The anxiety disorders are a prevalent mental health problem in older age with a considerable impact on quality of life. Until recently there have been few longitudinal studies on anxiety in this age group, consequently most of the evidence to date has been cross-sectional in nature.

We undertook a literature search of Medline, PsycINFO, the Cochrane trials database and the TRIP medical database to identify longitudinal studies which would help elucidate natural history and prognosis of anxiety disorders in the elderly.

We identified 12 papers of 10 longitudinal studies in our Review. This represented 34,691 older age participants with 5,199 with anxiety disorders including anxious depression and 3,532 individuals with depression without anxiety. Relapse rates of anxiety disorders are high over 6 year follow-up with considerable migration to mixed anxiety-depression and pure depressive mood episodes. Mixed anxiety-depression appears to be a poorer prognostic state than pure anxiety or pure depression with higher relapse rates across studies. In community settings treatment rates are low with 7–44% of the anxious elderly treated on antidepressant medications.

To our knowledge this is the first Systematic Review of longitudinal trials of anxiety disorders in older people. Major longitudinal studies of the anxious elderly are establishing the high risk of relapse and persistence alongside the progression to depression and anxiety depression states. There remains considerable under-treatment in community studies. Specialist assessment and treatment and major public health awareness of the challenges of anxiety disorders in the elderly are required.

Four constructs are encompassed by the term “falls-related psychological concerns” (FrPC); “fear of falling” (FOF), “falls-related self-efficacy” (FSe), “balance confidence” (BC) and “outcome expectancy” (OE). FrPC are associated with negative consequences including physical, psychological, and social. Identifying factors associated with FrPC could inform interventions to reduce these concerns.

Sixty-two empirical papers relating to psychological factors associated with FrPC in community-dwelling older people (CDOP) were reviewed. Four levels of evidence were used when evaluating the literature: good, moderate, tentative, and none.

Evidence that anxiety predicted FOF, BC, and OE was tentative. Moderate evidence was found for anxiety predicting FSe. Good evidence was found for depression predicting FSe. Moderate evidence was found for depression predicting both FOF and BC. No evidence was found for depression predicting OE. Tentative evidence was found for FSe predicting depression. Good and moderate evidence was found for quality of life (QoL) being predicted by FOF and BC respectively. Tentative evidence was found for FSe predicting QoL. Moderate evidence was found for QoL predicting both FSe and BC. No evidence was found for QoL predicting FOF. Good and moderate evidence was found for activity avoidance/restriction (AA/AR) being predicted by FOF and FSe respectively. Tentative evidence was found for BC and OE predicting AA/AR, as well as for AA/AR predicting FOF. Moderate evidence for activity level (AL) predicting FOF was identified, however the evidence of this predicting FSe and BC was tentative. Evidence for FOF, FSe, and BC predicting AL was tentative as was evidence to suggest FOF predicted coping.

Mixed evidence has been found for the association of psychological factors in association with FrPCs. Future research should employ theoretically grounded concepts, use multivariate analysis and longitudinal designs.

The Geriatric Anxiety Inventory is a 20-item geriatric-specific measure of anxiety severity. While studies suggest good internal consistency and convergent validity, divergent validity from measures of depression are weak. Clinical cutoffs have been developed that vary across studies due to the small clinical samples used. A six-item short form (GAI-SF) has been developed, and while this scale is promising, the research assessing the psychometrics of this scale is limited.

This study examined the psychometric properties of GAI and GAI-SF in a large sample of 197 clinical geriatric participants with a comorbid anxiety and unipolar mood disorder, and a non-clinical control sample (N = 59).

The internal consistency and convergent validity with other measures of anxiety was adequate for GAI and GAI-SF. Divergent validity from depressive symptoms was good in the clinical sample but weak in the total and non-clinical samples. Divergent validity from cognitive functioning was good in all samples. The one-factor structure was replicated for both measures. Receiver Operating Characteristic analyses indicated that the GAI is more accurate at identifying clinical status than the GAI-SF, although the sensitivity and specificity for the recommended cutoffs was adequate for both measures.

Both GAI and GAI-SF show good psychometric properties for identifying geriatric anxiety. The GAI-SF may be a useful alternative screening measure for identifying anxiety in older adults.

The Geriatric Anxiety Scale (GAS; Segal et al. (Segal, D. L., June, A., Payne, M., Coolidge, F. L. and Yochim, B. (2010). Journal of Anxiety Disorders, 24, 709–714. doi:10.1016/j.janxdis.2010.05.002) is a self-report measure of anxiety that was designed to address unique issues associated with anxiety assessment in older adults. This study is the first to use item response theory (IRT) to examine the psychometric properties of a measure of anxiety in older adults.

