Global prevalence of anxiety symptoms and its associated factors in older adults: A systematic review and meta‐analysis

Abstract

This systematic review and meta‐analysis aimed to determine the global prevalence of anxiety symptoms in older adults and explore the associated factors contributing to anxiety symptoms in this population. A comprehensive search was conducted across multiple electronic databases to identify relevant studies published up to 27 February 2023. A total of 17 articles met the inclusion criteria. The global prevalence of anxiety symptoms among older adults was 28%. Subgroup analyses indicated the continent, and the assessment tools significantly affect the overall prevalence. The associated factors identified encompassed sociodemographic, health‐related, and psychosocial domains, with the female gender, physical comorbidities, and low social support and social isolation being the most reported associated factors for anxiety in each domain. The findings of this systematic review and meta‐analysis provide valuable insights into the global prevalence of anxiety symptoms in this population and the associated factors contributing to its occurrence.

Studies showed various prevalence rates of anxiety in older adults, yet the rate needs to be updated using the specified geriatric assessment tools. The global prevalence of anxiety symptoms among older adults was estimated to be 28%. Subgroup analyses indicated the geographical location, and the assessment tools significantly affect the overall prevalence.

1. INTRODUCTION

According to the most recent estimates, 1400 million and 2.1 billion adults will be older than 60 by 2030 and 2050, respectively, a noticeable spike from the current 756 million. ¹ , ² Aging greatly impacts mental and physical health and can be associated with psychiatric problems including depression and anxiety. ³ , ⁴

Anxiety disorders are among the most common mental health disorders across the globe, affecting 4% of the population, ⁵ and are commonly defined by excessive and enduring fear, anxiety, or avoidance of perceived threats and panic attacks. ⁶ Anxiety has been found to have a pervasive impact on health, being associated with the burden of disability, use of both mental health services and general health services, and memory impairment. ⁷ , ⁸ , ⁹ It is also reported that it can worsen the prognosis of major depressive disorders and is associated with adverse cardiac events. ¹⁰ , ¹¹ Therefore, anxiety can affect public health immensely, yet its prevalence among older adults has been conflicting within the literature. ¹² , ¹³ It is vital to determine its global prevalence among older adults because of their prevalence, ¹⁴ personal/social costs, ¹⁴ and for early diagnosis and effective treatment. ¹⁵

There are many limitations for anxiety diagnosis in older adults because of a variety of older adult settings such as community samples, psychogeriatric samples, residential aged care settings, and in‐home care recipients and health settings including cognitive impairment, Parkinson's disease, and chronic obstructive pulmonary disease (COPD). ¹⁶ Furthermore, expressions of anxiety may change with age, further complicating its diagnosis, ¹⁴ with a recent study finding older adults with anxiety have more sleep disturbances and a lower need for reassurance due to worries compared to younger adults with anxiety. ¹⁷ Lack of persistent anxiety symptoms, ¹⁸ higher rates of somatic symptoms, ¹⁸ and anxiety‐induced memory impairment ¹⁹ are even more challenges. These limitations may cause underdiagnosis of anxiety in older adults and prohibit them from receiving effective treatments, ¹⁴ and may have played a role in previous studies reporting lower rates of anxiety in older adults compared to younger adults. ¹⁸

In order to overcome these limitations and find the real prevalence of anxiety in older adults, several tools have been developed, one of which is the geriatric anxiety inventory (GAI), a self‐reporting or nurse‐administrating scale with 20 items that can be used to distinguish people with or without anxiety. ²⁰ GAI has the ability to address many of these limitations by minimizing the emphasis on somatic symptoms. ¹⁶ The geriatric anxiety scale (GAS) is another tool that has shown acceptable psychometric properties in older adults. ²¹ Both of these tools have also shown acceptable performance in older adults with impaired memory. ²²

As mentioned, early diagnosis and effective treatment of anxiety can play an immense role in improving quality of life and decreasing comorbidities. ²³ , ²⁴ , ²⁵ Therefore, we have conducted a systematic review and meta‐analysis to estimate the global prevalence of anxiety symptoms in older adults using GAI or GAS and to identify associated factors.

