Neuroticism and Longevity: Mortality Research, Contradictory Findings, and Healthy Neuroticism
Updated: Oct 8
Author: Ukrainian Psychological Hub · Published: September 30, 2026 · Editorial Policy
Does neuroticism affect longevity? The best current evidence says that higher neuroticism is associated with a slightly higher average risk of premature mortality, but the relationship is small, heterogeneous, and highly sensitive to what is measured, which population is studied, and which covariates are included. A 2026 preregistered systematic review and meta-analysis combined 158 personality–mortality effect sizes from 569,859 people, 5,997,667 person-years, and 43,851 deaths across four continents. For neuroticism, the pooled association was an increased mortality hazard, but its magnitude was modest and varied across studies (McGeehan et al., 2026).
That finding does not mean that a person with a high neuroticism score will die early, that neuroticism removes a fixed number of years from life, or that a hazard ratio can be translated directly into an individual life-expectancy forecast. Mortality research is population-level survival research. Neuroticism is a dimensional personality construct, not a disease, biomarker, or diagnosis. For the construct itself, see Neuroticism: What It Is in the Big Five Personality Model.
The apparent contradictions in this literature are real rather than merely rhetorical. Large studies have reported higher mortality risk, null associations, and in some adjusted analyses lower mortality risk. More recent work helps explain why. Broad neuroticism scores combine partially distinct facets; different instruments operationalize the trait differently; age and health status can moderate associations; health behaviors and medical burden can attenuate estimates; and statistical adjustment can materially change the question being answered. The phrase “healthy neuroticism” adds another layer: it has been used for a proposed interaction between neuroticism and conscientiousness, yet a preregistered 12-cohort analysis found no evidence that this combination protects longevity (Turiano et al., 2020).
Neuroticism and longevity: the short answer
At the broad-trait level, the current evidence leans toward a small adverse association between higher neuroticism and mortality. In the 2026 meta-analysis, 46 neuroticism effect sizes produced a pooled hazard ratio of 1.03 with a 95% confidence interval of 1.02–1.04 and a prediction interval of 1.00–1.06. Between-study heterogeneity was modest but statistically detectable. Analyses designed to probe small-study bias attenuated the pooled estimate slightly rather than eliminating it (McGeehan et al., 2026).
The number 1.03 is easy to overinterpret. A hazard ratio is a relative rate from a survival model, not a statement that neuroticism shortens every person’s life by 3%, not a probability that a particular person will die, and not a fixed biological effect. Scaling also matters: mortality studies have used standardized and nonstandardized personality scores, different questionnaires, and different model specifications. The 2026 review explicitly found weaker pooled effects in studies using nonstandardized scoring than in those using standardized scoring (McGeehan et al., 2026).
The practical scientific conclusion is therefore precise: higher neuroticism is a small population-level predictor of mortality in the aggregate literature, while the size and sometimes even the direction of the association depend on measurement, sample composition, health context, and analytic decisions.
What neuroticism means in mortality research
Neuroticism is a broad personality dimension describing relatively stable individual differences in proneness to negative emotion, distress, threat sensitivity, frustration, worry, and emotional reactivity. In many Big Five models, emotional stability is the opposite pole of the same broad dimension. Some contemporary instruments instead use the label negative emotionality. These labels overlap conceptually, but scores from different models are not automatically interchangeable.
This distinction is especially important in longevity research because studies do not all use the same personality instrument. Some cohorts use Five-Factor Model measures; others use Eysenck instruments; UK Biobank studies commonly rely on a short Eysenck-derived neuroticism scale. A score from the NEO PI-R, a short Big Five inventory, the BFI-2 Negative Emotionality domain, an IPIP scale, or an Eysenck measure is meaningful only in the context of that instrument, its scoring rules, and its population. A raw score or percentile is not a universal biological dose of neuroticism.
