INT’L. J. AGING AND HUMAN DEVELOPMENT, Vol.

66(4), 283-305, 2008

AGE-RELATED DIFFERENCES IN WORRY AND

RELATED PROCESSES*

PAUL BASEVITZ
Gilmour Psychological Services, Ottawa, Canada

DOLORES PUSHKAR
JUNE CHAIKELSON

MICHAEL CONWAY
Centre for Research in Human Development,
Concordia University, Montréal, Québec, Canada
CONNIE DALTON
Royal Ottawa Hospital, Ottawa, Canada

ABSTRACT

This study investigated the hypothesis that older adults would show age-related
reductions in the tendency to worry in both their retrospective accounts and
through cross-sectional age comparisons with a sample of younger adults. We
also sought to determine whether age differences would be evident in psycho-
logical processes associated with a tendency to worry in general adult samples
(intolerance of uncertainty and beliefs in the functional value of worry).
Support was found for the hypothesized age-related reduction in worry in both
retrospective reports among the older adults and cross-sectional age compari-
sons. Older adults were also found to report less intolerance of uncertainty and

*The research was supported by a research grant from the Fonds pour la Formation de Chercheurs et
L’Aide à la Recherche (FCAR).

283
Ó 2008, Baywood Publishing Co., Inc.
doi: 10.2190/AG.66.4.b
http://baywood.com
284 / BASEVITZ ET AL.

less belief in the functional value of worrying compared to younger adults.

These findings suggest that worry-proneness is reduced in late adulthood and
that a greater ability to tolerate uncertainty in life and to see less value in
worrying may partially account for this. Developmental and contextual
changes that may account for these findings are considered.

Over the last two decades, researchers have become increasingly interested in the
underlying functions and processes of worry, leading to a substantial accumulation
of knowledge in this area (e.g., Borkovec, 1994; Borkovec & Newman, 1998;
Brown, O’Leary, & Barlow, 2001; Dugas, Gagnon, Ladouceur, & Freeston, 1998;
Wells, 1999). Although worry in the general adult population has been studied
extensively, relatively little is known about worry among older adults. The exist-
ing research on worry in late adulthood has begun to explore worry themes
that may be common in this population, along with their relative frequency and
some psychosocial correlates of worry among older adults (Wisocki, 1994). How-
ever, the knowledge-base on worry in this population lags behind the state of the
research on worry in the general adult population.
Conceptual distinctions between worry and anxiety have traditionally defined
anxiety as a broader construct, which includes somatic features (muscle tension,
palpitations), a cognitive component (e.g., fear, worry), and a behavioral aspect
(e.g., escape or avoidant response) (Zebb & Beck, 1998). In contrast, worry has
been conceptualized more narrowly as a thought, about the possibility of one
or more potentially negative but as of yet uncertain future events, and asso-
ciated with feelings of anxiety (MacLeod, Williams, & Bekerian, 1991). Although
some research suggests that non-chronic worry may have some adaptive value in
leading to goal-directed activity and problem solving (e.g., Davey, Hampton, Far-
rell, & Davidson, 1992), chronic worrying seems to be associated with thwarted
attempts to problem solve and in fact, leads to further exacerbation of anxiety
(Davey, 1994) and difficulty focusing attention (Borkovec, Robinson, Pruzinsky,
& Dupree, 1983). Although worrying is a normal psychological phenomenon and
most individuals worry to some degree (Dupuy, Beaudoin, Rhéaume, Ladouceur,
& Dugas, 2001; Tallis, Davey, & Capuzzo, 1994), trait or pathological worry has
also been associated with a variety of negative features including a heightened
state of anxiety, depression, interference with task-based performance, negative
affect, and low self-esteem (e.g., Meyer, Miller, Metzger, & Borkovec, 1990).
Interestingly several studies have suggested that worry is a relatively infrequent
phenomenon among older adults (Skarborn & Nicki, 1996; Watari & Brodbeck,
2000; Wisocki, 1988; Wisocki, Handen, & Morse, 1986) and some evidence has
accumulated indicating that the frequency of worrying further decreases in late
adulthood (Babcock, Laguna, Laguna, & Urusky, 2000; Doucet, Ladouceur,
Freeston, & Dugas, 1998; Powers, Wisocki, & Whitbourne, 1992; Skarborn &
Nicki, 2000). This may be seen as paradoxical given that certain environmental
 AGE-RELATED DIFFERENCES / 285

conditions in late adulthood are likely to elicit feelings of worry (Borkovec, 1988).
Specifically, many older adults experience less controllable negative life events
such as declining health in self and others and reduced and/or fixed financial
resources. In addition, many older people spend a considerable amount of their
time alone and at home (Baltes, Wahl, & Schmid-Furstoss, 1990), conditions
regarded as having relatively low environmental demand when fewer distractions
are likely to occur. Tallis et al., (1994) in a study of the experience of worrying in
adults, found that worrying is most likely to occur in the evenings, at night and
early mornings, times when older adults are most likely to be alone and not dis-
tracted by other people. Consequently, the finding that worry is a relatively infre-
quent phenomenon among older people may be viewed as surprising.
Recent research on aging processes has emphasized that despite age-related
losses and problems, older people are capable of continued psychological devel-
opment that facilitates better socioemotion regulatory processes such that the
experience of negative affect decreases with age. Carstensen, Pasupathi, Mayr
and Nesselroade (2000), for example, has developed Socioemotional Selectivity
Theory (SST) which postulates that future time perspective is a particularly
salient dimension in explaining age-related changes, resulting in better emotion
regulation processes, especially of negative emotions. Given increasingly limited
time, older people have been found to selectively regulate social contact so as to
emphasize interactions with close others and have also learned to reduce contact
with peripheral others who may be experienced as less emotionally rewarding.
Both these processes help maintain and increase positive affect and reduce the
experience of negative affect. Other psychological processes that predispose
individuals to worry may also be developmentally linked with aging, resulting in
better emotion regulation and consequently, decreased worry among older people
(Blanchard-Fields, 2007).
Given the paucity of research on worry in older adults, this study sought to
further investigate the hypothesized age-related reduction in the experience of
worry, to explore possible explanations for this finding, and to determine whether
certain processes related to a heightened tendency to worry in the general pop-
ulation also extend to older adults.

