Aging & Mental Health, September 2003; 7(5): 334–341

ORIGINAL ARTICLE

Characterization of worry according to severity of anxiety in elderly

living in the community

I. MONTORIO, R. NUEVO, M. MÁRQUEZ, M. IZAL & A. LOSADA

Facultad de Psicologı´a, Universidad Auto´noma de Madrid, Spain

Abstract
The goals of this study were to analyse the characteristics of the phenomenon of worry in old age, to analyse differential
characteristics of worry in the elderly according to severity of anxiety, and to explore the discriminative ability of different
DSM-IV criteria for generalized anxiety disorder (GAD) to reveal the presence of this disorder. Starting from a randomized
sample of 97 individuals, those 85 representing three levels of severity of the anxiety were retained: 74 non-clinical, four with
sub-threshold anxiety and seven with GAD. They were assessed with different variables typically considered to be relevant
for the analysis and characterization of worry. Results indicate that the most remarkable differences in contents of worry as a
function of severity of anxiety occurred in the domains of worries about health and personal worries. Furthermore, elderly
people with high levels of anxiety worry more frequently, and about more issues, and perceive less control over their
worrying. Significant differences between groups in past and present orientation of worry were found. Worries were more
frequent in GAD and they were oriented mainly towards the present and minor everyday problems in both GAD and non-
clinical people. Moreover, worry about minor things together with the extent to which worry interferes in daily life were the
best discriminant variables for GAD, being better than the core DSM-IV GAD criteria. We conclude that this pattern of
results suggests that the potential of a specific worry to affect daily well-being and quality of life is strongly related to the
presence of a disorder.

Introduction of GAD among older people ranged from 0.7–7.1%.

Normal ageing often involves different behavioural,
cognitive, and physiological changes that may be
associated with an increase in objective stress sources
and provide additional sources of worry (Wisocki,
1988). Thus, behavioural slowing, concentration, or
memory problems, loss of relatives or friends and
health problems are frequent among elderly people.
However, worry levels and the proportion of people
with emotional pathology are not greater in the
elderly, and are even slightly lower than in the
general population (Wisocki, 1994; Montorio et al.,
2001). Uncontrollable and excessive worry is con-
sidered to be the main characteristic of generalized
anxiety disorder (GAD) in the DSM-IV (American
Psychiatric Association, 1994), and seems to play a
relevant role in most anxiety disorders (Brown,
Antony & Barlow, 1992) and mood disorders
(Barlow, Chorpita & Turovsky, 1996, Starcevic,
1995). Epidemiological data for anxiety disorders
among the elderly people indicate that GAD may be
a highly relevant disorder in this population.
According to Flint’s review (1994), prevalence rates
The Epidemiological Catchment Area survey found
an overall GAD prevalence of 4.6% among people
aged 65 years or over (Blazer, George & Hughes,
1991). Similar prevalence rates for anxiety disorders
have been found in studies analyzing Spanish
samples (Montorio et al., 2001). The study by
Montorio et al., (2001) analyzed a sample of older
people, finding that 11.3% of the sample matched
the diagnostic criteria for an anxiety disorder, 7.2%
of which matched GAD criteria.
Worry domains are among the most analyzed
aspects in the characterization of the worry phenom-
enon. With regard to elderly people, the pioneering
work of Patricia Wisocki and her research team has
been focused mainly on the field of worry domains.
They developed a questionnaire, the Worry Scale
(WS; Wisocki, Handen & Morse, 1986) aimed at
analyzing worry domains, which included three sub-
scales corresponding to different content areas:
financial, health and social condition. One of the
strengths of the WS was the method used for
selecting the items, which were extracted from a list
of fears typically associated with the ageing process.

Correspondence to: Ignacio Montorio, Facultad de Psicologı́a, Universidad Autónoma de Madrid, 28049 Madrid, Spain.
Tel: 34 1 397 4060. Fax: 34 1 397 5215. E-mail: ignacio.montorio@uam.es

Received for publication 30th July 2002. Accepted 21st December 2002.

