Differences in emotional personality traits and stress between sustained hypertension and normotension

Page 1

Differences in Personality Between SustainedHypertension, Isolated Clinic Hypertension

and Normotension

JESUS SANZ*, MARIA PAZ GARCIA-VERA, INES MAGAN,REGINA ESPINOSA and MARIA FORTUN

Department of Clinical Psychology, Universidad Complutense de Madrid, Spain

Abstract

The aim of this study was to determine whether there are differences in personality between

hypertension and normotension. Forty-two male patients with essential hypertension were

divided into two groups after self-assessment of blood pressure, 18 with sustained

hypertension and 24 with isolated clinic (white coat) hypertension, and were compared

with 25 men with normotension on Spielberger’s State-Trait Anxiety Inventory and the

Jenkins Activity Survey. In line with hypotheses, the sustained hypertensive group showed

higher levels of trait anxiety, Type A behaviour pattern, and hard-driving behaviours/

competitiveness than the normotensive group, whereas isolated clinic hypertensives

occupied an intermediate position between those two groups. Results provide

support to the hypothesised relationship between personality and hypertension and stress

the need of distinguishing sustained hypertension from isolated clinic hypertension.

Copyright # 2006 John Wiley & Sons, Ltd.

Key words: essential hypertension; isolated clinic hypertension; personality; anxiety;

Type A behaviour pattern

INTRODUCTION

The role of personality in the aetiology of essential hypertension has been a major area of

research at least since the 30s–40s of the 20th century (Jorgensen, Johnson, Kolodziej, &

Scheer, 1996). Meta-analytic reviews of cross-sectional and prospective studies have

supported the hypothesis that some personality traits, such as anxiety, anger, defensiveness,

affect expression and depression, are associated with essential hypertension and are

European Journal of Personality

Eur. J. Pers. 21: 209–224 (2007)

Published online 7 September 2006 in Wiley InterScience

(www.interscience.wiley.com) DOI: 10.1002/per.605

*Correspondence to: Dr J. Sanz, Departamento de Personalidad, Evaluacion y Psicologıa Clınica, Facultad dePsicologıa, Universidad Complutense de Madrid, Campus de Somosaguas, 28223 Madrid, Spain.E-mail: [email protected]

Contract/grant sponsor: Community of Madrid; contract/grant number: 06/0028/2002.Contract/grant sponsor: Spanish Ministry of Science and Technology; contract/grant number: BSO2003-08321.

Copyright # 2006 John Wiley & Sons, Ltd.

Received 11 May 2006

Accepted 11 July 2006

Page 2

predictors of hypertension development (Jorgensen et al., 1996; Rutledge & Hogan, 2002).

However, such associations are small and highly variable, and further studies are warranted

because inconsistent results have been frequently reported. Thus, after analysing the data

of 83 studies and 25 469 participants, Jorgensen et al. (1996) concluded that their ‘results

empirically substantiate the view that research in the personality correlates of BP (blood

pressure) is marked by substantial amount of inconsistency and confusion’ (p. 310). In fact,

such inconsistencies may easily explain that, in Jorgensen et al.’s meta-analysis,

differences in negative affectivity (including anxiety, anger or depression) between higher

and lower blood pressure groups only reached a mean weighted effect size (d ) of 0.07 that,

although statistically significant, did not meet the standard for a small difference in

function of Cohen’s (1988) conventions for effect sizes (d¼ 0.20).

For example, trait anxiety is one of the personality characteristics most often implicated

as being associated with hypertension, but evidence linking trait anxiety to hypertension

remains ambiguous. Several cross-sectional, case-control studies have found that

hypertensive patients exhibit higher levels of trait anxiety than individuals with normal

BP (Calvo Frances, Dıaz Palarea, Ojeda Ojeda, Ramal, & Aleman, 2001; Cicconetti et al.,

1998); however, other studies have failed to show those differences (Steptoe, Melville, &

Ross, 1982, 1984).

Likewise, the data indicating an association between Type A behaviour pattern and

hypertension are inconsistent. In general, there are as many cross-sectional, case-control

studies indicating that normotensive and hypertensive individuals do not differ

significantly in Type A behaviour pattern (Smyth, Call, Hansell, Sparacino, & Strodtbeck,

1978; Steptoe et al., 1984; Theorell et al., 1985) as there are studies showing that

hypertensives have higher levels of Type A behaviour pattern than normotensives

(Ekeberg, Kjeldsen, Eide, & Leren, 1990; Irvine, Garner, Craig, & Logan, 1991; Steptoe

et al., 1982).

The above-mentioned inconsistencies concerning the role of trait anxiety, Type A

behaviour pattern, and, in general, of personality in hypertension are not surprising since

much of the previous work has been methodologically limited by a number of factors,

including the inherent difficulty in measuring BP and, therefore, the troubles in

categorising individuals as having hypertension, and the difficulty in controlling some

possible confounding variables (e.g. the presence of other cardiovascular disorders).

The classification of participants into hypertensive and normotensive groups is a crucial

aspect of cross-sectional, case-control studies examining the relationship of personality

and hypertension. Most studies have used casual BP measurement taken in the clinic as the

standard for that classification. However, there is ample evidence that this standard may not

be a reliable estimation of individuals’ average level of BP over a prolonged time, and that

pressures measured outside the clinic, either by self-measurement or by ambulatory

monitoring, may improve the reliability and validity of such an estimation (Garcıa-Vera,

Labrador, & Sanz, 1999; O’Brien et al., 2003). Thus, persons categorised as having

hypertension on the basis of casual clinic measurements might not be so classified if BP

readings were obtained outside the clinic. In fact, it has been estimated that as many as

25%–30% of all diagnosed patients with hypertension may have isolated clinic (or office)

hypertension (also known as white-coat hypertension), a clinical condition characterised

by persistently elevated clinic BPs in a patient with normal daytime ambulatory or self-

measured BPs (Garcıa-Vera, Sanz, & Labrador, 2000; O’Brien et al., 2003). Since evidence

to date suggests that the cardiovascular risk associated with isolated clinic hypertension is

lower than that of sustained hypertension and similar, or slightly higher, than that of true

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

210 J. Sanz et al.

Page 3

normotension (Celis & Fagard, 2004; Pickering, 2005), the potential inclusion of

individuals with isolated clinic hypertension in the comparison between persons with

hypertension and persons with normotension is an important confounding factor that might

obscure their differences in personality, and might explain the inconsistency of previous

studies.

