Personality change predicts self-reported mental and physical health

TLDR: The findings have implications for understanding how changes in personality traits over time are related to health, and could be used to aid the development of effective health promotion strategies targeted to specific personality traits and birth cohorts.

Abstract: Objective Personality dimensions are known to predict mortality and other health outcomes, but almost no research has assessed the effects of changes in personality traits on physical and mental health outcomes. In this article, we examined the effects of changes in the Big Five personality dimensions on health as assessed by the Short Form Health Survey (SF-36). Method Respondents were 11,105 Australian adults aged 20–79 years (52.7% female). Latent difference score modeling was used to examine whether personality change over a 4-year period was associated with mental and physical health, and whether these effects were moderated by birth cohort. Results Increases in Conscientiousness and Extraversion were found to be associated with improved mental and physical health, whereas increased Neuroticism was linked with poorer health. The nature of these associations varied significantly by birth cohort. Conclusion The findings have implications for understanding how changes in personality traits over time are related to health, and could be used to aid the development of effective health promotion strategies targeted to specific personality traits and birth cohorts.

Figures

Table 2.

Figure 1. Latent difference score model examining the relationship between personality change and Physical Health. Latent difference scores were calculated for each of the five personality domains and examined simultaneously in each model. Sex and age were included as covariates, and this analytic approach was replicated for Mental Health. P1 = Parcel 1; P2 = Parcel 2; PF = Physical Functioning; RP = Role Physical; BP = Bodily Pain; GH = General Health.

Full text

*94@0<=4>C:1+:77:92:92*94@0<=4>C:1+:77:92:92
'0=0,<.3%97490'0=0,<.3%97490
,.?7>C:1(:.4,7(.409.0=&,;0<= ,.?7>C:1<>=(:.4,7(.409.0=?8,94>40=

&0<=:9,74>C.3,920;<0/4.>==071<0;:<>0/809>,7,9/;3C=4.,730,7>3&0<=:9,74>C.3,920;<0/4.>==071<0;:<>0/809>,7,9/;3C=4.,730,7>3
3<4=>:;30<#,200
*94@0<=4>C:1+:77:92:92
.8,200?:A0/?,?
&,><4.6"0,@09
?=><,74,9,>3:74.*94@0<=4>C
;30,@09?:A0/?,?
"0:940##4770<
*94@0<=4>C:1+:77:92:92
70:9408?:A0/?,?
:77:A>34=,9/,//4>4:9,7A:<6=,>3>>;=<:?:A0/?,?==;,;0<=
&,<>:1>30/?.,>4:9:88:9=,9/>30(:.4,7,9/03,@4:<,7(.409.0=:88:9=
'0.:8809/0/4>,>4:9'0.:8809/0/4>,>4:9
#,2003<4=>:;30<0,@09&,><4.6",9/#4770<"0:940#&0<=:9,74>C.3,920;<0/4.>==071
<0;:<>0/809>,7,9/;3C=4.,730,7>3
,.?7>C:1(:.4,7(.409.0=&,;0<=

3>>;=<:?:A0/?,?==;,;0<=
'0=0,<.3%974904=>30:;09,..0==49=>4>?>4:9,7<0;:=4>:<C1:<>30*94@0<=4>C:1+:77:92:92:<1?<>30<491:<8,>4:9
.:9>,.>>30*%+"4-<,<C<0=0,<.3;?-=?:A0/?,?