2. METHODS

This systematic review and meta‐analysis adhered to the Preferred Reporting Items for Systematic Reviews and Meta‐analyses (PRISMA) guidelines. ²⁶ The study's protocol had been previously registered in PROSPERO (CRD42023425147) and was carried out following the Cochrane Handbook for Systematic Reviews of Interventions (Table S1).

2.1. Search strategy and selection criteria

Two independent investigators (AS and KJ) conducted a comprehensive search for relevant publications on February 27, 2023. The reports were then retrieved from Medline (via PubMed), Embase, Web of Science, and Scopus databases. The keywords used in database searches were as follows: (“elderly” OR “older adult” OR “geriatric” OR “senior”) AND (“geriatric anxiety scale” OR “geriatric anxiety inventory”) AND (“prevalence” OR “incidence” OR “percentage”) (Table S2), and were adapted as necessary for each database. No language or geographic restrictions were imposed on our searches. Moreover, a manual search of relevant reviews was carried out to find additional reports. The same two investigators then independently screened the titles and abstracts of potentially eligible studies and examined the full text of articles passing the initial screening. If there were a discrepancy between the two reviewers, a consensus would be reached via discussion.

2.2. Inclusion and exclusion criteria

The inclusion criteria according to the PICOS acronym were made as follows: Participants (P): older adults using study‐defined criteria. Intervention (I): not applicable; Comparison (C): not applicable; Outcomes (O): reporting the prevalence of anxiety symptoms assessed by the geriatric anxiety inventory (GAI) or geriatric anxiety scale (GAS); and Study design (S): epidemiological or cross‐sectional studies. Studies on patient populations who had major physical or psychological conditions or were under special circumstances (e.g., prisoners) were excluded. If more than one publication based on the same dataset was published, only the one with the largest sample and/or more complete report was included.

2.3. Data extraction

The data extraction was independently conducted using a standardized extraction form by two different investigators (SR and IM). Study characteristics that were extracted were as follows: the first author, year of publication, country, study type, number of participants, population, duration of the survey, sampling method, percentage of the male gender, mean age, type of measurement tool (GAI or GAS), the cut‐off score for assessing anxiety symptoms, percentage suffering from anxiety symptoms, and variables associated with anxiety symptoms.

2.4. GAI and GAS

The geriatric anxiety inventory is a valid and reliable 20‐question measurement tool devised by Pachana et al. in 2007. Initially, the GAI was formulated with 60 sample items from a wide array of other scales used to measure anxiety symptoms, and then 20 of the items most indicative of the anxiety level were chosen to be included in the final questionnaire. These items were chosen to reflect the primary domains covered in existing inventories: fearfulness, worry, meta‐worry (i.e., worry about worry), cognitions about anxiety, somatic symptoms of anxiety, anxious mood, and anxiety sensitivity. These types of items were common across all or almost all extant scales and broadly reflect anxiety symptomatology without being overly specific to any one type of anxiety disorder. The respondents answer the questions in an “agree/disagree” format, and a cut‐point of 10 positive responses is usually used for the detection of Geriatric Anxiety Disorder (GAD). ²⁰

The geriatric anxiety scale is a 30‐item questionnaire used to screen for somatic, cognitive, and affective anxiety symptoms among older adults. Individuals are asked to convey how often they have experienced each symptom in the last week, using a 4‐point Likert scale ranging from 0 (not at all) to 3 (all of the time), with higher scores indicating higher levels of anxiety. The GAS total score is based on the first 25 items. The additional 5 content items assess areas of anxiety that are of significant clinical concern. This measure has not been defined with a clear cut‐off point by the developers. ²¹

2.5. Quality assessment

We utilized the Joanna Briggs Institute (JBI) Critical Appraisal Tool for Prevalence Studies to evaluate the quality of the included reports. ²⁷ , ²⁸ The JBI Critical Appraisal Tool for Prevalence Studies is a systematic method designed to evaluate the quality of prevalence studies, assisting in critically assessing the validity, reliability, and applicability of prevalence studies in healthcare research. The tool consists of a set of 9 questions that cover key aspects of prevalence studies, including the framing and method of sample selection, adequacy of the size and description of the sample, whether the data analysis was conducted with sufficient coverage of the identified sample, whether the condition was identified with valid, standard, and reliable ways for all participants, whether the analysis was appropriate, and whether the response rate was adequate. For the purposes of our study, an adequate sample size for the quality assessment was calculated as 92 using the Naing et al. ²⁹ formula with an expected proportion of 40% and a precision of 10%. Studies that had either none or one “No” as responses to the questions were rated as good quality.

