Background

Falls have emerged as a major health issues, and health care is paying increasing attention to them as the primary cause of illness and early mortality among older adults. At least one in three older adults experience a fall each year, and fear of falling (FOF) causes more falls [1]. Negative fall-related consequences, such as fear of falling, increase the likelihood of disability and declining quality of life in relation to health [2]. FOF has been defined as the low perceived self-efficacy at avoiding falls in daily activities [3]. FOF is prevalent in older adults and may lead to several negative health outcomes such as changing gait [4], restricting daily activities [5], developing depression [5], increasing the risk of falls [6], and negatively impacting quality of life [2]. Earlier studies revealed that the 8-year mortality rate was nearly 14% greater among older adults with FOF than among those without FOF, and 16% greater among older adults with limited activities than among those without restricted activities [7]. In particular, avoiding certain situations can lead to fear of falling, which exacerbates poor physical health, balance concerns, and social isolation [8].

Fear of falling is common among older adults regardless of they have a history of falls. A study showed that the prevalence of FOF ranged from 3 to 85% among community-dwelling older adults with a history of falls [9]. However, different individuals and those from other nations reported the prevalence of FOF in different ways. FOF also occurs frequently among hospitalized patients [10, 11], especially among those who have undergone total knee arthroplasty [12], who have a hip fracture [13], and who have diabetic neuropathy [14]. Similarly, more than 50% of people with Parkinson’s disease had high levels of FOF, while close to 30% had moderate levels [15]. Previous studies demonstrated that there was significant variation in the incidence of FOF among older adults living in communities of different nationalities., ranging from 9.26% to 83.33% in the USA [16, 17], from 22.31% to 86.46% in Japan [18, 19], from 38.84% to 96.70% in Korea [20, 21], from 37.03% to 51.38% in Spain [22, 23], and from 44.56% to 86.71% in Turkey [24, 25]. At present, various tools are available to assess FOF, including the single question “Are you afraid of falling?”, the Falls Efficacy Scale (FES), the Falls Efficacy Scale-International (FES-I), the Short Falls Efficacy Scale International (SFES-I), the Modified fall efficacy scale(MFES) and the Activities-Specific Balance Confidence Scale (ABC). Moreover, earlier studies showed that similar results were obtained utilizing a single-question tool and various structured questionnaire instruments [26].

Age, female sex, balance, living alone, chronic illnesses, and psychiatric issues were all risk factors for FOF. People with neurological disease and a history of falls in the previous 6 months were more than twice as likely to have FOF, while those with depression were more than six times more likely to have FOF [27]. Compared with nonfrail older adults, those with a frail physical condition and lack of daily activity had a greater than three times greater risk of experiencing FOF, and both female sex and depression found to be independent predictors of FOF [28]. A systematic review revealed that FOF in stroke patients was closely correlated with female sex, impaired balance ability, decreased mobility, a history of falls and walking aids, and decreased weight, which may present more challenges for impaired balance ability and excessive safety awareness of life circumstances and daily activities, increasing the risk of FOF and leading to the development of psychological stress [29]. Evidence has shown that gait variability, such as slowing gait speed, shorter strides, and widening strides, is similarly associated with FOF; however there are no conclusive findings from brain imaging to date [4]. Cognitive impairment, which was confirmed to be a predictor of having a significant effect on a high level of FOF, could increase the risk of accepting high FOF by three times in the older adults with low and moderate levels of social support caused by limited social activity, lack of family support and an aging-unfriendly environment [30]. Additionally, chemotherapy might make cancer patients more feeble and reduce their postural and limb stability because taxanes and platinum drugs harm muscle and peripheral nerve tissue, which reduces the efficacy of falls [31].

In conclusion, FOF is complicated by people’s physical, psychological, and social support systems, and its occurrence varies widely among studies involving different subjects, tools, and nations. Notably, the global prevalence of FOF is currently unknown. Therefore, this study includes a systematic review and meta-analysis to address the limitations of previous studies by estimating the global prevalence of FOF among older adults and fully exploring its risk factors for further developing precise interventions to systematically manage FOF.