A large sample of older adults (n = 581; mean age = 72.32 years, SD = 7.64 years, range = 60 to 96 years; 64% women; 88% European American) completed the GAS. IRT properties were examined. The presence of differential item functioning (DIF) or measurement bias by age and sex was assessed, and a ten-item short form of the GAS (called the GAS-10) was created.

All GAS items had discrimination parameters of 1.07 or greater. Items from the somatic subscale tended to have lower discrimination parameters than items on the cognitive or affective subscales. Two items were flagged for DIF, but the impact of the DIF was negligible. Women scored significantly higher than men on the GAS and its subscales. Participants in the young-old group (60 to 79 years old) scored significantly higher on the cognitive subscale than participants in the old-old group (80 years old and older).

Results from the IRT analyses indicated that the GAS and GAS-10 have strong psychometric properties among older adults. We conclude by discussing implications and future research directions.

The Geriatric Anxiety Inventory (GAI) is a recently developed scale aiming to evaluate symptoms of anxiety in later life. This 20-item scale uses dichotomous answers highlighting non-somatic anxiety complaints of elderly people. The present study aimed to evaluate the psychometric properties of the Brazilian Portuguese version GAI (GAI-BR) in a sample from community and outpatient psychogeriatric clinic.

A mixed convenience sample of 72 subjects was recruited for answering the research protocol. The interview procedures were structured with questionnaires about sociodemographic data, clinical health status, anxiety, and depression previously validated instruments, Mini-Mental State Examination, Mini International Neuropsychiatric Interview, and GAI-BR. Twenty-two percent of the sample were interviewed twice for test–retest reliability. For internal consistency analyses, the Cronbach's α test was applied. The Spearman correlation test was applied to evaluate the test–retest GAI-BR reliability. A ROC (receiver operating characteristic) curve study was made to estimate the GAI-BR area under curve, cut-off points, sensitivity, and specificity for the Generalized Anxiety Disorder diagnosis.

The GAI-BR version showed high internal consistency (Cronbach's α = 0.91) and strong and significant test–retest reliability (ρ = 0.85, p < 0.001). It also showed moderate and significant correlation with the Beck Anxiety Inventory (ρ = 0.68, p < 0.001) and the State-Trait Anxiety Inventory (ρ = 0.61, p < 0.001) showing evidence of concurrent validation. The cut-off point of 13 estimated by ROC curve analyses showed sensitivity of 83.3% and specificity of 84.6% to detect Generalized Anxiety Disorder (DSM-IV).

GAI-BR has demonstrated very good psychometric properties and can be a reliable instrument to measure anxiety in Brazilian elderly people.

Although fear of falling is prevalent among older adults recovering from hip fracture, current instruments are inadequate due to focus on specific situations and measurement of self-efficacy rather than fear.

The authors revised and tested a form of the Fear of Falling Questionnaire with three groups of older adults: 405 recovering from hip fracture, 89 healthy community dwelling, and 42 with severe fear of falling. Test-retest reliability was evaluated in a subsample of 16 hip fracture patients. Internal consistency was compared across all groups. Construct validity was established through factor analysis, convergent validity with a measure of fall-related self-efficacy, and discriminant validity with measures of depression and affect.

A revised two-factor, six-item scale appears to have adequate psychometric properties. Scores were lower in the healthy comparison group relative to the hip fracture and fear of falling groups. Cronbach's α ranged from 0.72–0.83, with test-retest reliability of 0.82. Correlations with a measure of fall-related self-efficacy were moderate for the hip fracture group (0.42) and high with the healthy comparison (0.68) and fear of falling (0.70) groups. Correlations with depression and negative and positive affect were low to moderate.

The Fear of Falling Questionnaire – Revised shows promise as a self-report measure of fear of falling, and is one of the first to be tested in older adults recovering from hip fracture. Advantages are that it is global rather than situation-specific and measures fear rather than self-efficacy. Future research on this scale is recommended in other older adult samples for whom fear of falling is relevant.

Research on the behavioral correlates of anxiety in older adults is sparse. The aim of this study was to explore the association of anxiety with behavioral patterns defined by health, activity, emotional and social variables.

A convenience sample of 395 older adults completed measures of health, activity, emotions, social variables and experiential avoidance. Cross-sectional data were analysed using cluster analysis.

Five clusters were identified: active healthy, healthy, active vulnerable, lonely inactive and frail lonely. Participants in the active healthy and healthy clusters showed the highest scores on health variables (vitality and physical function), and adaptive scores on the rest of variables. They also reported the lowest scores on anxiety and included the lowest number of cases with clinically significant anxiety levels. Active vulnerable showed high scores on social support, leisure activities and capitalization on them but low scores in vitality and physical functioning. Participants in the lonely inactive cluster reported the highest mean score in experiential avoidance and high scores on boredom and loneliness, and low scores on social support, leisure activities capitalizing on pleasant activities and health variables. Frail lonely represent a particularly vulnerable profile of participants, similar to that of lonely inactive, but with significantly lower scores on health variables and higher scores on boredom and hours watching TV.