2.6. Data synthesis and analysis

We utilized the random effect model to combine data on the global prevalence of anxiety symptoms among older adults. This approach was necessary because of the variations in demographics and characteristics across the studies. Number of events and total participants were the only accepted data entry forms for the meta‐analysis. A p < 0.05 was considered significant. To measure the heterogeneity between studies, we employed the I ² statistics. If the I ² value exceeded 75%, it indicated significant heterogeneity. In case of significant heterogeneity, we performed sub‐group analysis and meta‐regressions with a p < 0.1 being considered significant. The subgroup analyses were based on different assessment tools, the country's income level according to the World Bank Country and Lending Groups classification, type of publication (whether it is a conference abstract or an article), setting (healthcare or community), continent of residence, and the quality of the study. Sensitivity analysis was conducted using the leave one out method. Additionally, we performed a meta‐regression to examine the relationship between the prevalence and various study covariates, such as publication date and the total sample size. Outlier studies were identified using Baujat's plot, which shows the contribution of each study toward overall heterogeneity. ³⁰ We evaluated publication bias using funnel plot visualization and the Egger test. All statistical analyses and graphical representations were carried out using R software using the meta package, ³¹ except for the sub‐group analysis which was conducted using StataCorp. 2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP.

3. RESULTS

3.1. Study selection

A comprehensive search of electronic databases yielded a total of 400 articles. After removing duplicates and conducting title and abstract screening, 38 articles were considered for full‐text assessment. ³ , ⁵ , ⁷ , ⁹ , ¹⁰ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁹ , ²⁰ , ²¹ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ Finally, 17 articles met the inclusion criteria and were included in the systematic review and meta‐analysis (Figure 1). ¹² , ¹³ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ , ⁴² , ⁴³ , ⁴⁴ , ⁴⁵ , ⁴⁶ , ⁴⁷ , ⁴⁸ , ⁴⁹ , ⁵⁰ , ⁵¹

FIGURE 1.

PRISMA flow diagram. *Databases searched are mentioned in methods; **excluded after title/abstract screening.

3.2. Characteristics of included studies

The included studies were conducted in various countries across different continents, providing a diverse representation of older adult populations. The sample sizes ranged from 58 to 897, with a median sample size of 274. Five studies used GAS, ¹² , ³⁷ , ⁴⁰ , ⁴⁵ , ⁴⁸ 6 used GAI, ¹³ , ³⁸ , ⁴¹ , ⁴² , ⁴⁴ , ⁴⁹ and 6 used the short form of GAI. ³⁹ , ⁴³ , ⁴⁶ , ⁴⁷ , ⁵⁰ , ⁵¹ Four of the effect sizes were conference abstracts. ¹² , ³⁷ , ³⁹ , ⁴⁵ The detailed results of the characteristics are presented in Table 1.

TABLE 1.

Characteristics of the included studies.