ISSN 1360-7863 print/ISSN 1364-6915 online/03/05334–08 ß Taylor & Francis Ltd

DOI: 10.1080/1360786031000150694

A later work, based on discussion groups with older
people, led the authors to propose an extended
questionnaire with 88 items, covering the following
domains of worry: financial, health, social, personal,
family, and world concerns (Wisocki, 1994).
Available evidence points out that older people in
general may worry about health, family and keeping
their independence (Wisocki, 1994; Diefenbach,
Stanley & Beck, 2001), whereas social worries,
which systematically appear as the main factor in
anxiety and high levels of worry among younger
adults (e.g., Eysenck, 1992), may be less relevant in
late adulthood (Diefenbach et al., 2001). From a
clinical perspective, some studies have pointed out
the relevance of worry about health for discriminat-
ing emotional disorders and GAD in the elderly
(Wisocki, 1994; Person & Borkovec, 1995; Doucet
et al., 1998; Diefenbach et al., 2001).
Other facets of the worrying process, apart from
domains, have been studied in the characterization of
worry according to the severity of anxiety or the
presence of GAD. In the elderly, there are data about
the temporal orientation of worry. Thus, worrying
may be oriented towards the present in the general or
normal older population and towards the future in
older people with higher levels of worry (Powers,
Wisocki & Whitbourne, 1992), whereas, among
younger adults, worrying is markedly oriented
towards the future, independently of the degree of
worrying (Borkovec, 1994). Nevertheless, there is
still a considerable lack of empirical data on the
characterization of worry in the elderly. Research has
yet to identify the discriminative characteristics of
worrying among older people as a function of the
severity of anxiety. In the general population,
however, there is a large body of research in this
area. For instance, it has been found that GAD
patients worry about a wider range of situations
(Craske et al., 1989) and that they also worry more
about minor things (such as being late for an
appointment or having to fix the car) (e.g.,
Sanderson & Barlow, 1990). Brown, O’Leary and
Barlow (1993) found that GAD people spent more
time per day time worrying, had less realistic worries
and perceived less control over their worries. Also,
one recent study suggests that, although worrying
about current problems may not discriminate GAD
patients from people with a different anxiety dis-
order, high levels of worrying about the future may
be a distinctive characteristic of GAD (Dugas et al.,
1998).
The main goal of this work was to analyse
characteristics of worry in a representative sample
of older adults according to severity of anxiety. The
differential characteristics of worry between three
groups corresponding to three levels of severity of
worrying are analyzed and described. Specifically,
this work had the following aims: (1) to analyse,
according to severity of worry, several relevant
characteristics of worry: the frequency of worrying,
Worry and anxiety in the elderly 335
its temporal orientation, the most frequent contents
of worry, perceived control over worry, worries about
minor things, the interference of worry in daily life,
the percentage of time per day that people spend
worrying and the time elapsed since the person
began considering worry as a problem in his or her
life; (2) to identify which diagnostic criteria for GAD
are the best discriminants for the presence of this
disorder. In short, this study analyses how the main
dimensions usually applied for characterizing worry
as a clinical phenomenon among younger adults
work in older adults.
Methods
Participants
Participants were randomly selected, stratifying by
age (age groups: 65–69, 70–74, 75–79, 80–84 and
85–90 years) and sex (male and female), from the
electoral register of a town near Madrid (Spain). In
all, 186 persons were selected, but only 109 agreed to
participate in the initial interview. Therefore, the
proportion of participation was 58.6%. Among the
109 subjects initially selected, 12 individuals were
not interviewed because of suspected cognitive
impairment according to their scores in the Mini-
Mental State Examination (25 or less in the Spanish
version by Lobo et al., 1979).
Variables
Each subject was assessed with different variables
typically considered to be relevant for the analysis
and characterization of worry. Variables and instru-
ments used for their assessment are as follows:
. Contents of worry were assessed with the Spanish
version of the Worry Scale for Older Adults—
Revised (WSOA—R) (Montorio et al., 1998). The
WSOA—R (Wisocki, 1994) is an extension of the
original Worry Scale (Wisocki et al., 1986), a 35-
item questionnaire designed to assess the fre-
quency of financial, health and social worries in
older adults. The extended version is composed of
88 items covering six domains of worry among
elderly people (financial, health, social, family,
personal and world). For the Spanish version, five
items were removed because they were not
applicable to a Spanish population. Also, the
seven-item world sub-scale was removed because
it has been found to have a poor capacity for
discriminating between different levels of severity,
probably due to a ceiling effect related to its
susceptibility to social desirability effects, as was
proven empirically with young adults in a study
with a similar scale (Tallis, Eysenck & Mathews,
1992). Hence, the version used in this study is a
336 I. Montorio et al.
76-item questionnaire (rated on a 0–4 Likert scale)
that assesses the number and frequency of different
kinds of worries. Worry contents are grouped in
five sub-scales or worry domains, namely: financial
(9 items), health (20), social (14), personal (18), and
family (15). Reliability analysis of the sub-scales
shows that all of them have good internal
consistency (Cronbach’s alpha ranging from
0.78–0.90), and analysis of factorial structure
suggests that they are appropriately assessing the
construct they aim to evaluate (Nuevo, 2001).
. Characteristics of generalized anxiety disorder
were assessed with the Worry and Anxiety Ques-
tionnaire (WAQ; Dugas et al., 1995). The Worry
and Anxiety Questionnaire is a self-report designed
to identify the presence of the DSM-IV diagnostic
criteria for GAD, using Likert scales (range: 0–8).
It also includes items about other relevant aspects
of GAD, such as level of interference of worry in
daily life (0–8), level of worry about minor things
(0–8), number of usual worries or percentage of
day spent worrying. The Worry and Anxiety
Questionnaire has been found to have good
capacity for discriminating GAD (Dugas &
Ladouceur, 1997). The Spanish adaptation used
in the present work has good internal consistency
(Cronbach’s alpha ¼ 0.86) and good convergent
validity with measures of trait-anxiety (trait version
of STAI; Spielberger et al., 1970) and trait-worry
(PSWQ; see below) (Nuevo, 2001).
. Temporal orientation of worry was assessed using
three items rated on a Likert scale (5 points) which
assessed the extent to which people’s worries focus
on past, present or future situations.
. Trait-worry or the general tendency to worry was
measured with the Pennsylvania State Worry
Questionnaire (PSWQ; Meyer et al., 1990), which
is a 16-item scale assessing an individual’s ten-
dency to worry. Each item is rated on a Likert scale
(1–5). The PSWQ has been validated in several
studies with both clinical and non-clinical samples
(e.g., Brown et al., 1992) and with elderly people
(e.g., Beck, Stanley & Zebb, 1996). As regards
elderly people in Spain, the Spanish version of the
PSWQ used in this work (Nuevo, Montorio &
Ruiz, 2002) has shown adequate psychometric
properties, with an internal consistency of 0.95
(Cronbach’s alpha) and good convergent validity
(Nuevo, 2001). Scores on the PSWQ are assumed
to be good indicators of the severity of worrying,
and are used in this study to identify, together with
the data from the clinical interview, the presence of
sub-threshold anxiety.
. A scheduled interview based on DSM-IV criteria
for anxiety and mood disorders was used for
clinical diagnosis. This interview included an
exhaustive clinical assessment and classification
and several questions regarding health information
and consumption of medicines, in order to rule
out, as far as possible, the possibility of any
disorder caused by disease or drugs. All subjects
were diagnosed with this interview according to
DSM-IV criteria for anxiety and mood disorders.
Procedure
Potential participants were first contacted by mail,
and then phoned to arrange an appointment with the
interviewer. All instruments were applied by specif-
ically trained interviewers within an individual inter-
view format. The order of presentation was the same
as described above. With the purpose of overcoming
problems typically presented by the use of self-
reports with older people, and as suggested by the
relevant literature, printed cards with the different
response options were used in the application of the
instruments, thus facilitating understanding and
minimizing the induction or deduction of answers
(Montorio & Izal, 1999). The role of the interviewer
was only to read the questions and record the
answers. It was not necessary to provide additional
explanations for items.
Analysis
Several statistical analyses were conducted in accor-
dance with the different goals of the study. With the
aim of analyzing specific characteristics of worry as a
function of diagnosis, several one-way analyses of
variance (ANOVA) were carried out, taking clinical
diagnosis (GAD, sub-threshold anxiety and non-
clinical subjects) as independent variables and
different variables assessed with the Worry and
Anxiety Questionnaire (WAQ) as dependent;
domains of worry, temporal orientation of worrying
and other central features of GAD: number of
worries, level of worrying about minor things,
perceived lack of control over worries, interference
in daily life, time elapsed since the beginning of the
problem with worrying and percentage of the day
spent worrying. Finally, to assess the different
contributions of the DSM-IV criteria for GAD in
the discrimination of the presence of GAD, a
discriminant analysis was performed including two
groups: people with GAD and people without any
clinical diagnosis. Independent variables were those
items from the Worry and Anxiety Questionnaire
that assess DSM-IV diagnostic criteria for GAD, as
well as the level of worry about minor things and
level of interference of worry in daily life—variables
found to be closely associated with the presence of
GAD (Dugas & Ladouceur, 1997).
Results
First, with regard to the epidemiological data of the
sample, 12 elderly subjects who presented high
 Worry and anxiety in the elderly 337