&0<=:9,74>C.3,920;<0/4.>==071<0;:<>0/809>,7,9/;3C=4.,730,7>3&0<=:9,74>C.3,920;<0/4.>==071<0;:<>0/809>,7,9/;3C=4.,730,7>3
-=><,.>-=><,.>
%-50.>4@0%-50.>4@0&0<=:9,74>C/4809=4:9=,<069:A9>:;<0/4.>8:<>,74>C,9/:>30<30,7>3:?>.:80=-?>,78:=>
9:<0=0,<.33,=,==0==0/>300110.>=:1.3,920=49;0<=:9,74>C><,4>=:9;3C=4.,7,9/809>,730,7>3
:?>.:80=9>34=,<>4.70A00B,8490/>300110.>=:1.3,920=49>30424@0;0<=:9,74>C/4809=4:9=:9
30,7>3,=,==0==0/-C>30(3:<>:<80,7>3(?<@0C(#0>3:/#0>3:/'0=;:9/09>=A0<0
?=><,74,9,/?7>=,20/C0,<=108,70",>09>/4110<09.0=.:<08:/07492A,=?=0/>:0B,8490
A30>30<;0<=:9,74>C.3,920:@0<,C0,<;0<4:/A,=,==:.4,>0/A4>3809>,7,9/;3C=4.,730,7>3,9/
A30>30<>30=00110.>=A0<08:/0<,>0/-C-4<>3.:3:<>'0=?7>='0=?7>=9.<0,=0=49:9=.409>4:?=90==,9/
B><,@0<=4:9A0<01:?9/>:-0,==:.4,>0/A4>348;<:@0/809>,7,9/;3C=4.,730,7>3A30<0,=49.<0,=0/
$0?<:>4.4=8A,=74960/A4>3;::<0<30,7>3)309,>?<0:1>30=0,==:.4,>4:9=@,<40/=4294E.,9>7C-C-4<>3
.:3:<>:9.7?=4:9:9.7?=4:9)30E9/492=3,@048;74.,>4:9=1:<?9/0<=>,9/4923:A.3,920=49;0<=:9,74>C><,4>=
:@0<>480,<0<07,>0/>:30,7>3,9/.:?7/-0?=0/>:,4/>30/0@07:;809>:10110.>4@030,7>3;<:8:>4:9
=><,>0240=>,<20>0/>:=;0.4E.;0<=:9,74>C><,4>=,9/-4<>3.:3:<>=
!0CA:</=!0CA:</=
.3,92030,7>3;3C=4.,7809>,7<0;:<>0/=071;<0/4.>=;0<=:9,74>C
4=.4;7490=4=.4;7490=
/?.,>4:9D(:.4,7,9/03,@4:<,7(.409.0=
&?-74.,>4:90>,47=&?-74.,>4:90>,47=
#,2000,@09&"#4770<"#&0<=:9,74>C.3,920;<0/4.>==071<0;:<>0/809>,7,9/
;3C=4.,730,7>3 :?<9,7%1&0<=:9,74>C
)34=5:?<9,7,<>4.704=,@,47,-70,>'0=0,<.3%974903>>;=<:?:A0/?,?==;,;0<=

1
Personality Change Predicts Self-Reported Mental and Physical Health
Christopher A. Magee
1,2
, Patrick C. L. Heaven
3
and Leonie M. Miller
1
1
School of Psychology, University of Wollongong.
2
Centre for Health Initiatives, University of Wollongong
3
Australian Catholic University
Corresponding Author: Dr Christopher Magee, School of Psychology, University of
Wollongong, NSW 2522, Australia. Email: cmagee@uow.edu.au; Ph: + 61 2 4221 3716;
Fax: +61 2 4221 4163
Acknowledgements
This paper uses unit record data from the Household, Income and Labour Dynamics in
Australia (HILDA) Survey. The HILDA Project was initiated and is funded by the Australian
Government’s Department of Families, Housing, Community Services and Indigenous
Affairs (FaHCSIA), and is managed by the Melbourne Institute of Applied Economic and
Social Research (Melbourne Institute). The findings and views reported in this paper,
however, are those of the authors and should not be attributed to either FaHCSIA or the
Melbourne Institute.

2
Abstract
Objective Personality dimensions are known to predict mortality and other health outcomes,
but almost no research has assessed the effects of changes in personality traits on physical
and mental health outcomes. In this article, we examined the effects of changes in the Big
Five personality dimensions on health as assessed by the Short Form Health Survey (SF-36).
Method Respondents were 11,105 Australian adults aged 20–79 years (52.7% female).
Latent difference score modeling was used to examine whether personality change over a 4-
year period was associated with mental and physical health, and whether these effects were
moderated by birth cohort.
Results Increases in Conscientiousness and Extraversion were found to be associated with
improved mental and physical health, whereas increased Neuroticism was linked with poorer
health. The nature of these associations varied significantly by birth cohort.
Conclusion The findings have implications for understanding how changes in personality
traits over time are related to health, and could be used to aid the development of effective
health promotion strategies targeted to specific personality traits and birth cohorts.
Keywords: Personality change, self-reported health, latent difference score modeling.