Author year	Study design	Country/income level	Total participants mean age (SD) male (%)	Population pre or post pandemic	Version of questionnaire	Adjusted variables associated with anxiety symptoms
Bendixen et al. 41	Cross‐sectional	Norway/High	473 75.1 (8.0) 31.9	Patients of 5 geriatric psychiatry departments in Oslo, Norway Pre‐pandemic	GAI	Being Female = standard beta of 0.119
Bekic et al. 40	Cross‐sectional	Croatia/High	159 71.26 (6.13) 35.84	Patients of a general medicine practice in eastern Croatia Pre‐pandemic	GAS	NA
Dow et al. 13	Cross‐sectional	Australia/High	87 76.95 (6.51) 34	Chinese senior groups in Melbourne Pre‐pandemic	GAI	NA
Ribeiro et al. 46	Cross‐sectional	Portugal/High	97 101.1 (1.5) 11.30	Centenarians living in Oporto and Guarda, Portugal Pre‐pandemic	GAI‐SF	Income inadequacy for medical expenses = unadjusted OR of 3.083 2 or more current diseases = unadjusted OR = 4.909/adjusted OR = 4.417 Bad self‐perception of health = unadjusted OR = 9.697/adjusted OR = 6.028 Acceptable self‐perception of health = unadjusted OR = 3.717/adjusted OR = 3.272 Loneliness = unadjusted OR = 3.450
Andrade et al. 38	Cross‐sectional	Brazil/Upper‐middle	380 67.2 (6.3) 24.7	60 years and older, living in Brazil, and have access to the Internet Post‐pandemic	GAI	Being female OR = 2.14 physically inactive OR = 4.29 Physically active 1–3 times/week OR = 3.48 Never or almost never receive support from family/friends OR = 2.13 Schooling for 8 years or less OR = 1.26
Agarwal et al. 12	Cross sectional	India/Lower‐middle	110 NR NR	Medically ill, ≥60 years attending outdoor of medicine department, ELMCH, Lucknow Pre‐pandemic	GAS	NR
Atti et al. 39	Cross‐sectional	Italy/High	462 NR NR	75+ living in Faenza (Northern Italy) Pre‐pandemic	GAI‐sf	NR
Das et al. 42	Cross‐sectional	India/Lower‐middle	92 75% between 60 and 70 64.10	60 years and above, residing in two sectors of Chandigarh city 20 , 37 for at least a year with an Everyday Abilities Scale for India (EASI) score not exceeding zero Post‐pandemic	GAI‐Hindi	NR
Forlani et al. 43	Cross‐sectional	Italy/High	366 46.8 NR	Community‐dwelling older adults in Faenza Pre‐pandemic	GAI‐sf	Being female OR = 2.8 Depression OR = 5.6 per rank Physical morbidity OR = 3.5
Gluyas et al. 44	Cohort study	Australia/High	88 78.52 (7.11) 35	Consecutive patients admitted to two subacute aged care and rehabilitation wards in Melbourne, Australia+ at least 60 years old + an MMSE score of at least 24 Pre‐pandemic	GAI	NR
Mnif et al. 45	Cross sectional	Tunisia/Lower‐middle	58 74.2 NR	Primary care patients in Sfax Post‐pandemic	GAS‐10	Negative correlation: Time spent outside, quality of life Positive correlation: number of chronic somatic pathologies and the number of treatments taken
Not provided 37	Cross sectional	India/Lower‐middle	196 NR NR	Above 60 years of age living in old age home of tier three city of Western India Pre‐pandemic	GAS	Social stress Spearman r = 0.3073
Russell et al. 47	Cross‐sectional	Australia/High	274 65.1 (10.8) 34.3	Aboriginal and Torres Strait Islander residents Pre‐pandemic	GAI‐SF	Age (rho = −0.163) Prior head injury The incontinence scale (rho = 0.251) Pain scale scores (rho = 0.284)
Sharma et al. 48	Cross‐sectional	Nepal/Lower‐middle	245 78.3 (6.3) 47.8	Senior citizens living in Tansen Municipality, Palpa Post‐pandemic	GAS‐10	Widowhood (OR = 3.002) Living alone Inability to read and write (OR = 3.673) Physical dependency (OR = 6.698) Presence of physical health problem‐Depression (Spearman's Rho (rs) =0.615)
Sirin et al. 49	Cross‐sectional	Turkey/Upper‐middle	278 72.64 (6.323) 43.9	Older adults in the central district of Ankara Post‐pandemic	GAI	Being female (OR = 2.275) Financial loss (OR = 2.310) Knowledge about COVID‐19 (OR = 3.046) Belief that COVID‐19 cannot be successfully controlled (OR = 2.503) Education level (OR = 0.238) Hobby (OR = 0.456) Following news about COVID‐19 (OR = for every 1 h increase the odds of GAD by 1.188)
Welzel et al. 50	Cohort study	Germany/High	897 86.8 (3.02) 34	Older primary care patients in six study centers across Germany Pre‐pandemic	GAI‐SF	Recent loss (OR = 1.91) Depressive symptoms (OR = 9.26) Higher Mini‐Mental State Examination (OR = 0.89)
Xie et al. 51	Cross‐sectional	China/Upper‐middle	753 72.8 (6.0) 46.3	Primary care patients in Wuhan, China Pre‐pandemic	GAI‐SF	Being female OR = 1.85 Poor economic status OR = 2.31 Fair and poor family relationship versus good, OR = 1.85 Hypertension OR = 2.01 Chronic gastric ulcer OR = 6.82 Parkinson's disease OR = 7.83