co-morbidity among disorders, mood disorders and
anxiety disorders other than GAD were deemed
unsuitable for the purposes of this study. Thus, the
final sample was made up of 85 subjects, 74 of whom
did not match the diagnostic criteria for any
emotional disorder; seven matched the criteria for
GAD and four were diagnosed as having sub-
threshold anxiety. The criteria for classifying parti-
cipants as having sub-threshold anxiety was that they
did not match the criteria for any anxiety or mood
DSM-IV disorder, but reported significant clinical
discomfort, scoring 50 or more in the Spanish
version of the Penn State Worry Questionnaire
(Nuevo et al., 2002). This criterion was taken
from Brown et al., (1992), who found that every
anxiety group they assessed (with the exception of
simple phobia) had a mean score of 50 or higher. Of
the final sample, 52.4% were female, 65.5% were
married, 32.1% were widowed, and 2.4% were
unmarried. Average educational level (years of
school attendance) was 9.4 years (SD ¼ 5.5), with a
range of 0–25. Mean age for the total sample was
73.3 (SD ¼ 5.7, range ¼ 65–90), 73.4 (SD ¼ 5.9,
range ¼ 65–90) for the 74 non-clinical subjects, 73
(5.7, 68–81) for the four sub-threshold subjects, and
72.1 (3.7, 66–77) for the seven GAD subjects.
The results of the series of ANOVAs conducted to
examine differences in worry contents as a function
of diagnosis are presented in Table 1. Significant
differences in worry contents between clinical groups
were found. To further analyse specific differences,
post-hoc contrasts were carried out; Scheffé test when
homoscedasticity could be assumed and Games-
Howell when not. Results indicate that people with
GAD have higher frequency scores than non-clinical
subjects in every domain of worry ( p < 0.01).
Moreover, GAD subjects score higher than people
with sub-threshold anxiety in health worries
( p < 0.01) and personal worries ( p < 0.05). No
significant differences in worry contents were found
between non-clinical and sub-threshold anxiety
groups.
Results from the ANOVA with diagnosis as the
independent variable and temporal orientations as
dependent variables are shown in Table 2.
Significant differences between groups in past and
present (but not future) orientation of worry were
found. Post-hoc analysis indicated that GAD people
have higher scores than the non-clinical group in
past-oriented worries ( p < 0.01). No significant
differences were found between these two groups
and the sub-threshold anxiety group. As regards
present-oriented worry, GAD ( p < 0.01) and sub-
threshold anxiety ( p < 0.05) groups scored signifi-
cantly higher than the non-clinical group, and non-
significant differences were observed between GAD
and sub-threshold groups.
One-way ANOVAs were carried out with clinical
diagnosis as the independent variable and several
variables extracted from the Worry and Anxiety
Questionnaire (WAQ) and representatives of DSM-
IV criteria for GAD as the dependent variables.
Specifically, these were: differences in number of
usual worries, level of worry about minor things,
perceived lack of control over worries, level of
interference of worries in daily life, time elapsed
since the person started considering worry as a
problem and percentage of daily time spent worry-
ing. Results can be seen in Table 3. Significant
differences were found in the global comparisons.
According to post-hoc analysis, people with GAD had
significantly higher means than non-clinical subjects
in every analyzed variable ( p < 0.01). Likewise, the
GAD group scored significantly higher than the sub-
threshold anxiety group in worry about minor things
and time elapsed since they started considering
worry as a problem ( p < 0.05). Sub-threshold and
non-clinical groups were found to differ significantly,
with the sub-threshold anxiety group showing higher
means in level of interference of worries in daily life