On the other hand, many studies examining the potential relationship between

personality characteristics and hypertension have not carefully excluded participants with

other cardiovascular disorders. There is a wide literature providing evidence that

personality traits such as anxiety, Type A behaviour pattern, depression and anger are also

associated with other cardiovascular disorders, especially with coronary heart disease

(Smith & Ruiz, 2002). Therefore, the presence of persons with coronary heart disease in the

normotension group, in the hypertension group or in both groups, could significantly

confuse the relationship between personality and hypertension, and could also explain the

inconsistency of previous studies on this topic.

The aim of this study was to examine whether there are differences in trait anxiety and

Type A behaviour pattern between patients with hypertension and persons with

normotension, in a manner that addresses the concerns described above. Thus, we used

self-measured BPs to distinguish between sustained hypertension and isolated clinic

hypertension, and, second, participants with other cardiovascular disorders were excluded.

On the other hand, an important line of research that tries to resolve the confusing state of

affairs concerning the relationship between Type A behaviour pattern and cardiovascular

disorders is based on the idea, and related empirical findings, that not all components of the

Type A behaviour pattern may be related to cardiovascular disorders, and perhaps only

some components of the multidimensional Type A behaviour pattern are ‘toxic’ (Booth-

Kewley & Friedman, 1987; Dembroski, MacDougall, Costa, & Grandits, 1989; Lyness,

1993). Therefore, in addition to examine differences in a global measure of Type A

behaviour, a secondary aim of this study was to examine whether there are differences

between hypertensives and normotensives in hard-driving behaviours/competitiveness, the

only Type A component that has been found to discriminate between persons with

cardiovascular disorders and healthy persons in previous studies with Spanish samples

(Jenkins, Zyzanski, & Rosenman, 1992). Based on data from the above-mentioned

literature, we hypothesised that sustained hypertensives would be more anxious, more

Type A, and more hard-driving/competitive than normotensives, whereas patients with

isolated clinic hypertension would occupy an intermediate position between sustained

hypertensives and normotensives.

METHOD

Participants

Forty-three male patients diagnosed as having essential hypertension, by their family

physicians, volunteered to participate in a research on behavioural treatment of essential

hypertension (Garcıa-Vera, Labrador, & Sanz, 1997). Potential patients were considered

for the present study if: (a) the diagnosis was confirmed by three readings of either diastolic

BP above 90 mmHg or systolic BP above 140 during three consecutive casual BP

measurement sessions occurring over a 2- or 3-month period in the clinic; (b) they did not

suffer severe psychiatric disturbances; and (c) they did not have other cardiovascular

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

Personality in hypertension 211

Page 4

disorders. One patient did not return the personality questionnaires, and therefore the final

patient sample was composed of 42 patients with essential hypertension. Based upon the

average self-measured BPs (see procedure) and the upper limit of normal self-measured

BPs proposed by the European Society of Hypertension (O’Brien et al., 2003), patients

were divided into two groups: an isolated clinic hypertension group (self-measured BPs

<135/85 mmHg and clinic BPs � 140/90 mmHg; n¼ 24) and a sustained hypertension

group (self-measured BPs � 135/85 mmHg and clinic BPs � 140/90 mmHg; n¼ 18).

A group of men with normotension was selected from a larger sample of 268 adults

recruited by a ‘snowball’ technique in which psychology undergraduates invited their

relatives to voluntarily participate in a research on personality and high BP. For the present

study, we selected all men out of that sample that met the following criteria: (a) they

reported not having hypertension, not having a history of hypertension, and not taking

antihypertensive medication; (b) their normotension was confirmed by an average clinic

BP below 140/90 mmHg and an average self-measured BP below 135/85 mmHg (see

procedure); (c) they reported not having heart troubles, atherosclerosis or circulation

troubles, and not having had a stroke and (d) their ages ranged from 26 to 69 years

(precisely the age range of the hypertensive patients). Twenty-five men fulfilled these

criteria and formed the normotension group. A break down of demographic and clinical

characteristics of the sample of participants for the two hypertension groups and the

normotension group is shown in Table 1.

Apparatus and materials

BP measurement

For patients, clinic BP readings were taken with a RIESTER NOVA mercury

sphygmomanometer. On the other hand, each patient was given an OMRON HEM-

403C digital BP monitor to self-record his/her BP and heart rate. This semi-automatic

device uses an oscillometric method to measure both BP and heart rate. For normal

participants, clinic BP readings were taken with an OMRON 705IT digital BP monitor, and

this same device was given to them to self-record their BP and heart rate. The OMRON

705IT is an automatic device that also uses an oscillometric method to measure both BP

and heart rate. A number of studies support the accuracy of these two digital devices to

measure BP (El Assaad, Topouchian, & Asmar, 2003; Imai et al., 1989; Garcıa-Vera et al.,

1999).

Psychological measurements

State-Trait Anxiety Scale (STAI; Spielberger, Gorsuch, & Lushene, 1970). The STAI

comprises separate 20-item self-report scales for measurement of two distinct anxiety

concepts: trait anxiety and state anxiety. In the present study, we used the trait anxiety

scale. High scores in this scale indicate a greater tendency to respond to situations

perceived as threatening with anxiety. We used the Spanish version of the STAI

(Spielberger, Gorsuch, & Lushene, 1988). Although this version uses for each item a 0–3

response scale instead of the 1–4 response scale of the original, this change does not affect

the reliability or validity of the STAI. For example, with adult samples from the Spanish

general population, Cronbach’s alpha coefficients ranging between 0.84 and 0.86 have

been reported for the Trait Anxiety scale, indicating adequate internal consistency

(Spielberger et al., 1988). In the present sample of participants, the Cronbach’s alpha

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

212 J. Sanz et al.

Page 5

Tab

le1

.D

emo

gra

ph

ican

dcl

inic

alch

arac

teri

stic

so

fsu

stai

ned

hy

per

ten

sive,

iso

late

dcl

inic

hy

per

ten

sive

and

no

rmo

ten

sive

par

tici

pan

ts

Mea

sure

Gro

up

so

fp

arti

cip

ants

p-v

alu

e

12

3

AN

OV

Aover

all

Stu

den

t’st-

test

Su

stai

ned

hy

per

ten

sive

(n¼

18

)Is

ola

ted

clin

ich

yp

erte

nsi

ve

(n¼

24

)N

orm

ote

nsi

ve

(n¼

25

)1

ver

sus

21

ver

sus

32

ver

sus

3

Ag

e(y

ears

)4

5.5

(9.1

)4

5.1

(9.1

)4

3.3

(12

.1)

n.s

.M

easu

rem

ent

atcl

inic

Sy

sto

lic

blo

od

pre

ssu

re(m

mH

g)

15

1.7

(9.3

)1

48

.9(7

.8)

12

0.0

(10

.0)

<0

.00

1n

.s.