3.3. Global prevalence of anxiety symptoms in older adults

The meta‐analysis revealed that the global prevalence of anxiety symptoms among older adults was estimated to be 28% (95% CI: 21%–35%; Prediction interval: 4%–61%; I ²: 96%; Tau2 = 0.0242; p < 0.0001) (Figure 2). This indicates a significant burden of anxiety in this population.

FIGURE 2.

Results of meta‐analysis for the prevalence of anxiety in older adults.

3.4. Factors associated with anxiety symptoms in older adults

Eleven of the included studies investigated factors associated with anxiety in older adults and were included in our systematic review, ³⁷ , ³⁸ , ⁴¹ , ⁴³ , ⁴⁵ , ⁴⁶ , ⁴⁷ , ⁴⁸ , ⁴⁹ , ⁵⁰ which identified several factors. These factors can be categorized into the following domains and are available in Table 2.

TABLE 2.

Summary of associated factors with anxiety in older adults identified among the studies.

Associated factor	Reference
	51	50	49	48	47	37	45	43	38	46	41
Advancing age
Female gender
Widowhood
Low income
MMSE score
Number of medications
Co‐morbid Depression
Physical morbidity
Physical Comorbidities
Loneliness
Low Social support

Note: Red, significant positive association after adjustment; Blue, nonsignificant association; Yellow, significant positive association which disappeared after adjustment; Green, significant negative association; Black, not assessed in the study.

3.4.1. Sociodemographic factors

Advanced age was associated with a lower risk of anxiety symptoms in one study, ⁴⁷ yet 4 other studies assessing age found no significant association. The female gender has also been found to correlate with a higher prevalence of anxiety symptoms compared to the male gender in 5 studies, ³⁸ , ⁴¹ , ⁴³ , ⁴⁹ , ⁵¹ yet 4 studies found no association. Widowhood was found to increase anxiety in one study of 245 older adults, yet 3 separate studies have found no association. Lower socioeconomic status was also found to be associated with an increased risk of anxiety symptoms in one study, ⁵¹ yet two other studies have not found a significant association. The study by Ribeiro et al., however, found a significant association between income adequacy for medical costs and anxiety in older adults (OR = 3.08). ⁴⁶

3.4.2. Health‐related factors

Higher intakes of prescription drugs have been found to increase anxiety in older adults in 2 studies, although the correlation disappeared in one study after adjustment and a separate study found no association at all. Older adults with chronic conditions such as cardiovascular diseases, diabetes, hypertension, and respiratory disorders were more likely to experience anxiety symptoms, ⁴⁵ , ⁴⁶ , ⁵⁰ , ⁵¹ yet one study found no association between the two. ⁴⁹ Previous head injury has also been reported to increase anxiety scores among older adults. ⁴⁷

Decreased physical functioning and mobility limitations were associated with higher anxiety levels in all 3 studies on the matter, ⁴³ , ⁴⁸ , ⁴⁹ while the study by Andrande et al. found older adults who are less physically active have an increased odds of anxiety. ³⁸

Two studies indicated a positive association between cognitive impairment and anxiety symptoms in older adults ⁵⁰ , ⁵¹ and higher intakes of alcohol were associated with higher rates of anxiety in a study by Forlani et al. ⁴³

3.4.3. Psychosocial factors

Mini‐mental status examination scores and anxiety prevalence in older adults were assessed in 3 studies, of which only 1 found a significant negative association. ⁵⁰ Higher MADRS scores have also been associated with higher rates of anxiety in older adults in a study of 473 older adults, yet only 1 study has investigated the matter. ⁴¹ Loneliness and social isolation have been attributed to anxiety in 2 of 4 studies investigating the matter, yet low social support increases anxiety in older adults in all 3 studies conducted on the matter.