TABLE 1. Results from ANOVAs with clinical diagnosis about domains of worry
TOTAL Non-clinical Sub-threshold anxiety GAD
Mean SD Mean SD Mean SD Mean SD
Economy (Range: 0–36) 2.9 3.7 2.5 3.4 5.3 6.2 5.7 3.7 F ¼ 3.4; p < 0.05
Social (0–56) 5.3 5.8 4.5 5.2 9 8.7 11.4 6.7 F ¼ 6.2; p < 0.01
Personal (0–72) 10.0 9.5 8.2 7.7 13.5 7.3 26.3 12.1 F ¼ 16.2; p < 0.01
Family (0–60) 11.8 10.2 10.5 9.7 18 10.2 22.4 8.7 F ¼ 5.7; p < 0.01
Health (0–80) 15.3 5.8 12.1 8.4 17.8 6.7 35.7 6.7 F ¼ 24.5; p < 0.01

TABLE 2. Results from ANOVAs with diagnosis about temporal orientation of worry
TOTAL Non-clinical Sub-threshold anxiety GAD
Mean SD Mean SD Mean SD Mean SD
Past 1.7 1.1 1.6 1.0 2.3 1.5 2.9 1.1 F ¼ 5.4; p < 0.01
Present 2.6 1.2 2.4 1.1 4 0.8 3.9 0.7 F ¼ 9.8; p < 0.01
Future 1.9 1.2 1.8 1.1 1.8 1.0 2.9 1.5 F ¼ 2.6; NS
338 I. Montorio et al.
TABLE 3. Results from ANOVAs with diagnosis about WAQ items
TOTAL Non-clinical Sub-threshold anxiety GAD
Mean SD Mean SD Mean SD Mean SD
Number of worries 2.4 1.3 2.3 1.2 2.8 1.5 4.1 1.2 F ¼ 7.6; p < 0.01
Worry about minor things 1.7 2.3 1.3 1.8 2.8 2.2 6 2.3 F ¼ 21.0; p < 0.01
Range: (0–8)
Lack of perceived control 2.4 2.4 1.9 2.2 4.5 1.7 6 1.2 F ¼ 13.7; p < 0.01
Range: (0–8)
Interference in daily life 1.5 1.9 1.0 1.3 4.5 1 5.3 2.0 F ¼ 42.8; p < 0.01
Time since worry is a problem 6.4 17.0 1.9 9.3 21.8 31.6 43.4 22.1 F ¼ 39.9; p < 0.01
(months)
Percentage of day spent 19.7 20.0 15.9 17.9 39.3 14.2 49 13.8 F ¼ 14.1; p < 0.01
worrying

TABLE 4. Standardized coefficients of canonical discriminant functions (items from WAQ about
presence of GAD)
Items from Worry and Anxiety Questionnaire Function 1
Do your worries seem excessive or exaggerated? 0.034
Days bothered by excessive worry in the past six months 0.235
Difficulty to control worries 0.035
Restlessness or feeling keyed up or on edge 0.056
Being easily fatigued 0.029
Difficulty concentrating or mind going blank 0.292
Irritability 0.199
Muscle tension 0.232
Sleep disturbance 0.269
Extent of interference of worry or anxiety in life 0.535
Worry about minor things 0.617