Anxiety in older adults is not only linked to poor health, but also to dysfunctional social behavior, loneliness, boredom and experiential avoidance. Maladaptive profiles of older adults with regard to these variables have been identified.

The recent addition of hoarding disorder (HD) to the Diagnostic and Statistical Manual of Mental Health Disorders, 5th edition, has highlighted the dearth of information about the demographic, sociologic, and medical predictors of HD severity, particularly in older adults. Although there have been several previous studies examining the characteristics of older adults with HD, and one investigation of psychiatric correlates of hoarding symptom severity in non-clinical older adults, there has been little investigation about which characteristics predict hoarding symptom severity in older adults with HD.

Participants were 71 older adults who were enrolled for one of the two studies of HD at the VA San Diego Healthcare System between January 2010 and January 2014.

There were multiple differences in the predictive ability of patient characteristics between the more cognition-related symptoms of HD and the more concrete measure of clutter, including gender-based differences and anxiety severity. Further, married participants were more likely to report lower hoarding severity, and there was no significant relationship between hoarding severity and intervention attempts or hoarding and reported falls in the past three years.

Multiple predictive factors have been presented, which may result in further studies to investigate possible predictive differences in cognition and clutter symptoms of HD. Future studies should examine the possibility of the predictive factors also identified to be moderators of treatment outcomes.

Comorbid anxiety disorders are common in late-life depression and negatively impact treatment outcome. This study aimed to examine personality characteristics as well as early and recent life-events as possible determinants of comorbid anxiety disorders in late-life depression, taking previously examined determinants into account.

Using the Composite International Diagnostic Interview (CIDI 2.0), we established comorbid anxiety disorders (social phobia (SP), panic disorder (PD), generalized anxiety disorder (GAD), and agoraphobia (AGO)) in 350 patients (aged ≥60 years) suffering from a major depressive disorder according to DSM-IV-TR criteria within the past six months. Adjusted for age, sex, and level of education, we first examined previously identified determinants of anxious depression: depression severity, suicidality, partner status, loneliness, chronic diseases, and gait speed in multiple logistic regression models. Subsequently, associations were explored with the big five personality characteristics as well as early and recent life-events. First, multiple logistic regression analyses were conducted with the presence of any anxiety disorder (yes/no) as dependent variable, where after analyses were repeated for each anxiety disorder, separately.

In our sample, the prevalence rate of comorbid anxiety disorders in late-life depression was 38.6%. Determinants of comorbid anxiety disorders were a lower age, female sex, less education, higher depression severity, early traumatization, neuroticism, extraversion, and conscientiousness. Nonetheless, determinants differed across the specific anxiety disorders and lumping all anxiety disorder together masked some determinants (education, personality).

Our findings stress the need to examine determinants of comorbid anxiety disorder for specific anxiety disorders separately, enabling the development of targeted interventions within subgroups of depressed patients.

Centenarians’ psychological well-being is presently of great interest in psychogeriatric research but little is known about factors that specifically account for the presence of clinically relevant anxiety symptoms in this age group. This study examined the presence of anxiety and its predictors in a sample of centenarians and aims to contribute to a better understanding of anxiety determinants in extreme old age.

We examined how socio-demographic, health, functional, and social factors contribute to the presence of clinically significant anxiety symptoms in centenarians recruited from two Portuguese centenarian studies. The Geriatric Anxiety Inventory – Short Form (GAI-SF) was used to assess anxiety symptoms.

A total of 97 centenarians (mean age 101.1 years; SD = 1.5 years; range = 100–108) with no/minor cognitive impairment were included. Clinically significant anxiety symptoms (GAI-SF ≥3) were present in 45.4% (n = 44) of the sample. Main predictive factors included worse health perception, higher number of medical conditions, financial concerns related to medical expenses (income inadequacy) and loneliness.

Results suggest that along with health status (subjective and objective), income inadequacy related to medical expenses and feeling lonely may predispose centenarians to clinically significant anxiety and be important to their overall well-being. Further research is needed on the repercussions of clinical anxiety in centenarians’ quality of life and on co-morbid conditions (e.g. depression) at such advanced ages.

Worry is experienced by many older adults, yet our understanding of the emotional experience of late-life worry is poor as findings regarding older adults are inferred from findings of studies conducted with young adults. In the present study, we aimed to characterize age differences in affect, self-reported arousal, and physiological arousal experienced during worry.

Fifty-three young (M = 21.4, SD = 2.6 years) and 55 older community-dwelling adults (M = 69.1, SD = 8.1 years) participated in an experimental induction of worry or pleasant/neutral recall. Measures collected included: Penn State Worry Questionnaire (PSWQ), worry intensity item, Multiple Affect Adjective Checklist-Revised (MAACL-R), Self-Assessment Maniken arousal item, and heart rate. Standardized residual scores were calculated to represent change from baseline for self-report and psychophysiological measures.