<0

.00

1<

0.0

01

Dia

sto

lic

blo

od

pre

ssu

re(m

mH

g)

99

.7(7

.1)

99

.1(6

.9)

71

.3(8

.1)

<0

.00

1n

.s.

<0

.00

1<

0.0

01

Sel

f-m

easu

rem

ent

ath

om

e/w

ork

Sy

sto

lic

blo

od

pre

ssu

re(m

mH

g)

13

9.4

(9.0

)1

23

.3(6

.1)

11

9.5

(7.8

)<

0.0

01

<0

.00

1<

0.0

01

n.s

.D

iast

oli

cb

loo

dp

ress

ure

(mm

Hg

)9

1.0

(4.3

)7

8.6

(4.6

)7

2.6

(6.2

)<

0.0

01

<0

.00

1<

0.0

01

<0

.00

1H

eart

rate

(bea

ts/m

in)

76

.6(1

0.0

)6

8.6

(7.9

)6

8.7

(7.7

)<

0.0

06

<0

.00

1<

0.0

12

n.s

.D

ura

tio

no

fh

yp

erte

nsi

on

(mo

nth

s)7

7.9

(68

.3)

61

.6(6

7.7

)0

n.s

.T

akin

gan

tihyper

tensi

ve

med

icat

ions

(%)

61.1

%83.3

%0%

n.s

.�

� p-v

alu

eo

fch

i-sq

uar

ete

st;

ns,

no

tsi

gn

ifica

nt

(p>

0.0

5).

Dat

aar

em

ean

s(SD

bet

wee

npar

enth

eses

)unle

ssoth

erw

ise

indic

ated

.

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

Personality in hypertension 213

Page 6

coefficient for the STAI was 0.87, similar to those obtained in the Spanish normative

samples.

Jenkins Activity Survey, Form C (JAS; Jenkins, Zyzanski, & Rosenman, 1979). The JAS

is a 52-item multiple-choice indicator of Type A-related behaviours. Its four scales give an

overall measurement of Type A behaviour pattern and specific measurements of three of its

components (hard-driving behaviours/competitiveness, speed and impatience and job

involvement), although we only used the Type A behaviour pattern scale (Type A-JAS) and

the hard-driving behaviours/competitiveness scale (Factor H-JAS). We used the Spanish

version of the JAS (Jenkins et al., 1992) whose psychometric properties are similar to those

showed by the original one, with Cronbach’s alpha coefficients of internal consistency of

0.86 for the Type A-JAS scale and of 0.87 for Factor H-JAS scale (Jenkins et al., 1992).

Surprisingly, in the participant sample of this study, the Cronbach’s alpha coefficients of

the Factor H-JAS and Type A-JAS scales were low (0.46 and 0.61, respectively), although

the latter exceeded the threshold of 0.60 proposed by Nunnally (1978) as an acceptable

reliability level for research purposes.

Procedure

Before a patient was referred to Garcıa-Vera et al.’s (1997) study and over a 2- or 3-month

period, three clinic measurements of BP (each measurement being the average of 2–3 clinic

BP readings) were taken by two experienced nurses in the health centre that patients

attended regularly. For patients, clinic BP was the average of those three measurements.

After being referred by their physicians, patients were invited to a two-session individual

assessment of their hypertension. After giving written informed consent for participation,

in the first session patients completed an interview that assessed several variables related to

their hypertension problem (e.g. duration of hypertension, previous treatments). Then,

patients were carefully instructed on how to self-measure and self-record BP readings

correctly, and carried out several practice trials. Patients were given self-recording sheets

that included reminders on how to carry out self-measurement of BP correctly. Patients

were asked to self-measure their BP and heart rate on three occasions per day for 16 days:

two times at home (when getting up in the morning and before bedtime) and one at work. In

each occasion, they obtained one reading. For patients, self-measured BP was the average

of those 48 home and work BP readings. This average allowed to obtain reliable estimates

of BP, with dependability coefficients (G�, analogous to reliability coefficients) higher than

0.90 (Garcıa-Vera & Sanz, 1999). In fact, in this sample, the intra-class correlations (ICC)

for a one-way random effects model (McGraw & Wong, 1996) were 0.98 both for systolic

and for diastolic BP self-measurements. Finally, patients received the JAS to be completed

at home. The study therapist reviewed the specific instructions of the JAS with patients, and

these were asked to schedule a time during the following week to complete the instrument

in a quiet room, free from interruption, ‘so that you can give your entire concentration to fill

out the questionnaire’. The second session of assessment was held 8 days after the first one.

After a psychophysiological assessment, patients received the STAI, and instructions for

BP self-measurement and for completing questionnaires at home were reviewed.

Data from normotensive participants were taken from a larger research on personality

and hypertension. After giving written informed consent for participation, participants

completed an interview that assessed their cardiovascular health and several risk factors

associated with hypertension, and also filled out several personality questionnaires,

including the trait scale of the STAI and the Type A behaviour pattern and hard-driving

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

214 J. Sanz et al.

Page 7

behaviours/competitiveness scales of the JAS. In addition, six clinic BP readings were

taken by the experimenter with a 2-minute interval between readings. For normotensive

participants, clinic BP was the average of those six BP readings. The ICCs for systolic and

diastolic clinic BP readings were 0.95 and 0.97, respectively. Then, participants were

carefully instructed on how to self-measure and self-record BP readings correctly, and they

carried out several practice trials. These participants were asked to self-measure their BP

two consecutive times (with a 2-minute interval between measurements) on three

occasions per day (on the same occasions as the self-measurement of the patients) for

6 days (36 BP measures in total). In addition, participants were also given self-recording

sheets that included reminders on how to carry out self-measurement of BP correctly. For

normotensive participants, self-measured BP was the average of those 36 home and work

BP readings. This average allowed to obtain reliable estimates of BP, with G� coefficients

higher than 0.90 (Sanz, Garcıa-Vera, Risco, & Magan, 2006). In fact, in this normotensive

sample, the ICCs were 0.97 and 0.98 for systolic and diastolic BP self-measurements,

respectively.

