Incidence and multidimensional predictors of occasional and recurrent falls among Malaysian community‐dwelling older persons

The global prevalence of falls was 5,186 (4,622–5,849) per 100,000 people in year 2017 [1]. Falls can lead to debilitating consequences, including injury-related hospitalization, disability and death [2]. Moreover, fall-related injuries in older persons result in increased healthcare costs. The estimated medical costs of fatal and non-fatal falls among older persons in the US was reported to be around 50 billion dollars [3]. Population-wide falls prevention programs that effectively improve the overall health and reduce social care burden associated with fall-related complications are, therefore, urgently required [4].

Several risk factors have been identified to be associated with falls in older persons. Females and those who are older in age are at higher risk of falls [5]. In the psychosocial context, older persons who live alone, have fewer years of education, or who experience loneliness and depression are more likely to be at risk of falls [6]. Medical illnesses such as arthritis, diabetes, chronic kidney diseases and stroke are among the other risk factors for falls in older persons [6, 7]. Physical frailty, cognitive decline and other geriatric syndromes also contribute to the occurrence of falls in older persons [5].

Declined physical function below a certain threshold, manifesting as muscle weakness and loss of balance are likely to lead to falls in older persons as well [8]. In addition, older persons with cognitive impairment are twice more likely to fall in comparison to those without cognitive impairment [9]. Nutritional deficiencies, such as vitamin D, calcium and inadequate protein intake are linked to osteoporosis, fractures and sarcopenia and, as a result, increase both the risk of falls and its consequences [10]. Environmental factors, for example, poor lighting, clutter in and around the house and inadequate footwear are often contributors to the risk of falls among older persons [5].

It is noteworthy that most risk factors of falls studied among community-dwelling older persons were performed via cross-sectional studies and were inconclusive. In addition, physical, nutritional, biochemical and biomarker status were not examined together within a single study. As falls are multifactorial in nature, identifying the multidimensional predictors of falls risk may be beneficial for early falls prevention strategies in community-dwelling older persons. The objective of this study was to determine the incidence of falls and identify predictors of occasional and recurrent falls. This was done in the social, medical, physical, nutritional, biochemical and cognitive dimensions among community-dwelling older Malaysians.

Three hundred and nine (17.53 %) participants reported fall occurrence at an 18-month follow-up, of whom 85 (27.51 %) had two or more falls. The incidence rate for occasional and recurrent falls was 8.47 and 3.21 per 100 person-years, respectively. The age-specific incidence rates of both occasional and recurrent falls are as shown in Table 1. Overall, the incidence rate of falls was independent of increasing age. The ≥ 75 age group had the highest incidence rate for occasional falls (9.09 per 100 person-years). While the 65–69 years age group had the highest incidence rate for recurrent falls (3.91 per 100 person-years).

Table 2 depicts the baseline characteristics of non, occasional and recurrent fallers. Occasional fallers were more likely (p < 0.05) to be women (60.27 %), single (3.13 %), have falls history (30.36 %), have higher depression scale scores (2.95 ± 2.42); lower fat-free mass (35.64 ± 7.07 kg) and skeletal muscle mass (18.82 ± 4.21 kg); lower intake of energy (1,583.99 ± 436.79 kcal/day), carbohydrate (209.95 ± 67.82 g/day) and zinc (3.38 ± 1.60 mg/day); lower hemoglobin levels (13.64 ± 1.85 g/L), scored lower in MMSE (22.62 ± 4.82), MoCA (18.35 ± 5.82) and digit symbol (4.73 ± 2.50); scored lower in chair stand (9.60 ± 3.28 times) and dominant handgrip muscle strength (20.82 ± 7.27 kg) tests. Recurrent fallers were more likely (p < 0.05) to be women (68.24 %), living alone (17.65 %), have falls history (42.35 %), with a medical history of stroke (4.71 %) and joint pain (36.47 %); higher depression scale score (3.35 ± 2.63); lower fat-free mass (35.17 ± 7.90 kg) and skeletal muscle mass (18.51 ± 4.70 kg); higher fat mass (26.32 ± 10.16 kg) and percentage of body fat (41.89 ± 10.26 %); lower hemoglobin levels (13.42 ± 2.13 g/L); scored lower in chair stand (8.83 ± 3.07 times), TUG (12.84 ± 5.28 seconds) and dominant handgrip muscle strength (19.57 ± 6.69 kg) tests.