Neuroticism is also distinct from negative affect measured at a moment in time, from trait anxiety, from current anxiety or depressive symptoms, from a screening result, and from a clinical disorder. A person can score high on neuroticism without meeting criteria for an anxiety disorder, depressive disorder, OCD, PTSD, or any other diagnosis. The broader clinical relationship belongs to Neuroticism and Mental Health: Why the Trait Is a Transdiagnostic Risk Factor; mortality studies should not be read as diagnostic studies.
What the 2026 mortality meta-analysis found
The 2026 synthesis by McGeehan and colleagues is the most comprehensive current summary of longitudinal Big Five mortality evidence. It was preregistered and pooled 158 effect sizes from 569,859 people across four continents. Neuroticism was associated with increased mortality; conscientiousness and extraversion were associated with lower mortality; agreeableness showed no overall association; and the openness result did not withstand small-study-bias adjustment (McGeehan et al., 2026).
The neuroticism effect was statistically reliable but small
For neuroticism, the pooled hazard ratio was 1.03. A trim-and-fill analysis reduced the estimate to 1.02, with a prediction interval that extended from 0.99 to 1.05. Other small-study-bias analyses continued to support a positive underlying association. This combination matters: the meta-analysis supports a real average relationship while simultaneously showing that the expected effect in a new study can be weak and may approach the null (McGeehan et al., 2026).
Age moderated the association
The same review found stronger neuroticism–mortality associations in younger samples than in older samples. In follow-up subgroup analyses, the pooled hazard ratio was approximately 1.03 in samples younger than 65 and approximately 1.01 in samples older than 65. This does not imply that neuroticism is unimportant in later life; it shows that the size of the observed association differs by baseline age and perhaps by selection, competing risks, accumulated disease, survivor effects, or other age-linked factors (McGeehan et al., 2026).
Health-related adjustment weakened the effect
Studies that adjusted for health behaviors, biological markers, or comorbidity had weaker pooled neuroticism effects. That pattern is consistent with several possibilities: part of the association may operate through health behavior or disease pathways; some covariates may confound both personality and mortality; and some adjustments may sit downstream of personality and therefore remove part of the pathway researchers are trying to estimate. Statistical attenuation alone does not identify which explanation is correct (McGeehan et al., 2026).
Why did earlier studies report contradictory results?
Before the 2026 synthesis, the field contained large, credible studies pointing in different directions. Those disagreements are scientifically informative because they reveal how much the result depends on study design.
A 2013 individual-participant meta-analysis found no robust neuroticism effect
Jokela and colleagues pooled 76,150 adults from seven cohorts with 3,947 deaths. After mutual adjustment for all Big Five traits plus demographic covariates, low conscientiousness was the only broad trait with a consistent mortality association; neuroticism was not a robust predictor across populations (Jokela et al., 2013). This study became an important reason to treat claims about neuroticism and longevity cautiously.
A 15-study coordinated analysis found higher neuroticism linked to mortality
A later coordinated analysis of 15 international longitudinal datasets, representing 44,094 participants, found that higher neuroticism and lower conscientiousness, extraversion, and agreeableness were consistent mortality predictors across studies. Smoking explained a small part of the neuroticism association, and baseline age and country helped explain variation across cohorts (Graham et al., 2017).
UK Biobank produced both risk and apparently protective estimates
In 321,456 UK Biobank participants, Gale and colleagues found that a one-standard-deviation increase in neuroticism was associated with a 6% higher all-cause mortality hazard after adjustment for age and sex. After broader covariate adjustment, especially self-rated health, the association reversed and higher neuroticism was associated with lower mortality. The lower mortality association appeared primarily among people who rated their health as fair or poor, and a worried-vulnerable facet was also linked to lower mortality (Gale et al., 2017).
A reversal after adjustment should not be translated into the claim that neuroticism protects people from death. Self-rated health is closely related to current health, symptom perception, mood, and future mortality. Adjusting for such a variable changes the estimand and may remove confounding, remove part of a causal pathway, or introduce conditioning effects. The finding is valuable precisely because it shows how fragile a simple “neuroticism is harmful” or “neuroticism is protective” narrative can become.
