenia and bipolar disorder

Sensitivity analyses reinforced the causal link between schizophrenia and bipolar disorder. Effect estimates from both MR-Egger (OR: 1.629, 95% CI: 1.20–2.20, p = 0.002) and weighted median (OR: 1.612, 95% CI: 1.46–1.78, p = 4.5 × 10⁻²⁰) analyses remained significant and were largely consistent with the corresponding values from the IVW analysis (Figure 3(a)). The MR-Egger intercept test (Figure 3(a)) and I² statistics (Table S3) indicated no evidence of heterogeneity or pleiotropy. Although MR-Presso detected global pleiotropy, no outlier SNPs were identified, suggesting that the detected pleiotropy was not due to individual SNP effects (Figure 3(a)). The individual SNP-effect forest plot visually demonstrates consistent directional effects across variants (Figure 3(b)), whereas the leave-one-out sensitivity analysis confirmed the result stability upon iterative exclusion of individual instruments (Figure 3(c)). Steiger analysis indicated that the instrumental variables had a significantly greater impact on schizophrenia than on bipolar disorder, confirming the expected direction of causality (Table S5). Overall, these findings support a clear causal relationship between schizophrenia and bipolar disorder.

Figure 3. — Sensitivity analyses of the causal association between schizophrenia and the risk of bipolar disorder. (a) Forest plot of MR sensitivity analyses (MR-Egger, weighted median, and MR-Presso) assessing schizophrenia’s causal effect on bipolar disorder. (b) Individual SNP-effect forest plot, where blue markers represent β estimates with 95% CI for each SNP and the red diamond denotes the pooled IVW estimate with its 95% CI and (c) leave-one-out sensitivity analysis showing β estimates with 95% CI upon iterative exclusion of individual SNPs, with the red horizontal line indicating the overall IVW estimate. Statistical significance was defined as p < 0.05. MR: Mendelian randomization; SNP: single-nucleotide polymorphism; CI: confidence interval; IVW: inverse variance-weighted.

Schizophrenia and ASD

All sensitivity analyses supported the robustness of the IVW results (Figure 4). Both the MR-Egger (OR: 1.269, 95% CI: 1.11–1.24, p = 0.03) and weighted median (OR: 1.005, 95% CI: 1.001–1.002, p = 5.6 × 10⁻⁵) approaches aligned with the IVW regression results for causal estimates (Figure 4(a)). The MR-Egger intercept test suggested the absence of pleiotropic bias (Figure 4(a)), and I² statistics revealed negligible heterogeneity (Table S3). Although MR-Presso identified overall horizontal pleiotropy and outlier SNPs, correction for these outliers still supported the causal impact of schizophrenia on ASD with statistical significance (outlier correction p = 7.2 × 10⁻⁹) (Figure 4(a)). The individual SNP-effect forest plot demonstrated consistent effect directions (Figure 4(b)), whereas the leave-one-out analysis confirmed result stability across iterative exclusions (Figure 4(c)). Steiger analysis confirmed the directionality of the causation (Table S5). Therefore, we conclude that there is a definitive causal relationship between schizophrenia and ASD.

Schizophrenia and AUD

Comprehensive sensitivity analyses robustly affirmed the causal effect of schizophrenia on AUD risk (Figure 5). MR-Egger regression yielded a borderline significant positive effect (OR: 1.231, 95% CI: 0.995–1.523, p = 0.056), whereas the weighted median estimator demonstrated stronger evidence for causality (OR: 1.157, 95% CI: 1.077–1.241, p = 5.72 × 10⁻⁵) (Figure 5(a)). The absence of directional pleiotropy was confirmed by a nonsignificant MR-Egger intercept (p = 0.495), with additional support from low I² values (43.1%, Table S3), indicating negligible heterogeneity. MR-Presso detected global horizontal pleiotropy (p = 0.016) and identified one outlier SNP (Figure 5(a)). Crucially, after outlier correction, the causal estimate remained statistically significant with enhanced effect size (outlier correction p = 2.18 × 10⁻⁷). Variant-level assessments reinforced result consistency. The individual SNP-effect distribution in the forest plot revealed homogeneous directional effects without influential variants (Figure 5(b)), while leave-one-out sensitivity testing demonstrated remarkable stability of causal estimates across sequential exclusions (Figure 5(c)). Steiger directionality analysis confirmed the correct causal orientation (Table S5). Therefore, these multimodal sensitivity examinations confirm a definitive causal relationship between schizophrenia and AUD.