Variables that were significant in the univariate test were entered into binary logistic regression. The univariate predictors of occasional and recurrent falls are as shown in Table 3. These variables were then further analyzed using multivariate logistic regression analysis (Table 4). Being female (OR: 1.57; 95 % CI: 1.04–2.36), being single (OR: 5.31; 95 % CI: 3.36–37.48), having history of falls (OR: 1.86; 95 % CI: 1.19–2.92) higher depression scale scores (OR: 1.10; 95 % CI: 1.02–1.20), lower hemoglobin levels (OR: 0.90; 95 % CI: 0.81-1.00) and lower chair stand test scores (OR: 0.93; 95 % CI: 0.867-1.00) appeared as occasional falls predictors in this model [χ² (df = 8, N = 1763) = 13.38, p = 0.100 with 86.60 % accuracy]. While, having history of falls (OR: 2.74; 95 % CI: 1.45–5.19), stroke (OR: 8.57; 95 % CI: 2.12–34.65), higher percentage of body fat (OR: 1.04; 95 % CI: 1.01–1.08) and lower scores in chair stand test (OR: 0.87; 95 % CI: 0.77–0.97) were identified as predictors of recurrent falls [χ² (df = 8, N = 1763) = 10.82, p = 0.212 with 94.70 % accuracy].

We identified the incidence rate and multidimensional falls risk predictors at 18-months follow-up among 1,763 community-dwelling older Malaysians. The incidence rate of occasional and recurrent falls observed in our study was 8.47 and 3.21 per 100 person-years respectively, with no increase with age. Falls prevalence of 15–18 % [18, 19] and 27 % over a six-months follow-up period [20] had previously been reported among community-dwelling older Malaysians. The prevalence of recurrent falls is reported as 8.3 % [21].

Despite the apparent association between age and decline in both physical and cognitive functions [5], our results showed that advancing age did not predict the incidence of occasional and recurrent falls. In an age-specific population, the incidence of falls was not age-dependent, as opposed to the prevalence of falls [1]. This suggests that the number of falls does not increase with increasing age. It may be possible that cumulatively, the number of falls among older people in any given period of time could be observed due to the consistent addition of new cases. Decline in memory with age might also be another reason that may have affected the ability of older persons to recall falls incidences.

Occasional falls could be accidental and are usually associated with extrinsic factors [22, 23]. In comparison, recurrent falls among older persons are commonly related to multifactorial intrinsic factors suggesting a more complex risk model. Our study findings showed that the identified predictors for both occasional and recurrent falls were different, with the exception of having history of falls and taking longer to complete the chair stand test. Having lower muscle strength and experiencing falls in the past appeared as robust predictors of both occasional and recurrent falls in community-dwelling older persons. Being female, single, having higher depression scale score and lower hemoglobin levels were predictors of occasional falls. While, predictors of recurrent falls included a history of stroke and having higher percentage of body fat.

A history of falls appeared to be the predictor of both occasional and recurrent falls. These findings are consistent with previous reports where having a history of falls was found to be a significant predictor of subsequent falls [24]. Furthermore, our study results showed that older women had higher risk of falls as compared to men. Older women having a higher risk of falls has been well established in the literature. Previously, the English Longitudinal Study of Ageing (ELSA) involving 1994 men and 2357 women had reported higher prevalence of falls among women (29.1 %) compared to men (23.5 %) [2]. The number of women admitted to hospital due to falls increased every year from 15,000 in year 2001 to 20,000 in 2009 [25]. It is also plausible that women have a greater loss in bone mineral density due to menopause and this could be associated with decline in muscle strength and falls.

Depression is associated with occasional falls among community-dwelling older persons as reported previously [26]. Depressive symptoms may affect older persons’ mobility and executive function [27]. The causal relationship between depression and falls was not fully explained by adjustment for the medical comorbidities, nutritional, physical and biochemical factors. Use of medications was not adjusted within this study, which may account for the increased risk of falls among individuals with symptoms of depression [28]. Depression in older persons has been attributed to structural brain changes that interfere with cortical-subcortical circuits, basal-ganglia and limbic networks, which in turn affect postural stability leading to the occurrence of falls [29]. Furthermore, antidepressants have been associated with single and recurrent falls [29, 30]. The proposed mechanisms for antidepressant-related falls include orthostatic hypotension, dizziness, compromised vision and mental confusion [30].