TABLE 5. Structure matrix; discriminant analysis (items from WAQ about presence of GAD)
Items from Worry and Anxiety Questionnaire Function 1
Extent of interference of worry or anxiety in life 0.662
Worry about minor things 0.538
Restlessness or feeling keyed up or on edge 0.417
Difficulty to control worries 0.398
Irritability 0.396
Difficulty concentrating or mind going blank 0.395
Days bothered by excessive worry in the past six months 0.388
Sleep disturbance 0.362
Muscle tension 0.320
Do your worries seem excessive or exaggerated? 0.281
Being easily fatigued 0.255

( p < 0.01), in time elapsed since they started
considering worry as a problem ( p < 0.01) and in
percentage of daily time spent worrying ( p < 0.05).
Finally, discriminant analysis was carried out,
entering all independent variables together and
taking into account the size of each group. The aim
of this was to analyse the different weights of each
DSM-IV criterion for GAD (assessed with WAQ) in
the discrimination between non-clinical and GAD
participants. With this descriptive goal, the different
size of groups does not affect the results.
Discriminant analysis revealed a discriminant func-
tion with a canonical correlation of 0.805, equivalent
to a Wilks’ Lambda of 0.352 (2 ¼ 76.71 with 11
degrees of freedom; p 4 0.01). Tables 4 and 5 show,
respectively, standardized canonical coefficients of
the discriminant function and structure matrix.
Group centroids in discriminant function were
4.355 for the GAD group and 0.412 for the non-
clinical group. The discriminant function correctly
classifies 100% of cases. Results of the discriminant
analysis indicate that level of interference of worries
in daily life and level of worry about minor things are
the best predictors for the presence of GAD. Other
criteria, such as perceived lack of control over
worries or how much they are considered to be
excessive or exaggerated, present low discriminative
capacity for GAD. Among the six symptoms related
to GAD in the DSM-IV criteria, level of irritability or
impatience associated with worry has the greatest
discriminative capacity for GAD, that is, the highest
correlation with the discriminant function.