Schizophrenia and depression

The effect estimates from MR-Egger and weighted median analyses were very similar to those of the IVW, albeit with small magnitudes (Figure 6(a)). Notably, MR-Egger’s effect estimates were close but did not reach statistical significance (Figure 6(a)). The MR-Egger intercept test (p = 0.479) and low I² values (34.1%, Table S3) indicated no substantial directional pleiotropy or heterogeneity. MR-Presso identified both horizontal pleiotropy and outlier SNPs; however, after removing these outliers, the causal estimate of schizophrenia on depression remained significant (outlier correction p = 1.06 × 10⁻⁴) (Figure 6(a)). Supporting these findings, the individual SNP-effect forest plot revealed generally consistent effect directions without dominant influential variants (Figure 6(b)), and the leave-one-out sensitivity analysis showed stable causal estimates across sequential exclusions (Figure 6(c)). Steiger analysis did not suggest reverse causation (Table S5). Considering the somewhat inconsistent results above, we still conclude that schizophrenia has a modest causal effect on depression.

Finally, no association between schizophrenia and either anorexia nervosa, anxiety disorders, ADHD, or OCD was observed based on MR-Egger and weighted median analyses (Figures S2 to S5). Consequently, we believe that there is no compelling evidence supporting a causal association between schizophrenia and these disorders.

Reverse MR analysis to assess the effect of seven psychiatric disorders on schizophrenia

To evaluate potential reverse causation, we performed bidirectional MR analyses with schizophrenia as the outcome and seven psychiatric disorders as exposures (OCD was excluded due to insufficient instrumental SNPs). MR-Egger suggested a putative causal effect of bipolar disorder on schizophrenia (OR: 1.51, 95% CI: 1.32–1.74, p = 3.6 × 10⁻⁹), but significant intercept evidence indicated horizontal pleiotropy potentially confounding this association. Crucially, the primary IVW method showed no supportive evidence (Figure 7). Anxiety disorders demonstrated a nominally significant IVW association that proved unstable in sensitivity analyses (Figure 7). For the remaining five disorders, IVW or complementary methods (MR-Egger/weighted median) revealed no significant reverse causal effects (Figure 7). Collectively, these analyses provide no robust evidence for reverse causation between the examined psychiatric disorders and schizophrenia.

Figure 7. — Reverse MR analyses evaluating the causal effects of seven psychiatric disorders on schizophrenia. Statistical significance was defined as p < 0.05. MR: Mendelian randomization.

Discussion

This study is the first to utilize MR to elucidate causal associations between schizophrenia and multiple psychiatric disorders. Our principal findings suggest that a genetically increased risk of schizophrenia substantially elevates the risk of bipolar disorder, ASD, and AUD. Additionally, we observed a causal association between schizophrenia and depression, although the effect size was modest. In contrast, no compelling evidence was found for causal relationships between schizophrenia and anxiety disorders, ADHD, anorexia nervosa, or OCD. A unique contribution of this study is its demonstration from a genetic epidemiology perspective of the causal relationships between schizophrenia and other major psychiatric disorders, offering new insights into the mechanisms and management of comorbid mental illnesses.

Although extensive literature documents phenotypic comorbidities between schizophrenia and disorders including bipolar disorder, ASD, and AUD, our MR study provides compelling causal evidence establishing schizophrenia as a genetic driver of these conditions. This shift from correlation to causation demands changes in clinical practice. Proactive comorbidity screening should be implemented—for example, using the Young Mania Rating Scale even in schizophrenia patients without overt bipolar disorder symptoms. Earlier preventive interventions are also justified. Integrating medications such as naltrexone or acamprosate into initial antipsychotic regimens for high-risk individuals is preferable to delaying treatment until AUD diagnosis. These findings fundamentally reorient management from reactive to pre-emptive strategies.

Historically, symptom-based psychiatric diagnostic systems have led to overlapping diagnoses and inconsistencies in clinical management. Ongoing research reveals shared biological processes between schizophrenia and other psychiatric diseases. For instance, transcriptomic evidence has identified several pathways related to both schizophrenia and bipolar disorder, such as dysfunctions in immune function.¹⁹ Neuroimaging studies have shown spatial correlation between the neuroimaging epicenters of schizophrenia and bipolar disorder.²⁰ Similar structural and functional brain changes, such as decreased gray matter volume in the limbic–striato–thalamic circuit and reduced activity in primary sensory-motor areas, have been observed in both schizophrenia and ASD.²¹ These common biological processes underscore the interconnected nature of psychiatric disorders but do not imply causality. Our results encourage further exploration of the unidirectional neurobiological mechanisms by which schizophrenia may precipitate ASD and bipolar disorder, informing potential interventions.