Longer UK Biobank follow-up again found higher mortality risk
A 17-year UK Biobank analysis of 491,323 participants and 43,400 deaths found that a one-standard-deviation increase in neuroticism was associated with a 10% higher all-cause mortality hazard after adjustment for age, sex, and ethnicity. The association was substantially attenuated after socioeconomic variables, health behaviors, chronic conditions, and depression were added; in the fully adjusted model the hazard ratio was 1.02 (Deason et al., 2025). The same cohort can therefore yield very different-looking numbers depending on follow-up duration, item construction, covariates, and the scientific question.
Another UK Biobank analysis found disease risk but lower mortality in later disease states
Huang and colleagues analyzed 483,916 UK Biobank participants using an Eysenck neuroticism measure. Higher neuroticism was associated with a range of incident diseases and with greater transition risk from healthy status to a first neuroticism-related disease and from disease to multimorbidity. Yet after extensive adjustment, moderate-to-high neuroticism was associated with lower all-cause mortality, and high neuroticism was associated with a lower transition rate from multimorbidity to death. The depressed-affect cluster looked adverse in health-to-disease transitions, whereas a sensitivity-to-environmental-stress-and-adversity cluster contributed more to the lower mortality association (Huang et al., 2025).
This result should be read as evidence of heterogeneity across health stages and components of neuroticism, not as proof that the trait is biologically protective. It comes from one very large cohort with specific measurements and model choices. The 2026 cross-study meta-analysis still finds a small adverse average association at the broad-trait level.
Broad neuroticism can hide opposite facet-level associations
A broad neuroticism score combines multiple tendencies that need not have identical relationships with health behavior or mortality. A 2023 systematic review of six prospective cohorts included 335,715 participants and found that several facets—vulnerability, cynicism, pessimism, anxious tendencies, and depressive tendencies—had been associated with increased mortality. Other facet constructions, including inadequacy and worried-vulnerable patterns, were reported as protective in specific studies. The authors concluded that broad-trait aggregation may mask important personality–health relationships (Butler et al., 2023).
Facet labels deserve careful measurement-specific reading. “Vulnerability” in a NEO instrument, a “worried-vulnerable” component extracted from UK Biobank items, and a worry cluster in an Eysenck-derived scale are not identical constructs merely because their names sound related. The mortality literature therefore cannot support a universal rule such as “worry is healthy” or “anxiety shortens life.”
This also explains why broad-trait averages and item-level results can coexist. A total neuroticism score may show a small harmful average association while one subset of items is weakly protective, another subset is more strongly adverse, and many items are close to null. Aggregation answers a different question from facet analysis.
What does “healthy neuroticism” mean?
“Healthy neuroticism” is a research hypothesis, not a diagnosis and not an established personality type that guarantees protection. In one influential formulation, the idea proposes that the consequences of neuroticism depend on conscientiousness. A person high in both traits might be vigilant about threats yet also organized enough to turn concern into preventive action. In statistical terms, the strong version of this hypothesis predicts a neuroticism-by-conscientiousness interaction.
The longevity test did not support the interaction
Turiano and colleagues tested the hypothesis in a preregistered coordinated integrative analysis of 12 cohort studies with 44,702 participants. Higher conscientiousness consistently predicted lower mortality hazard. Neuroticism showed a less consistent pattern. Most importantly, no study provided statistical evidence for the predicted neuroticism × conscientiousness interaction, and the pooled analysis did not support the idea that high conscientiousness makes high neuroticism protective for longevity (Turiano et al., 2020).
This is a useful boundary for interpreting the phrase. The concept may still be investigated for specific behaviors, disease contexts, facets, or alternative moderators, but mortality evidence does not justify telling a person that being both highly neurotic and highly conscientious is a longevity advantage. The dedicated profile question belongs to a separate reserved article in this cluster; this page addresses healthy neuroticism only where it intersects with mortality research.
Conscientiousness matters, but it does not turn neuroticism into a protective trait
Across personality and longevity research, conscientiousness has been one of the most reproducible broad-trait predictors. The 2013 individual-participant meta-analysis identified low conscientiousness as the clearest broad-trait mortality risk marker across seven cohorts (Jokela et al., 2013). The 2020 healthy-neuroticism analysis found an overall conscientiousness hazard ratio of 0.88 while failing to detect the predicted interaction with neuroticism (Turiano et al., 2020). The 2026 meta-analysis also found conscientiousness associated with reduced mortality risk (McGeehan et al., 2026).