Age-Related Decreases in Worrying

Earlier studies examining worry frequency among older adults were done using
a checklist of common worries in late adulthood (Wisocki et al., 1986; Wisocki,
1988). Although this offered some interesting data suggesting that worry occurs
relatively infrequently among older adults, conclusions based solely on these data
were limited by the scope of the worries assessed by the checklist and by the fact
that comparisons were not made between younger and older individuals. More
recently, a number of studies have assessed worry among older adults on the Penn
State Worry Questionnaire (PSWQ; Meyer et al., 1990), a standardized measure of
286 / BASEVITZ ET AL.

the general tendency to worry, and found older adults tend to have lower scores
(Babcock et al., 2000; Tobey, 1996; Watari & Brodbeck, 2000) compared with
normative data for this scale (Molina & Borkovec, 1994). Some studies have
employed the PSWQ to assess differences across different adult age samples
(Babcock et al., 2000; Doucet et al., 1998; Hunt, Wisocki, & Yanko, 2003; Skar-
born & Nicki, 2000). Babcock et al. (2000) for example, showed that older indi-
viduals had lower scores on the PSWQ relative to a sample of undergraduate
students. Age differences on this measure were also found in a Canadian study
comparing middle-aged adults with a mean age of 55 to a group of retired older
individuals with a mean age of 68 (Skarborn & Nicki, 2000). Their findings
showed that the retired group worried less than the non-retired group. Finally, in
another Canadian study, Doucet et al. (1998) found evidence for less worrying in
later adulthood on both the PSWQ and on the number of worries reported in an
open-ended report of worry themes. In that study, adults aged 75 and over were
found to worry less than those aged 55 to 74.
Another study also showed that older people generally worry less than younger
people, except for specific content themes that were more relevant for older adults
(Hunt et al., 2003). Specifically, the authors reported that although college stu-
dents scored higher on a general measure of trait worry, compared to older adults,
the older adults did score higher on scales tapping worry about health, family and
world issues. As Hunt et al. state, it is not surprising that older adults score higher
on scales where the content is more relevant to them than to younger students who
are unlikely to be concerned about major health issues and the well being of
non-existent children in a future world. Of greater interest is the finding that older
people are less likely to engage in frequent and more uncontrollable non-
productive worries as exemplified in the measure of trait worry.
Although longitudinal data would be needed to determine whether or not the
finding of reduced worry among seniors reflects a developmental shift or a cohort
effect, when taken together, these data raise intriguing questions as to what might
account for this finding. Although situations requiring choice and assumption of
responsibility can generate more cognitive effort resulting in better judgements
(Schwarz & Clore, 1996), these situations can also produce negative affect. Con-
textual factors in later adulthood result in fewer demands and choices required by
the older person, reducing potentially stressful situations requiring decision-
making and responsibility. In general, the more stable life situation in the retire-
ment years, where future possibilities are more limited, including opportunities to
increase income when finances are generally more fixed, and where freedom from
work and familial responsibilities is greater, may be perceived as removing pos-
sibilities and stimuli for worry and reducing negative affect. For example, Skar-
born and Nicki’s (2000) data showing age-related differences in worrying after
retirement may be interpreted as suggesting that reduced responsibility that
coincides with retirement may partially explain the observed age differences in
worrying. Similarly, Doucet et al.’s (1998) finding that worries about the future,
 AGE-RELATED DIFFERENCES / 287

work roles, and financial matters were more frequently mentioned by middle-aged
adults, and that worries about family and friends were less frequently mentioned
after the age of 75, are consistent with experiencing the personal environment as
less demanding and stressful. It is unlikely, however, that age-related changes in
worry are solely explained by situational factors, given that legitimate causes for
concern for older adults continue to exist. Although some situational changes in
aging are normative, (e.g., retirement and reduced family responsibilities as chil-
dren become adults) individual differences in important dispositional traits help
shape cognitive and affective reactions to salient contingencies.
An emerging field of research has shown that improvements in emotion regu-
lation occur through the adult years, a process that seems to continue into late
adulthood (Baltes et al., 1990; Mroczek & Kolarz, 1998) and which has led
researchers to investigate how older adults are able to maintain relatively high
levels of well-being and reduce the experience of negative affect, despite the losses
that typically occur in late adulthood (Gross, Carstensen, Tsai, Skorpen & Hsu,
1997; Carstensen et al., 2000). Reis and Pushkar Gold (1993) developed an
interactive model examining how dispositional traits linked with positive and
negative affect interact with situational changes in retirement to determine
appraisal and affective reactions in aging. They hypothesized that lower levels of
psychological processes related to neuroticism would lead to more positive
appraisals of life context factors in aging and to better affective outcomes. Recent
research supports this approach with the finding that neuroticism declines with
age (Terracciano, McCrae, Brant, & Costa, 2005), suggesting that develop-
mental changes in psychological processing linked to dispositional traits can
reduce worry among older people.

Age-related Differences in Psychological Processes

Associated with Worry
The literature on emotion and aging suggests that reduced worry in late
adulthood could reflect an enhanced ability to regulate emotion. Such a process
would necessitate an increasing need to notice the negative effects of excessive
worrying, and to take active measures to reduce this experience. However, it is
important to consider what specific internal factors might change and foster a
reduced tendency to worry. Two such factors that have been shown in the general
adult population to be associated with a tendency to worry include the tendency to
be intolerant of uncertainty and beliefs in the functional value of worrying (Dugas
et al., 1998).

Intolerance of Uncertainty

Intolerance of uncertainty is described as a cognitive bias affecting how people

attend and respond to their environment (Dugas, Hedayati, Karavidas, Buhr,
Francis, & Phillips, 2005). Specifically, intolerance of uncertainty manifests itself
288 / BASEVITZ ET AL.

by an excessive tendency to find ambiguous situations stressful because events

and outcomes can not be fully known, predicted or guaranteed. Individuals highly
intolerant of uncertainty believe that unexpected events should be avoided and are
excessively concerned by the idea that a negative event might occur, even when
this possibility is minute (Dugas et al., 2005). However, it is impossible to avoid
uncertainty or ambiguity in life, which means that individuals who are severely
intolerant of uncertainty would be constantly faced with situations that they
experience as anxiety provoking. Intolerance of uncertainty has been viewed as a
central process in conceptualizations of pathological worrying (Dugas et al., 1998)
and is also an important target for clinical intervention in the treatment of Gen-
eralized Anxiety Disorder (Dugas & Ladouceur, 2000; Ladouceur et al., 2000),
characterized by excessive and uncontrollable worry and anxiety (American
Psychiatric Association, 1994).
There are several reasons to suspect that intolerance for uncertainty decreases
with age. If as Carstensen’s (2006) socioemotional selectivity theory suggests, older
individuals do select environments that minimize negative emotional experiences,
then surely such environments would have a greater element of stability and cer-
tainty, potentially making intolerance of uncertainty a less salient dimension in the
daily lives of older adults. In addition, if as some research indicates (Carstensen
et al., 2000; Gross et al., 1997), adults do develop better emotion regulation skills as
they age, this process could involve an increased ability to tolerate uncertainty when
faced with potential threat. Repeated life experience with ambiguous situations may
lead to the recognition that uncertainty is sometimes inevitable and to the
development of the skill of being able to act in such situations. Finally, the life stage
of older adults may simply make intolerance of uncertainty less relevant. Namely,
the future is more bounded and many previously uncertain life situations such as
career development are usually in the past. Determining whether intolerance of
uncertainty also changes with age would provide evidence suggesting that
intra-psychic factors do in fact change in late adulthood and could contribute to an
explanatory framework for age-related reductions in worry.