Furthermore, substantial evidence suggests shared genetic factors between schizophrenia and other psychiatric disorders.^22,23 In such cases, MR analyses are susceptible to pleiotropy, where the same genetic variations equally affect two different psychiatric disorders, potentially leading to false-positive causal inferences. To address this, we employed various sensitivity analyses to mitigate this possibility. Notably, MR-Egger and MR-Presso did not detect pleiotropy, reinforcing our main conclusions. Furthermore, methods such as Steiger analysis have confirmed that the same SNPs differentially impact schizophrenia and bipolar disorder; reverse MR analyses also did not reveal any causal effects of ASD or bipolar disorder on schizophrenia, affirming that our causal inferences are robust against pleiotropic biases.

From the perspective of disease onset, it might seem intuitive that ASD increases the risk of schizophrenia. Research indicates that many psychiatric disorder diagnoses are often made within approximately 3 years prior to a diagnosis of an ASD.²⁴ However, our results did not show a causal effect of ASD on schizophrenia but rather showed the opposite. Indeed, longitudinal observational studies on schizophrenia have shown that many patients are diagnosed with ASD after being diagnosed with schizophrenia.²⁵

On a pathophysiological level, there is currently no direct evidence proving that schizophrenia causes bipolar disorder and ASD. Here, we offer some hypotheses. First, many studies indicate that schizophrenia patients harbor genetic variations associated with both bipolar disorder and ASD.^26,27 Notably, the 22q11.2 deletion, identified as the most significant genetic risk factor for schizophrenia,²⁸ also enhances susceptibility to ASD.²⁹ If future studies confirm that these genetic variations have a more pronounced impact on schizophrenia than on ASD, it would lend support to the hypothesis that schizophrenia increases the risk of ASD. Second, schizophrenia has been associated with structural and functional neural anomalies, such as altered brain volume³⁰ and dynamic changes in hippocampal gray matter and overall white matter.^31,32 These neurobiological changes could form the biological basis for increased susceptibility to ASD or bipolar disorder. Finally, although schizophrenia is likely influenced by familial and social environments, it inevitably also affects these factors, thereby exerting an impact on other psychiatric disorders.

This study has several limitations. First, the classification of schizophrenia, bipolar disorder, and ASD still primarily relies on symptomatic diagnoses,³³ which can lead to diagnostic overlap and may exaggerate causal associations. However, reverse MR analyses found no causal effects of bipolar disorder or ASD on schizophrenia, suggesting minimal bias. Second, most study participants were of European descent, potentially limiting the generalizability of results to other populations. The genetic architecture of schizophrenia in East Asian populations differs markedly from that in Europeans.³⁴ Finally, the limited availability of robust instrumental variables for ASD and anorexia nervosa, due to the small size of the existing GWAS datasets, constrains our ability to assess their reverse causal effects on schizophrenia effectively.

Conclusion

Our study robustly demonstrates a causal effect of a genetically elevated risk of schizophrenia on the risks associated with ASD and bipolar disorder. Additionally, although we observed a causal effect of schizophrenia on depression, its impact appears to be of limited practical significance. Our findings enhance the understanding of the interrelationships among psychiatric disorders and offer novel insights into the clinical diagnosis and management of psychiatric comorbidities.

Supplemental Material

sj-pdf-1-imr-10.1177_03000605251369855 - Supplemental material for Causal associations between schizophrenia and other psychiatric disorders: A Mendelian randomization study

sj-pdf-1-imr-10.1177_03000605251369855.pdf^{(710.6KB, pdf)}

Supplemental material, sj-pdf-1-imr-10.1177_03000605251369855 for Causal associations between schizophrenia and other psychiatric disorders: A Mendelian randomization study by Yin Zhou, Yuxiao Chen, Pengli Wang, Kejing Zhang and Yili Zhang in Journal of International Medical Research

Acknowledgments

We employed artificial intelligence tools (DeepSeek) solely for language polishing during manuscript preparation. The authors retained full oversight of content generation, data interpretation, and critical editing throughout this process.