Methods

Design

This systematic review and meta-analysis was conducted according to the Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines and was registered in the International Database of Prospectively Registered Systematic Reviews (ROSPERO: CRD42022358031).

Search methods

A systematic search of the literature was performed in PubMed, Web of Science, EMBASE, and the Cochrane Library from the inception of the database until August 20, 2022, and updated to September 2, 2023. The search strategies were developed using a combination of MeSH terms and free words with Boolean operators. The details were as follows: (“Aged”[Mesh] OR “older” OR “older adult” OR “older” OR “elderly” OR “the aged”) AND (“fear of falling”) AND (“influence factors” OR “risk factors”) (see PubMed search strategies in Supplementary Material 1) and manually searched methods were used to check potential studies.

Inclusion and/or exclusion criteria

After removing duplicate studies, two reviewers independently reviewed studies based on the title and abstract and subsequently screened the literature based on full-text via the inclusion and exclusion criteria. The inclusion criteria for the studies were as follows: 1) participants were at least 60 years old; 2) the prevalence or risk factors for FOF were reported; 3) the study design was observational study, including cohort, case-control and cross-sectional studies; and 4) the study was published in the English language. The exclusion criteria were as follows: 1) full-text could not be obtained, or 2)incomplete or erroneous data.

Data extraction

Data were extracted independently by two researchers with the following variables: first author name, publication year, country, type of study instrument, subject, age, female ratio, sample size, prevalence of FOF, quality of studies and risk factors for FOF. If FOF was assessed using kinds of different instruments, the prevalence of FOF was extracted according to the results of the eligible studies. All disagreements were resolved by discussion between two researchers, and a third researcher was consulted if needed.

Quality assessment

Two researchers independently assessed the possibility of bias, and disagreements were settled by discussion or consultation with a third researcher. The quality of the case-control and cohort studies was assessed by using the Newcastle–Ottawa Scale [32], which has 8 items and a total score ranging from 0 to 9, and scores ranging from 0 to 3, 4 to 6 and 7 to 9 indicated low, medium and high quality, respectively. The quality of cross-sectional studies was assessed by using the instrument Agency for Healthcare Research and Quality (AHRQ) [33] with a total scores ranging from 0 to 11, and scores ranging from 0 to 3, 4 to 7 and 8 to 11 indicating low, medium, and high quality, respectively. The detailed items of the AHRQ and the Newcastle–Ottawa Scale are shown in Supplementary Material 2.

Data analysis

Stata 12.0 software was used to analyze all the data, and odds ratios (ORs) were calculated as the effect size meanwhile 95% confidence intervals (CIs) were provided. Heterogeneity was tested by I2, and I2 > 50% indicated high heterogeneity while I2 < 50% indicated low heterogeneity [34]. Pooled effect size was analyzed using a fixed-effects model if I2 < 50%. Subgroup and sensitivity analysis were used to analyze the sources of heterogeneity if I2 > 50%, and subsequently, a random effects model was used. To assess risk factors for FOF, the odds ratios (ORs) and 95% CIs, which reported the association between risk factors and FOF in eligible studies, were extracted to estimate the pooled effect size in meta-analyses using a fixed-effects model if I2 < 50%, or a random effects model if I2 > 50%. Publication bias was assessed through funnel plots, Egger’s test and Begg’s test [35].

Results

Study process

A total of 3452 studies were retrieved from databases and manual searches, and 1491 duplicate studies were eliminated. A total of 277 studies that met the inclusion criteria were selected after screening titles and abstracts, 124 of which were excluded after screening the full text: 16 did not publish in English, 67 did not report the prevalence of FOF, and 41 did not provide the full text. Finally, 153 studies with 200,033 participants from 38 countries were included and analyzed (shown in Fig. 1).

Fig. 1
figure 1

PRISMA flow diagram of literature search and selection of included studies for meta-analysis

Full size image

Characteristics of the included studies

The characteristics of the 153 studies, 64(41.83%) of which were of medium quality and 89(58.17%) of which were of high quality, are summarized in Table 1. The publication years of all studies ranged from 1994 to 2023, and there were 200,033, with 112,697(56.34%)female. The details of the 153 studies are shown in Supplementary Material 3.