Co‐morbid depression also had a significant positive association with anxiety in all 4 studies investigating the matter, yet co‐morbid dementia, mania, and psychosis were found to have a negative association with anxiety in older adults in one study and have not been investigated in any other study. ⁴¹ Anxiolytic and antidepressant usage increases anxiety in older adults, yet this finding is also limited to a singular study. ⁴¹ Major life events, such as recent loss, were identified as potential triggers for the incidence of anxiety symptoms in two studies. ⁴⁹ , ⁵⁰

3.5. Heterogeneity and publication bias

Publication bias was assessed using the funnel plot and Egger test. There was no indication of publication bias (p = 0.19). (Figure S1).

Significant heterogeneity was observed among the included studies (I ² = 96%, p < 0.01), indicating substantial variability in the prevalence estimates. As there was unexplained heterogeneity, we used the Baujat plot and identified 2 studies as outliers ⁴¹ , ⁵⁰ (Figure S2).

3.6. Quality assessment

The overall quality of the included studies varied, with most studies demonstrating moderate to good quality. However, a few studies had limitations in terms of sample size and peer‐reviewed domain (Table S3).

3.7. Subgroup analysis, meta‐regression, and sensitivity analysis

Subgroup analyses were performed based on income level, assessment tool, quality of studies, setting (healthcare or community), continent of residence, and type of publication to explore potential sources of heterogeneity. The results showed a significant effect of geographical location (Figure 3). Meta‐regression was conducted to examine the influence of covariates, such as study year and sample size, on the prevalence estimates. The results of our meta‐regression did not show a significant relationship between the prevalence and the total sample size of the study (p = 0.26) or study year (p = 0.85). The results of sensitivity analysis showed robustness of findings, with the prevalence ranging between 22% and 27%. (Figure S3).

FIGURE 3.

Subgroup analysis results showing significant between group differences when dividing studies based on their: Type of study; setting; income; location; quality, and assessment tool.

4. DISCUSSION

The present systematic review and meta‐analysis aimed to investigate the global prevalence of anxiety in older adults and explore the associated factors contributing to anxiety in this population. The findings provide valuable insights into the burden of anxiety among older adults worldwide and shed light on important factors that contribute to its occurrence. The estimated global prevalence of anxiety among older adults is 25%, indicating that approximately one in four older adults may experience anxiety. This finding highlights the significant burden of anxiety within this population, underscoring the importance of addressing and providing support for older adults affected by anxiety. The factors associated with anxiety in older adults can be categorized into several domains, including sociodemographic, health‐related, and psychosocial factors. Advanced age was consistently identified as a risk factor for anxiety, suggesting that older individuals may be more susceptible to experiencing anxiety symptoms. This finding aligns with previous research highlighting the challenges and transitions that accompany the aging process, such as declining health and loss of social support systems. ⁵² Sociodemographic factors such as gender and socioeconomic status also emerged as potential contributors to anxiety in older adults. ⁵³ Females were found to have a higher prevalence of anxiety compared to males, which may be attributed to biological, psychological, and social factors. Additionally, lower socioeconomic status was associated with an increased risk of anxiety, potentially because of limited access to resources and support networks. Health‐related factors played a significant role in the occurrence of anxiety among older adults. The presence of chronic medical conditions, such as cardiovascular diseases, diabetes, and respiratory disorders, was consistently associated with a higher likelihood of experiencing anxiety. ⁵³ The burden of managing chronic illnesses, along with the associated physical limitations and psychological distress, may contribute to the development of anxiety symptoms. Functional impairment and cognitive decline were also identified as risk factors for anxiety, indicating the interplay between physical and mental health in older adults. Psychosocial factors, including social isolation, loneliness, and major life events, were found to be influential factors in the development of anxiety. Older adults who reported low social support or lived alone were more likely to experience anxiety symptoms. Major life events, such as bereavement or retirement, were also identified as triggers for anxiety. Additionally, the responsibility of being a caregiver for a spouse or family member increased the risk of anxiety in older adults.