Beliefs in the Functional Value of Worry

Research and theoretical models of worry also suggest that chronic worriers
and anxious individuals tend to harbour erroneous beliefs about the functional
value of worrying (Davey, et al., 1996; Dugas et al., 1998; Wells, 1999; Wells &
Carter, 1999). A compilation of such beliefs has been derived via theoretical
formulations of worrying and through clinicians’ experiences in working with
individuals with high levels of worry (Freeston, Rhéaume, Letarte, Dugas, &
Ladouceur, 1994). Examples of such beliefs include: beliefs that worrying will
lead to effective problem solving or increased control, and superstitious beliefs
that the individual would be more vulnerable if they do not worry (Freeston
et al., 1994). Such beliefs have been shown to be stronger among high worriers
 AGE-RELATED DIFFERENCES / 289

(Freeston et al., 1994; Ladouceur, Blais, Freeston, & Dugas, 1998), and were
found to correlate with worry even after controlling for general negative mood
states (Freeston et al., 1994).
Older individuals may have learned to harbor fewer erroneous beliefs about the
functional value of worrying for various reasons. Most obviously, an accumu-
lation of life experience may have taught these individuals that most worrisome
fears are not realized (Borkovec & Newman, 1998). Their experiences may have
also taught them that worry does not lead to the generation of effective solutions to
life difficulties (Szabó & Lovibond, 2002). Finally, as was described above, life
circumstances in later adulthood often reflect a more established life position and a
more bounded future. In addition to facilitating reductions in worry, such an
environment may also be accompanied by a downward shift in beliefs about the
functional value of worry due to greater emotional distance from worry-related
affect, thus allowing erroneous beliefs to fade.

THE PRESENT STUDY

The purpose of this study was to further investigate the hypothesis that there are
age-related reductions in the experience of worry in late adulthood and to explore
possible underlying mechanisms. Two different methodologies were used to better
assess the possibility of age-related changes in the tendency to worry. First, older
adults were asked to retrospectively recall whether or not they perceived a change
in their tendency to worry relative to when they were younger, and if they did
perceive changes they were asked to speculate on why these changes might have
occurred. Second, samples of older and younger adults were compared on a stan-
dardized measure of trait worry and on measures related to the tendency to worry
(i.e., intolerance of uncertainty and beliefs in the functional value of worrying). In
accordance with previous research it was predicted that older adults would per-
ceive themselves as having a reduced tendency to worry. It was also predicted that
older adults would show lower intolerance of uncertainty and fewer beliefs in the
functional value of worrying. The latter hypotheses were derived from the litera-
ture on emotion and aging, which suggests greater emotional regulation skills
among older adults and increased importance of minimizing negative emotional
experiences. The hypothesis that older individuals would harbor fewer beliefs in
the functional value of worry compared to younger adults is also derived from the
possibility that through an accumulation of life experiences, older individuals may
have learned that feared outcomes associated with worry rarely occur (Borkovec
& Newman, 1998). Examination of age differences in such processes would also
provide evidence that the hypothesized age-related reduction in worrying may not
be solely attributable to external circumstances such as a general diffusion of
responsibilities. Although longitudinal data are ultimately needed to determine the
extent to which such retrospective perceptions of past worrying are accurate and
also that the perceived change reflects a developmental shift rather than a cohort
290 / BASEVITZ ET AL.

effect, it was hoped that converging evidence from this multi-method approach
would provide more convincing evidence for this potentially age-related phenom-
enon. To our knowledge, this was the first study to examine the association
between worry and these related processes among older adults. Consequently,
prior to assessing for age differences, it was also important to assess whether the
associations between worry, intolerance of uncertainty, and beliefs about the
functional value of worry also generalize to older adults.
METHOD

Participants
Older Sample

The sample of older adults consisted of 113 community-dwelling adults aged 65

and over, recruited from lists of people who had previously participated in research
at the Adult Development and Aging laboratory at Concordia University (n = 106)
and via word of mouth (n = 7). Those who were participants in previous studies
were originally recruited from community organizations and through advertise-
ments in local newspapers. Two participants were dropped from the sample due to
cognitive difficulties or troubled mental status, leaving 111 participants avail-
able for data analysis. The participants ranged in age from 65 to 92 (M = 74.18,
SD = 6.13), and 61% were female. The participants were well educated (M = 14.3
years of education, SD = 3.63), most (72%) reported themselves to be in good or
very good health, and the majority (86%) reported being financially very comfort-
able to fairly comfortable. Approximately 48% of the participants were married,
29% were widowed, 16% were either divorced or separated, and 7% were single.
Younger Sample

The younger sample consisted of 106 university students who were recruited
from an on-campus recruitment booth for general research in psychology (n = 66),
from undergraduate classes (n = 32), and via word of mouth (n = 8), at Concordia
University. They ranged in age from 19 to 37 (M = 24.18, SD = 4.22), and 59%
were female. Most of the undergraduates (83%) were full-time students, while the
remaining 17% were engaged in part-time studies. Most (77%) of the participants
reported being in good health, and 88% were single. Their perceived financial
status varied widely, with the majority of participants centred around the mid-
point (“not bad, I can manage”) on a 7-point financial rating scale.

Measures
Demographic Variables

Information on the demographic variables of age, marital status, years of

education, were provided for the older sample only as the younger samples were
all undergraduate students. Gender, perceived health, and perceived adequacy of
 AGE-RELATED DIFFERENCES / 291

finances were used to describe both samples. Years of education were opera-
tionalized as the highest level of educational attainment based on the educational
system in Quebec. Perceived health and financial status were measured with single
items that were rated by participants, using a visual analogue rating scale. Per-
ceived health was measured with a single 5-point scale ranging from “very poor”
to “very good.” Perceived financial status was measured with a single 7-item scale
ranging from “very difficult, I can’t manage at all” to “very comfortable, I can
afford everything that I need or want.”

Changes in Worrying Interview

Retrospective reports of perceived changes in worrying among older adults

were measured using a brief interview. The participants were asked the following
questions about their experience of worrying:
For the next part, I’d like you to think back to when you were a young adult (in
your 20s or 30s) (pause). Generally, do you perceive any changes in the
frequency of your worries now, compared to when you were younger? Have
there been changes in how often you worry?