Table 1 Characteristics of included studies (N = 153)
Full size table

Global prevalence of FOF

The global prevalence of FOF among older adults widely ranged from 6.96% to 90.34% in the 153 studies. The overall prevalence of FOF was 47.80% [95% CI: 47.7%–48.0%], with high heterogeneity (χ2 = 50,648.15, I2 = 99.7%, p < 0.001). A random effects model was then constructed, and the results showed that the overall prevalence of FOF was 49.60% [95% CI:45.9%–53.2%, I2 = 99.7%, p < 0.001] (as shown in Supplementary Material 3).

Subgroup analysis

The subgroup analysis by region, country, subject and instrument used is shown in Table 2. The estimates of the pooled prevalence of FOF were higher in Africa and Asia than in other regions, at 56.80% and 52.90%, respectively; were higher in developing countries (53.40%) than in developed countries (46.7%); and were higher in patients (52.20%) than in community residents (48.40%). Moreover, there is a difference in the prevalence of FOF among different instruments.

Table 2 Subgroups analyses by regions, countries, subjects and instruments
Full size table

Risk factors for FOF

A total of thirty-eight risk factors for FOF were analyzed, and twenty-eight of them were significant significantly associated FOF (p < 0.05), including demographic characteristics (e.g., female sex, age (70–84 years), low education level, living alone, high BMI, etc.), physical function(e.g., using walking aid, frailty status, poor perceived health, Timed Up and Go test results (abnormal), balance problems), chronic diseases(e.g., diabetes, hearing impairment, visual impairment, body pain, dizziness, number of chronic diseases, etc.) and mental problems (e.g., anxiety and depression), while ten of them were not (p > 0.05), as shown in Table 3.

Table 3 Pooled risk factors of fear of falling(FOF)
Full size table

Publication bias and sensitivity analysis

Begg’s test (z = 1, p = 0.320) and Egger’s test (t = 15.34, p < 0.001) revealed the potential publication bias of the included literature, and the funnel plot showed a asymmetry (shown in Fig. 2). However, the sensitivity analysis of this finding was robust (shown in Fig. 3).

Fig. 2
figure 2

Funnel plot for assessing publication biases

Full size image
Fig. 3
figure 3

Sensitivity analysis of global prevalence of FOF

Full size image

Discussion

This study was the first systematic review and meta-analysis to analyze the global prevalence of FOF among older adults, and to fully explore its potential risk factors. A total of 153 studies involving 200,033 participants from 38 countries revealed that the prevalence of FOF ranged widely from 6.96% to 90.34%, which was lower than that reported (ranging from 22.5% to 100%) among hip fracture patients [168], and the pooled global prevalence of FOF was high at 49.60%, which was similar to the results (44.6%) of previous research [25]. Subgroup analysis revealed that the pooled prevalence of FOF was higher in Africa and Asia than in other regions, higher in developing countries than in developed countries, and higher in patients than in community residents. In addition, twenty-eight potential risk factors were found to be significantly associated with FOF, mainly including demographic characteristics, physical function, chronic diseases and mental problems, which was the same as that reported in earlier studies [29, 169].