These are separate statistical findings. A main effect for conscientiousness does not imply that neuroticism becomes beneficial when conscientiousness is high. Interaction claims require direct interaction evidence.
Neuroticism, life expectancy, and disability-free life
Mortality is not the only outcome relevant to longevity. A person can live longer with substantial disability, or have fewer disability-free years despite similar total lifespan. An individual-participant meta-analysis of 131,195 people from 10 cohorts examined both life expectancy and disability-free life. Lower emotional stability—conceptually the opposite pole of neuroticism—was associated with shorter life expectancy primarily among the lowest 15% of the emotional-stability distribution, while associations with disability extended more broadly. The authors estimated that, under a counterfactual model in which the excess risk associated with low emotional stability were absent, population life expectancy would be about 0.4 years longer and disability-free life about 2.4 years longer (Jokela et al., 2020).
Those numbers are model-based population estimates, not a calculator for individuals. They depend on the study populations, the assumed counterfactual, and the way emotional stability was modeled. They also show why “How many years does neuroticism take off your life?” is the wrong individual-level question. The literature estimates average associations and population scenarios, not a personal countdown.
How could neuroticism be related to mortality?
Several pathways are plausible, and none currently functions as a single proven mechanism. Neuroticism is associated with stress appraisal and emotional reactivity, and stress processes may shape behavior and health over long periods. For that mechanism-specific evidence, see Neuroticism and Stress: Perceived Stress, Reactivity, and Recovery. Mortality studies also point to health behavior, socioeconomic conditions, chronic disease, depressive symptoms, sleep, inflammation, blood pressure, and other intermediate variables.
Health behaviors explain some, not all, of the association
The 15-study coordinated analysis found that smoking statistically mediated a small portion of the neuroticism–mortality relationship (Graham et al., 2017). A systematic review of mediating pathways found that smoking, inflammation biomarkers, blood pressure, and sleep had been tested as pathways linking personality traits to mortality, but only seven studies met inclusion criteria and mediation results varied substantially across cohorts and methods (Grogan et al., 2024).
The evidence therefore supports a multi-pathway model more than a single mechanism. Some pathways may be behavioral, some social, some medical, some affective, and many are correlated with one another. Observational mediation also does not automatically prove a causal chain.
Mental health is related but cannot be substituted for neuroticism
Neuroticism is associated with a broad range of mental-health symptoms and disorders at the population level. Those associations can contribute to long-term health burden, but neuroticism is not equivalent to depression, anxiety, or psychiatric distress. Mortality models that adjust for depressive symptoms often estimate a different residual relationship than models that do not. This is one reason the same trait can look more or less predictive across analyses.
Disease stage may matter
The 2025 Huang study illustrates a particularly important possibility: the association may differ across transitions from health to disease, from one disease to multimorbidity, and from multimorbidity to death (Huang et al., 2025). A trait can be associated with increased disease incidence and yet with lower mortality conditional on already having multiple diseases. Those statements answer different conditional questions and are not logically contradictory.
Why statistical adjustment can reverse the result
Mortality studies routinely adjust for age and sex, and many additionally adjust for education, deprivation, smoking, alcohol, physical activity, body mass index, chronic conditions, depression, biomarkers, or self-rated health. Adding covariates can reduce confounding, but it can also remove part of a pathway through which personality is associated with mortality.
Suppose higher neuroticism is associated with smoking, and smoking contributes to mortality. A model without smoking estimates a total association that can include this pathway. A model that adjusts for smoking asks something closer to whether neuroticism predicts mortality among people with similar smoking status. Neither coefficient is automatically “the true effect”; they estimate different quantities under different assumptions.