Statistical analyses

Demographic, clinic and personality variables were analysed by one-way ANOVAs with

participant group (sustained hypertensives, isolated clinic hypertensives and normoten-

sives) as between-subjects factor, except for the percentage of medicated patients that was

analysed by Pearson chi-square. When the variances of the dependent variables were not

equal across groups, we used the Welch test as a more robust and conservative alternative to

the usual F-test. When the ANOVA F-tests (or the Welch tests) revealed statistically

significant differences between groups, pairwise comparisons of the group means were

conducted by Student’s t-tests. For personality variables, those t-tests were computed with

Bonferroni adjusted p-values, and effect sizes for pairwise comparisons were measured by

Cohen’s d statistic (the difference between the means divided by the pooled standard

deviation).

RESULTS

Demographic and clinical characteristics

The demographic and clinical characteristics of the three participant groups (sustained

hypertensives, isolated clinic hypertensives and normotensives) are displayed in Table 1.

There were no significant differences between the three participant groups in age (Welch

statistic (2, 40.6)¼ 0.25, n.s.), but, as expected, the three groups significantly differed in

their BP and heart rate measures (all F-tests with p< 0.05). The subsequent t-tests revealed

that sustained and isolated clinic hypertensives did not significantly differ in clinic systolic

and diastolic BPs (both t-tests n.s.), but that both hypertensive groups showed higher levels

of clinic systolic and diastolic BPs than did the normotensive group (all t-tests with

p< 0.05). In contrast, the t-tests also revealed that the sustained hypertensive patients

showed higher levels of self-measured systolic and diastolic BPs and of self-measured

heart rate than did isolated clinic hypertensive and normotensive groups (all t-tests with

p< 0.05), and that these latter groups did not significantly differ in self-measured systolic

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

Personality in hypertension 215

Page 8

BP and heart rate (all t-tests n.s.), but they significantly differed in self-measured diastolic

BP [t (47)¼ 3.82, p< 0.001].

On the other hand, there were no significant differences between the sustained

hypertensive and the isolated clinic hypertensive groups in such clinical variables as

duration of hypertension [t (40)¼ 0.77, ns] or percentage of patients taking

antihypertensive medication [chi-square (1, N¼ 42)¼ 2.63, ns].

Personality characteristics

The personality test score means of the three participant groups are presented in Table 2.

Given that the presence of outliers can lead to errors and distortions when using either

parametric or nonparametric tests, we checked for the presence of personality scores

deviated three or more standard deviations from the mean. We did not found any outliers

and, therefore, we conducted one-way ANOVAs for each personality variable. Group

differences on all personality variables were statistically significant (all F-tests with

p< 0.05). Using Bonferroni adjusted p-values, the subsequent t-tests revealed a same

pattern of differences between groups for all personality measures: sustained hypertensives

significantly showed higher levels of trait anxiety, Type A behaviour pattern and hard-

driving behaviours/competitiveness than did the normotensives (all t-tests with p< 0.05),

but both participant groups did not significantly differ from the isolated clinic hypertensive

group. Thus, although isolated clinic hypertensives tended to show lower levels of trait

anxiety, Type A behaviour pattern and hard-driving behaviours/competitiveness than did

the sustained hypertensives (see Table 2), these differences did not reach statistical

significance (all t-tests n.s.). Likewise, although isolated clinic hypertensives tended to

show higher levels of the above-mentioned personality characteristics than did the

normotensives (see Table 2), these differences did not also reach statistical significance (all

t-tests n.s.).

Interestingly, the personality differences between sustained hypertensives and

normotensives not only were statistically significant, but also they were large in terms

of effect size, with d statistics ranged from 0.83 for trait anxiety to 0.90 and 0.92 for Type A

behaviour pattern and hard-driving behaviours/competitiveness, respectively.

Given that isolated clinic hypertensives and normotensives significantly differed in self-

measured diastolic BP, we carried out ANCOVAs with isolated clinic hypertensives versus

normotensives as between-subjects factor and self-measured diastolic BP as covariate, for

each personality variable. The ANCOVAs revealed similar results to those obtained with

the t-tests.

Both JAS scales (Type A-JAS and Factor H-JAS) were significantly intercorrelated

(r¼ 0.60, p< 0.001), whereas the trait anxiety subscale of the STAI did not correlate

significantly with the Type A-JAS (r¼�0.05, n.s.), but it did with the Factor H-JAS

(r¼ 0.25, p< 0.04). Given these significant correlations, to determine whether the

differences in trait anxiety, Type A behaviour pattern and hard-driving behaviours/

competitiveness found in the comparison of sustained hypertensives with normotensives

could be explained by only a subset of these variables, a stepwise discriminant analysis was

conducted. The stepwise discriminant analysis selects the variable that accounts for the

largest amount of variance in defining group membership and, then, includes in the

discriminant function each subsequent variable only if it contributes to additional unique

variance between groups. The stepwise discriminant analysis provided a canonical

discriminant function that significantly distinguished between sustained hypertensives and

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

216 J. Sanz et al.

Page 9

Tab

le2

.P

erso

nal

ity

char

acte

rist

ics

of

sust

ain

edh

yp

erte

nsi

ve,

iso

late

dcl

inic

hy

per

ten

sive

and

no

rmo

ten

sive

par

tici

pan

ts

Mea

sure

Gro

up

so

fp

arti

cip

ants

p-v

alu

e

12

3

AN

OV

Aover

all

Stu

den

t’st-

test�

Su

stai

ned

hy

per

ten

sive

(n¼

18

)

Iso

late

dcl

inic

hy

per

ten

sive

(n¼

24

)N

orm

ote

nsi

ve

(n¼

25

)1

ver

sus

21

ver

sus

32

ver

sus

3

Sta

te-T

rait

An

xie

tyS

cale

(ST

AI)

—T

rait

An

xie

ty2

4.8

(11

.0)

19

.5(9

.0)

16

.9(7

.8)

<0

.02

4n

.s.

<0

.00

7n

.s.

