Abstract
Introduction: Although observational data suggest a relationship between headache and smoking, there remain questions about causality. Smoking may increase headache risk, individuals may smoke to alleviate headaches, or smoking and headache may share common risk factors. Mendelian randomization (MR) is a method that uses genetic variants as instruments for making causal inferences about an exposure and an outcome.
Aims and Methods: First, we conducted logistic regression of observational data in UK Biobank assessing the association between smoking behaviors (smoking status, cigarettes per day amongst daily smokers, and lifetime smoking score) on the risk of self-reported headache (in the last month and for more than 3 months). Second, we used genetic instruments for smoking behaviors and headache (identified in independent genome-wide association studies [GWAS]) to perform bidirectional MR analysis.
Results: Observationally, there is a weak association between smoking behavior and experiencing headache, with increased cigarettes per day associated with increased headache risk. In the MR analysis, genetic liability to smoking initiation and lifetime smoking increased odds of headache in the last month but not odds of headaches lasting more than 3 months. In the opposite direction, there was weak evidence for higher genetic liability to headaches decreasing the chance of quitting.
Conclusions: There was weak evidence for a partially bidirectional causal relationship between smoking behaviors and headache in the last month. Given this relationship is distinct from smoking heaviness, it suggests headache and smoking may share common risk factors such as personality traits. Implications: Using MR, this study addresses the uncertainty regarding the observed relationship between headache and smoking. There was evidence for weak causal effects of smoking initiation and lifetime smoking (but not smoking heaviness) on likelihood of experiencing headache in the last month, but not over a prolonged period of more than 3 months. Those with higher genetic liability to headaches were also less likely to successfully stop smoking. This partially bidirectional causal relationship distinct from smoking heaviness suggests that observed associations are unlikely due to biological effects of tobacco smoke exposure and may be explained by shared personality traits.
| Original language | English |
|---|---|
| Pages (from-to) | 903-912 |
| Number of pages | 10 |
| Journal | Nicotine and Tobacco Research |
| Volume | 26 |
| Issue number | 7 |
| Early online date | 13 Sept 2023 |
| DOIs | |
| Publication status | Published - 31 Jul 2024 |
Data Availability Statement
Researchers can apply for access to UK Biobank, details are available at https://www.ukbiobank.ac.uk/. Genome-wide summary statistics are publicly available, details found in the referenced publications.Acknowledgements
This research has been conducted using data from UK Biobank, a major biomedical database (www.ukbiobank.ac.uk – Project 9142). We want to acknowledge the participants and investigators of the FinnGen study. Thank you to Dr Rachael Trow-Poole for her contributions, and thanks to Mark Adams and Weihua Meng for sharing the headache summary statistics.Funding
SH, HMS, GDS, MRM, and REW are part of the Medical Research Council Integrative Epidemiology Unit at the University of Bristol (MC_UU_00032/01, MC_UU_00032/07). REW is supported by a postdoctoral fellowship from the Norwegian South-Eastern Regional Health Authority (2020024). KL is a PhD fellow funded by the Gatsby Foundation. MRM is supported by the NIHR Biomedical Research Centre at the University Hospitals Bristol NHS Foundation Trust and the University of Bristol. The views expressed in this publication are those of the authors and not necessarily those of the NHS, the NIHR, or the Department of Health and Social Care.
ASJC Scopus subject areas
- General Medicine