If changes were perceived, the participants were asked: “why do you think you
worry (less/more) now compared to when you were younger?”
The interviews were audio-taped and were subsequently transcribed for later
coding. The data were coded by the first author using a bottom-up (data driven)
approach. First, all of the responses were read and a coding scheme was developed
that included the full range of responses. After the coding system was finalized,
each response was classified. Perceived changes or stability in worrying were
coded as follows: increased, decreased, no perceived change, or ambivalent
response. One participant’s reponses were not coded due to a recording error.
When increases in worrying were perceived, four general reasons for this
change emerged from the data: 1) a perception of more available time to worry
(labeled more time); 2) a perception that worries in late adulthood involve more
serious issues (labeled more serious concerns) which are less controllable and
which were perceived as having more serious consequences. This theme included
specific references to increased worrying due to the challenges of aging such as
health problems or a loss of independence. Other coded themes included: 3) a
perception of increased responsibility to others including family members and
children (labeled increased responsibility); and 4) a realization of a need to be
prudent and aware of potential danger (labeled awareness of danger), which
contrasted with perceptions of themselves as more care-free in earlier adulthood.
For the perceived decreases in worrying, two broad categories became evident:
1) perceived changes in self; and 2) perceived changes in life circumstances.
Examples of changes in self included: a perception of reduced worry due to better
emotion regulation skills, having learned that worry is a waste of time and that
292 / BASEVITZ ET AL.

worrisome concerns rarely materialize, having a better perspective on what is

important along with a more adaptive orientation to problems, and an increased
ability to disengage when faced with an uncontrollable situation. Examples of
changes in life circumstances included: having fewer work-related problems,
perceiving a more time-limited future, fewer familial responsibilities, and feeling
more financially secure and settled. It is recognized that these broad categories
may be inextricably linked. The decision of which category to place a response
was based on whether the participant referred to something inside the self, outside
the self, or both.
In recognition that there may be multiple reasons for changes in worrying, no
limits were placed on the number of coded reasons for perceived changes in
worrying. For example, it was possible to report both perceived changes in self and
perceived changes in life circumstances, if that is what the participant stated. To
gain an estimate of the reliability of these codes, a second coder was first trained on
the coding system and then independently coded 23 interviews (21% of the
interviews). The inter-rater agreement for the first question about whether changes
were perceived in the frequency of worrying, was 96%. For the reasons for
changes in worrying, where more than one response was possible, the inter-rater
agreement was 78% for the number of reasons provided by each respondent. When
the same content was coded, the inter-rater agreement was 74%.

Trait Worry

The Penn State Worry Questionnaire (PSWQ; Meyer et al., 1990) was used as a
measure of trait worry. This questionnaire consists of 16 items which measure the
general tendency to worry, difficulty controlling worries, and extent of disturb-
ance associated with worrying. The items are rated on a 5-point scale with possible
scores ranging from 16 to 80. The PSWQ is a widely used measure with excellent
psychometric properties for adults and older adults (Beck, Stanley, & Zebb, 1995;
Meyer et al., 1990; Molina & Borkovec, 1994; Stanley, Novy, Bourland, Beck, &
Averill, 2001). In the present study, internal consistency coefficients of .91 and .93
were obtained for the older and younger samples, respectively.

Intolerance of Uncertainty

The Intolerance of Uncertainty Scale (IUS; Buhr & Dugas, 2002; Freeston et al.,
1994) is a 27-item questionnaire measuring a general tendency to find uncer-
tain situations stressful and upsetting, to have difficulty acting when faced with
uncertainty, and to believe that unexpected events are unfair and should be
avoided. Each of the 27 items on the IUS are rated on a 5-point Likert scale, with
possible scores ranging from 27 to 135. The IUS has shown excellent internal
consistency (a = .94) and a five week test-retest reliability coefficient of .74 (Buhr
& Dugas, 2002). Although correlated with other measures of mental health, the
IUS shows a particularly strong association with trait worry, predicting a unique
 AGE-RELATED DIFFERENCES / 293

share of the variance (Buhr & Dugas; 2002; Dugas et al., 2005; Freeston et al.,
1994). The IUS was also shown to be a pivotal variable in discriminating between
individuals with and without Generalized Anxiety Disorder (Dugas et al., 1998).
In further support of its construct validity, the IUS has also been shown to be
associated with a tendency to seek out further evidence prior to making a decision
when faced with an ambiguous behavioral task (Ladouceur, Talbot, & Dugas,
1997). In the present study, the internal consistency coefficients were .91 for each
of the younger and older samples.

Beliefs about Worry

Beliefs about the functional value of worrying were assessed using the Why
Worry Scale (WWS; Freeston et al., 1994), a 20-item scale measuring various
beliefs based on the authors’ clinical experience with worriers and with patients
who met DSM-III-R criteria for GAD. Only items that differentiated between
specific clinical and somatic criteria were retained in the scale. Each item is rated
on a 5-point Likert scale with total possible scores ranging from 20 to 100.
Worry-related beliefs on the WWS include: beliefs that worrying will lead to
solution generation, increased control, and a more efficient route to dealing with
life difficulties. The scale also includes beliefs that worry will somehow avert neg-
ative outcomes. The WWS has displayed good internal consistency, (a = .87;
Freeston et al., 1994). In terms of construct validity, the WWS was able to
discriminate between adults with and without Generalized Anxiety Disorder
(Dugas et al., 1998; Freeston et al., 1994), and has been shown to account for
unique variance in trait worry (Freeston et al., 1994). In the present study, inter-
nal consistency coefficients for the WWS were .88 for each of the younger and
older samples.

PROCEDURE
Both younger and older participants were contacted by telephone and were
provided with further information about the study. If they remained interested, an
individual appointment for testing was arranged. At the time of testing, the partici-
pants were provided with an overview of the study and questions were answered
prior to getting informed consent.
Individual testing of the older participants occurred in the context of a larger
study on life reflections among older adults. The participants completed the
PSWQ and WWS measures at the laboratory whereas the IUC was completed in
the take-home package that was returned by mail in a pre-paid addressed envelope.
The demographics measures and the changes in worrying interview were always
completed at the laboratory. The return rate for the take-home packages was 95%.
The younger adults were all tested at the Adult Development and Aging Labor-
atory at Concordia University and all measures were completed at the laboratory.
294 / BASEVITZ ET AL.

RESULTS

Data Screening
Prior to conducting analyses, the data were screened for missing values, the
presence of outliers, and for skewed distributions. A square root transfor-
mation was applied to the WWS to reduce the impact of skewness (Tabachnick &
Fidell, 1996).