Jen

kin

sA

ctiv

ity

Su

rvey

(JA

S)

Ty

pe

A2

72

.3(5

9.0

)2

33

.1(7

9.8

)2

07

.1(8

2.9

)<

0.0

33

n.s

.<

0.0

09

n.s

.H

ard

-Dri

vin

g/c

om

pet

itiv

efa

cto

r(F

acto

rH

)1

35

.6(3

1.7

)1

20

.6(2

9.0

)1

06

.6(3

1.3

)<

0.0

14

n.s

.<

0.0

04

n.s

.

� On

lyp-v

alu

essi

gn

ifica

nt

afte

rB

on

ferr

on

ico

rrec

tio

n(0

.05

/3¼

0.0

16

7)

are

dis

pla

yed

;n

.s.,

no

tsi

gn

ifica

nt

afte

rB

on

ferr

on

ico

rrec

tio

n(p>

0.0

16

7).

Dat

aar

em

ean

s(SD

bet

wee

np

aren

thes

es)

un

less

oth

erw

ise

ind

icat

ed.

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

Personality in hypertension 217

Page 10

nomortensives (Wilks’ lambda¼ 0.63; chi-square¼ 16.48, p< 0.001), allowed one to

correctly classified the 75.6% of the cases, and only included two variables: hard-driving

behaviours/competitiveness and trait anxiety. The standardised coefficients of the

canonical discriminant function indicated that hard-driving behaviours/competitiveness

(0.81) accounted for larger proportion of the variance than trait anxiety (0.58). The fact that

the second variable selected for entry was trait anxiety indicated that the association of

higher trait anxiety with sustained hypertension was not entirely explained by the

significant but weak correlation between the trait anxiety scores and the hard-driving

behaviours/competitiveness scores (r¼ 0.25).

On the other hand, we found low internal consistency indices for the two JAS scales.

Therefore, it may be that the observed between-group differences in these scales provides a

misestimation of the true association of hypertension with Type A behaviour pattern and

hard-driving behaviours/competitiveness. To obtain more reliable measures, we eliminated

the JAS items with low correlations with the total adjusted score of the corresponding scale,

and created two measures of Type A behaviour pattern and hard-driving behaviours/

competitiveness with good internal consistency indices (equal or above the standard of

0.70 for good internal consistency; Nunnally, 1978). Thus, four items from the original

Type A-JAS and nine items from the original Factor H-JAS were eliminated, and both the

resultant 17-item Type A scale and the resultant 11-item Factor H scale reached Cronbach’s

alpha coefficients of 0.70. With these two new scales, we repeated all the statistical

analyses. The results were almost identical to those obtained with the original JAS scales.

DISCUSSION

The aim of this study was to examine whether there are differences in trait anxiety, Type A

behaviour pattern and hard-driving behaviours/competitiveness (a component of Type A

behaviour pattern), between hypertension and normotension, especially when hyperten-

sion is diagnosed based on both clinic and self-measured BPs and, therefore, isolated clinic

hypertension can be detected and controlled. Our first hypothesis was that, in comparison

with normotensives, true hypertensives (sustained hypertensives) could be characterised by

higher levels of those personality traits. Our results revealed statistically significant

differences between sustained hypertensives and normotensives in trait anxiety, Type A

behaviour pattern and hard-driving behaviours/competitiveness, with higher scores in these

personality traits for sustained hypertensives. In fact, these differences in personality

between normotensives and sustained hypertensives were large in function of Cohen’s

(1988) conventions for effect sizes (d> 0.80).

Based on data that indicate that persons with isolated clinic hypertension have a worse or

equal cardiovascular prognosis than normotensives, but a better one than those with

sustained hypertension (Celis & Fagard, 2004; Pickering, 2005), we also hypothesised that

isolated clinic hypertensives would occupy an intermediate position between sustained

hypertensives and normotensives in trait anxiety, Type A behaviour pattern and hard-

driving behaviours/competitiveness. However, this hypothesis was not fully supported by

our results. Although, isolated clinic hypertensives showed higher levels of those

personality characteristics than normotensives, and they also showed lower levels of those

personality characteristics than sustained hypertensives, all these between-group

differences were not statistically significant.

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

218 J. Sanz et al.

Page 11

In sum, our results support our first hypothesis concerning differences in personality

between normotension and hypertension, and are consistent with those obtained in

previous studies that also support the role of trait anxiety and Type A behaviour pattern in

hypertension (Calvo Frances et al., 2001; Cicconetti et al., 1998; Ekeberg et al., 1990;

Irvine et al., 1991; Steptoe et al., 1982).

Of course, as mentioned in the introduction of this article, there are studies that have

failed to show differences in trait anxiety and Type A behaviour pattern between persons

with normotension and patients with essential hypertension (Smyth et al., 1978; Steptoe

et al., 1982, 1984; Theorell et al., 1985). However, most of these studies (Smyth et al.,

1978; Steptoe et al., 1982, 1984; Theorell et al., 1985) did not used, along with clinic BPs,

out-of-clinic BPs (self-measured or ambulatory BPs) to diagnose hypertension and,

therefore, they were not able to detect and control isolated clinic hypertension. As our

results indicate, the levels of trait anxiety and Type A behaviour pattern of isolated clinic

hypertensives are not statistically distinguishable from those of sustained hypertensives

and normotensives, occupying an intermediate position between these latter groups.

Therefore, the differences in personality between normotension and hypertension could be

obscured by the unnoticed inclusion of isolated clinic hypertensives in the sustained

hypertensive group.

Although sustained hypertensives significantly differed from normotensives on all

personality variables examined in the present study, the results of the discriminant analysis

indicated that trait anxiety and hard-driving behaviours/competitiveness (a component of

Type A behaviour pattern) were the most important personality characteristics to

discriminate between sustained hypertension and normotension, and even though there was

some correlation between the measures of both personality traits, these contributed

independently to that discrimination.

In sum, our results suggest that sustained hypertensives are more prototypical of the

‘hypertensive personality’ than patients whose BP is different in different settings (isolated

clinic hypertension). This difference could explain why research literature on the relations

between personality and hypertension has yielded mixed findings and, although suggestive,

it is far from conclusive. Further studies on this issue should, therefore, distinguish

different types of hypertensive patients based on the observed discrepancies between clinic

and nonclinic measures of BP, and examine differentially not only patients with isolated

clinic and sustained hypertension, but also subjects with normal clinic BP but high self-

measured or ambulatory BPs (which has been termed ‘isolated clinic normotension’,

‘white coat normotension’ or ‘masked hypertension’).