Discussion
The main goal of this work was to analyse the
characteristics of worry in the elderly, focusing on
examining the presence of differences in patterns of
worrying as a function of its severity. The most
frequent contents of worry in the total sample were
related mainly to family and personal health.
However, the most remarkable differences in contents
of worry as a function of the severity of anxiety,
represented here by the three diagnostic groups
analyzed (GAD, sub-threshold and non-clinical),
were found to occur in the domain of worries about
health (physical threat or danger) and personal
worries (maintenance of independence and personal
autonomy). Post-hoc analyses showed that there were
significant differences between the GAD and sub-
threshold groups only in these two domains. Some
studies have already pointed out the relevance of
worry about health in the elderly and its capacity to
discriminate emotional disorders and GAD in this age
group (Wisocki, 1994; Person & Borkovec, 1995;
Doucet et al., 1998; Diefenbach et al., 2001).
Diefenbach et al., (2001) provide two possible expla-
nations: (1) older people with GAD may have experie-
nced a larger number of minor physical symptoms
than older people without the disorder; (2) worries
about health may have been differentially reinforced
in older people with GAD with respect to younger
people with GAD. In any case, the elderly seem
especially susceptible to anxiety about failing health,
since it is a situation that threatens to reduce the
control they have or perceive to have over their lives
(Rodin & Timko, 1992). In this sense, worries about
health and personal worries refer to the aspects most
directly threatening daily well-being in the elderly.
Worries about family constituted the main domain
of worry among the non-clinical group, but they
appeared less relevant for clinical groups. This
suggests that worry about the welfare and well-
being of the family is not experienced by older adults
as such an aversive emotional experience as worry
about health. Both frequent worries about the family
and their weaker association with clinical problems
may be explained by socio-cultural norms of family
responsibility. It has recently been suggested that one
of the factors that fosters adaptation in elderly people
is the opportunity to express their needs with regard
to emotional generativity (Carstensen, 1995). In this
line, the Socio-emotional Selectivity Theory
(Carstensen, 1993; 1995; Fung et al., 1999) postu-
lates that the perception of time as limited, which
usually accompanies the ageing process, leads to a
progressive increase in the relevance of goals related
to intimacy and significant personal relations, to the
detriment of goals related to achievement or success
in the external world. Hence, family relations will
have preponderance over other kinds of social
relationships, as they are more likely to provide
positive emotional experiences and feelings of social
Worry and anxiety in the elderly 339
embeddedness. If, as suggested, worries about the
well-being of the family are normatively accepted,
they would not be perceived as intrusive or strange
thoughts; consequently, the person would be quite
accustomed to these kinds of thoughts, and would
not try to avoid or control them. This scenario
represents a central feature of most etiological
models for GAD (e.g., Barlow, 1991).
With regard to temporal orientation of worry in the
general adult population, it has been found that
worry is basically focused on situations that may
happen in the future. However, according to Powers
et al., (1992) elderly people in general focus their
worries basically on present or past situations, and
this may be an adaptive strategy for controlling stress
in their lives. In contrast, elderly people with
excessive levels of worry, who are frequently diag-
nosed with GAD, might worry about future situa-
tions, possibly about what they have not yet been able
to do and may likely have no time left to do (Powers
et al., 1992). Data in our work are consistent with
this hypothesis as far as non-clinical people are
concerned, but not so with regard to GAD patients.
Hence, in this study, the entire sample, including
people with GAD and sub-threshold anxiety, worry
mainly about present situations, a finding that would
be contrary to Power’s postulate. Even more at odds
with Powers et al., (1992) is the fact that it is only in
future-oriented worries where there are no significant
differences as a function of diagnosis. It seems
reasonable, given the perception of time as limited
in ageing and the relevance of present orientation in
the process of affect optimization or maximization of
positive emotions (Carstensen, Isaacowitz & Charles,
1999), that both clinical anxiety and worry in the
elderly tend to focus on analysis of the extent to
which events can affect their daily life problems,
threatening their expectations of emotional gener-
ativity. Thus, what may be dependent on the severity
of anxiety in the elderly is not the temporal
orientation of worry, but rather the things people
worry about, and the differential capacity assigned
(by culture, society or individual) to these different
contents to interfere with daily life and decrease both
quality of life and the likelihood of experiencing
positive emotions.
The series of ANOVAs (Table 4) provides addi-
tional information about the relationship between
worry and GAD in old age. In general, these data
indicate that, as the level of anxiety—defined by
diagnosis—increases, so does the variety of worry
contents, the tendency to worry about minor things,
the perception of interference of worry in their lives,
the time per day spent worrying, the likelihood that
they consider worrying as a problem in their lives and
the amount of time elapsed since they started
considering worrying as problem. In an inverse
fashion, the increment in severity of anxiety is related
to decreases in perceived control regarding the onset