Older adults had lower trait worry and worry intensity at baseline. A significant age by induction type interaction was found for the MAACL-R subscales of anxiety, depression, hostility, and positive affect. Compared with young adults, older adults experienced smaller changes in emotions in response to the worry induction than in the recall induction. For both worry and recall inductions, older adults exhibited less change in self-reported arousal and interbeat intervals from baseline compared with young adults.

Findings from the present study illuminate both similarities and differences in the experience of worry for older and young adults. This study provides preliminary evidence for the characterization of late-life worry as generating less anxiety than worry during young adulthood.

Little is known about the occurrence of psychotic or quasi-psychotic experiences in older people with anxiety disorders.

We used a cross-sectional national probability sample of community-residing individuals to investigate the prevalence and correlates of delusion-like experiences in older people with DSM-IV anxiety disorders. The 2007 Australian National Survey of Mental Health and Well-being (NSMHWB) included 1,905 persons between the ages of 65 and 85 years. Anxiety disorder diagnoses were established using the Composite International Diagnostic Interview (CIDI v3). Participants were asked about three types of delusion-like experiences: thought control or interference, special meaning, and special powers. We used multivariate logistic regression to examine the relationship between a 12-month history of any anxiety disorder and the presence of these delusion-like experiences, adjusting for several potential confounders.

Eighty-two of 1,905 (4.3%) older people met criteria for an anxiety disorder over the previous 12 months. Of these, six reported delusion-like experiences, whereas the prevalence of these experiences among older people without anxiety disorder was 26/1,822 (7.3% vs. 1.4%; χ2 = 16.5; p = 0.000). In a logistic regression model, male gender (OR 0.38; p = 0.019), separated marital status (OR 4.86; p = 0.017), and the presence of anxiety disorder (OR 5.33; p = 0.001) were independently associated with delusion-like experiences, whereas MMSE (Mini-Mental State Examination) score, general medical conditions and affective disorder were not.

In this cross-sectional study, self-reported delusion-like experiences occurred at increased prevalence among community-residing older persons with anxiety disorder. More work is needed to clarify the nature and significance of these findings.

Anxiety disorders are assumed to increase suicide risk, although confounding by comorbid psychiatric disorders may be one explanation. This study describes the characteristics of older patients with an anxiety disorder who died by suicide in comparison to younger patients.

A 15-year national clinical survey of all suicides in the UK (n = 25,128). Among the 4,481 older patients who died by suicide (≥ 60 years), 209 (4.7%) suffered from a primary anxiety disorder, and 533 (11.9%) from a comorbid anxiety disorder. Characteristics of older (n = 209) and younger (n = 773) patients with a primary anxiety disorder were compared by logistic regression adjusted for sex and living arrangement.

Compared to younger patients, older patients with a primary anxiety disorder were more often males and more often lived alone. Although 60% of older patients had a history of psychiatric admissions and 50% of deliberate self-harm, a history of self-harm, violence, and substance misuse was significantly less frequent compared to younger patients, whereas physical health problems and comorbid depressive illness were more common. Older patients were prescribed significantly more psychotropic drugs and received less psychotherapy compared to younger patients.

Anxiety disorders are involved in one of every six older patients who died by suicide. Characteristics among patients who died by suicide show severe psychopathology, with a more prominent role for physical decline and social isolation compared to their younger counterparts. Moreover, treatment was less optimal in the elderly, suggesting ageism. These results shed light on the phenomenon of suicide in late-life anxiety disorder and suggest areas where prevention efforts might be focused.

Quality of life (QOL) is lower in older adults with generalized anxiety disorder (GAD). QOL generally improves following cognitive-behavioral treatment for GAD. Little is known, however, about additional variables predicting changes in QOL in older adults with GAD. This study examined predictors of change in QOL among older participants in a randomized clinical trial of cognitive behavioral therapy (CBT) for GAD, relative to enhanced usual care (EUC).

Hierarchical multilevel mixed-model analyses were used to examine inter-individual and intra-individual factors that predicted QOL over time. Predictors were categorized into treatment, personal and clinical characteristics.

QOL improved over time, and there was significant variability between participants in change in QOL. Controlling for treatment condition, baseline general self-efficacy, baseline social support, within-person variation in worry and depression and average levels of depression across different time points predicted changes in QOL.

QOL has increasingly been used as an outcome measure in treatment outcome studies to focus on overall improvement in functioning. Attention to improvement in symptoms of depression and worry, along with psychosocial variables, such as social support and self-efficacy, may help improve QOL in older adults with GAD.

This study was a secondary study of data from a randomized clinical trial (NCT00308724) registered with clinical.trials.gov.