A one-year prospective investigation of Type D personality and self-reported physical health

TL;DR: Although the relationships appear to be primarily driven by NA, the theory of a stress-related mechanism potentially underpinning the Type D-health relationship is supported and contributes to the literature continuing to highlight Type D personality as a risk factor for negative health outcomes.

Abstract: Objective: Type D personality is characterised by negative affectivity (NA) and social inhibition (SI), and is often associated with poorer physical and psychological health. However, the underlyin...

Summary

Introduction

• Type D personality is characterised by the interaction of negative affectivity (NA) and social inhibition (SI) (Denollet, 2000).
• Existing research has established Type D as a predictor of poorer physical health, including increased somatic symptoms, general health complaints and immune related illnesses (Condén, Leppert, Ekselius, & Åslund, 2013; Stevenson & Williams, 2014; Williams & Wingate, 2012).
• Prospective examination is required to further understand the mechanisms underpinning the now well-documented relationship.
• It has also been criticised for not accurately representing the interactive effect of SI and NA, and typologies generated from two continuous variables in this way, have been criticised (Coyne et al., 2011).
• In light of the current Type D literature it is hypothesised that: i) Type D personality will be linked to increased reporting of physical symptoms ii) Stress, anxiety and depression will play a mediating role in the relationships between Type D and physical symptoms iii).

Participants and Procedure

• An online questionnaire based study was conducted with 535 healthy individuals (18-65 years).
• Participants were recruited online via social media platforms, student participation pools and via email within the researcher’s institution.
• Exclusion criteria stipulated that individuals with a history of psychological health issues, known chronic or immune related illnesses, and diagnosed sleep disorders should refrain from taking part (including those diagnosed in the year between baseline and follow up).
• Informed consent was gained online via a multiple-choice selection and an option to provide an email address (to be contacted for the follow-up) was given at baseline.

Type D scale- 14

• The CHIPS is a list of 33 common physical symptoms (e.g. ‘back pain’; ‘diarrhoea’) rated on a 5-point Likert scale ranging from (0) ‘not been bothered by the problem’ to (4) ‘extremely bothered by the problem’ (during the previous 2 weeks).
• The total score is the sum of the responses on the 33 items (possible score range 0-132).
• Each cluster was considered individually within the current study, with the exception of haemorrhagic symptoms, given the low internal consistency of this factor (Allen et al., 2017).

Perceived Stress Scale

• Subjective stress was measured using the Perceived Stress Scale (PSS; Cohen, Kamarck, & Mermelstein, 1983).
• The PSS is a 10-item scale (α= 0.85) that assesses how respondents have experienced and dealt with stressful situations in the past month.
• Response choices are on a 5-point Likert scale and range from (0) “never” to (4) “very often” and a number of items are reverse scored.
• Scores are calculated by summing the 10 item ratings, and range from 0 to 40, with higher scores indicating higher levels of perceived stress.

Hospital Anxiety and Depression Scale

• Levels of anxiety and depression were examined using the Hospital Anxiety and Depression Scale (HADS; Zigmond & Snaith, 1983).
• The HADS comprises 14 items with answers coded between 0 and 3 (positively worded items are reversed scored).
• 7 items measure anxiety (α= 0.83) and 7 items measure depression (α=0.82), a separate score is derived for each scale with higher scores indicating higher levels of anxiety and depression, respectively.

Social Readjustment Rating Scale

• The Social Readjustment Rating Scale (SRRS; Holmes & Rahe, 1967) assessed the number of stressful life events experienced over the past year.
• The scale comprises 42 items weighted by impact (e.g. 100 points for death of a spouse, 11 points for minor violations of the law).
• The sum of the weighted items is calculated to give a total score for each participant.

Retrospective Health Questions

• A number of retrospective health questions were included in the follow-up.
• Example responses were categorical including scales of frequency (e.g. ‘never’ to ‘often’) and perceptions of quality (e.g. ‘very good’ to ‘very poor’).

Treatment of data

• At baseline, all 535 participants completed the DS14, CHIPS and HADS; however, 11 participants did not complete the PSS.
• The remainder of the participants’ data was included in analyses.
• Mediation analyses using the PROCESS macro for SPSS (Hayes & Preacher, 2013) were then used to examine if perceived stress, anxiety, depression, or stressful life events mediated the relationship between Type D and physical symptoms at follow up.
• The indirect effect shows the indirect relationship between Type D personality and CHIPS scores via each mediator (i.e. path a*b).

Temporal stability

• Pearson’s correlations analyses demonstrated strong correlations between baseline and follow-up scores; (SI; r= .857**, NA; r=.837**, Type D; r=.828**) indicating the DS14 exhibited excellent test-retest reliability, and stability over the year period.
• For the purpose of the current study Type D scores and categorisation determined at baseline were used for the remaining analyses.
• Test-retest correlations of SI, NA, Type D, anxiety, depression, perceived stress and total physical symptoms between baseline and follow up can be observed in table 2. TABLE 2 HERE.