Retrospective Reports of Changes in Worrying

The focus of this set of analyses was to assess the extent to which older indi-
viduals report changes in their worrying, and where applicable, their perceived
reasons for the changes. Figure 1 graphically presents the participants’ responses
to the changes in worrying question. As predicted, the largest proportion of the
participants (46.4%) perceived decreases in their frequency of worrying relative to
when they were younger, whereas a sizeable minority (24.5%) of the participants
reported increased worry. An additional 24.5% of respondents perceived no
change in their frequency of worrying over time. Finally a very small minority of
respondents reported either ambivalence about this issue (3.6%) or were not able
to clearly answer the question (.9%).

Figure 1. Percentage perceiving increased, decreased, and stability in

worry frequency (n =110).
 AGE-RELATED DIFFERENCES / 295

In an effort to determine what might underlie the perceived changes in worry,

the participants who reported changes were also asked why this change had
occurred. Four participants who reported decreased worry and one participant
who reported increased worry were not available for this analysis due to responses
that were too vague to be coded, reducing the number of participants to 47
and 26 for decreases and increases respectively. Additionally, because the par-
ticipants were free to provide more than one reason, the individuals included in
each category sometimes overlapped, producing a sum of the percentage break-
down that exceeded 100.
The percentage of participants who reported each reason for change is presented
in Table 1. For those who reported increased worry, no dominant reason for this
change was evident. Between 30.7% and 38.5% reported a perceived increase in
worry due to either a perception of more serious concerns in their later years, an
increased sense of responsibility, or a heightened awareness of danger. A smaller
proportion of the participants (15.4%) reported increased worry due to having
more time to think about worrisome concerns.
For those who reported reduced worry, the majority of respondents (65.9%)
indicated that they worried less due to perceived change in life circumstances such
as decreased levels of responsibility. However, a substantial minority of partici-
pants (42.6%) attributed their decreases in worrying to changes in self.

Age-Differences in Worry, Intolerance of Uncertainty,

and Beliefs about Worrying
Consistent with previous studies on the topic it was predicted that older adults
would be found to exhibit less trait worry compared with younger adults. Fur-
thermore, it was predicted that similar findings would also be evident on processes
related to worry; namely, in intolerance of uncertainty and in their beliefs about the

Table 1. Reasons for Perceived Changes in Worrying

% reporting n

Reasons for increase (n = 26):

Increased sense of responsibility 38.5% 10
More serious concerns 34.6% 9
Heightened awareness of danger 30.7% 8
More time 15.4% 4

Reasons for decrease (n = 47):

Change in self 42.6% 20
Changes in life circumstances 65.9% 31
296 / BASEVITZ ET AL.

functional value of worrying. Because this was the first study to assess these latter
two constructs in relation to worry among older adults, it was also important to
verify whether these constructs were associated with trait worry in both older and
younger samples.

Correlations among the Variables

The correlations between the PSWQ, WWS, and IUS, along with demographic
variables for both younger and older adults are presented in Table 2. Importantly,
trait worry was associated with both the WWS and with the IUS in both samples,
and the correlations were of similar magnitude in each sample. In the older sample,
age, gender, and years of education were not significantly associated with trait
worry, nor with intolerance of uncertainty and beliefs about worrying. Only health
was significantly correlated, in a negative direction with beliefs about worry for
the older participants. In the undergraduate sample, women tended to score higher
on trait worry and older students tended to have less belief in the functional value
of worry. Also in the younger sample, poorer perceived health was associated with
a greater tendency to worry, to be more intolerant of uncertainty, and to harbor
more beliefs about the functional value of worrying. In both samples, perceived
financial status was not associated with the tendency to worry.
Prior to conducting the principle age comparison, age differences on the demo-
graphic variables of health and financial status were conducted. Interestingly,
there were no differences between the older (M = 4.08, SD = .69) and younger
adults (M = 4.04, SD = .80) on perceived health, t(215) = .43, n.s., but there
was a significant age difference on perceived financial status, with older adults
(M = 5.56, SD = 1.01) reporting a better financial situation than did younger adults
(M = 4.41, SD = 1.37), t(215) = 7.08, p < .001, h2 = .19.

Age Comparison Analyses

The PSWQ, IUS, and WWS showed a strong, but non-redundant pattern of
correlation in both age groups, and were consequently entered into a multivariate
analysis of variance (MANOVA) to test for age differences. A test of the assump-
tion of homogeneity of variance-co-variance matrices yielded a satisfactory result.
Using Tabachnick and Fidell’s recommended alpha level of .001 to evaluate the
significance of Mahalanobis’ distance, no multivariate outliers were detected.
The means for the older and younger adults on the PSWQ, IUS, and WWS,
are presented in Table 3. Wilks’ Lamda indicates an overall significant multi-
variate difference between age groups on the combined dependent measures,
F(3, 313) = 26.97, p < .001, with h2 = .28. Because the variables were correlated, a
Roy-Bargmann stepdown analysis was planned to determine the relative contri-
bution of each measure to the multivariate effect. This analysis is in effect an anal-
ysis of covariance which allows the specification of the amount each measure
contributes to variance in the comparisons between the age groups. The dependent
 AGE-RELATED DIFFERENCES / 297

Table 2. Correlations Among PSWQ, IUS, WWS, and Demographic Variables

Younger sample (n = 106)
PSWQ IUS WWS (sqrt)
Demographics:
Age –.12 –.18 –.22*
Gendera .29** .09 .01
Health ratingb –.28** –.27** –.26**
Finances ratingb –.07 –.03 –.06
Worry-related questionnaires:
PSWQ
IUS .56***
WWS (sqrt) .54*** .53***

Older sample (n = 111)

PSWQ IUS WWS (sqrt)
Demographics:
Age .01 .08 –.04
Gendera .05 –.04 –.06
Years education –.07 –.08 –.02
Health ratingb –.17 –.07 –.27**
Finances ratingb –.12 –.07 .04
Worry-related questionnaires:
PSWQ
IUS .56***
WWS (sqrt) .52*** .42***
a
Malegender coded as 1; female gender coded as 2.
b
Higher scores indicate better health and a better financial situation.
*p < .05, **p < .01, ***p < .001.

variable that was entered in the first step is tested for group differences in a uni-
variate analysis with each subsequent step testing a different dependent variable to
see whether it significantly contributes to the explained variance, after controlling
for the variables that had been entered in previous steps. For the stepdown
analysis, the PSWQ was entered on the first step, followed by the IUS on step 2,
and the WWS on step 3. This order of entry followed the theoretical rationale that
trait worry would be the primary source of variance in the age comparisons fol-
lowed by intolerance of uncertainty and beliefs about the functional value of
298 / BASEVITZ ET AL.