The problem becomes even more delicate with self-rated health. Self-rated health can reflect underlying disease, symptoms, mood, vigilance, access to care, and perceptions of bodily functioning. In the 2017 UK Biobank analysis, adjustment for self-rated health was central to the reversal from higher to lower mortality hazard associated with neuroticism (Gale et al., 2017). Interpreting that reversal causally requires assumptions the observational analysis cannot fully verify.
Why age can change the neuroticism–mortality association
Age can modify observed associations for several reasons. The 2026 meta-analysis found stronger neuroticism effects in younger samples (McGeehan et al., 2026), and the 2025 17-year UK Biobank study likewise reported stronger associations among relatively younger participants (Deason et al., 2025).
One interpretation is that personality-linked behavior and stress processes have more room to accumulate before competing disease processes dominate. Another is selection: people who reach older ages while participating in longitudinal research are not a random sample of everyone born in their cohort. Survivor effects, cohort differences, measurement properties, disease prevalence, and cause-of-death composition can all change the observed association. Age moderation is therefore evidence that context matters, not proof of one specific mechanism.
Can neuroticism predict how long one person will live?
No mortality study cited here can provide a clinically valid individual lifespan prediction from a neuroticism score. A person’s lifespan is influenced by age, sex, genetics, medical conditions, smoking, alcohol use, physical activity, socioeconomic conditions, health care, accidents, infections, environmental exposures, social relationships, and many other factors. Personality contributes only a small amount of information relative to that full system.
A high score also has no universal meaning without the instrument. A NEO PI-R T-score, a BFI-2 Negative Emotionality score, an Eysenck score, and a short IPIP score have different scales, norms, item content, and psychometric assumptions. Mortality coefficients from one measure should not be pasted onto another measure as if they were doses of the same substance.
Even a statistically significant hazard ratio can have limited predictive value for an individual. Association, explanation, and prediction are different scientific tasks.
Does neuroticism cause earlier death?
The literature supports prospective association more strongly than simple causal attribution. Longitudinal designs establish that personality was measured before later mortality, but they do not eliminate confounding, measurement error, shared causes, reverse processes that begin before baseline, or the possibility that health changes personality before death becomes observable.
The 2026 meta-analysis shows that adjusting for health behaviors, biomarkers, and comorbidity weakens the association (McGeehan et al., 2026). The 2024 mediation review finds candidate pathways but inconsistent mediation evidence (Grogan et al., 2024). These findings are compatible with causal pathways, shared causes, and combinations of both. They do not justify saying that neuroticism itself directly causes premature death.
Is neuroticism ever protective for longevity?
Some individual analyses do report lower mortality at higher neuroticism levels under specific conditions. Gale and colleagues found lower adjusted mortality primarily among people reporting fair or poor health and for a worried-vulnerable facet (Gale et al., 2017). Huang and colleagues found lower mortality in fully adjusted models and a lower transition from multimorbidity to death, driven more by a sensitivity-to-environmental-stress-and-adversity cluster than by depressed affect (Huang et al., 2025).
Those findings support moderator and facet research. They do not establish a general protective effect. The 2026 meta-analysis still finds a small average association with higher mortality, and the preregistered 2020 analysis did not support the neuroticism × conscientiousness version of “healthy neuroticism” for longevity (McGeehan et al., 2026; Turiano et al., 2020).
What mortality research can and cannot tell you
Mortality research can tell us whether personality scores measured in large cohorts predict differences in survival rates over time, whether associations replicate across cohorts, how large the average effect is, which facets or moderators may matter, and whether adjustment for candidate pathways changes the estimate.
It cannot tell us that a high-neuroticism individual is destined to die younger, that worrying is biologically protective, that emotional stability guarantees longevity, or that a particular personality score should be medically treated. It also cannot replace assessment of actual health conditions. Personality is one layer of risk architecture, not a diagnosis.
For broader outcome research in this cluster, the Work and Outcomes branch is anchored by Neuroticism at Work: Stress, Behavior, and Workplace Experience. The dedicated life-satisfaction and physical-health pages remain separate canonical intents and will be linked here only after their English pages are live. This preserves the distinction between mortality/longevity, subjective well-being, and physical-health outcomes.