Anxiety, depression and stress

• As shown in Table 3, Type D participants reported significantly greater levels of anxiety, depression, and perceived stress at baseline (N=535) compared with non-Type D participants.
• These findings were supported by Pearson’s correlational analyses, which indicated significant positive correlations between SI, NA, and total Type D (NA × SI) scores with anxiety, depression, and perceived stress at baseline.

Physical symptoms

• Type D participants reported significantly more physical symptoms than non-Type D participants at baseline across all symptom clusters.
• Pearson’s correlational analyses indicated significant positive correlations between SI scores (at baseline) and physical symptoms, with the exception of gastrointestinal symptoms.

Mediating pathways

• All symptom clusters at baseline significantly predicted the respective cluster at follow up (all p values ≤ .001).
• As shown in table 4, Type D personality was found to be a significant predictor of metabolic, cold/flu, and gastrointestinal symptoms (path c) one year later when baseline scores were controlled (total effect).
• The indirect effect (a*b) of Type D on metabolic symptoms (whereby the bootstrapped confidence interval for the indirect effect did not include 0) via both anxiety (BaC CI [.0017, .0070]) and stressful life events (BaC CI [.0001, .0020]) was significant.
• The direct effect (path c’) became nonsignificant when considered through the anxiety pathway (p=.130) indicating full mediation, and remained significant through the stressful life events pathway (p=.015) indicating partial mediation.
• No mediating effects of, depression, anxiety and perceived stress were observed.

Perceived general health

• Ds were more likely to rate their current general health as ‘fair’, their health over the past year as ‘not very good’, and be dissatisfied’ with their current health status.
• Ds were also less likely to provide ratings of ‘very good’ for these questions.
• There were no differences in how participants compared their current health status to one year earlier (X² (4) =7.299, p=.099).

Frequency of illnesses

• Ds were more likely to have felt unwell or run-down ‘frequently’ and were less likely to have ‘never’ suffered a non-serious illness or taken time off work.
• Type D category did not relate to the frequency of participants suffering an illness requiring prescription medication; X² (4) =2.749, p=.65.

Healthcare utilisation

• Type D was significantly associated with seeking medical information (without visiting a medical professional) (X² (4) =15.444, p=.003).
• Type D individuals were more likely to have ‘never’, or ‘once or twice’ sought medical information.
• (See Table S2 for responses to each retrospective health question for Type D and non-Type Ds).

Discussion

• The current study is the largest to date to i) consider the relationship between Type D and physical symptoms in the general population, and ii) investigate differences in aspects of retrospective health and healthcare utilization in relation to Type D status.
• This also indicates that the relationships between Type D and physical symptoms exist across both the dichotomous and dimensional conceptualisations of Type D. However, in the regression models controlling for SI and NA, Type D was only found to be predictive of total symptoms, sympathetic/cardiac, muscular, metabolic, vasovagal, and headache symptoms.
• This should also be considered when interpreting these findings, particularly with respect to the predictive utility of the Type D construct.
• Further, as stress is associated with various health outcomes, it is likely this theory could explain the observed associations with other symptom clusters.
• In summary, the current study has provided evidence that Type D personality may predict particular symptoms which are often associated with heightened stress.

Full text


 !"#$$%
%&  '  (#$  )  $#    '"$  $#  *  +#&#  ,
-./0!*$$*00."0 1*2330"//4
+#(#,5
6  $77*&7*37330//4* *143/.
8$77*&7*37330//4* *143/.9
(  %  :  :  '      
$77***77$7.0117
6%#%$!
6%#;$*$#&<&'
#%!7$#&:*&$
''$$#$'&%$#
''$#"'"$=$$
:  $  $    &  $%          '  &$
&%:#$76&$&*(
&#:#*5##
':'$'$#&* ('$#
%$77***7$*
(#>'=$%'
%:$$*(7'
$% '  $ % $; :  $
#?*!

1
A one-year prospective investigation of Type D personality and self-reported physical health
Authors: Sarah F. Allen PhD
a
, Mark A. Wetherell PhD
b
, & Michael A. Smith PhD
bc
a
Department of Health Sciences, Faculty of Science, University of York, UK. sarah.allen@york.ac.uk
b
Department of Psychology, Faculty of Health and Life Sciences, Northumbria University, UK
c
Faculty of Health and Medical Sciences, University of Western Australia, Australia
Running Head: Type D and Physical Health
Key words: Type D; health; physical symptoms; stress; anxiety.
Words: 6900
References: 79
Number of tables: 4
Number of figures: 1
Number of appendices: 1
Supplementary tables: 2
COI: No conflicts of interest.
Funding: The project was undertaken as part of a Northumbria University funded PhD studentship.
*Corresponding author: Dr Sarah F. Allen. Department of Health Sciences, Faculty of Science, University
of York, UK. sarah.allen@york.ac.uk. +44(0)1904321949