Table 3. Means Scores on PSWQ, IUS, and WWS in Younger

and Older Samples
Younger sample (n = 106) Older sample (n = 111) F**
Measure M SD Range M SD Range
PSWQ 47.27 13.58 16-79 37.53 11.78 16-71 31.02*

IUS 62.24 16.77 27-106 49.03 12.68 27-88 14.55*

WWS 43.27 12.97 23-82 30.62 9.21 20-60 26.62*

*p < .001
**Note: F values are based on group differences on the first step-down analysis with the
PSWQ entered first, followed by the IUS and the WWS.

worrying, which are assumed to be a central underlying process variable in con-

ceptualizations of worry. This order of entry also allowed for the assessment of
whether the IUS and the WWS would account for additional variance in predicting
group differences, after controlling for variance already accounted for by the
PSWQ. As predicted, the PSWQ made a significant contribution in predicting age
differences, with older adults showing lower scores on this measure, F(1, 215) =
31.02, p < .001, h2 = .13. Interestingly, after controlling for variance already
accounted for by the PSWQ, the IUS made a significant contribution to the age
differences, with the older adults scoring lower on this measure relative to the
younger adults, F(1, 214) = 14.55, p < .001, h2 = .06. Similarly, on the third
step, after controlling for variance accounted for by both the PSWQ and the
IUS, the WWS made a significant contribution to predicting age differences,
again, with older adults scoring lower on this measure relative to younger adults,
F(1, 213) = 26.62, p < .001, h2 = .11.
Because the IUS and WWS each contributed unique explained variance in pre-
dicting age differences, it was possible that it was these constructs and not worry
per se that were the primary source of the observed age-related changes in worry-
ing. Consequently, a second step-down analysis was conducted, this time with
PSWQ scores entered last. When the IUS was entered on the first step of this
analysis, age differences on this variable emerged, F(1, 215) = 43.12, p < .001,
h2 = .17. After controlling for variance associated with the IUS, the WWS contri-
buted significantly to the explained variance associated with the age differences,
F(1, 214) = 31.93, p < .001, h2 = .13. Interestingly, age differences on the PSWQ
were not significant after controlling for the IUS and the WWS on earlier steps,
F(1, 213) = .03, n.s.
 AGE-RELATED DIFFERENCES / 299

DISCUSSION
The major purpose of this study was to further investigate whether older indi-
viduals experience reductions in worrying, and to better understand factors asso-
ciated with this hypothesized change. The findings provide evidence in support of
an age-related decrease in worrying, and build on an emerging literature showing
similar results (Babcock et al., 2000; Doucet et al., 1998; Powers et al., 1992;
Skarborn & Nicki, 2000). The study has also provided a novel contribution to this
literature through retrospective accounts of the older participants, which showed
that the dominant response to an inquiry about changes in worry frequency was a
perceived reduction. Although longitudinal data are ultimately required to draw
the conclusion that these findings reflect a bona fide developmental shift rather
than a cohort effect or a report bias, the converging evidence from these two
methodologies underscores the possibility of a developmental shift. Such a shift
would also be consistent with longitudinal data from the research literature on
emotion and aging, which suggests that the experience of negative affect decreases
with age partly due to an increased ability to minimize negative emotional states in
later life (Carstensen et al., 2000; Gross et al., 1997; Turk-Charles, Reynolds, &
Gatz, 2001).

Reasons for Perceived Changes in Worry

It was important to better understand why this change might occur.
Accordingly, the older participants were asked for their impressions about the
reasons underlying their perceived changes in worry. For those who perceived a
decrease, most reported that this reflected a change in life circumstances where
they had fewer work-related and familial responsibilities, and felt more settled in
life. However, a significant minority of those who reported decreased worry also
reported that this decrease reflected an intra-psychic shift. Such accounts are
consistent with the notion that situational factors only partially account for this
change.
Although a perceived reduction in worry was the dominant response, it is
noteworthy that a significant minority of respondents (25%) reported that the fre-
quency of their worrying had increased relative to when they were younger and
another 25% of respondents noted that their worry frequency had not changed. The
participants provided a variety of reasons for this perceived increase, which
included but were not limited to the idea that worries in later life reflect more
serious concerns such as those associated with the constraints of aging. In fact,
only 9 individuals reported an increase in worrying due, in part or whole, to this
reason. Other reasons for perceived increases in worry that were reported with
similar frequency were perceptions of increased responsibility in later life and a
heightened awareness of danger, relative to their younger years. The findings
suggest that only a minority of relatively healthy seniors experience heightened
and increasing levels of worry due to factors linked with the declines and
300 / BASEVITZ ET AL.

constraints associated with aging, and contrast with common stereotypes that
many older adults experience worry due to these factors.

Age Differences in Psychological Processes

Associated with Worry

A second and more indirect approach to better understand why older individuals
show decreases in their levels of worry was to determine whether other psycho-
logical processes that are associated with worry, also differ with age. Both longi-
tudinal and cross-sectional data indicate some decline in the personality trait of
neuroticism (Caspi, Roberts & Shiner, 2005). For example, data from the Balti-
more Longitudinal Study of Aging indicate a modest normative decline in later
adulthood on neuroticism (Terracciano et al., 2005). Such a decline may reflect
reduced stress associated with the cessation or lesser salience of social roles, such
as work and parenting, in older years. Alternatively, intrinsic maturation reflecting
genetic effects may cause a decrease in neuroticism. Whatever the causal proces-
ses underlying reduced neuroticism scores, lower levels of the facets compos-
ing neuroticism, e.g., self-consciousness, anxiety, vulnerability, could lead to
reduced worry.
The theoretical framework outlined by socioemotional selectivity theory sug-
gested that the increasing saliency of emotional regulatory goals might lead to
several shifts in coping with emotions. Other aspects of increased emotional stabil-
ity that occurs with developmental maturity in older years (Caspi et al., 2005)
could predict a greater ability to tolerate uncertainty. This study predicted that
seniors would be more tolerant of uncertainty, relative to younger adults. It was
also predicted that older individuals would have learned through their experiences
to perceive less functional value in worrying. We also sought to determine whether
these features would be associated with a tendency to worry among both younger
and older adults.
As expected, both intolerance of uncertainty and beliefs in the functional value
of worry were associated with trait worry in both the younger and the older
samples. Age differences in both intolerance of uncertainty and beliefs about
worry emerged in the expected direction, even after trait worry was co-varied.
Interestingly, once age differences on intolerance of uncertainty and beliefs about
worry were statistically accounted for, the age difference on trait worry was no
longer statistically significant. If this finding does emerge in future research as a
replicable and reliable effect, it suggests that having difficulty tolerating uncer-
tainty and harboring erroneous beliefs about the functional value of worrying
predisposes people to worry (Dugas et al., 1998) and consequently, developmental
shifts in worrying might be due in part to prior changes in these areas. Such a
proposition is further supported by research showing that changes in intolerance of
uncertainty were found to precede changes in worry in a therapeutic context
(Dugas & Ladouceur, 2000).
 AGE-RELATED DIFFERENCES / 301

In contrast to the pattern of significant associations among the worry-related

measures for both older and younger adults, the absence of significant correlation
coefficients between the worry-related measures with demographic characteristics
for the older sample in comparison with the pattern found for the younger sample
is striking. Within the younger sample, age was negatively correlated with worry,
possibly reflecting that undergraduates learn over time that most worrisome fears
are not realized (Borkovec & Newman, 1998). The findings that female under-
graduates showed greater levels of trait worry supports research on gender differ-
ences in worry for young adults (Stavosky & Borkovec, 1988). Poorer subjective
health was significantly linked with all three worry-related measures for the
younger sample. Since the experience of ill health is atypical for young adults,
presumably health concerns and problems may act as a trigger to generate worries
and concerns about the significance of symptoms of ill health.