Author contributions: Zhou Yin was responsible for conceptualization, data curation, funding acquisition, investigation, supervision, validation, writing the original draft, and review and editing of the manuscript. Chen Yuxiao contributed to investigation, methodology, software development, funding acquisition, and validation of the study. Pengli Wang was involved in methodology, software development, and data visualization. Zhang Kejing was involved in software development and data visualization. Zhang Yili performed formal analysis, contributed to software development, and performed data visualization. All authors approved the final version of the manuscript.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.

Funding: Yin Zhou was supported by the National Natural Science Foundation of China (Grant Number: 82101792). Yuxiao Chen was supported by the National Natural Science Foundation of China (Grant Number: 82400278).

ORCID iD: Yin Zhou https://orcid.org/0009-0002-5179-3445

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.

Supplemental material

Supplemental material for this article is available online.

References

1.Kahn RS, Sommer IE, Murray RM, et al. Schizophrenia. Nat Rev Dis Primers 2015; 1: 15067. [DOI] [PubMed] [Google Scholar]
2.GBD 2019 Mental Disorders Collaborators. Global, regional, and national burden of 12 mental disorders in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Psychiatry 2022; 9: 137–150. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Jeon SM, Lee DY, Cha S, et al. Psychiatric comorbidities and schizophrenia in youths with attention-deficit/hyperactivity disorder. JAMA Netw Open 2023; 6: e2345793. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Nourredine M, Gering A, Fourneret P, et al. Association of attention-deficit/hyperactivity disorder in childhood and adolescence with the risk of subsequent psychotic disorder: a systematic review and meta-analysis. JAMA Psychiatry 2021; 78: 519–529. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Zhang R, Larsen JT, Kuja-Halkola R, et al. Familial co-aggregation of schizophrenia and eating disorders in Sweden and Denmark. Mol Psychiatry 2021; 26: 5389–5397. [DOI] [PubMed] [Google Scholar]
6.Krieger I, Grossman-Giron A, Comaneshter D, et al. The co-occurrence of autistic spectrum disorder and schizophrenia: a nationwide population-based study. J Psychiatr Res 2021; 138: 280–283. [DOI] [PubMed] [Google Scholar]
7.Laursen TM, Agerbo E, Pedersen CB. Bipolar disorder, schizoaffective disorder, and schizophrenia overlap: a new comorbidity index. J Clin Psychiatry 2009; 70: 1432–1438. [DOI] [PubMed] [Google Scholar]
8.Li W, Yang Y, An FR, et al. Prevalence of comorbid depression in schizophrenia: a meta-analysis of observational studies. J Affect Disord 2020; 273: 524–531. [DOI] [PubMed] [Google Scholar]
9.Abdullah HM, Azeb SH, Hwang MY, et al. Comorbidity in schizophrenia: conceptual issues and clinical management. Focus (Am Psychiatr Publ) 2020; 18: 386–390. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Shnayder NA, Novitsky MA, Neznanov NG, et al. Genetic predisposition to schizophrenia and depressive disorder comorbidity. Genes (Basel) 2022; 13: 457. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Chisholm K, Lin A, Abu-Akel A, et al. The association between autism and schizophrenia spectrum disorders: a review of eight alternate models of co-occurrence. Neurosci Biobehav Rev 2015; 55: 173–183. [DOI] [PubMed] [Google Scholar]
12.Xie C, Xiang S, Shen C; ZIB Consortium et al. A shared neural basis underlying psychiatric comorbidity. Nat Med 2023; 29: 1232–1242. [DOI] [PMC free article] [PubMed] [Google Scholar]
13.O'Donnell CJ, Sabatine MS. Opportunities and challenges in Mendelian randomization studies to guide trial design. Jama Cardiol 2018; 3: 967. [DOI] [PubMed] [Google Scholar]