How to interpret a neuroticism score in light of longevity research
A personality score is best interpreted descriptively: it locates a person relative to the reference distribution used by that instrument. It can help describe patterns of emotional reactivity or negative emotionality. It does not diagnose a medical condition or produce a lifespan estimate.
If a score is high, the mortality literature does not imply that the trait itself must be “cured.” The more actionable variables in longevity research are ordinary health determinants that can be assessed directly: smoking, sleep, physical activity, blood pressure, chronic disease management, access to care, and clinically significant mental-health symptoms. The 2024 systematic review of mediating pathways reinforces that the mechanisms connecting personality and mortality are heterogeneous rather than reducible to one personality target (Grogan et al., 2024).
Likewise, a low neuroticism score should not be interpreted as a health guarantee. Lower reported distress can coexist with medical risk, delayed help-seeking, or other vulnerabilities. Personality scores provide context; they do not substitute for medical evidence.
Frequently asked questions
Does neuroticism shorten lifespan?
On average, higher neuroticism is associated with a slightly higher mortality hazard in the current meta-analytic literature. The 2026 meta-analysis estimated a pooled hazard ratio of 1.03, but effects varied across studies and were weaker after some health-related adjustments (McGeehan et al., 2026). This cannot be converted into a fixed number of years lost for an individual.
How much does neuroticism affect mortality?
The average broad-trait association is small. Study-specific estimates range from null to harmful and, in some adjusted subgroups or models, apparently protective. The magnitude depends on the personality measure, age, population, follow-up, health status, and statistical model.
Can high neuroticism help someone live longer?
Some cohorts report lower mortality under specific conditions, but the evidence does not support a general rule that high neuroticism improves longevity. The strongest current synthesis points to a small average increase in mortality risk, with important heterogeneity (McGeehan et al., 2026).
What is healthy neuroticism?
In mortality research, “healthy neuroticism” commonly refers to the hypothesis that high neuroticism may have better health consequences when conscientiousness is also high. A preregistered 12-cohort study found no evidence that this neuroticism × conscientiousness combination was protective for longevity (Turiano et al., 2020).
Is conscientiousness more important for longevity than neuroticism?
Conscientiousness has shown especially consistent mortality associations across several large studies and meta-analyses. That does not create a universal ranking of personality traits for health, because effect sizes depend on outcome and population, but the longevity evidence for conscientiousness has generally been more stable than the older neuroticism literature (Jokela et al., 2013; McGeehan et al., 2026).
Does worry protect health?
The evidence does not support a universal protective effect of worry. A worried-vulnerable facet was associated with lower mortality in one UK Biobank analysis, while systematic facet research has found both adverse and protective patterns depending on the facet and measure (Gale et al., 2017; Butler et al., 2023).
Is neuroticism the same as anxiety?
No. Neuroticism is a broad personality trait. Anxiety can refer to a temporary state, a narrower dispositional tendency such as trait anxiety, a symptom, or a clinical disorder. These constructs are related but should not be treated as interchangeable.
Does high neuroticism mean a person has a mental disorder?
No. High neuroticism is not a psychiatric diagnosis. It is a dimensional personality score. Diagnosis depends on condition-specific symptoms, duration, distress, impairment, exclusions, and clinical criteria.
Why can two neuroticism studies reach opposite conclusions?
They may use different personality instruments, age groups, follow-up periods, countries, covariates, facet definitions, health-status restrictions, causes of death, or survival models. The 2017 and 2025 UK Biobank findings show how covariate sets and analytic framing can materially change the estimated relationship even within a very large cohort (Gale et al., 2017; Huang et al., 2025).
Can a neuroticism test predict how long I will live?
No. Personality instruments are not validated as individual lifespan calculators. Their scores can describe trait variation, while longevity depends on many medical, behavioral, social, genetic, and environmental factors.
Conclusion
Neuroticism and longevity have produced one of the more instructive examples of why personality–health research requires careful interpretation. Early pooled work found no robust broad-trait association; coordinated international studies found higher mortality risk; UK Biobank analyses have produced both adverse and apparently protective estimates depending on covariates, facets, and health state; and systematic facet work shows that different components of neuroticism may point in different directions.