As we found the prevalence of anxiety to be high among older adults relative to other age groups, ⁵ we recommend that regular, community‐based interventions such as screenings be conducted to detect and treat this condition in its infancy. In addition, the heterogeneity observed in the prevalence estimates across studies highlights the need for further exploration of contextual factors and methodological differences that may contribute to this variation. These analyses provided additional insights into the variations observed across studies, which may help tailor interventions and policies to specific populations. For instance, it is advisable to screen for and detect physical comorbid conditions, as they have been shown to increase the risk of anxiety. Additionally, it is important to bolster the social support systems of older adults, design and develop programs they can participate in after retirement, and develop assistance programs that alleviate the responsibility that some bear for other family members. And finally, as the results of our subgroup analysis showed the higher prevalence of anxiety symptoms in countries with lower income levels, national and international policies can be tailored to allocate the appropriate resources to each country. Additionally, as we did not find significant differences in subgroups of income, assessment tool, study quality, and publication type, these factors can be given lower priority in terms of informing and influencing policies and interventions.

According to The Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2019, depression and anxiety were two of the most disabling disorders, both ranked among the top 25 leading causes of global burden in 2019. ⁵⁴ In 2020, before the COVID‐19 pandemic, the global prevalence of anxiety disorders was 3824·9 per 100,000 population, which is equivalent to 298 million (256–348) people, but after the COVID‐19 pandemic, the estimated global prevalence of anxiety disorders was 4802·4 (4108·2–5588·6), equivalent to 374 million (320–436) people. ⁴⁸ Social isolation as a result of the pandemic influenced the older adult population immensely; complications in having routine physical and mental healthcare; fear caused by uncertainty of life; and social distancing have led to mental disorders in older adults, including anxiety. ⁵⁵ Although we did not observe trends that are indicative of higher postpandemic anxiety levels among older adults, further investigation is warranted to address the effect of the pandemic on anxiety levels in this population. Anxiety can be compounded with other mental disorders, including depression and cognitive impairment. ⁵⁶ The result of a systematic review done by Horestein et al. has shown that in most cases anxiety leads to increased health utilization which can cause increased healthcare costs and can lead to a strain on healthcare resources. ⁵⁷ Also, anxiety can result in economic burden because of misdiagnosis, undertreatment, medical treatment costs, indirect workplace costs, mortality costs, and prescription drug costs Indirect costs include lost production and income losses because of absences and disability which totally can cost $46.6 billion. ⁵⁸ Furthermore, in some cases, anxiety can lead to delayed, irregular, or inconsistent use of healthcare services. ⁵⁷

It is important to note that several medical conditions with similar symptoms, including hyperthyroidism, pheochromocytoma, hyperparathyroidism, arrhythmia or obstructive pulmonary diseases, temporal lobe epilepsy, or transient ischemic attacks, should be ruled out. Also, other psychiatric disorders (major depressive disorder, bipolar disorder) and the use of substances or substance withdrawal should be excluded. ³⁶

Therrien and Hunsley conducted one of the first systematic reviews on this topic. ³³ Although their study does not directly provide prevalence estimates, it offers valuable insights into the commonly used measures for assessing anxiety in this population. The findings of their review can be compared and complemented with our study, as they provide a foundation for understanding the measurement tools used in assessing anxiety among older adults. Creighton et al. conducted a systematic review specifically exploring the prevalence of anxiety among older adults in nursing homes and residential aged care facilities. Their study focused on a specific setting, which differs from our global perspective. ⁵⁹ However, their findings can still be relevant for comparison, as they provide insights into anxiety prevalence within a specific population of older adults receiving institutional care. By comparing their findings, which show a prevalence of 3.2% to 20% among older adults in these settings, we can examine potential differences in the prevalence rate of anxiety according to the residency status. Recently, Yan et al. conducted a systematic review and meta‐analysis specifically focused on the prevalence of depressive and anxiety symptoms among Chinese older adults during the COVID‐19 pandemic. ³² They showed among different groups, including the general population, those affected by chronic illnesses, and individuals experiencing the COVID‐19 situation, the prevalence rates of anxiety symptoms were found to be 23%, 85%, and 14%, respectively. These results showed the significant effect of chronic disease on the overall prevalence of anxiety in this population.