2
Abstract
Objective: Type D personality is characterised by negative affectivity and social inhibition, and is
often associated with poorer physical and psychological health. However, the underlying mechanisms
are unclear and the literature lacks longitudinal assessment. We aimed to prospectively examine the
relationships between Type D and physical symptoms, in addition to retrospective health.
Design: An online questionnaire-based study (N=535) with a one-year follow-up (N=160) was
conducted with healthy individuals (18-65years). Type D was assessed as a categorical and dimensional
construct.
Main Outcome Measures: Participants completed Type D scale-14, Hospital Anxiety and
Depression Scale, Cohen-Hoberman Inventory of Physical Symptoms and Perceived Stress Scale at both
phases. Retrospective health questions and Social Readjustment Rating Scale were completed at follow-
up.
Results: Type D was related to cardiac/sympathetic, metabolic, vasovagal, muscular, and
headache symptoms at baseline. At follow-up stressful events and anxiety mediated the relationships
between Type D and particular symptoms. Type Ds were more likely to report poorer health, and
increased minor illnesses, work absences, and medical-information-seeking.
Conclusions: Type D is associated with stress-related symptoms. Although the relationships are
primarily driven by NA, this supports the theory of a stress-related mechanism. These findings
contribute to the literature highlighting Type D as a risk factor for poor health.

3
Introduction
Type D personality is characterised by the interaction of negative affectivity (NA) and social
inhibition (SI) (Denollet, 2000). Type D individuals have the tendency to experience negative emotions
including dysphoria, anger, anxiety, hostility, and general distress across situations and time, whilst also
inhibiting the expression of these emotions in social situations due to fear of rejection or disapproval
(Denollet, 1998b; Mols & Denollet, 2010a).
Since its initial proposal the prognostic validity of the Type D construct has been demonstrated in
cardiac patient populations. However, there is accumulating evidence that Type D may be an important
risk factor for poor health in other illness groups (Mols & Denollet, 2010a), as well as in otherwise
‘healthy’ individuals (Smith et al., 2018; Williams & Wingate, 2012), that is, individuals who are free
from any chronic conditions. Existing research has established Type D as a predictor of poorer physical
health, including increased somatic symptoms, general health complaints and immune related illnesses
(Condén, Leppert, Ekselius, & Åslund, 2013; Stevenson & Williams, 2014; Williams & Wingate, 2012).
Early estimates suggest that Type D personality was prevalent in 20% of the general population
(Denollet, 2005). However, recent studies have estimated prevalence of up to 42.8% (Booth & Williams,
2015), which further exemplifies the importance of researching the health effects of Type D in ‘healthy’
populations
Type D has also been associated with increases in anxiety, depression, somatisation (Michal,
Wiltink, Grande, Beutel, & Brähler, 2011), maladaptive stress reactivity (Habra, Linden, Anderson, &
Weinberg, 2003; Howard & Hughes, 2013; Kelly-Hughes, Wetherell, & Smith, 2014; O’Leary, Howard,
Hughes, & James, 2013), dysfunctional coping strategies, lower social support (Williams & Wingate,
2012), and adverse health behaviours (Booth & Williams, 2015) in the general population. Accordingly,
these are all factors which may potentially mediate the relationship between Type D and physical health
(Howard, Hughes, & James, 2011; Williams & Wingate, 2012). The relationship between Type D and
health in the general population is beginning to receive more attention, and in a recent study, has been
found to be mediated by anxiety and perceived stress (Smith et al., 2018). However, prospective

4
examination is required to further understand the mechanisms underpinning the now well-documented
relationship.
Type D personality is traditionally assessed categorically, with individuals scoring above a
particular threshold on both SI and NA being classified as Type D (Denollet, 2005). Although this
approach is useful, it has also been criticised for not accurately representing the interactive effect of SI
and NA, and typologies generated from two continuous variables in this way, have been criticised
(Coyne et al., 2011). Consequently, Ferguson et al., (2009) has recommended that Type D may be better
conceptualised as a dimensional variable. Therefore, in line with previous studies (e.g. Stevenson &
Williams, 2014), Type D will be considered as both a categorical and a continuous variable within the
current study.
Given previous findings linking Type D to a range of health outcomes including; poor prognosis in
heart disease patients (Kupper & Denollet, 2007); cancer survivors (Mols, Denollet, Kaptein, Reemst, &
Thong, 2012) and other clinical populations (Mols & Denollet, 2010a) as well as increased physical
symptoms (Smith et al., 2018; Williams & Wingate, 2012), it appears necessary to assess the extent to
which Type D personality may predict physical health over time. Moreover, there is a notable lack of
longitudinal evidence to support the predictive value of Type D personality on health in the general
population. This makes it difficult to reliably infer cause and effect, and limits the capacity to investigate
potential mediating mechanisms (Maxwell & Cole, 2007). Therefore, a longitudinal assessment of the
associations between Type D personality and physical symptoms, in addition to aspects of general
health status and healthcare utilization, is warranted.
In light of the accumulating evidence with respect to the role of stress and distress in the Type D-
health relationship (e.g. Smith et al., 2018) this study will examine the potential mediating effects of
stress, anxiety and depression. The current study aims to contribute to our understanding of the
pathways underpinning the relationship between Type D personality and the manifestation of physical
symptoms. In light of the current Type D literature it is hypothesised that:
i) Type D personality will be linked to increased reporting of physical symptoms