The 2026 preregistered meta-analysis now provides the strongest overall summary: higher neuroticism is associated with a small increase in mortality hazard on average, with stronger effects in younger samples and weaker effects in studies that adjust for health-related factors (McGeehan et al., 2026). That population-level association remains too small and context-dependent to support deterministic claims about any individual lifespan.
The “healthy neuroticism” idea does not resolve the contradiction. High conscientiousness is associated with lower mortality, but the preregistered 12-cohort longevity analysis did not find the predicted neuroticism × conscientiousness interaction (Turiano et al., 2020). The most defensible conclusion is therefore dimensional and probabilistic: neuroticism is one modest correlate of mortality risk within a much larger system of health, behavior, social context, and disease processes.
Related Articles
References
Butler, J. L., et al. (2023). Neuroticism facets and mortality risk in adulthood: A systematic review and narrative synthesis. Journal of Psychosomatic Research, 175, 111500. DOI: 10.1016/j.jpsychores.2023.111500. PubMed
Deason, K. G., Luchetti, M., Karakose, S., Stephan, Y., O’Súilleabháin, P. S., Hajek, A., Sutin, A. R., & Terracciano, A. (2025). Neuroticism, loneliness, all-cause and cause-specific mortality: A 17-year study of nearly 500,000 individuals. Journal of Affective Disorders, 368, 274–281. DOI: 10.1016/j.jad.2024.09.077. PubMed
Gale, C. R., Čukić, I., Batty, G. D., McIntosh, A. M., Weiss, A., & Deary, I. J. (2017). When is higher neuroticism protective against death? Findings from UK Biobank. Psychological Science, 28(9), 1345–1357. DOI: 10.1177/0956797617709813. PubMed
Graham, E. K., et al. (2017). Personality predicts mortality risk: An integrative data analysis of 15 international longitudinal studies. Journal of Research in Personality, 70, 174–186. DOI: 10.1016/j.jrp.2017.07.005. PubMed
Grogan, C. S., Turiano, N. A., Habenicht, A., McGeehan, M., & O’Súilleabháin, P. S. (2024). Personality traits and mediating pathways to mortality risk: A systematic review. Health Psychology, 43(3), 214–224. DOI: 10.1037/hea0001335. PubMed
Huang, S., Zhang, Y., Ma, L., Wu, B., Feng, J., Cheng, W., & Yu, J. (2025). Neuroticism is associated with future disease and mortality risks. Chinese Medical Journal, 138(11), 1355–1366. DOI: 10.1097/CM9.0000000000003503. PubMed
Jokela, M., Batty, G. D., Nyberg, S. T., Virtanen, M., Nabi, H., Singh-Manoux, A., & Kivimäki, M. (2013). Personality and all-cause mortality: Individual-participant meta-analysis of 3,947 deaths in 76,150 adults. American Journal of Epidemiology, 178(5), 667–675. DOI: 10.1093/aje/kwt170. PubMed
Jokela, M., Airaksinen, J., Virtanen, M., Batty, G. D., Kivimäki, M., & Hakulinen, C. (2020). Personality, disability-free life years, and life expectancy: Individual participant meta-analysis of 131,195 individuals from 10 cohort studies. Journal of Personality, 88(3), 596–605. DOI: 10.1111/jopy.12513. PubMed
McGeehan, M., Sutin, A. R., Gallagher, S., Terracciano, A., Turiano, N. A., Ahern, E., Kirwan, E. M., Luchetti, M., Graham, E. K., & O’Súilleabháin, P. S. (2026). Personality and mortality risk: A systematic review and meta-analysis of longitudinal data. Journal of Personality and Social Psychology, 130(4), 786–813. DOI: 10.1037/pspp0000577. PubMed
Turiano, N. A., et al. (2020). Is healthy neuroticism associated with longevity? A coordinated integrative data analysis. Collabra: Psychology, 6(1), 33. DOI: 10.1525/collabra.268. PubMed