or suppression of worry. These results, similar to
340 I. Montorio et al.
those found with younger adults in previous studies,
permit us to make the following comments by way of
summary: (1) the number of contents of worry,
greater in GAD patients, may be a determinant of
clinical severity and not only a consequence of it.
People may shift quickly from one content to another
(Bados, 1999), thus increasing frequency of worries,
but, above all, increasing the subjective sensation of
intensity, intrusiveness and loss of control; (2) the
relevance of both worry about minor things and
perceived degree of interference of worry in daily life
strengthens the hypothesis that clinical severity of
worrying among these subjects is determined by the
extent to which they perceive that worry can affect
daily problems and matters. Immediacy gains rele-
vance with advancing age. It is worth mentioning
that among older adults, as among younger ones,
level of worrying about minor things is a good
predictor of the presence of GAD, and helps to
discriminate it from the rest of anxiety disorders
(Sanderson & Barlow, 1990). Hence, while more
than 70% of GAD older people worried about minor
things to a high degree (6, 7 or 8), only 4% of non-
clinical elderly did so; (3) average percentages of
time per day spent worrying, 15.9% for non-clinical
and 49% for GAD people, match approximately
those indicated by Craske et al., (1989), which were
respectively, 18.2% and 60.7%. These percentages
show, on the one hand, that time per day spent
worrying is another good indicator of the severity of
anxiety in the elderly and, on the other hand, that
worry is not exclusively a pathological phenomenon,
since people without problems of anxiety also spend
a considerable portion of the day worrying. As
Dupuy et al., (2001) suggest, the assessment of
daily time spent worrying may be a good procedure
for distinguishing between excessive and normal
worry and establishing the threshold for what can be
considered as clinical improvement. In this regard,
the percentage (15.9% ) of the day non-clinical
subjects spent worrying in this study may be used as
a norm for elderly people without disorder, and as a
reference point for future studies with this popula-
tion. In addition, this percentage lends support to the
notion that the goal in the clinical treatment of GAD
should be the reduction of worry, not its elimination
(Dupuy et al., 2001).
Finally, with regard to testing the predictive
capacity of items assessing diagnostic criteria for
GAD in DSM-IV, it seems that level of interference
of worry in daily life and level of worrying about
minor things are the best variables discriminating
GAD. Relevance of worry about minor things
suggests that severity of anxiety in the elderly may
be strongly determined by the extent to which people
perceive that future potential dangers will affect their
everyday life. Moreover, this finding is consistent
with the current conception of anxiety as a future-
oriented mood state that serves to anticipate a
potential threat or danger (Di Bartolo, Brown &
Barlow, 1997). As regards the high capacity shown
by level of interference of worry in daily life for
discriminating GAD in older adults, it seems to be a
logical consequence of the fact, discussed before,
that worries in this age group are principally focused
on present situations, which can somehow affect
their daily life and make it difficult for them to fulfil
their needs of emotional generativity.
It should be pointed out that the scope of
conclusions that can be drawn from the results of
this study is limited. It must be stated explicitly that
this work is conceptualised as an approximation
to the description of worries and anxiety and that,
therefore, more studies are needed that go deeper
into analysis of the different aspects of worry (some
of which have been reviewed here). The main
drawback with the present study is the small size of
its clinical and sub-threshold samples. This seriously
limits the scope of conclusions regarding the severity
of worry. Furthermore, the proportion of participa-
tion was slightly low (58.6%), and although this is a
common problem in research on the elderly popula-
tion, it also restricts the generalizability of the
conclusions. In any case, and taking into account
this problem, the randomized sampling method
employed in the present work permits us to suggest
that the results obtained with the non-clinical sample
constitute a suitable description of non-pathological
worry in community-dwelling elderly people.
To summarize, this work suggests that the most
usual ‘non-pathological worries’ in the elderly are
focused on the situation and well-being of their
family and on minor everyday problems, that is, they
are oriented towards the present rather than towards
the past or future. People with high levels of anxiety
differ from non-clinical people in that they worry
more about their own health and the maintenance of
their independence (and worrying about these
contents could foster chronification of anxiety in
the elderly), they worry more frequently, and about
more issues, and they perceive less control over their
worrying. However, it seems that there are no
differences in temporal orientation of worry as a
function of the severity of anxiety; although worries
are more frequent in GAD, they are also oriented
towards the present and minor everyday problems.
Furthermore, the potential of a specific worry to
affect daily well-being and quality of life seems to be
strongly related to the presence of this disorder.
References
AMERICAN PSYCHIATRIC ASSOCIATION. (1994). Diagnostic and
statistical manual of mental disorders, (4th edition).
Washington, DC: American Psychiatric Association.
BADOS, A. (1999). Trastorno de Ansiedad Generalizada. In
M.A. VALLEJO (Ed.), Manual de Terapia de Conducta
(pp. 219–269). Madrid: Dyckinson.
BARLOW, D.H. (1991). The nature of anxiety: anxiety,
depression, and emotional disorders. In R.M. RAPEE &

