Substantial evidence indicates that physical activity (PA) improves symptoms, cognitive function, and quality of life in patients with schizophrenia.1 Some studies suggest a protective effect of PA against schizophrenia/psychosis risk itself, although current evidence is inconclusive.2 Here, using mendelian randomization (MR) and its multivariable extension (MVMR), we have examined the association between PA (exposure) and schizophrenia risk (outcome). Likewise, we have investigated the potential pleiotropic role of body mass index (BMI), a common confounder in studies involving PA, in this interplay.
Instrumental variables (IVs) for the main exposures in our study were extracted from summary data of UK Biobank genome-wide–association study (GWAS) on accelerometer-based PA (minimum n = 90 667; maximum n = 91 105) and self-reported PA (minimum n = 261 055; maximum n = 377 234).3,4 For these GWAS, summary statistics with and without BMI correction were obtained (see the Table for a list of PA phenotypes). Likewise, IVs for schizophrenia (40 675 cases and 64 643 controls) and BMI (n = 339 224) were extracted from their respective GWAS.5,6 Whenever possible, exposure IVs were selected among genome-wide associated variants. All original GWAS investigations were conducted with ethics committee approval. The UK Biobank studies received approval from the National Health Service National Research Ethics Service. Written informed consent was obtained from participants.
TwoSampleMR (https://github.com/MRCIEU/TwoSampleMR) was used to generate a list of linkage disequilibrium–independent IVs for each PA exposure and extract them from schizophrenia risk (outcome). LDlinkR (https://github.com/CBIIT/LDlinkR/) was implemented to find proxies with an r2 greater than 0.80 for those IVs not available in the outcome. Exposure and outcome data were harmonized. Horizontal pleiotropy was evaluated using MR-PRESSO (https://github.com/rondolab/MR-PRESSO), leading to removal of outlier IVs. The MR main analyses and sensitivity analyses were run using TwoSampleMR. The MVMR analyses, where 1 exposure (BMI) potentially mediates the association between the exposure of primary interest (PA) and the outcome (schizophrenia risk), were run in a similar fashion.
No association between PA and schizophrenia risk was observed in any of our analyses (Figure and Table). Univariate analysis with and without BMI correction provided evidence of the association of self-reported moderate/vigorous PA with increased schizophrenia risk (inverse variance–weighted and weighted median P <.05). Similar results were obtained with MVMR using BMI as covariate. Overall activity showed a similar trend in the univariate analysis, but the association was no longer significant after BMI correction.
Sensitivity analyses suggested that horizontal pleiotropy (Egger intercept P value >.05), heterogeneity (Cochran Q P >.05), or individual SNP effects (leave-one-out analyses, data not shown) were not likely to confound the results obtained for moderate/vigorous PA.
Our results suggest that PA might not have preventive effects for schizophrenia. On the contrary, moderate/vigorous self-reported PA seems to increase schizophrenia risk, results that are difficult to align with current evidence.1 Interestingly, the most beneficial effects of PA in clinical studies are found on negative symptoms, especially cognitive dysfunction, and to a lesser extent on positive symptoms, although this is still an active area of research.1 Because positive symptoms usually drive the diagnosis of samples included in schizophrenia GWAS, we hypothesize that we are (1) missing the association of PA with the cognitive/negative symptom domain, and (2) capturing a factor closely related to intense physical exercise (perhaps stress-related or personality traits) that worsens the symptomatology of psychosis. In addition, we identified BMI as a relatively modest confounder in our analyses, probably due to the properties of MR analysis.
The potential causal associations we report, or lack thereof, should be interpreted with caution given the limitations of MR and the limited number of valid IVs that can be extracted from current PA GWAS. The potential implications of our results for disease prevention policies warrant the validation of these findings in well-powered cohort studies.
Corresponding Author: Sergi Papiol, PhD, Department of Psychiatry and Psychotherapy, University Hospital, LMU Munich, Germany, Nussbaumstraße 7, 80336 Munich, Germany (sergi.papiol@med.uni-muenchen.de).
Accepted for Publication: October 21, 2020.
Published Online: December 9, 2020. doi:10.1001/jamapsychiatry.2020.3946
Author Contributions: Drs Falkai and Papiol had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Papiol, Schmitt, Rossner, Falkai.
Acquisition, analysis, or interpretation of data: Papiol, Maurus, Schulze.
Drafting of the manuscript: Papiol.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Papiol.
Obtained funding: Schulze.
Administrative, technical, or material support: Schmitt, Rossner.
Supervision: Schmitt, Rossner, Falkai.
Conflict of Interest Disclosures: Dr Schmitt has been an honorary speaker for TAD Pharma and Roche and a member of advisory boards for Roche. Dr Falkai has received grants and served as consultant, advisor, or continuing medical education speaker for the following entities: Abbott, GlaxoSmithKline, Janssen, Essex, Lundbeck, Otsuka, Recordati, Gedeon Richter, Servier, and Takeda as well as the German Ministry of Science and the German Ministry of Health. No other disclosures were reported.
Funding/Support: This research was funded by the following grants from the Deutsche Forschungsgemeinschaft: Klinische Forschergruppe 241: TP1 (SCHU 1603/5-1; BI576/5-1) and PsyCourse (SCHU 1603/7-1; FA241/16-1). Further funding was received from the German Federal Ministry of Education and Research (BMBF) through the research network on psychiatric diseases ESPRIT (grant number 01EE1407E). Drs Schmitt, Maurus, Rossner and Falkai are supported by the Else Kröner-Fresenius Foundation.
Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
1.Stubbs
B, Vancampfort
D, Hallgren
M,
et al. EPA guidance on physical activity as a treatment for severe mental illness: a meta-review of the evidence and Position Statement from the European Psychiatric Association (EPA), supported by the International Organization of Physical Therapists in Mental Health (IOPTMH).
Eur Psychiatry. 2018;54:124-144. doi:
10.1016/j.eurpsy.2018.07.004PubMedGoogle ScholarCrossref 3.Klimentidis
YC, Raichlen
DA, Bea
J,
et al. Genome-wide association study of habitual physical activity in over 377,000 UK biobank participants identifies multiple variants including CADM2 and APOE.
Int J Obes. 2018;42(6):1161-1176. doi:
10.1038/s41366-018-0120-3Google ScholarCrossref 5.Locke
AE, Kahali
B, Berndt
SI,
et al; LifeLines Cohort Study; ADIPOGen Consortium; AGEN-BMI Working Group; CARDIOGRAMplusC4D Consortium; CKDGen Consortium; GLGC; ICBP; MAGIC Investigators; MuTHER Consortium; MIGen Consortium; PAGE Consortium; ReproGen Consortium; GENIE Consortium; International Endogene Consortium. Genetic studies of body mass index yield new insights for obesity biology.
Nature. 2015;518(7538):197-206. doi:
10.1038/nature14177PubMedGoogle ScholarCrossref 6.Pardiñas
AF, Holmans
P, Pocklington
AJ,
et al; GERAD1 Consortium; CRESTAR Consortium. Common schizophrenia alleles are enriched in mutation-intolerant genes and in regions under strong background selection.
Nat Genet. 2018;50(3):381-389. doi:
10.1038/s41588-018-0059-2PubMedGoogle ScholarCrossref