Furthermore, from a clinical viewpoint, there is the need for taking BP readings away

from the medical environment, by self-measurement of BP in the home or work place or by

ambulatory BP measurement using automated devices, in order to identify isolated clinic

hypertension, especially in those cases where the probability of isolated clinic hypertension

is high (clinic systolic BP 140–159 mmHg or diastolic BP 90–99 mmHg, female sex, non-

smokers, hypertension of recent onset, limited number of clinic BP measurements;

O’Brien et al., 2003). Given the high incidence and low risk profile of isolated clinic

hypertension, the consequences of failing to identify this condition are considerable,

because life-long treatment may be prescribed unnecessarily, and if antihypertensive

medication is given to people whose usual BP is normal, they may be made unwell by the

adverse effects of medication. Moreover, given that our results suggest that ‘toxic’

personality traits are more present in sustained hypertension than in isolated clinic

hypertension, it may be that antihypertensive psychological treatments aimed to change

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

Personality in hypertension 219

Page 12

those personality traits could only benefit to sustained hypertensives, and could be

unnecessary for isolated clinic hypertensives.

On the other hand, it has been hypothesised that patients with isolated clinic

hypertension would be generally more anxious and that it would be their anxiety associated

with clinic visit that would be responsible for elevated clinic BPs (Pickering & Friedman,

1991). We found that isolated clinic hypertensives and sustained hypertensives did not

differ significantly in scores on a well-standardised, widely-used, psychometric test

measuring trait anxiety (the trait anxiety scale of the STAI; see Table 2). Although a

nonsignificant trend was found among isolated clinic hypertensives to show higher levels

of trait anxiety than normotensives, a nonsignificant trend was also found among isolated

clinic hypertensives to show lower levels of trait anxiety than sustained hypertensives.

Therefore, it is clearly safest to conclude that trait anxiety does not play a major role in the

phenomenon of isolated clinic hypertension in light of previous literature showing no

group differences on different anxiety measures (Schneider, Egan, Johnson, Drobny, &

Julius, 1986), including the measure employed in the present study (Gerardi, Blanchard,

Andrasik, & McCoy, 1985; Siegel, Blumenthal, & Divine, 1990).

In conclusion, our results do not support the hypothesis that isolated clinic hypertension

is related to trait anxiety. However, our data do not preclude the possibility that trait anxiety

is related to the white-coat effect. Although isolated clinic hypertension is frequently

confounded with the white-coat effect, they are different phenomena. The white-coat effect

refers to an alerting reaction of the patient to the measurement of BP in the clinic

environment that can be quantified by continuous BP monitoring before and during the

physician’s visit. On the other hand, isolated clinic hypertension is a clinical condition

characterised by persistently elevated clinic BP and normal daytime ambulatory or self-

measured BPs. For a long time, it was assumed that this condition could reflect the

persistence over time of an alerting reaction to the physician’s visit. However, results from

several studies indicate that the difference between office and ambulatory or self-measured

BP values does not correlate with the BP rise induced in the patient by the physician’s visit

(Lantelme, Milon, Vernet, & Gayet, 2000; Palatini et al., 2003). In fact, these results are the

reason for the recommendation, given by the World Health Organization-International

Society of Hypertension (1999), that the condition characterised by persistently high clinic

BP and by persistently normal daytime ambulatory or self-measured BP values should be

named ‘isolated clinic (or office) hypertension’ instead of using the appealing, but

misleading, term ‘white coat hypertension’ (see also Garcıa-Vera et al., 2000). Therefore, it

may be that the white-coat effect could be partially explained by an exaggerated pattern of

anxiety (trait anxiety) to novel or stressful stimuli, whereas isolated clinic hypertension

could be explained by other multiple factors including, for example, factors affecting both

the BP measured in the physician’s office and the BP recorded in real-life conditions.

Future studies should test those hypotheses to determine the conditions and mechanisms

underlying the white-coat effect and isolated clinic hypertension, and, in any case, these

two conditions need to be separately considered to avoid further misunderstandings that

have often occurred in previous literature.

Finally, there are some limitations of this study that should be considered. First, this is a

cross-sectional study and, therefore, its design does not allow to decide whether trait

anxiety, Type A behaviour pattern and hard-driving behaviours/competitiveness are

causing hypertension or vice versa. Further studies should use prospective designs to test

whether, in normotensives, high levels of trait anxiety, Type A behaviour pattern and hard-

driving behaviour/competitiveness predict later incidence of hypertension. Second, this

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

220 J. Sanz et al.

Page 13

study focused on men and, obviously, this may limit the generalisation of our findings to

women. Therefore, further research is needed to replicate our findings with samples of

hypertensive women. This need is more apparent in the light of findings suggesting that: (1)

hypertension is a growing problem in women, particularly after the fifth to sixth decade of

life (Oparil & Miller, 2005); (2), although no one group seems to be exempt from isolated

clinic hypertension, the probability increases in women (O’Brien et al., 2003) and (3) the

personality traits may be differently related to hypertension in men and women. For

example, although Type A males show fairly consistent cardiovascular responsiveness to

stressful challenges or competition, the findings in women are more equivocal (Matthews

& Haynes, 1986). In a related vein, Type A men, but not women, show elevated

cardiovascular responses during a challenging reaction time test; however, Type A women

show greater cardiovascular increases during a verbally challenging interaction

(MacDougall, Dembroski, & Krantz, 1981). Finally, Type A seems to have different

personality correlates in men and women. For example, women show higher correlation

between Type A and guilt, and men between Type A and suspiciousness (McCann,

Woolfolk, Lehrer, & Schwarcz, 1987). However, there are also a number of empirical

reasons to suppose that personality traits are not differently related to hypertension in men

and women. Perhaps, the most compelling reason is that previous meta-analytic studies on

the relationship between personality and hypertension have found that gender was

unrelated to the magnitude and direction of effect sizes reflecting the association between

personality traits and blood pressure (Jorgensen et al., 1996; Rutledge & Hogan, 2002).