Overall, this study revealed that the global prevalence of FOF among older adults was high. One important reason was the increasing aging of the global population, which increased the prevalence of FOF among older adults. The WHO’s Aging and Health Report showed that in older adults, falls, as one of the common health conditions associated with aging, could lead to major public health problems and socioeconomic burdens [170], and FOF, as a fall-related mental problem, could increase older adults’ risk of falls, and these two factors could form a vicious cycle. Unlike in general older adults, the prevalence of FOF among patients after hip fracture tended to decrease within 4 weeks, approximately 12 weeks and over 12 weeks, at 50% to 100%, 47% to 59%, and 23% to 50%, respectively [168]. Moreover, the findings of this study revealed that the prevalence of FOF in Africa and Asia was high. Vo, et al. [169] reported that FOF among older adults Southeast Asia ranged from 21.6% to 88.2%, which was probably explained by a social environment that was unfriendly toward older adults, population aging, unbalanced economic conditions and a lack of familial support throughout urbanization. A previous study revealed that in developing countries, the high prevalence of FOF might be caused by low levels of education that prevent people from successfully managing FOF on their own, health caused by chronic diseases, inability to participate in social activities, and insufficient medical resources to properly manage both physical and psychological concerns [49], and inadequate service systems might have an impact on people’s ability to self-manage poor coping skills. Notably, although developed countries have better economic conditions, access to health care, educational opportunities, and social services than developing countries, some of them have high rates of FOF, such as the USA [16], Spain [28], Korea [20], and Japan [19]. It is likely that unhealthy lifestyles and diet habits lead to abnormal BMIs [54]. Ercan [171] reported that obesity could impact individuals’ posture and lead to balance problems, and obese females have higher FOF, higher activity restriction, and lower activity confidence than obese males. Earlier evidence demonstrated that older women with a high waist circumference had three times more likely to develop FOF than were those with a low waist circumference, which could alter the body’s center of gravity, further impair postural stability, and contribute to FOF [155]. However, another study showed that although BMI could slightly influence on body swing on unstable surfaces, obesity was not associated with FOF [171]. Furthermore, James [42] attempted to investigate the effect of the English language on fear of falling among Mexican-Americans in USA, but the results showed that not speaking or understanding English did not increase the incidence of FOF among those less than 80 years old, but it could affect activity restriction, to some extent.

Compared with community-dwelling residents, the prevalence of FOF was higher among those with chronic disease, especially patients with hip fracture [172], knee osteoarthritis [12], diabetes [173], etc. Previous studies have shown a high incidence of fear of falling in patients with hip fractures who underwent surgery involving knee replacement, total hip replacement or spinal surgery [37, 39], and FOF and cognitive impairment had a stronger impact on functional rehabilitation than did pain and depression [172]. In diabetic patients, symptoms of peripheral neuropathy, such as pain, feeling of ant walking, freezing, and burning, eventually impeded their ability to move and increased their likelihood of experiencing fear of falling [14]. Chronic pain has also been confirmed to m increase individuals’ susceptibility to FOF, and it plays a mediating role between FOF and poor physical performance [108]. Moreover, a qualitative study revealed that FOF gave patients with Parkinson’s disease (PD) a sense of insecurity, vulnerability and danger in daily activities, and when facing PD-related symptoms, such as rigidity, gait freezing or balance problems, positive emotions would help them successfully cope with FOF [174]. However, cardiopulmonary pattern (hypertension) and cognitive impairment were not significantly associated with FOF in this study, which was not consistent with previous studies [30], in which the relationship between cognitive impairment and FOF decreased due to the effect of high social support. In addition, the meta-analysis of risk factors in this study indicated that regardless of the number of chronic diseases, they negatively effected on FOF. On the one hand, multiple comorbidities can affect the multiorgan function of older adults, and on the other hand, due to a reduction in metabolic function, the side effects of treating this disease with multiple drugs can negatively impact on health.Therefore, multidisciplinary cooperation, including rehabilitation, pharmacy, nutrition, psychology, etc., can help to prevent and reduce FOF among older adults.