Summary, Limitations, and Direction for

Future Research

Taken together, the findings from both age comparisons and retrospective
recall suggest a reduced general tendency to worry among older individuals and
shed light on why this might occur. This is also the first study to determine
that intolerance of uncertainty and beliefs in the functional value of worrying,
processes that are strongly correlated with trait worry in the general population,
have the same pattern of association with worry in older adults. This suggests
that models of chronic worry among the general population may extend to older
adults as well. Clearly however, further research would be needed before drawing
such conclusions.
Several limitations are noteworthy in interpreting these findings. Namely, this
study was based on a sample of relatively healthy community-dwelling seniors
who volunteered for research, a factor which likely reflects an active engaged
lifestyle. Furthermore, the comparison sample of younger adults consisted of a
convenience sample of university students. Consequently, the findings may not
generalize to less healthy or active individuals and to younger individuals who are
less educated. Additionally, as was highlighted above, the age differences in worry
were based on cross-sectional comparisons and on retrospective recall, which may
have been affected by cohort effects and by recall biases which potentially could
limit the accuracy of recall for worry in the past. Accordingly, longitudinal
research is required prior to drawing stronger conclusions regarding the possibility
of age-related reductions in worrying and the temporal sequence of the proposed
worry process variables in facilitating such a change. It would be particularly
intriguing for longitudinal research to assess worry and related psychological
processes before and after developmental milestones such as retirement. Such
a design, which would assess worry before and after a major life transition
that is often associated with a reduction in many responsibilities, may help to
302 / BASEVITZ ET AL.

disentangle the extent to which these changes are influenced by intra-psychic

and situational factors.
Although this research is at an early state of development, the finding that intol-
erance of uncertainty and beliefs in the functional value of worry were strongly
associated with trait worry in older individuals offers direction for clinicians who
work with highly anxious seniors. These constructs are strong predictors of worry
in the general adult population (Dugas et al., 1998), and have been integral com-
ponents of a treatment program for individuals with Generalized Anxiety Disorder
(Ladouceur et al., 2000). The initial findings that such associations generalize to
older adults suggests that clinical efforts focused on helping chronically worried
seniors to better tolerate uncertainty and to more realistically appraise erroneous
beliefs in the functional value of worrying would be helpful in improving the
quality of affective life for these seniors. In addition to these clinical implications,
it is hoped that further research on this topic will help to shed light on processes
underlying developmental changes in the experience of worry.

REFERENCES

American Psychiatric Association (1994). Diagnostic and statistical manual of mental

disorders (4th ed.). Washington, DC: Author.
Babcock, R. L., Laguna, L. B., Laguna, K. D., & Urusky, D. A. (2000). Age differences in
the experience of worry. Journal of Mental Health and Aging, 6, 227-235.
Baltes, M. M., Wahl, H.-W., & Schmid-Furstoss, U. (1990). The daily life of elderly Ger-
mans: Activity patterns, personal control, and functional health. Journal of Geron-
tology: Psychological Sciences, 45, 173-179.
Beck, J. G., Stanley, M. A., & Zebb, B. J. (1995). Psychometic properties of the Penn State
Worry Questionnaire in older adults. Journal of Clinical Geropsychology, 1, 33-42.
Blanchard-Fields, F. (2007). Everyday problem solving and emotion: An adult develop-
mental perspective. Current Directions in Psychological Research, 16, 26-31.
Borkovec, T. D. (1988). Comments on “Worry as a phenomenon relevant to the elderly.”
Behavior Therapy, 19, 381-383.
Borkovec, T. D. (1994). The nature, functions, and origins of worry. In G. C. L. Davey and
F. Tallis (Eds.), Worrying: Perspectives on theory, assessment and treatment
(pp. 5-33). Chichester, England: Wiley.
Borkovec, T. D. & Newman, M. G. (1998). Worry and generalized anxiety disorder. In A.
S. Bellack, M. Hersen (Series Eds.), and P. Salkovskis (Vol. Ed.), Comprehensive
clinical psychology: Vol. 6. Adults: Clinical formulation & treatment (pp. 439-459).
New York: Elsevier Science.
Borkovec, T. D., Robinson, E., Pruzinsky, T., & DuPree, J. A. (1983). Preliminary
exploration of worry: Some characteristics and processes. Behaviour Research and
Therapy, 21, 9-16.
Brown, T. A., O’Leary, T. A., & Barlow, D. H. (2001). Generalized anxiety disorder. In
D. H. Barlow (Ed.), Clinical handbook of psychological disorders: A step by step
treatment manual (3rd ed., pp. 154-208). New York: Guilford.
 AGE-RELATED DIFFERENCES / 303