14.Trubetskoy V, Pardinas AF, Qi T; Schizophrenia Working Group of the Psychiatric Genomics Consortium et al. Mapping genomic loci implicates genes and synaptic biology in schizophrenia. Nature 2022; 604: 502–508. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol 2013; 37: 658–665. [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Bowden J, Davey SG, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol 2015; 44: 512–525. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Bowden J, Davey SG, Haycock PC, et al. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol 2016; 40: 304–314. [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Verbanck M, Chen CY, Neale B, et al. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet 2018; 50: 693–698. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Sahu A, Chowdhury HA, Gaikwad M, et al. Integrative network analysis identifies differential regulation of neuroimmune system in schizophrenia and bipolar disorder. Brain Behav Immun Health 2020; 2: 100023. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Jiang Y, Palaniyappan L, Luo C; ZIB Consortium et al. Neuroimaging epicenters as potential sites of onset of the neuroanatomical pathology in schizophrenia. Sci Adv 2024; 10: eadk6063. [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Jin X, Zhang K, Lu B, et al. Shared atypical spontaneous brain activity pattern in early onset schizophrenia and autism spectrum disorders: evidence from cortical surface-based analysis. Eur Child Adolesc Psychiatry 2024; 33: 2387–2396. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Hindley G, Bahrami S, Steen NE, et al. Characterising the shared genetic determinants of bipolar disorder, schizophrenia and risk-taking. Transl Psychiatry 2021; 11: 466. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Smeland OB, Bahrami S, Frei O, et al. Genome-wide analysis reveals extensive genetic overlap between schizophrenia, bipolar disorder, and intelligence. Mol Psychiatry 2020; 25: 844–853. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Hsu TW, Chu CS, Tsai SJ, et al. Diagnostic progression to schizophrenia: a nationwide cohort study of 11 170 adolescents and young adults with autism spectrum disorder. Psychiatry Clin Neurosci 2022; 76: 644–651. [DOI] [PubMed] [Google Scholar]
25.Krebs MD, Themudo GE, Benros ME, et al. Associations between patterns in comorbid diagnostic trajectories of individuals with schizophrenia and etiological factors. Nat Commun 2021; 12: 6617. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Rees E, Creeth H, Hwu HG, et al. Schizophrenia, autism spectrum disorders and developmental disorders share specific disruptive coding mutations. Nat Commun 2021; 12: 5353. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Rees E, Kendall K, Pardinas AF, et al. Analysis of intellectual disability copy number variants for association with schizophrenia. JAMA Psychiatry 2016; 73: 963–969. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Marshall CR, Howrigan DP, Merico D; CNV and Schizophrenia Working Groups of the Psychiatric Genomics Consortium et al. Contribution of copy number variants to schizophrenia from a genome-wide study of 41,321 subjects. Nat Genet 2017; 49: 27–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Nehme R, Pietilainen O, Artomov M, et al. The 22q11.2 region regulates presynaptic gene-products linked to schizophrenia. Nat Commun 2022; 13: 3690. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Toulopoulou T, Van Haren N, Zhang X, et al. Reciprocal causation models of cognitive vs volumetric cerebral intermediate phenotypes for schizophrenia in a pan-European twin cohort. Mol Psychiatry 2015; 20: 1386–1396. [DOI] [PubMed] [Google Scholar]
31.Brosch K, Stein F, Schmitt S, et al. Reduced hippocampal gray matter volume is a common feature of patients with major depression, bipolar disorder, and schizophrenia spectrum disorders. Mol Psychiatry 2022; 27: 4234–4243. [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Cetin-Karayumak S, Di Biase MA, Chunga N, et al. White matter abnormalities across the lifespan of schizophrenia: a harmonized multi-site diffusion MRI study. Mol Psychiatry 2020; 25: 3208–3219. [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Pearlson GD. Etiologic, phenomenologic, and endophenotypic overlap of schizophrenia and bipolar disorder. Annu Rev Clin Psychol 2015; 11: 251–281. [DOI] [PubMed] [Google Scholar]
34.Lam M, Chen CY, Li Z; Genetic REsearch on schizophreniA neTwork-China and the Netherlands (GREAT-CN) et al. Comparative genetic architectures of schizophrenia in East Asian and European populations. Nat Genet 2019; 51: 1670–1678. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

sj-pdf-1-imr-10.1177_03000605251369855 - Supplemental material for Causal associations between schizophrenia and other psychiatric disorders: A Mendelian randomization study

sj-pdf-1-imr-10.1177_03000605251369855.pdf^{(710.6KB, pdf)}

Data Availability Statement

The data underlying this article will be shared on reasonable request to the corresponding author.