3
Personality Change Predicts Self-Reported Mental and Physical Health
Longitudinal research has demonstrated that personality traits are associated with a
range of health outcomes, including self-reported health, hypertension, obesity, mental
illness, and early mortality (e.g., Bogg & Roberts, 2004; Friedman et al., 1993; Hampson &
Friedman, 2008; Kern & Friedman, 2008; Kubzansky, Martin, & Buka, 2009; Martin,
Friedman, & Schwartz, 2007). In particular, Neuroticism (N) predicts poorer health, whereas
Conscientiousness (C) appears to have positive effects on health; findings for Extraversion
(E), Agreeableness (A), and Openness to Experience (O) have been less conclusive.
Most longitudinal research in this area has examined whether personality traits
assessed at a single time point predict health outcomes several years later (e.g., Friedman
et al., 1993, 1995; Hampson, Goldberg, Vogt, & Dubanoski, 2006; Shipley, Weiss, Der,
Taylor, & Deary, 2007). This approach raises a number of important theoretical issues
because it assumes that, while health has the potential for change over time, personality traits
remain largely stable. The stability of personality is consistent with traditional
conceptualizations of personality traits as reflecting enduring patterns of thoughts, feelings,
and behaviors (Roberts, Walton, & Viechtbauer, 2006). However, recent empirical and
theoretical work suggests that although personality traits are relatively stable over time, they
have potential for change and continued development during the adult life span (e.g., Roberts
& Mroczek, 2008). As elaborated below, personality change could reflect an interaction of
maturational and social/environmental influences (e.g., Roberts et al., 2006).
Personality change may have implications for understanding the relationships
between personality and health. This is because variations in traits such as N and C over time
have the potential to affect health either directly (e.g., through physiological changes) or
indirectly (e.g., via changes in health behaviors or social support), but this is not captured in

Frequently asked questions

Q1. What have the authors contributed in "Personality change predicts self-reported mental and physical health" ?

In this article, the authors examined the effects of changes in the Big Five personality dimensions on health as assessed by the Short Form Health Survey ( SF-36 ). 

Q2. What is the effect of increased N on health?

Increased N may also contribute to poorer health through reduced social support, less adaptive coping stylesbehaviors such as smoking, alcohol consumption, and physical inactivity (Lahey, 2009). 

Q3. What is the effect of increased C on mental health?

Increases in C over time may lead to improved health by promoting health-enhancing behaviors (e.g., physical activity, medication adherence, healthy eating) and reducing health-compromising behaviors such as smoking and alcohol consumption (Bogg & Roberts, 2004; Courneya & Hellsten, 1998). 

Q4. What could be the main benefits of increased C?

Increased C could also facilitate the achievement of important life goals (e.g., career success) and social functioning, which could benefit overall well-being (Hayes & Joseph, 2003). 

Q5. What subscales were used to reflect physical health?

In the analyses, the Physical Functioning, Role Physical, Bodily Pain, and General Health subscales were used to reflect Physical Health (Ware et al., 1993/2000). 

Q6. What could lead to increased socialization and thus increased social support?

Increased C could lead to health-enhancing behaviors, whereas E could lead to increased socialization and hence increased social support, all of which could benefit health. 

Q7. What is the effect of personality scarring on future health?

This scarring has consequences for future health since increases in N over time could lower one's resilience and tolerance to stressors (Ormel,emotional and physiological stress responses that may compromise their resilience to stressors and increase vulnerability to recurrent health events. 

Q8. What were the three subscales used to assess mental and physical health?

Self-reported mental and physical health were assessed as latent variables, withPhysical Health reflected by Physical Functioning, Role Physical, Bodily Pain, and General Health and Mental Health reflected by Social Functioning, Role Emotional, Mental Health, and Vitality. 

Q9. What is the common explanation for the decline of O in adulthood?

O increases in young adulthood, stabilizes somewhat during middle adulthood, and then declines in older age (Roberts & Mroczek, 2008; Specht et al., 2011). 

Q10. What could be done to improve the relationship between personality and health?

In particular, use of techniques such as growth mixture modeling to examine distinct trajectories of personality and health could provide additional and important insights into these relationships.