D.H. BARLOW (Eds.), Chronic anxiety: generalized anxiety
disorder and mixed anxiety-depression (pp. 1–28). New
York: Guilford Press.
BARLOW, D.H., CHORPITA, B.F. & TUROVSKY, J. (1996).
Fear, panic, anxiety, and disorders of emotion. Nebraska
Symposium on Motivation, 43, 251–328.
BECK, J.G., STANLEY, M.A. & ZEBB, B.J. (1996).
Characteristics of generalized anxiety disorder in older
adults: a descriptive study. Behaviour Research and
Therapy, 34, 225–234.
BLAZER, D., GEORGE, L.K. & HUGHES, D. (1991). The
epidemiology of anxiety disorders: an age comparison. In
C. SALZMAN & B.D. LEBOWITZ (Eds.), Anxiety in the
elderly: treatment and research (pp. 17–30). New York:
Springer.
BORKOVEC, T.D. (1994). The nature, functions, and
origins of worry. In G.C.L. DAVEY & F. TALLIS (Eds.),
Worrying: perspectives on theory, assessment, and treatment
(pp. 5–33). Chichester, UK: Wiley.
BROWN, T.A., ANTONY, M.M. & BARLOW, D.H. (1992).
Psychometric properties of the Penn State Worry
Questionnaire in a clinical anxiety disorders sample.
Behaviour Research and Therapy, 30, 33–37.
BROWN, T.A., O’LEARY, T.A. & BARLOW, D.H. (1993).
Generalized anxiety disorder. In D.H. BARLOW (Ed.),
Clinical handbook of psychological disorders (pp. 137–189).
New York: Guilford Press.
CARSTENSEN, L.L. (1993). Motivation for social contacts
across the life span: a theory of socio-emotional selectiv-
ity. Nebraska Symposium on Motivation, 40, 209–254.
CARSTENSEN, L.L. (1995). Evidence for a life-span theory
of socio-emotional selectivity. Current Directions in
Psychological Science, 45, 151–156.
CARSTENSEN, L.L., ISAACOWITZ, D.M. & CHARLES, S.T.
(1999). Taking time seriously: a theory of socio-
emotional selectivity. American Psychologist, 54, 165–181.
CRASKE, M.G., RAPEE, R., JACKEL, L. & BARLOW, D.H.
(1989). Qualitative dimensions of worry in DSM-III-R
generalized anxiety disorder subjects and non-anxious
controls. Behaviour Research and Therapy, 27, 397–402.
DI BARTOLO, P.M., BROWN, T.A. & BARLOW, D.H. (1997).
Effects of anxiety on attentional allocation and task
performance: an information processing analysis.
Behaviour Research and Therapy, 35, 1101–1111.
DIEFENBACH, G.J., STANLEY, M.A. & BECK, J.G. (2001).
Worry content reported by older adults with and without
generalized anxiety disorder. Aging & Mental Health, 5,
269–275.
DOUCET, C., LADOUCEUR, R., FREESTON, M.H.& DUGAS, M.J.
(1998). Thèmes d’inquiétudes et tendance à s’inquiéter
chez les aı̂nés. Canadian Journal on Aging, 17, 361–371.
DUGAS, M.J., FREESTON, M.H., LACHANCE, S., PROVENCHER,
M. & LADOUCEUR, R. (1995). The Worry and Anxiety
Questionnaire: initial validation in non-clinical and
clinical samples. Paper presented at the World Congress
of Behavioural and Cognitive Therapies. Copenhague,
Dinamarca.
DUGAS, M.J., GAGNON, F., LADOUCEUR, R. & FREESTON,
M.H. (1998). Generalized anxiety disorder: a prelimin-
ary test of a conceptual model. Behaviour Research and
Therapy, 36, 215–226.
DUGAS, M.J. & LADOUCEUR, R. (1997). Análisis y
tratamiento del trastorno por ansiedad generalizada. In
V.E. CABALLO (Ed.), Manual para el tratamiento cognitivo-
conductual de los trastornos psicolo´gicos (V.1; pp. 211–240).
Madrid: Siglo XXI.
DUPUY, J.B., BEAUDOIN, S., RHÉAUME, J., LADOUCEUR, R. &
DUGAS, M.J. (2001). Worry: daily self-report in clinical
and non-clinical populations. Behaviour Research and
Therapy, 39, 1249–1255.
Worry and anxiety in the elderly 341
EYSENCK, K.M. (1992). Anxiety: the cognitive perspective.
Hove, UK: LEA
FLINT, A.J. (1994). Epidemiology and comorbidity of
anxiety disorders in the elderly. American Journal of
Psychiatry, 151, 640–649.
FUNG, H.H., CARSTENSEN, L.L. & LUTZ, A.M. (1999).
Influence of time on social preferences: implications for
life-span development. Psychology and Aging, 14, 595–604.
LOBO, A., EZQUERRA, J., GÓMEZ, F.B., SALA, J.M. & SEVA,
A. (1979). El Mini-Examen Cognoscitivo: un test
sencillo y práctico para detectar alteraciones intelec-
tuales en pacientes médicos. Actas Luso-Españolas de
Neurologı´a Psiquiátrica y Cientı´fica, 7, 189–202.
MEYER, T.J., MILLER, M.L., METZGER, R.L. & BORKOVEC,
T.D. (1990). Development and validation of the Penn
State Worry Questionnaire. Behaviour Research and
Therapy, 28, 487–495.
MONTORIO, I. & IZAL, M. (Eds.) (1999). Intervencio´n
psicolo´gica en la vejez. Aplicaciones en el ámbito clı´nico y
de la salud. Madrid: Sı́ntesis.
MONTORIO, I., NUEVO, R., IZAL, M. & LOSADA, A. (1998).
La Escala de Preocupaciones para Personas Mayores:
utilidad para la evaluación de la ansiedad generalizada.
Paper presented at V Congreso Nacional de Psicologı́a
Conductual, Vitoria.
MONTORIO, I., NUEVO, R., LOSADA, A. & MÁRQUEZ, M.
(2001). Prevalencia de trastornos de ansiedad y depre-
sión en una muestra de personas mayores residentes en
la comunidad. Mapfre Medicina, 12, 19–26.
NUEVO, R. (2001). Ansiedad en la vejez. Prevalencia y
caracterización del fenómeno de las preocupaciones.
Unpublished doctoral dissertation, Universidad
Autónoma de Madrid.
NUEVO, R., MONTORIO, I. & RUIZ, M.A. (2002).
Aplicabilidad del Inventario de Preocupación de
Pensilvania (PSWQ) a población de edad avanzada.
Ansiedad y Estre´s, 8, 57–172.
PERSON, D.C. & BORKOVEC, T.D. (1995). Anxiety dis-
orders among the elderly: patterns and issues. Paper
presented at the 103rd Annual Meeting of the American
Psychological Association, New York.
POWERS, C.B., WISOCKI, P.A. & WHITBOURNE, S.K. (1992).
Age differences and correlates of worrying in young and
elderly adults. The Gerontologist, 32, 82–88.
RODIN, J. & TIMKO, C. (1992). Sense of control, aging, and
health. In M.G. ORY & R.P. ABELES (Eds.), Aging, health
and behaviour (pp. 174–206). Thousand Oaks: Sage.
SANDERSON, W.C. & BARLOW, D.H. (1990). A description
of patients diagnosed with DSM-III-R generalized
anxiety disorder. The Journal of Nervous and Mental
Disease, 178, 588–591.
SPIELBERGER, C.D., GORSUCH, R.L. & LUSHENE, R.E.
(1970). STAI manual for the State-Trait Anxiety
Inventory. Palo Alto, CA: Consulting Psychologists Press.
STARCEVIC, V. (1995). Pathological worry in major depres-
sion: a preliminary report. Behaviour Research and
Therapy, 33, 55–56.
TALLIS, F., EYSENCK, M. & MATHEWS, A. (1992). A
questionnaire for the measurement of non-pathological
worry. Personality and Individual Differences, 13, 161–168.
WISOCKI, P.A. (1988). Worry as a phenomenon relevant to
the elderly. Behavior Therapy, 19, 369–379.
WISOCKI, P.A. (1994). The experience of worry among
the elderly. In G.C.L. DAVEY & F. TALLIS (Eds.),
Worrying. Perspectives on theory, assessment, and treatment
(pp. 247–261). Chichester: Wiley.
WISOCKI, P.A., HANDEN, B. & MORSE, C.K. (1986). The
Worry Scale as a measure of anxiety among home bound
and community active elderly. The Behavior Therapist, 5,
91–95.