Third, for some comparisons (between sustained hypertensives and isolated clinic

hypertensives, and between isolated clinic hypertensives and normotensives), it may be

that the present study had low statistical power since the three participant groups were

relatively small. Post-hoc analysis of statistical power revealed that, for the

comparisons between sustained hypertensives and isolated clinic hypertensives,

statistical power indices were 0.45 for hard-driving behaviour/competitiveness, 0.51

for trait anxiety, and 0.55 for Type A behaviour pattern. For comparisons between

isolated clinic hypertensives and normotensives, statistical power indices were 0.28 for

trait anxiety, 0.29 for Type A behaviour pattern, and 0.46 for hard-driving behaviour/

competitiveness. These data suggest that further studies should recruit between 84

and 124 participants to have a probability of 80% of correctly rejecting the null

hypothesis of absence of personality differences between sustained hypertensives and

isolated clinic hypertensives, whereas they should recruit between 120 and 318

participants for having a probability of 80% of correctly rejecting the null hypothesis

of absence of personality differences between isolated clinic hypertensives and

normotensives.

On the other hand, further studies would also benefit from the use of a more

comprehensive overarching theoretical framework in conceptualising the relationship

between psychological traits and hypertension. This meta-model would provide a matrix of

interrelated conceptual personality domains to guide research on the relationships between

personality traits and hypertension and to improve conceptual clarity. One such promising

framework is the five-factor model (or Big Five), arguably the most popular and influential

contemporary taxonomy of personality traits. This model suggest that the five dimensions

of neuroticism or emotional stability, extraversion, openness to experience or culture,

agreeableness and conscientiousness summarise and integrate the majority of personality

traits (McCrae & Costa, 1997; Sanz, Silva, & Avia, 1999). Interestingly enough, a number

of studies have already tested the personality dimensions of neuroticism and extraversion

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

Personality in hypertension 221

Page 14

for association with hypertension (Hozawa et al., 2002; Kohler, Scherbaum, Richter, &

Bottcher, 1993).

ACKNOWLEDGEMENTS

Preliminary analyses based on a subsample of participants of this study were presented as a

poster in the 17th Annual Conference of the European Health Psychology Society (Kos,

Greece; September 24–27, 2003). We thank the other members of the Complutense

University research group on personality and hypertension (Francisco Calabozo, Drissa

Delkader, Marıa Esther Garcıa, Almudena Gomez, Marıa Gomez, Beatriz Huertas, Amaia

Izquierdo, Jose Enrique Lopez, Carolina Munoz, Patricia Pastor, Carolina Perandones,

Elvira Polo, Mercedes Risco and Coral Rojo) for their assistance on the assessment of

normotensive participants.

REFERENCES

Booth-Kewley, S., & Friedman, H. S. (1987). Psychological predictors of heart disease: Aquantitative review. Psychological Bulletin, 101, 343–362.

Calvo Frances, F., Dıaz Palarea, D., Ojeda Ojeda, B., Ramal, J., & Aleman, S. (2001). Diferencias enansiedad, depresion, estres y apoyo social entre sujetos hipertensos y normotensos [Differences inanxiety, depression, stress, and social support between hypertensive and normotensive subjects].Ansiedad y Estres, 7, 203–213.

Celis, H., Fagard, R. H., (2004). White-coat hypertension: A clinical review. European Journal ofInternal Medicine, 15, 348–357.

Cicconetti, P., Thau, F., Bauco, C., Bianchi, A., Fidente, D., Vetta, F., Ettorre, E., & Marigliano, V.(1998). Psychological assessment in the elderly with new mild systolic hypertension. Archives ofGerontology and Geriatrics Suppl., 6, 79–82.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: LawrenceErlbaum Associates.

Dembroski, T. M., MacDougall, J. M., Costa, P. T., & Grandits, G. A. (1989). Components of hostilityas predictors of sudden death and myocardial infarction in the multiple risk factor interventiontrial. Psychosomatic Medicine, 51, 514–522.

Ekeberg, O., Kjeldsen, S. E., Eide, I., & Leren, P. (1990). Childhood traumas and psychosocialcharacteristics of 50-year-old men with essential hypertension. Journal of PsychosomaticResearch, 34, 643–649.

El Assaad, M. A., Topouchian, J. A., & Asmar, R. G. (2003). Evaluation of two devices for self-measurement of blood pressure according to the international protocol: The Omron M5-I and theOmron 705IT. Blood Pressure Monitoring, 8, 127–133.

Garcıa-Vera, M. P., Labrador, F., & Sanz, J. (1997). Stress management training for essentialhypertension:a controlled study. Applied Psychophysiology and Biofeedback, 22, 261–283.

Garcıa-Vera, M. P., Labrador, F., & Sanz, J. (1999). Comparison of clinic, home self-measured, andwork self-measured blood pressures. Behavioral Medicine, 25, 13–22.

Garcıa-Vera, M. P., & Sanz, J. (1999). How many self-measured blood pressure readings are neededto estimate hypertensive patients’ ‘‘true’’ blood pressure? Journal of Behavioral Medicine, 22, 93–113.

Garcıa-Vera, M. P., Sanz, J., & Labrador, F. J. (2000). Proposal for a system of assessment anddiagnosis of essential hypertension in primary health services. Psychology in Spain, 4, 35–54.

Gerardi, R. J., Blanchard, E. B., Andrasik, F., & McCoy, G. C. (1985). Psychological dimensions of’office hypertension’. Behaviour Research and Therapy, 23, 609–612.

Hozawa, A., Ohkubo, T., Tsuji, I., Kikuya, M., Matsubara, M., Suzuki, T., Nagai, K., Kitaoka, H.,Arai, Y., Hosokawa, T., Satoh, H., Hisamichi, S., & Imai, Y. (2002). Relationship between

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

222 J. Sanz et al.

Page 15

personality and self-measured blood pressure value at home: The Ohasama study. ClinicalExperimental Hypertension, 24, 115–123.

Imai, Y., Abe, K., Sasaki, S., Minami, N., Munakata, M., Sakuma, H., Hashimoto, J., Sekino, H.,Imai, K., & Yoshinaga, K. (1989). Clinical evaluation of semiautomatic and automatic devices forhome blood pressure measurement: Comparison between cuff-oscillometric and microphonemethods. Journal of Hypertension, 7, 983–990.

Irvine, J., Garner, D. M., Craig, H. M., & Logan, A. G. (1991). Prevalence of type A behavior inuntreated hypertensive individuals. Hypertension, 18, 72–78.

Jenkins, C. D., Zyzanski, S. J., & Rosenman, R. H. (1979). Jenkins Activity Survey. New York: ThePsychological Corporation.