Buhr, K., & Dugas, M. J. (2002). The intolerance of uncertainty scale: Psychometric
properties of the English version. Behaviour Research and Therapy, 40, 931-945.
Caspi, A., Roberts, B. W., & Shiner, R. L. (2005) Personality development: Stability and
change. Annual Revue of Psychology, 56, 453-484.
Carstensen, L. L. (2006). The influence of a sense of time on human development. Science,
312, 1913-1915.
Carstensen, L. L., Pasupathi, M., Mayr, U., & Nesselroade, J. R. (2000). Emotional experi-
ence in everyday life across the adult life span. Journal of Personality and Social Psy-
chology, 79, 644-655.
Davey, C. G. L. (1994). Pathological worrying as exacerbated problem-solving. In C. G. L.
Davey and F. Tallis (Eds.), Worrying: Perspectives on theory, assessment and treat-
ment (pp. 247-261). Chichester, England: Wiley.
Davey, G. C. L., Hampton, J., Farrell, J., & Davidson, S. (1992). Some characteristics of
worrying: Evidence for worrying and anxiety as separate constructs. Personality and
Individual Differences, 13, 133-147.
Davey, C. G. L., Tallis, F., & Capuzzo, N. (1996). Beliefs about the consequences of worry-
ing. Cognitive Therapy & Research, 20, 499-520.
Doucet, C., Ladouceur, R., Freeston, M. H., & Dugas, M. J. (1998). Thèmes d’inquiétudes
et tendance à s’inquiéter chez les aînés. La Revue Canadienne du vieillissement, 17,
361-371.
Dugas, M. J., Gagnon, F., Ladouceur, R., & Freeston, M. H. (1998). Generalized anxiety
disorder: A preliminary test of a conceptual model. Behaviour Research and Therapy,
36, 215-226.
Dugas, M. J., Hedayati, M., Karavidas, A., Buhr, K., Francis, K., & Phillips, N. (2005).
Intolerance of uncertainty and information processing: Evidence of biased recall and
interpretation. Cognitive Therapy and Research, 29, 57-70.
Dugas, M. J., & Ladouceur, R. (2000). Treatment of GAD: Targeting intolerance of uncer-
tainty in two types of worry. Behavior Modification, 24, 635-657.
Dupuy, J. B., Beaudoin, S., Rhéaume, J., Ladouceur, R. & Dugas, M. (2001). Worry: Daily
self-report in clinical and non-clinical populations. Behaviour Research and Therapy,
39, 1249-1255.
Freeston, M. H., Rhéaume, J., Letarte, H., Dugas, M. J., & Ladouceur, R. (1994). Why do
people worry? Personality and Individual Differences, 17, 791-802.
Gross, J. J., Carstensen, L. L., Tsai, J., Skorpen, C. G., & Hsu, A. Y. C. (1997).
Emotion and aging: Experience, expression, and control. Psychology and Aging, 12,
590- 599.
Hunt, S., Wisocki, P., & Yanko, J. (2003). Worry and use of coping strategies among older
and younger adults. Anxiety Disorders, 17, 547-560.
Ladouceur, R., Talbot, F., & Dugas, M. J. (1997). Behavioral expressions of intolerance of
uncertainty in worry. Behavior Modification, 21, 355-371.
Ladouceur, R., Blais, F., Freeston, M. H., & Dugas, M. J. (1998). Problem-solving and
problem orientation in generalized anxiety disorder. Journal of Anxiety Disorders, 12,
139-151.
Ladouceur, R., Dugas, M. J., Freeston, M. H., Léger, E., Gagnon, F., & Thibodeau, N.
(2000). Efficacy of a cognitive-behavioral treatment for generalized anxiety disorder.
Journal of Consulting and Clinical Psychology, 68, 957-964.
304 / BASEVITZ ET AL.

MacLeod, A. K., Williams, J. M., & Bekerian, D. A. (1991). Worry is reasonable: The role
of explanation in pessimism about future personal events. Journal of Abnormal
Psychology, 100, 478-486.
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.
Molina, S., & Borkovec, T. D. (1994). The Penn State Worry Questionnaire: Psychometric
properties and associated characteristics. In G. C. L. Davey and F. Tallis (Eds.),
Worrying: Perspectives on theory, assessment and treatment (pp. 247-261). Chi-
chester, England: Wiley.
Mroczek, D. K., & Kolarz, M. (1998). The effect of age on positive and negative affect: A
developmental perspective on happiness. Journal of Personality and Social Psy-
chology, 75, 1333-1349.
Powers, C. B., Wisocki, P. A., & Whitbourne, S. K. (1992). Age differences and correlates
of worrying in younger and older adults. The Gerontologist, 32, 82-88.
Reis, M. & Pushkar Gold, D. (1993). Retirement, personality and life satisfaction: A review
and two models. Journal of Applied Gerontology, 12, 261-282.
Schwarz, N., & Clore, G. L. (1996). Feelings as phenomenal experiences. In E. T. Higgens
and A. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 433-
465). New York: Guilford Press.
Skarborn, M., & Nicki, R. (1996). Worry among Canadian seniors. International Journal of
Aging and Human Development, 43, 169-178.
Skarborn, M., & Nicki, R. (2000). Worry in pre- and post-retirement persons. International
Journal of Aging and Human Development, 50, 61-71.
Stanley, M. A., Novy, D. M., Bourland, S. L., Beck, J. G., & Averill, P. M. (2001).
Assessing older adults with generalized anxiety: A replication and extension. Behavior
Research and Therapy, 39, 221-235.
Stavosky, J. M., & Borkovec, T. D. (1988). The phenomenon of worry: Theory, research,
treatment and its implications for women. Women and Therapy, 6, 77-95.
Szabo, M., & Lovibond, P. F. (2002). The cognitive content of naturally occurring worry
episodes. Cognitive Therapy & Research, 26, 167-177.
Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd ed.). New York:
Harper Collins.
Tallis, F., Davey, G. C. L., & Capuzzo, N. (1994). The phenomenology of non-pathological
worry: A preliminary investigation. In G. C. L. Davey and F. Tallis (Eds.), Worrying:
Perspectives on theory, assessment and treatment (pp. 61-89). Chichester, England:
Wiley.
Terracciano, A., McCrae, R. R., Brant, L. J., & Costa, Jr., P. T. (2005). Hierarchical linear
modeling analyses of the NEO-PI-R scales in the Baltimore Longitudinal Study of
Aging. Psychology and Aging, 20, 493-506.
Tobey, B. K. (1996). A comparison of worry in the psychiatric and community elderly
(Doctoral dissertation, Allegheny University, 1996). Dissertation Abstracts Interna-
tional, 58, 1-B.
Turk-Charles, S., Reynolds, C. A., & Gatz, M. (2001). Age-related differences and change
in positive and negative affect over 23 years. Journal of Personality and Social Psy-
chology, 80, 136-151.
 AGE-RELATED DIFFERENCES / 305

Watari, K. F., & Brodbeck, C. (2000). Culture, health, and financial appraisals: Comparison
of worry in older Japanese Americans and European Americans. Journal of Clinical
Geropsychology, 6, 25-39.
Wells, A. (1999). A cognitive model of generalized anxiety disorder. Behavior Modi-
fication, 23, 526-555.
Wells, A. & Carter, K. (1999). Preliminary tests of a cognitive model of generalized anxiety
disorder. Behaviour Research and Therapy, 37, 585-594.
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 and F.
Tallis (Eds.), Worrying: Perspectives on theory, assessment and treatment (pp.
247-261). Chichester, England: Wiley.
Wisocki, P. A., Handen, B., & Morse, C. K. (1986). The worry scale as a measure of anxiety
among homebound and community active elderly. The Behavior Therapist, 5, 91-95.
Zebb, B. J., & Beck, J. G. (1998). Worry versus anxiety: Is there really a difference? Behav-
ior Modification, 22, 45-61.

Direct reprint requests to:

Dolores Puskar
Centre for Research in
Human Development
Concordia University
7141 Sherbrooke Street West
Montréal, Québec
Canada H4B 1R6
e-mail: Dolores.Pushkar@Concordia.ca