Lower hemoglobin levels increased the risk of occasional falls. Potential mechanisms linking the age-associated decline in hemoglobin and falls include fatigue, reduced muscle strength and muscle quality. The decline in oxygen delivery is attributed to the reduction of hemoglobin levels, whereby hemoglobin functions as an oxygen carrier to skeletal muscles, leading to a reduction in muscle function and declining mobility. This finding was in agreement with the three-year Longitudinal Aging Study Amsterdam demonstrating frequent episodes of falls among older persons with anemia as compared to their non-anemic counterparts [31]. The presence of lower hemoglobin may also reflect underlying nutritional deficiencies or chronic conditions affecting hemoglobin production. Others include undetected causes of hemoglobin loss due to medical conditions such as peptic ulcer disease or malignancy and medications, including ulcer-inducing drugs and those that inhibit marrow function. However, these factors had not been fully accounted for within this study.

The chair stand test, a measure of lower extremity muscle strength, has been demonstrated to be beneficial in determining fall risk [32]. Older persons who had lower chair stand test scores were reported to be associated with a higher risk of fall-related injuries [33]. Moreover, lower extremity weakness was reported to increase the odds of occasional and recurrent falls in older persons [34] since it was associated with abnormal gait, loss of balance, declined mobility, flexibility and functional performance [35]. Besides, strengthening of lower limb muscles has been reported to be effective in preventing falls in older persons [35]. Similarly, we have demonstrated that muscle strength was associated with falls among Malaysian community-dwelling older persons in our earlier pooled data findings [36].

Our study results also showed that the risk of recurrent falls at 18 months was increased among older persons with higher percentage of body fat. One of the probable reasons for this relationship could be due to declined lower extremity muscle strength following excess adiposity, which could affect postural stability and balance [37]. Excess adipose tissue accumulation may also lead to dynapenic obesity, a condition linked to a decline in muscle strength, loss of muscle mass and sarcopenic obesity [38, 39]. Recent studies have also linked adiposity with low-grade inflammation, which not only predisposes individuals to osteoarthritis and dementia but also increases the risk of sarcopenia and osteoporosis [40]. As a result, it may lead to impaired mobility and balance and consequently, increase fall risk in older persons. In an observational study involving 164,737 participants between the ages of 19 to 106 years, older persons with obesity had the odd ratio of 1.10 and 1.12 for one fall and two or more episodes of falls respectively [41].

Increased body fat predisposes individuals to underlying medical conditions such as diabetes, hypertension, heart disease, hypercholesterolemia and stroke. Stroke also has been identified as one of the major risk factors of falls and recurrent falls [42]. Stroke survivors tend to develop fear of falls, which is associated with physical and functional decline, decreased quality of life, impaired social interaction, depression and anxiety [43, 44]. Depressive symptoms and loss of dynamic balance have been demonstrated to increase the risk of falls among stroke survivors [45, 46].

In our longitudinal study, being single (unmarried) was a risk factor of occasional falls during the 18th-months follow-up. Single older persons were commonly identified as living alone, experiencing loneliness, depressive symptoms and poor health [47, 48]. Similarly, these characteristics were associated with increased risk of falls in older persons. Older persons living alone may be having declined physical fitness because of limited participation in physical activity [49, 50]. Subsequently, having an increased risk of fall-related injuries, mortality and morbidity [49, 50].

One of the main limitations of this study is that we obtained retrospective history of falls subjectively. Older persons may have difficulties recalling their fall events in the past 18-months, causing under or over-reporting of the frequency of falls.

In conclusion, the incidence rates of occasional and recurrent falls in community-dwelling older Malaysian were 8.47 and 3.21 per 100 person-years respectively. Being female, single, having history of falls, higher depression scale, lower hemoglobin levels and lower chair stand test score remained independent predictors of occasional falls. Having history of falls, being a stroke survivor, higher percentage of body fat and lower chair stand test score appeared as recurrent falls predictors in this model. In this study, we identified predictors for prospective falls using a comprehensive, multidomain approach. Further studies are required to determine the value of individualized interventions informed by such an approach to risk factors identification.