Using the Geriatric Anxiety Inventory (GAI) as the sole measurement tool for estimating the prevalence of anxiety in older adults offers several advantages. First, the GAI is specifically designed to assess anxiety symptoms in this population, ensuring its relevance and accuracy. Second, it has been extensively validated, demonstrating good psychometric properties such as reliability and validity. Its comprehensive assessment covers a wide range of anxiety symptoms, including both cognitive and somatic domains. Additionally, the GAI is sensitive to subclinical anxiety, capturing individuals who may not meet diagnostic criteria but still experience significant symptoms. Furthermore, its use promotes comparability and standardization across studies, facilitating meta‐analyses and systematic reviews for a more comprehensive understanding of anxiety prevalence in older adults. Overall, the GAI provides a specialized, valid, reliable, and comprehensive assessment, enhancing our knowledge of anxiety in older populations. ²⁰ , ⁶⁰ Despite the numerous advantages of GAI, many of the included studies utilizing it have administered an unvalidated translated version to the subjects. It is recommended that in future studies, psychometric evaluations of the translated tool be conducted before administration. Evidence supports that GAI and GAI‐SF both have validation for being a good predictor of anxiety in older adults and are also adequate for detecting and screening anxiety in both clinically disordered and nonclinical older adults. ¹⁶ The present study is the first systematic review and meta‐analysis that has determined the prevalence of anxiety in older adults using the specific tools designed for this population (GAI/GAS).

Several limitations should be acknowledged when interpreting the findings of this systematic review. First, the majority of included studies were cross‐sectional, limiting the ability to establish causal relationships between anxiety and associated factors. Longitudinal studies are warranted to further explore the temporal relationship between anxiety and its contributing factors. Second, publication bias and language restrictions may have influenced the findings, as studies with significant results and those published in specific languages may be more likely to be included. Efforts were made to mitigate this bias by conducting a comprehensive search and including studies from diverse regions. Third, most older persons included in the studies were either residents of older adult care facilities or were hospital patients, which may not be representative of the more general older populace, as there is a higher overall prevalence of anxiety disorders among older aged care residents. ⁵⁹ Lastly, the sample size for our review was roughly 4200, which is not enough to interpret any useful findings for the over 1 billion older populace across the world.

Despite these limitations, this systematic review and meta‐analysis provide a comprehensive overview of the global prevalence of anxiety in older adults and its associated factors. The results have important implications for healthcare professionals, policymakers, and researchers involved in older adults' mental health. The identification of factors contributing to anxiety can inform the development of targeted interventions and strategies to prevent and manage anxiety symptoms in older adults. Future research should focus on longitudinal studies, standardized measurement tools, and culturally sensitive approaches to further understand the complex nature of anxiety in this population and improve the well‐being of older adults worldwide.

AUTHOR CONTRIBUTIONS

Arman Shafiee, Abolfazl Abdollahi: Conceptualization, Project Administration, Data curation, Writing—Original Draft, Writing—Review and Editing, Visualization. Kyana Jafarabady, Arman Shafiee: Validation, Resources, Methodology, Software, Formal analysis, Writing—Original Draft. Ida Mohammadi: Writing—Original Draft. Shahryar Rajai: Data curation.

FUNDING INFORMATION

This study did not receive funding, grant, or sponsorship from any individuals or organizations.

CONFLICT OF INTEREST STATEMENT

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript.

ETHICS STATEMENT

Not applicable.

CONSENT FOR PUBLICATION

Not applicable.

Supporting information

Data S1:

Shafiee A, Mohammadi I, Rajai S, Jafarabady K, Abdollahi A. Global prevalence of anxiety symptoms and its associated factors in older adults: A systematic review and meta‐analysis. J Gen Fam Med. 2025;26:116–127. 10.1002/jgf2.750

DATA AVAILABILITY STATEMENT

The data has not been previously presented orally or by poster at scientific meetings. All data has been presented in the manuscript.