Demographic characteristics (etc., age, female sex, low education level, living alone, history of falls) are the well-known, significant factors of FOF among older adults. Birhanie [49] reported that compared with individuals aged 60 to 70 years, those aged more than 70 years were four times more likely develop FOF, which was similar to our results. Previous studies have shown that females had a greater risk of fear of falling than men do, and a decrease in estrogen among older women could cause osteoporosis, bone hyperplasia and a decrease in limb muscle mass, further leading to a weakened musculoskeletal system [49]. Additionally, those with back pain, mental health conditions and neurological disorders are more likely to develop chronic disease [175]. Thiamwong [91] reported that the female sex and low education were closely associated with fear of falling, and the latter was important for preventing individuals from engaging in FOF education and learning how to prevent it. Moreover, according to a previous investigation, nearly 80% of older adults with a history of falling had a high fear of falling, especially among those who were over 85 years old, for whom nearly 95% of the participants were adults [7]. Frankenthal [84] indicated that the prevalence of FOF (69.8%) among people with a history of falls was higher than that among people without a history of falls (41.4%). Notably, living alone was also a significant factor for FOF, while being unmarried not. Older adults who lived alone had no assistance in daily activities and had no else help in dangerous situations. However, interestingly, De Roza [9] reported that older adults who were married status had greater FoF than those who were never married, which could be explained by the fact that those who never married might have developed great independence at an early age. In addition, because of the great independence, older adults who unmarried might have greater psychological resilience and better ability to cope well with FOF. Furthermore, we also found that the low social support was not significantly related to FOF in this study. Dierking [176] noted that social support had both positive and negative effects on people’s health, and that familial conflict could increase the risk of FOF, but friend support had a positive effect on preventing FOF. Hence, we suggest that actively addressing family conflicts and more social networks should be considered in FOF prevention programs.

Physical function, such as using walking aid, frailty, dependent daily activities, and balance problems, was significantly related to FOF, which was consistent with the findings of Gadhvi [168]. Birhanie [49] showed that older adults who used walking aids were fourteen timed more likely to developing FOF than those who did not use them. De Roza [9] noted that older adults who used quad sticks had greater FOF than did those who used umbrellas or walking sticks and that the use of walking aids was closely related to frailty, which subsequently impacted FOF. Furthermore, previous studies had reported that FOF and its related activity restriction were associated with impaired gait, balance problems, frailty, sarcopenia, depressive symptoms, and mortality [5, 144, 177]. Moreover, consistent with a previous study [49, 147], depression and anxiety were found to be the most common, significant psychological risk factors for FOF in this study. A meta-analysis by Gambaro et al. [6], revealed that FOF might play a mediating role between depression and falls. In addition, social culture and attitudes regarding aging-related changes were found to be strongly associated with FOF [178]. For example, one of the important reasons for the high prevalence of FOF among Korean older adults was the use of public transportation, such as buses or subway [48]. Therefore, we suggest that in addition to improving physical function, increasing balance confidence and changing the incorrect cognition of FOF, social infrastructure (e.g., walking paths, public transportation), home environments(e.g., using automated LED lighting), and social service policies to prevent and reduce FOF in older adults should be considered to increase the prevalence of FOF and create an age-friendly society.

This study also had several limitations. First, due to the involved observational studies, there might be some compounding factors, which might bias to the results. Notably, a large sample of 153 studies with 200,033 subjects from 38 countries could be advantageous for guaranteeing the consistency and universal applicability of the results. Second, high heterogeneity in this work was found, caused in part by the subjects from various nations, living conditions, cultures and lifestyles. Finally, only three studies from Africa were analyzed, probably because the work included only English studies, which may have left out some important evidence in other languages. Hence, more studies should pay more attention to FOF among older adults who speak different languages in the future.

Conclusion

This study as the first systematic review and meta-analysis provided substantial evidence that the global prevalence of FOF was high, and it was higher in developing countries than in developed countries, and higher in patients than in community residents. Twenty-eight potential risk factors, including demographic characteristics, physical function, chronic diseases and mental problems, were found a significant association with FOF. Policy-makers, health care providers and government officials should comprehensively evaluate the risk factors for FOF among older adults and formulate precise intervention measures to improve FOF based on the characteristics of different individuals. Firstly, multidisciplinary cooperation models should be established, including rehabilitation, psychology, pharmacology, etc, to help older patients normatively treat chronic diseases, strengthen drug safety management, and prevent drug abuse to reduce FOF. Secondly, a friendly living environment including improving exercise facilities and equipment and providing social support should be built to help older adults actively participate in social engagement. Finally, policy-makers should formulate the age-appropriate transformation system and intelligent health care system, optimize the health service model of older adults, actively develop the silver economy, and provide policy support and economic guarantee for promoting the physical and mental health of older adults.