[bibr1-03000605251369855] 1.Kahn RS, Sommer IE, Murray RM, et al. Schizophrenia. Nat Rev Dis Primers 2015; 1: 15067. [DOI] [PubMed] [Google Scholar]

[bibr2-03000605251369855] 2.GBD 2019 Mental Disorders Collaborators. Global, regional, and national burden of 12 mental disorders in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Psychiatry 2022; 9: 137–150. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr3-03000605251369855] 3.Jeon SM, Lee DY, Cha S, et al. Psychiatric comorbidities and schizophrenia in youths with attention-deficit/hyperactivity disorder. JAMA Netw Open 2023; 6: e2345793. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr4-03000605251369855] 4.Nourredine M, Gering A, Fourneret P, et al. Association of attention-deficit/hyperactivity disorder in childhood and adolescence with the risk of subsequent psychotic disorder: a systematic review and meta-analysis. JAMA Psychiatry 2021; 78: 519–529. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr5-03000605251369855] 5.Zhang R, Larsen JT, Kuja-Halkola R, et al. Familial co-aggregation of schizophrenia and eating disorders in Sweden and Denmark. Mol Psychiatry 2021; 26: 5389–5397. [DOI] [PubMed] [Google Scholar]

[bibr6-03000605251369855] 6.Krieger I, Grossman-Giron A, Comaneshter D, et al. The co-occurrence of autistic spectrum disorder and schizophrenia: a nationwide population-based study. J Psychiatr Res 2021; 138: 280–283. [DOI] [PubMed] [Google Scholar]

[bibr7-03000605251369855] 7.Laursen TM, Agerbo E, Pedersen CB. Bipolar disorder, schizoaffective disorder, and schizophrenia overlap: a new comorbidity index. J Clin Psychiatry 2009; 70: 1432–1438. [DOI] [PubMed] [Google Scholar]

[bibr8-03000605251369855] 8.Li W, Yang Y, An FR, et al. Prevalence of comorbid depression in schizophrenia: a meta-analysis of observational studies. J Affect Disord 2020; 273: 524–531. [DOI] [PubMed] [Google Scholar]

[bibr9-03000605251369855] 9.Abdullah HM, Azeb SH, Hwang MY, et al. Comorbidity in schizophrenia: conceptual issues and clinical management. Focus (Am Psychiatr Publ) 2020; 18: 386–390. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr10-03000605251369855] 10.Shnayder NA, Novitsky MA, Neznanov NG, et al. Genetic predisposition to schizophrenia and depressive disorder comorbidity. Genes (Basel) 2022; 13: 457. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr11-03000605251369855] 11.Chisholm K, Lin A, Abu-Akel A, et al. The association between autism and schizophrenia spectrum disorders: a review of eight alternate models of co-occurrence. Neurosci Biobehav Rev 2015; 55: 173–183. [DOI] [PubMed] [Google Scholar]

[bibr12-03000605251369855] 12.Xie C, Xiang S, Shen C; ZIB Consortium et al. A shared neural basis underlying psychiatric comorbidity. Nat Med 2023; 29: 1232–1242. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr13-03000605251369855] 13.O'Donnell CJ, Sabatine MS. Opportunities and challenges in Mendelian randomization studies to guide trial design. Jama Cardiol 2018; 3: 967. [DOI] [PubMed] [Google Scholar]

[bibr14-03000605251369855] 14.Trubetskoy V, Pardinas AF, Qi T; Schizophrenia Working Group of the Psychiatric Genomics Consortium et al. Mapping genomic loci implicates genes and synaptic biology in schizophrenia. Nature 2022; 604: 502–508. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr15-03000605251369855] 15.Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol 2013; 37: 658–665. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr16-03000605251369855] 16.Bowden J, Davey SG, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol 2015; 44: 512–525. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr17-03000605251369855] 17.Bowden J, Davey SG, Haycock PC, et al. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol 2016; 40: 304–314. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr18-03000605251369855] 18.Verbanck M, Chen CY, Neale B, et al. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet 2018; 50: 693–698. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr19-03000605251369855] 19.Sahu A, Chowdhury HA, Gaikwad M, et al. Integrative network analysis identifies differential regulation of neuroimmune system in schizophrenia and bipolar disorder. Brain Behav Immun Health 2020; 2: 100023. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr20-03000605251369855] 20.Jiang Y, Palaniyappan L, Luo C; ZIB Consortium et al. Neuroimaging epicenters as potential sites of onset of the neuroanatomical pathology in schizophrenia. Sci Adv 2024; 10: eadk6063. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr21-03000605251369855] 21.Jin X, Zhang K, Lu B, et al. Shared atypical spontaneous brain activity pattern in early onset schizophrenia and autism spectrum disorders: evidence from cortical surface-based analysis. Eur Child Adolesc Psychiatry 2024; 33: 2387–2396. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr22-03000605251369855] 22.Hindley G, Bahrami S, Steen NE, et al. Characterising the shared genetic determinants of bipolar disorder, schizophrenia and risk-taking. Transl Psychiatry 2021; 11: 466. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr23-03000605251369855] 23.Smeland OB, Bahrami S, Frei O, et al. Genome-wide analysis reveals extensive genetic overlap between schizophrenia, bipolar disorder, and intelligence. Mol Psychiatry 2020; 25: 844–853. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr24-03000605251369855] 24.Hsu TW, Chu CS, Tsai SJ, et al. Diagnostic progression to schizophrenia: a nationwide cohort study of 11 170 adolescents and young adults with autism spectrum disorder. Psychiatry Clin Neurosci 2022; 76: 644–651. [DOI] [PubMed] [Google Scholar]