Jenkins, C. D., Zyzanski, S. J., & Rosenman, R. H. (1992). Inventario de Actividad de Jenkins (FormaC) [Jenkins Activity Survey, Form C]. Madrid: TEA.

Jorgensen, R. S., Johnson, B. T., Kolodziej, M. E., & Scheer, G. E. (1996). Elevated blood pressureand personality: A meta-analytic review. Psychological Bulletin, 120, 293–320.

Kohler, T., Scherbaum, N., Richter, R., & Bottcher, S. (1993). The relationship between neuroticismand blood pressure reexamined: An investigation of a nonclinical sample of military conscripts.Psychotherapy and Psychosomatics, 60, 100–105.

Lantelme, P., Milon, H., Vernet, M., & Gayet, C. (2000). Difference between office and ambulatoryblood pressure or real white coat effect: Does it matter in term of prognosis? Journal ofHypertension, 18, 383–389.

Lyness, S. A. (1993). Predictors of differences between Type A and B individuals in heart rate andblood pressure reactivity. Psychological Bulletin, 114, 266–295.

MacDougall, J. M., Dembroski, T. M., & Krantz, D. S. (1981). Effects of types of challenge onpressor and heart rate responses in Type A and Type B women. Psychophysiology, 18, 1–9.

Matthews, K. A., & Haynes, S. G. (1986). Type A behavior pattern and coronary disease risk. Updateand critical evaluation. American Journal of Epidemiology, 123, 923–960.

McCann, B. S., Woolfolk, R. L., Lehrer, P. M., & Schwarcz, L. (1987). Gender differences in therelationship between hostility and the type A behavior pattern. Journal of Personality Assessment,51, 355–366.

McCrae, R. R., & Costa, P. T., Jr. (1997). Personality trait structure as a human universal. AmericanPsychologist, 52, 509–516.

McGraw, K. O., & Wong, S. P. (1996). Forming inferences about some intraclass correlationcoefficients. Psychological Methods, 1, 30–46.

Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill.O’Brien, E., Asmar, R., Beilin, L., Imai, Y., Mallion, J.-M., Mancia, G., Mengden, T., Myers, M.,

Padfield, P., Palatinin, P., Parati, G., Pickering, T., Redon, J., Staessen, J., Stergiou, G., &Verdecchia, P. on behalf of the European Society of Hypertension Working Group on BloodPressure Monitoring (2003). European Society of Hypertension recommendations for conven-tional, ambulatory and home blood pressure measurement. Journal of Hypertension, 21, 821–848.

Oparil, S., & Miller, A. P. (2005). Gender and blood pressure. Journal of Clinical Hypertension, 7,300–309.

Palatini, P., Palomba, D., Bertolo, O., Minghetti, R., Longo, D., Sarlo, M., & Pessina, A. C. (2003).The white-coat effect is unrelated to the difference between clinic and daytime blood pressure andis associated with greater reactivity to public speaking. Journal of Hypertension, 21, 545–553.

Pickering, T. G. (2005). Should white coat hypertension be treated? The Journal of ClinicalHypertension, 7, 550–553.

Pickering, T. G., & Friedman, R. (1991). The white coat effect: A neglected role for behavioral factorsin hypertension. In P. M. Mccabe, N. Schneiderman, T. M. Field, & J. S. Skyler (Eds.), Stress,coping and disease (pp. 35–49). Hillsdale, NJ: LEA.

Rutledge, T., & Hogan, B. E. (2002). A quantitative review of prospective evidence linkingpsychological factors with hypertension development. Psychosomatic Medicine, 64, 758–766.

Sanz, J., Garcıa-Vera, M. P., Risco, M., & Magan, I. (2006). How many self-measured blood pressurereadings are needed to estimate adults’ ‘‘true’’ blood pressure? manuscript submitted forpublication.

Sanz, J., Silva, F., & Avia, M. D. (1999). La evaluacion de la personalidad desde el modelo de los«Cinco Grandes»: El Inventario de Cinco Factores NEO de Costa y McCrae [Personalityassessment from the Big Five model: Costa and McCrae’s NEO Five Factor Inventory].

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

Personality in hypertension 223

Page 16

In F. Silva (Ed.), Avances en Evaluacion Psicologica [Advances in Psychological Assessment](pp. 171–234). Valencia: Promolibro.

Schneider, R. H., Egan, B. M., Johnson, E. H., Drobny, H., & Julius, S. (1986). Anger and anxiety inborderline hypertension. Psychosomatic Medicine, 48, 242–248.

Siegel, W. C., Blumenthal, J. A., & Divine, G. W. (1990). Physiological, psychological, andbehavioral factors and white coat hypertension. Hypertension, 16, 140–146.

Smith, T. W., & Ruiz, J. M. (2002). Psychosocial influences on the development and course ofcoronary heart disease: current status and implications for research and practice. Journal ofConsulting and Clinical Psychology, 70, 548–568.

Smyth, K., Call, J., Hansell, S., Sparacino, J., & Strodtbeck, F. L. (1978). Type A behavior pattern andhypertension among inner-city black women. Nursing Research, 27, 30–35.

Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). STAI. Manual for the State-Trait AnxietyInventory. Palo Alto, CA: Consulting Psychologist Press.

Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1988). STAI. Cuestionario de Ansiedad Estado-Rasgo: manual [STAI. Manual for the State-Trait Anxiety Inventory]. Madrid: TEA.

Steptoe, A., Melville, D., & Ross, A. (1982). Essential hypertension and psychological functioning:A study of factory workers. British Journal of Clinical Psychology, 21, 303–311.

Steptoe, A., Melville, D., & Ross, A. (1984). Behavioral response demands, cardiovascular reactivity,and essential hypertension. Psychosomatic Medicine, 46, 33–48.

Theorell, T., Hjindahl, P., Eriesson, F., Kallner, A., Knox, S., Perski, A., Svensson, J., Tidgren, B., &Waller, D. (1985). Psychosocial and physiological factors in relation to blood pressure at rest—astudy of Swedish men in their upper twenties. Journal of Hypertension, 3, 591–600.

World Health Organization-International Society of Hypertension Guidelines Subcommittee. (1999).1999 World Health Organization-International Society of Hypertension guidelines for the manage-ment of hypertension. Journal of Hypertension, 17, 151–183.

Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Pers. 21: 209–224 (2007)

DOI: 10.1002/per

224 J. Sanz et al.

Page 17