Could individual personality traits affect adherence to exercise as a public health intervention? A causal Mendelian randomization study using United Kingdom longitudinal survey data

Cherodian, R., Franklin, M., Gschwandtner, A., Hunter, R., Jewell, S. ORCID: https://orcid.org/0000-0002-4413-6618 and Kambhampati, U. ORCID: https://orcid.org/0000-0001-5906-2394 (2026) Could individual personality traits affect adherence to exercise as a public health intervention? A causal Mendelian randomization study using United Kingdom longitudinal survey data. Humanities and Social Sciences Communications. ISSN 2662-9992 (In Press)

Abstract/Summary

Background. Public health policy and clinical guidelines worldwide have stressed the importance of exercise for health and well-being. However, an increasing evidence-base shows that personality traits can impact individual lifestyle decisions, including exercise uptake and maintenance. Neuroticism is such a personality trait, associated with negative emotions linked to ≈20% of the United Kingdom (UK) population. Our aim is to determine if neuroticism has a causal effect on exercise. We also examine if this relationship is moderated by conscientiousness: the quality of wishing to do one's work or duty well and thoroughly. Subsequently, we discuss the policy and practice implications when recommending exercise as a public health intervention to those with such personality traits. Methods. To estimate causal effects, we use Mendelian Randomisation (MR) as an Instrumental Variable (IV) approach that utilises genetic data. A criticism of MR is the potential for genetic information to be a weak instrument, thus we use the Generalised Method of Moments based Continuously Updated Estimator (CUE) to tackle this problem. METADAC UK (www.metadac.ac.uk) provided access to the UK’s ‘Understanding Society’ longitudinal survey linked to genetic data for these analyses. Results. Our results indicate that increased neuroticism levels cause a decrease in exercise. While these results are intuitive, our analyses under IV assumptions confirm that this relationship is causal. We also find that conscientiousness might moderate the effect of neuroticism, such that conscientiousness can potentially fully offset the effect of neuroticism on exercise. Conclusion. Although our analyses are restricted to a sub-population with specific genetic structures associated with neuroticism, under the assumptions of IV analysis, our results can be interpreted as causal. Our results highlight the need to tailor exercise recommendations to personality traits to increase their efficacy, thus contributing to the reduction of the economic burden of ill health associated with less active lifestyles.