[bibr25-03000605251369855] 25.Krebs MD, Themudo GE, Benros ME, et al. Associations between patterns in comorbid diagnostic trajectories of individuals with schizophrenia and etiological factors. Nat Commun 2021; 12: 6617. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr26-03000605251369855] 26.Rees E, Creeth H, Hwu HG, et al. Schizophrenia, autism spectrum disorders and developmental disorders share specific disruptive coding mutations. Nat Commun 2021; 12: 5353. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr27-03000605251369855] 27.Rees E, Kendall K, Pardinas AF, et al. Analysis of intellectual disability copy number variants for association with schizophrenia. JAMA Psychiatry 2016; 73: 963–969. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr28-03000605251369855] 28.Marshall CR, Howrigan DP, Merico D; CNV and Schizophrenia Working Groups of the Psychiatric Genomics Consortium et al. Contribution of copy number variants to schizophrenia from a genome-wide study of 41,321 subjects. Nat Genet 2017; 49: 27–35. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr29-03000605251369855] 29.Nehme R, Pietilainen O, Artomov M, et al. The 22q11.2 region regulates presynaptic gene-products linked to schizophrenia. Nat Commun 2022; 13: 3690. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr30-03000605251369855] 30.Toulopoulou T, Van Haren N, Zhang X, et al. Reciprocal causation models of cognitive vs volumetric cerebral intermediate phenotypes for schizophrenia in a pan-European twin cohort. Mol Psychiatry 2015; 20: 1386–1396. [DOI] [PubMed] [Google Scholar]

[bibr31-03000605251369855] 31.Brosch K, Stein F, Schmitt S, et al. Reduced hippocampal gray matter volume is a common feature of patients with major depression, bipolar disorder, and schizophrenia spectrum disorders. Mol Psychiatry 2022; 27: 4234–4243. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr32-03000605251369855] 32.Cetin-Karayumak S, Di Biase MA, Chunga N, et al. White matter abnormalities across the lifespan of schizophrenia: a harmonized multi-site diffusion MRI study. Mol Psychiatry 2020; 25: 3208–3219. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bibr33-03000605251369855] 33.Pearlson GD. Etiologic, phenomenologic, and endophenotypic overlap of schizophrenia and bipolar disorder. Annu Rev Clin Psychol 2015; 11: 251–281. [DOI] [PubMed] [Google Scholar]

[bibr34-03000605251369855] 34.Lam M, Chen CY, Li Z; Genetic REsearch on schizophreniA neTwork-China and the Netherlands (GREAT-CN) et al. Comparative genetic architectures of schizophrenia in East Asian and European populations. Nat Genet 2019; 51: 1670–1678. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Causal associations between schizophrenia and other psychiatric disorders: A Mendelian randomization study

Yin Zhou

Yuxiao Chen

Pengli Wang

Kejing Zhang

Yili Zhang

Abstract

Objective

Methods

Results

Conclusion

Introduction

Methods

Study design

Figure 1.

Selection of genetic instruments for schizophrenia

Data sources for other psychiatric disorders

Statistical analyses

Results

Schizophrenia and other psychiatric disorders: principal results

Figure 2.

Sensitivity analysis

Schizophrenia and bipolar disorder

Figure 3.

Schizophrenia and ASD

Figure 4.

Schizophrenia and AUD

Figure 5.

Schizophrenia and depression

Figure 6.

Reverse MR analysis to assess the effect of seven psychiatric disorders on schizophrenia

Figure 7.

Discussion

Conclusion

Supplemental Material

Acknowledgments

Data availability

Supplemental material

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases