Association between weight and body composition changes with falls risk in the Malaysian Elders Longitudinal Research (MELoR) study

Abstract

Objective

Both changes in body composition and increased fall risk occur with increasing age. While weight management may be considered a component of falls prevention, the long-term consequences of changes in weight, however, remain uncertain. This prospective study aimed to evaluate the relationship between weight and body composition changes over 5 years with fall occurrence.

Design

Prospective cohort study.

Setting

Community-dwelling older adults interviewed at baseline (2013–2016) and follow-up (2020–2022) as a part of the Malaysian Elders Longitudinal Research study were included.

Participants

Participants who attended face-to-face follow-up visits.

Primary and secondary outcome measures

Fall occurrence over 12 months preceding the follow-up visit was determined. Anthropometric, bioimpedance analysis and physical performance measurements were obtained at both time points. Participants were categorised into three groups according to changes in weight and body composition using≥5% increase or decrease in weight to determine loss or gain.

Results

Of the 225 participants, aged 71.8±6.8 years, 128 (56.9%) were women. Weight gain was associated with increased fall risk at follow-up compared with stable weight (adjusted rate ratio, aRR (95% confidence interval, CI)=2.86 (1.02–8.02)) following adjustments for age and body mass index (BMI), but this relationship was attenuated by low baseline percentage lean body mass (%LBM) in women. The association was strenghtened after adjusting for age, BMI, and low muscle strength (aRR (95% CI)=2.89 (1.01–8.28)). Weight change did not influence falls risk in men. No difference was observed with changes in percentage body fat and %LBM over time with fall occurrence for both genders.

Conclusion

Lower baseline lean body mass influenced the relationship between weight gain and falls longitudinally. Interventions addressing low lean body mass should be considered in the prevention of weight-gain-related falls in older women.

Correspondence to Dr Maw Pin Tan; [email protected]

STRENGTHS AND LIMITATIONS OF THIS STUDY

• The Malaysian Elders Longitudinal Research study was a prospective cohort study conducted among adults aged 55 years and over residing in a low- and middle-income country, with physical assessments, including body mass index and bioimpedance analysis, obtained at two time points at least 5 years apart.

• Logistic regression analyses were conducted to determine whether the increased risk of falls associated with weight gain was accounted for by reduced lean body mass.

• The sample size was limited by a lower number of participants agreeing to attend face-to-face follow-up assessment during the COVID-19 pandemic due to infection control recommendations for those aged 60 years and over.

• Validated questionnaires on physical activity levels, which could explain how weight change contributes to the increased risk of falls, were not applicable during the pandemic lockdown.

Introduction

Falls are common among older adults, which often lead to adverse outcomes, such as fractures, head injuries, hospitalisations, decline in the overall physical functioning and mortality.^1–3 As individuals age, changes in body composition are expected to occur,⁴ and these include a decline in muscle mass and strength (known as sarcopenia), and changes in weight distribution and overall physical functioning, which negatively impacts balance and gait, increasing the likelihood of falls and fall-related injuries.^{4 5}

Increased body weight or obesity is a potential risk factor for various diseases and can further impair physical functioning, particularly walking ability and performance.^{1 6–8} Longitudinal follow-up of the Malaysian Elders Longitudinal Research (MELoR) study participants had suggested that individuals with obesity defined by body mass index (BMI) of 30 kg/m² and over were at higher risk of falls in the fifth year.⁹ However, as this wave was conducted exclusively through telephone interviews, there is a lack of information on changes in weight and body compositions, which suggests that a future study should determine the role of changes in BMI in the relationship between increased BMI and falls. Other studies have also found that higher BMI is associated with impaired mobility and physical function, which are key risk factors for falls among older adults.^{7 10 11} While sarcopenia is thought to play a role in the obesity–fall relationship, the relationship between changes in body weight and composition has not been recorded in previous studies.^{9 12 13}

It has been hypothesised that unfavourable consequences, such as falls, observed in older adults with higher BMI and higher levels of body fat may be attributable to the development of sarcopenia, which is accompanied by a reduction in lean body mass.^{14 15} In this context, weight loss occurs as a result of the increased risk of medical conditions associated with excess body weight.¹³ Hence, it is important to consider changes in body weight and body composition, in addition to cross-sectional measures, when examining negative health outcomes in older adults. The objective of this study was, therefore, to investigate the association between longitudinal changes in body weight and body composition with falls in older adults while determining the role of lean body mass and muscle strength in this potential relationship.

Methods

Study population

This study included participants from the MELoR study, which comprised community-dwelling adults aged 55 years and above recruited through stratified random sampling from three parliamentary constituencies in the Klang Valley of Malaysia between 2013 and 2016. A total of 1614 participants were recruited and assessed at baseline through a computer-assisted interview, and 1367 attended physical health checks. Survival data were obtained from the National Registry Department before participants were contacted for follow-up interviews and visits. The first telephone follow-up was conducted in 2019 and 2020, and participants were then selected and invited for face-to-face visits based on their BMI at baseline. The physical assessment follow-up occurred from 2020 to 2022.

Sociodemographic and medical history

Sociodemographic information, including age, gender, ethnicity and educational attainment, was obtained at baseline through computer-assisted interviews conducted at participants’ own homes, in their preferred languages by trained interviewers fluent in that language. Medical conditions were self-reported, obtained by asking participants if they had ever been told by a doctor, they had any of the conditions listed in the survey questionnaire.

History of falls and fear of falling

A fall was operationally defined as instances in which an individual unintentionally came to rest on the ground or a lower level, with or without the loss of consciousness.¹⁶ The participants were asked a similar question regarding their history of fall at both the baseline and follow-up, which was ‘Have you fallen in the past 12 months?’, with specific questions pertaining to the occurrence and frequency of falls.¹⁷ Additionally, the participants were also asked, ‘Are you afraid of falling?’.

Quality of life and cognitive assessment

The quality of life was assessed by a 12-item age-specific quality-of-life instrument, the control, autonomy, self-realisation and pleasure (CASP-12) questionnaire.^{18 19} In addition to the original English language version, the validated versions are also available in the traditional Mandarin Chinese and Bahasa Malaysia.^{19 20} Cognitive performance was conducted using locally validated tool, the Montreal Cognitive Assessment (MoCA). The MoCA test assessed cognitive abilities in the areas of executive function, language, visual construction, conceptual thinking, orientation, attention and focus and memory.²¹

Sarcopenia screening

Inquiries pertaining to the screening of sarcopenia and weight loss were included in the follow-up assessment of frailty. The Strength, Assistance in walking, Rising from a chair, Climbing stairs and Falls (SARC-F) questionnaire was used for case finding.^{5 22} A score of four or higher is indicative of the likelihood of sarcopenia and unfavourable outcomes.

Unintentional weight loss

Unintentional weight loss was defined as a reduction in body weight occurring within the preceding 12 months and not attributable to the anticipated outcome of therapeutic interventions for a diagnosed disease.^{23 24} Participants were asked, ‘Have you recently lost weight without trying?’.

Physical examination

Anthropometric measurements

The physical assessment commenced with the measurement of height and weight using a height stadiometer (SECA 220, Hamburg, Germany) and calibrated weighing scale (SECA 769, Hamburg Germany), respectively. Subsequently, BMI was computed by dividing the measured weight in kilograms (kg) by the square of the body height (m²). The measurement of waist circumference (WC) was acquired by assessing the circumference of the abdomen at the level of the iliac crest. The measurement of hip circumference was taken at the widest part of the participant’s buttock, followed by the calculation of waist-to-hip ratio (WHR).

Body composition

Percentage body fat (%BF) and lean body mass (%LBM) were measured through bioimpedance analysis using a body composition analyser (baseline: Quadscan 4000, BodyStat, UK; follow-up: TANITA DC-360, Japan). The body composition was automatically calculated through the manufacturers’ software, taken at two frequencies (Quadscan 4000: 5 Hz and 50 Hz; TANITA DC-360: 6.25 kHz and 50 kHz). Low %LBM was defined as the lowest sex-specific quintile.

Functional performance

Handgrip strength

Muscle strength was assessed using handgrip strength (HGS). Participants were given instructions to maintain an upright sitting position, with their elbow flexed at a 90° angle, while seated on a chair that provided both back and arm support. HGS was assessed three times in kilograms (kg) using a hand dynamometer (Patterson Medical/Samsons Preston, USA). They were instructed to grip or squeeze the dynamometer with their maximum strength. The correct HGS procedure was first demonstrated before executing the test. The average of three measurements obtained from the dominant hand was used for the study. Low muscle strength was defined as HGS<28 kg for men and<18 kg for women.⁵

Timed-up and go

The timed-up and go (TUG) test was carried out to examine gait and balance for fall risk assessment. Participants were instructed to get up from their seats, move independently or with a walking aid for 3 m at their normal walking pace, turn around and return to their seats. Before participants began the task, the TUG was initially demonstrated to them by the researcher. Using a digital stopwatch, the time interval from the instant their backs disengaged from the chair’s back until they were seated again was recorded in seconds (s).^{14 25}

Functional reach

Stability and dynamic balance were evaluated using the functional reach (FR). A 100-cm measuring tape was first attached to a wall, parallel to the floor at shoulder height. Participants were instructed to stand straight, place their shoulder at the tape’s zero position and raise their hand parallel to the tape. Participants were then instructed to bend forward as far as they could while holding their arm outstretched and at 90° of forward flexion, and the distance between the initial and final positions of the middle finger measured in centimetres was considered the FR.

Changes in body weight, body composition and functional performance

Changes in adiposity indices (body weight, BMI, WC, WHR, %BF and %LBM) and physical performance (TUG, HGS and FR) were determined between follow-up and baseline measurements for fallers and non-fallers (online supplemental table 1). To take into account common fluctuations in body weight, a reduction in 5% of the total body weight or greater was considered weight loss, while an increase in 5% of the total body weight or greater was considered weight gain, and body weight measurements within 5% of the original baseline were considered stable weight.^{26 27} Sensitivity analyses were carried out for continuous measures as well as 3%, 5% and 10% changes prior to the grouping, which indicated significant outcome for falls at 5% cut-off within the three groups. Therefore, the changes in weight and body composition were categorised into three groups: stable (less than 5% increase or decrease), loss (≥5% reduction) and gain (≥5% increase). The percentage change ((follow-up value − baseline value) × (100/baseline value)) between baseline and follow-up was used to calculate changes in body weight and body composition (%BF and %LBM).²⁸ A similar equation was applied for changes in muscle strength measured by HGS. Decline in muscle strength was defined as reduction of ≥5% compared with baseline performance, while changes within 5%, reduction or increment, at follow-up were characterised as stable muscle strength.²⁶

Sample size

Power calculation was performed using G*Power 3.1 programme (Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany).^{29 30} Assuming that a 2:1 ratio of non-fallers to fallers is expected, a sample size of 224 will provide 80% power to detect an effect size of 0.4, which is considered a moderate effect.

Statistical analysis

Data analysis was carried out using the Statistical Package for the Social Sciences Statistics 26.0 software (IBM, Armonk, NY, USA). The significance level was set at two-sided p<0.05. Descriptive statistics for all participants were presented as mean (standard deviation, SD) and frequencies, N (percentage, %), as appropriate. Paired t-test was carried out to compare mean for continuous variables, while the McNemar test was used to compare the frequencies for dichotomous variables measured at baseline and follow-up (table 1). Mean differences between faller and non-faller at baseline and follow-up were determined through the paired t-test to select the variables of physical changes for further regression analysis (online supplemental table 1). A gender-specific physical analysis by falls status at 6-year follow-up was presented in online supplemental table 2. The association among weight, body composition and muscle strength changes with falls was determined by rate ratios (RR) with 95% confidence interval (CI). The stable category for respective changes was assigned as the reference group in the regression analysis. Logistic regression adjustment was made in gender-specific weight change analysis for age (Model 1) and further adjustment for BMI (Model 2), low %LBM (Model 3) and low muscle strength (Model 4) at baseline. Variables selected for adjustment were to evaluate the pre-existing hypothesis that the risk of falls increases with age and BMI, and may be exacerbated by existing sarcopenia symptoms, such as low LBM and low muscle strength.

Table 1

Gender-specific characteristics at baseline and follow-up

Variables	Men (N=97)			Women (N=128)
	Baseline	Follow-up	P value	Baseline	Follow-up	P value
Age (years), mean (SD)	67.0 (5.8)	72.6 (6.1)	<0.001*	65.6 (6.2)	71.2 (6.6)	<0.001*
Secondary education and above, n (%)	95 (97.9%)		–	119 (93.0%)		–
Diabetes, n (%)	22 (23.4%)		–	17 (13.5%)		–
Hypertension, n (%)	41 (43.6%)		–	44 (34.9%)		–
High blood cholesterol, n (%)	41 (43.6%)		–	63 (50.0%)		–
Joint problem, n (%)	20 (21.5%)		–	22 (18.0%)		–
Anthropometric measure, mean (SD)
Height (m)	1.68 (0.1)	1.66 (0.1)	<0.001*	1.54 (0.1)	1.53 (0.1)	<0.001*
Weight (kg)	69.8 (12.6)	67.2 (12.2)	<0.001*	57.4 (10.5)	56.5 (10.8)	0.025*
BMI (kg/m2)	24.9 (4.3)	24.3 (4.1)	0.379	24.2 (4.6)	24.2 (4.8)	0.911
WC (cm)	92.1 (11.5)	92.6 (13.0)	0.609	83.9 (11.1)	84.2 (12.5)	0.705
WHR	0.930 (0.1)	0.937 (0.1)	0.379	0.851 (0.1)	0.858 (0.1)	0.419
Body composition, mean (SD)
%BF	25.6 (5.3)	23.2 (7.5)	0.001*	38.7 (6.1)	34.4 (8.1)	<0.001*
%LBM	74.4 (5.3)	76.8 (7.5)	0.001*	61.2 (5.9)	65.7 (8.1)	<0.001*
Functional performance, mean (SD)
HGS (kg)	32.5 (7.8)	28.1 (6.5)	<0.001*	20.8 (4.5)	18.4 (4.3)	<0.001*
TUG (s)	10.9 (2.3)	10.6 (3.1)	0.504	11.1 (2.6)	10.8 (3.9)	0.417
FR (cm)	29.7 (7.1)	29.8 (9.9)	0.952	26.1 (6.7)	27.1 (7.8)	0.179
Falls history, n (%)	9 (9.3%)	22 (22.7%)	0.015*	35 (27.3%)	45 (35.2%)	0.193
Fear of falling, n (%)	54 (56.3%)	47 (48.5%)	0.585	97 (77.0%)	83 (64.8%)	0.080
CASP-12, mean (SD)	29.0 (4.4)	30.7 (3.8)	0.003*	28.0 (5.2)	29.5 (4.9)	0.001*
SARC-F score≥4, n (%)	N/A	6 (6.2%)	–	N/A	17 (13.3%)	–
Unintentional weight loss, n (%)	N/A	14 (16.9%)	–	N/A	18 (15.9%)	–
MoCA score, mean (SD)	25.7 (2.8)	19.3 (2.1)	<0.001*	26.2 (2.7)	19.7 (2.2)	<0.001*

Values are in mean (SD) and n (%) as appropriate.

Asterisk (*) indicates significance at p<0.05.

%BF, percentage body fat; BMI, body mass index; CASP-12, 12 items of control, autonomy, self-realisation and pleasure; FR, functional reach; HGS, handgrip strength; %LBM, percentage lean body mass; MoCA, Montreal Cognitive Assessment; N/A, not available; SARC-F, Strength, Assistance in walking, Rising from a chair, Climbing stairs and Falls; TUG, timed-up and go; WC, waist circumference; WHR, waist-to-hip ratio.

Patient and public involvement statement

Members of the public were consulted in the development of the survey questionnaire through focus group discussions on the method of delivery as well as the content of the survey.³¹ The results of the study were also disseminated through social media, mass media and stakeholder meetings and public engagement events.³² During the stakeholder and public engagement events, results findings were interpreted and disseminated.

Results

Among the 1367 participants of the MELoR study with available BMI and physical assessment data at baseline, a total of 124 deaths (9.1%) and 150 deaths (11.0%) were reported by the end of 2019 and 2020, respectively. A total of 484 (35.4%) were lost to follow-up, which included 67 (4.9%) participants who declined to participate and 417 (30.5%) who could not be reached. The mean (SD) age of participants lost to follow-up was 68.6 (7.3) years. 283 (58.7%) women, 162 (33.7%) attained primary education and lower, 29 (6.1%) had heart disease, 261 (54.6%) had hypertension, 263 (55%) had high cholesterol and 147 (30.8%) had diabetes at baseline. Of the 733 (53.6%) participants who responded to the follow-up telephone calls, 225 participants (mean age of 66.2 (6.1) years at baseline and 71.8 (6.8) years at follow-up, 128 (56.9%) women) completed the physical assessments. 508 (69.3%) of those who responded to the prior telephone interview were unwilling to attend the physical health check due to government recommendations for older adults to practice shielding, despite reassurance of strict infection control measures. Figure 1 shows a flowchart of participants included in the follow-up study.

Figure 1. Flowchart of participants included in follow-up. BMI, body mass index; MELoR, Malaysian Elders Longitudinal Research.

Participants’ basic characteristics

Participants’ basic characteristics at baseline and follow-up were summarised according to gender in table 1. The difference in age between participants at baseline and follow-up check was 5.58 (1.3–1.6) years in both men and women. Overall, participants were observed to experience a reduction in height, weight, %BF and HGS at follow-up, while positive changes in %LBM and quality of life were observed at the follow-up health check. The history of at least one fall in the preceding 12 months was higher at follow-up compared with baseline in both men and women, with statistically significant changes observed among men. Quality of life by CASP-12 scores, however, was improved after 6 years. 10.2% of participants had SARC-F scores of ≥4, while 14.2% of participants reported unintentional weight loss at follow-up. Cognitive assessment using MoCA test showed a significant decrease in cognitive performance among men and women at follow-up.

Association between physical changes with falls

The gender-specific analysis for physical changes by falls status at 6-year follow-up is presented in online supplemental table 2. Logistic regression analysis showed that there was no significant association among body weight, body composition and muscle strength with fall occurrence in the 12 months preceding follow-up appointments for the overall population (table 2). Further gender-stratified analysis showed that only weight gain of 5% or greater among women was associated with falls at follow-up, RR (95% CI) = 2.81 (1.01–7.84), compared with the women who had stable weight.

Table 2

Unadjusted logistic regression between physical changes and falls by gender

Changes	N (%)	Falls, RR (95% CI)
		Overall	Men	Women
Weight
Stable	118 (52.4)	1.00 (Ref)	1.00 (Ref)	1.00 (Ref)
Loss	78 (34.7)	0.79 (0.41–1.49)	0.45 (0.16–1.29)	1.15 (0.50–2.64)
Gain	29 (12.9)	1.39 (0.60–3.25)	0.00 (0.00)	2.81 (1.01–7.84)*
Weight change (kg)**	225 (100)	1.02 (0.98–1.06)	1.00 (0.94–1.07)	1.02 (0.97–1.07)
%BF
Stable	43 (19.1)	1.00 (Ref)	1.00 (Ref)	1.00 (Ref)
Loss	111 (49.3)	0.98 (0.45–2.10)	1.83 (0.35–9.64)	0.81 (0.33–1.99)
Gain	38 (16.9)	0.82 (0.31–2.18)	2.03 (0.34–12.24)	0.64 (0.18–2.34)
BF change (%)**	192 (85.3)	1.00 (0.98–1.01)	1.00 (0.98–1.03)	0.99 (0.97–1.01)
%LBM
Stable	86 (38.2)	1.00 (Ref)	1.00 (Ref)	1.00 (Ref)
Loss	18 (8.0)	0.39 (0.11–1.47)	0.00 (0.00)	0.98 (0.21–4.58)
Gain	88 (39.1)	0.74 (0.39–1.41)	0.62 (0.20–1.97)	0.75 (0.34–1.67)
LBM change (%)**	192 (85.3)	1.01 (0.98–1.04)	1.00 (0.93–1.07)	1.00 (0.97–1.04)
HGS
Stable	37 (16.4)	1.00 (Ref)	1.00 (Ref)	1.00 (Ref)
Loss	157 (69.8)	1.15 (0.52–2.57)	0.63 (0.14–2.79)	1.64 (0.62–4.32)
Gain	31 (13.8)	1.29 (0.45–3.66)	0.17 (0.01–1.96)	3.00 (0.85–10.57)
HGS change (kg)**	225 (100)	1.00 (0.99–1.02)	0.99 (0.97–1.02)	1.01 (0.99–1.03)

Asterisk (*) indicates significance at p<0.05. ** per unit increase from baseline.

%BF, percentage body fat; HGS, handgrip strength; %LBM, percentage lean body mass; Ref, reference group; RR, rate ratio.

Adjustment models in multiple regression analyses (table 3) for weight changes between gender demonstrated that the association between weight gain and falls in women remained after adjustment for age and BMI at baseline (Model 2, adjusted rate ratio (aRR) (95% CI)= 2.86 (1.02–8.02)). Adjustment for low %LBM reduced the parameter estimate to 2.65 (0.89–7.84). The model was, however, not influenced by low muscle strength (Model 4, aRR (95% CI)= 2.89 (1.01–8.28)).

Table 3

Multiple logistic regression analyses between weight change and falls by gender

Gender	Variables	RR (95% CI)
		Crude	Model 1	Model 2	Model 3	Model 4
Men	Stable (N=44)	1.00 (Ref)	1.00 (Ref)	1.00 (Ref)	1.00 (Ref)	1.00 (Ref)
	Loss (N=26)	0.45 (0.16–1.29)	0.39 (0.13–1.18)	0.34 (0.11–1.07)	0.42 (0.14–1.22)	0.34 (0.11–1.07)
	Gain (N=8)	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)	0.00 (0.00)
	Age	–	1.05 (0.96–1.15)	1.08 (0.98–1.19)	1.04 (0.93–1.17)	1.08 (0.32–3.66)
	BMI	–	–	1.10 (0.97–1.24)	1.04 (0.90–1.20)	1.09 (0.96–1.242)
	Low %LBM	–	–	–	0.95 (0.20–4.58)	–
	Low HGS	–	–	–	–	1.08 (0.32–3.66)
Women	Stable (N=57)	1.00 (Ref)	1.00 (Ref)	1.00 (Ref)	1.00 (Ref)	1.00 (Ref)
	Loss (N=40)	1.15 (0.50–2.64)	1.21 (0.52–2.79)	1.24 (0.53–2.88)	1.06 (0.43–2.57)	1.18 (0.50–2.80)
	Gain (N=19)	2.81 (1.01–7.84)*	2.85 (1.02–7.99)*	2.86 (1.02–8.02)*	2.65 (0.89–7.84)	2.89 (1.01–8.28)*
	Age	–	0.97 (0.91–1.03)	0.97 (0.91–1.03)	0.98 (0.92–1.05)	0.94 (0.87–1.00)
	BMI	–	–	0.98 (0.90–1.06)	1.03 (0.91–1.16)	0.96 (0.89–1.05)
	Low %LBM	–	–	–	0.43 (0.10–1.80)	–
	Low HGS	–	–	–	–	3.00 (1.23–7.33)*

Asterisk (*) indicates significance at p<0.05. Model 1=Crude RR adjusted for baseline age; Model 2=Model 1 adjusted for baseline BMI; Model 3=Model 2 adjusted for baseline low %LBM; Model 4=Model 2 adjusted for baseline low HGS.

Low %LBM=sex-specific lowest %LBM quintile (Men≤70.1% and Women≤56.7%); Low HGS=Men <28 kg and Women<18 kg.

BMI, body mass index; HGS, handgrip strength; %LBM, percentage lean body mass; Ref, reference group; RR, rate ratio.

Discussion

The relationship between changes in body weight and body composition towards falls occurrence in a subpopulation of MELoR participants who attended the physical check-up was investigated at their fifth- to sixth-year follow-up visit. It was observed that gender-specific influences played a role in this relationship. The association between weight gain and falls in older women was attenuated by adjustment for a low baseline lean body mass of less than 56.7%, suggesting that reduced lean body mass accounted for some of the increase in risk of falls associated with weight gain.

Previous studies have demonstrated that individuals in developed countries generally experience changes in body weight and body composition as they age, with increased risk of developing sarcopenia or sarcopenic obesity.^{8 33 34} However, it is important to consider the differences in economic activities, healthcare services, lifestyle and cultures between developed and developing countries. These differences may have contributed to variations in body weight changes experienced by our older adults. The mandatory retirement age in Malaysia at the time of conception of the MELoR study was 55 years. It is possible that the start of retirement could have led to an increase in passive or sedentary lifestyle among the participants, which could have resulted in weight gain.

Reduced lean body mass is associated with reduced physical activity, sedentary behaviour or other adverse lifestyle factors.^{35 36} A sedentary lifestyle in older adults can result in a reduction in basal metabolic rate, which can contribute to weight gain and central fat distribution over time.^{4 6 37 38} Changes in body composition may lead to infiltration of fat into muscle and bone, resulting in reduced strength and physical function,^{1 8 39} with then subjects the individual to an increased risk of falls, fractures and functional decline. A previous study had suggested that older adults who experience changes in body composition are at a higher risk of falls.¹ Published data from the MELoR study proposed that increased fasting blood glucose mediates the relationship between sarcopenia and falls in individuals with obesity, suggesting the potential role of insulin resistance.⁹ Weight gain may also lead to increased insulin resistance making individuals more susceptible to reduced lean body mass. This finding supports a previous study, which found that a lower lean body mass mediates the relationship between higher BMI and falls¹⁴

The aforementioned gender-specific tendency emphasises the significance of taking into account the unique physiological and health-related variables that impact older adults, particularly with regards to body composition and the likelihood of experiencing falls. Age-related muscle atrophy, referred to as sarcopenia, is frequently accompanied by a reduction in lean body mass or fat-free mass, which includes both muscle and bone mass. This decline in lean body mass may substantially impact the individual’s capacity to engage in routine tasks and safely navigate their surroundings.⁴⁰ A simultaneous decrease in lean body mass with weight gain in older women may involve multifaceted interaction of various elements, potentially encompassing hormonal fluctuations, specifically the decline in oestrogen levels experienced during menopause, which could impact the distribution of adipose tissue and muscle mass in later life.⁴¹ The hormonal shift may have led to greater difficulties in weight maintenance among older women compared with men.

It is plausible that men exhibit specific physiological reactions to increase in body weight, such as a potential inclination to preserve a greater proportion of lean body mass or manifestation of an altered pattern of body fat distribution. The higher levels of testosterone in males frequently result in a greater ratio of muscle mass. However, there is an observable drop in growth hormone and insulin-like growth factors among men in later life.^{42 43} The decline of male sex steroids, such as testosterone and free testosterone, occurs at a slower rate in men than in female sex steroids in women.^44–46 Therefore, men may have a greater propensity to preserve their muscle mass even when experiencing weight increase, which could provide potential benefits in terms of metabolic function and overall well-being.

The results of our study seem to contradict those of the Health, Ageing and Body Composition (Health ABC) study, which included community-dwelling older adults aged 70–79 years old in the USA. In their study cohort, weight change led to a higher preservation of lean mass compared with fat mass.²⁸ They found that men are more likely to experience loss of lean mass in association with weight loss that was not fully regained with subsequent weight gain compared with women, while our study showed that men tended to maintain their weight and lean mass over the follow-up period. They also found that muscle strength is a more important marker for muscle quality in predicting mortality risk compared with muscle mass.⁴⁷ Another research group from the Health ABC study found that the loss of muscle strength occurs more rapidly than the loss of muscle mass, while gaining or maintaining lean mass does not halt the loss of muscle strength.⁴⁸ Conversely, the present study showed that poor muscle strength did not significantly impact falls among individuals with weight gain rather than low lean mass, suggesting that muscle mass is a better predictor of falls than muscle strength. Health ABC participants are older, which places them at a higher risk for weight loss. As people age, there is a gradual decline in muscle strength of approximately 12%–14% per decade in healthy adults aged 50 years and above, accompanied by a reduction in muscle mass.¹³ Although weight loss with increasing age is usually associated with sarcopenia, it is important to highlight that increase in adiposity may exacerbate sarcopenia, known as sarcopenic obesity. The role of %LBM and muscle strength in our findings might differ in the context of high body fat among the study participants. The %LBM cut-off of 70.1% for men and 56.7% for women, and the average %BF (men=25.6% and women=38%) in this study corresponded with obesity according to the World Health Organization (WHO) definitions of %BF over 25% for men and over 30% for women.⁴⁹ Asians are known to have higher body fat despite having low BMI, although previous studies have not been conducted exclusively in older adults.^{50 51}Increased adipose tissue potentially triggers muscle cell degradation and impairs muscle synthesis by the release of proinflammatory adipokines and cytokines, therefore accelerating muscle loss.^{12 52}

Our study was conducted in an urban population and, hence, may not be generalisable to the whole population of Malaysia. As validated physical activity questionnaires in Malaysia all included outdoor activity, physical activity levels could not be measured during the Coronavirus disease 2019 (COVID-19) pandemic. This then limited our ability to provide evidence for sedentary behaviour as the underlying mechanism for weight gain in our study population. Additionally, the lower follow-up rates for physical measurements, such as anthropometric and body composition assessments, were the result of the COVID-19 pandemic-related infection control recommendations. Interim measures of body composition were not available within 5- to 6-year period. Nevertheless, follow-up measurements obtained at least 5 years apart in a low- and middle-income country are extremely valuable considering the resource constraints in these settings. While a history of falls in the preceding 12 months may appear to be a temporal mismatch, this measure was selected based on the Prevention of Falls Network Europe (ProFaNE) falls taxonomy to provide a standardised measure of falls occurrence.¹⁷ Although recall bias may have occurred with self-reported falls, we had previously found a high degree of agreement between retrospective and prospective reports by comparing prospective diaries, retrospective recall and telephone interviews in a subpopulation of the MELoR study.⁵³ Future follow-up studies should consider the role of physical activity in the association between body weight and body composition changes with adverse outcomes in older adults. Moreover, laboratory measures should be included to explore potential hormonal and biochemical mechanisms.

Our findings revealed that weight gain, only in women, and not weight loss in our community-dwelling population aged 55 years at recruitment was associated with falls occurrence, contrary to our initial hypothesis. This finding is consistent with the growing concern over the rising incidence of obesity among the Malaysian population.^{54 55} Our study highlights the potential role of lean body mass in reducing weight-gain-related falls in women aged 55 years and older in a low- and middle-income population. Interventions aimed at increasing lean body mass may be a promising therapeutic target to prevent falls in this population.^{56 57}

Conclusion

An observed increase in body weight of at least 5% over a 6-year period was associated with an increased risk of at least one fall over the past 12 months at follow-up among community-dwelling women aged 55 years and over in a low- and middle-income country. The potential mediating role of low lean body mass points to potential lifestyle, hormonal and biochemical mechanisms. Future research should also evaluate the value of interventions targeting low lean body mass in reducing falls associated with weight gain among this population group, which will see an exponential increase in numbers over the next 2 decades.

This study utilized the baseline data from the MELoR study. We acknowledged all the MELoR investigators and the participants who involved throughout this prospective study.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

Ethics approval was obtained from the Universiti Malaya Medical Centre (UMMC) Medical Research Ethics Committee (MREC ID Number: 2019103-7897). Written consent was obtained from participants for longitudinal data collection prior to study inclusion.

Footnote

X @mawptan

Contributors NNAH and MPT were involved in writing the original draft and formal analysis. SM, MPT, PKM and MD contributed to conceptualization of the study. SM, A-VC, SBK, NNH, SHK, SPKK, PKM and MD were involved in manuscript review and editing. SBK, NNH, A-VC, SPKK, SM and MPT obtained the funding for the study and contributed to the methodology. MPT will act as guarantor. All authors have read and approved the final draft of the manuscript.

Funding This study was supported by grants from the Ministry of Higher Education Malaysia; the Long-Term Research Grant Scheme (LRGS/1/2019/UM/01/1) and the Fundamental Research Grant Scheme (FRGS/1/2019/SKK02/UM/01/1).

Competing interests None declared.

Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.

Provenance and peer review Not commissioned; externally peer reviewed.

Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

References

¹ Xu C, Ebeling PR, Scott D. Body Composition and Falls Risk in Older Adults. Curr Geri Rep 2019; 8: 210–22. doi:10.1007/s13670-019-00294-6

² Watanabe JH. Medication Use, Falls, and Fall-Related Worry in Older Adults in the United States. Consult Pharm 2016; 31: 385–93. doi:10.4140/TCP.n.2016.385

³ Reimann H, Ramadan R, Fettrow T, et al. Interactions Between Different Age-Related Factors Affecting Balance Control in Walking. Front Sports Act Living 2020; 2: 94. doi:10.3389/fspor.2020.00094

⁴ Asselt D, Groot L. Aging and Changes in Body Composition. Food Aging Pop Elsevier 2017; 171–84. doi:10.1016/B978-0-08-100348-0.00008-1

⁵ Chen L-K, Woo J, Assantachai P, et al. Asian Working Group for Sarcopenia: 2019 Consensus Update on Sarcopenia Diagnosis and Treatment. J Am Med Dir Assoc 2020; 21: 300–7. doi:10.1016/j.jamda.2019.12.012

⁶ Strasser B. Physical activity in obesity and metabolic syndrome. Ann N Y Acad Sci 2013; 1281: 141–59. doi:10.1111/j.1749-6632.2012.06785.x

⁷ Ma W, Liu Y, Wu N, et al. Obesity, Even in the Metabolically Healthy, Increases the Risk of Poor Physical Performance: A Cross-Sectional Study of Older People in a Chinese Community. Clin Interv Aging 2021; 16: 697–706. doi:10.2147/CIA.S302167

⁸ Paranhos Amorim DN, Nascimento D da C, Stone W, et al. Body composition and functional performance of older adults. Osteoporos Sarcopenia 2022; 8: 86–91. doi:10.1016/j.afos.2022.04.002

⁹ Hashim NNA, Mat S, Myint PK, et al. Increased body mass index is associated with sarcopenia and related outcomes. Eur J Clin Invest 2023; 53: e13874. doi:10.1111/eci.13874

¹⁰ Hooker ER, Shrestha S, Lee CG, et al. Obesity and Falls in a Prospective Study of Older Men: The Osteoporotic Fractures in Men Study. J Aging Health 2017; 29: 1235–50. doi:10.1177/0898264316660412

¹¹ Mitchell RJ, Lord SR, Harvey LA, et al. Associations between obesity and overweight and fall risk, health status and quality of life in older people. Aust N Z J Public Health 2014; 38: 13–8. doi:10.1111/1753-6405.12152

¹² Kalinkovich A, Livshits G. Sarcopenic obesity or obese sarcopenia: A cross talk between age-associated adipose tissue and skeletal muscle inflammation as a main mechanism of the pathogenesis. Ageing Res Rev 2017; 35: 200–21. doi:10.1016/j.arr.2016.09.008

¹³ Ligibel JA, Schmitz KH, Berger NA. Sarcopenia in aging, obesity, and cancer. Transl Cancer Res 2020; 9: 5760–71. doi:10.21037/tcr-2019-eaoc-05

¹⁴ Kioh SH, Mat S, Kamaruzzaman SB, et al. Does Lower Lean Body Mass Mediate the Relationship Between Falls and Higher Body Mass Index in Asian Older Persons? J Aging Phys Act 2020; 28: 426–33. doi:10.1123/japa.2019-0011

¹⁵ Öztürk ZA, Türkbeyler İH, Abiyev A, et al. Health-related quality of life and fall risk associated with age-related body composition changes; sarcopenia, obesity and sarcopenic obesity. Intern Med J 2018; 48: 973–81. doi:10.1111/imj.13935

¹⁶ Montero-Odasso M, van der Velde N, Martin FC, et al. World guidelines for falls prevention and management for older adults: a global initiative. Age Ageing 2022; 51: 1–36. doi:10.1093/ageing/afac205

¹⁷ Lamb SE, Becker C, Gillespie LD, et al. Reporting of complex interventions in clinical trials: development of a taxonomy to classify and describe fall-prevention interventions. Trials 2011; 12: 1–8. doi:10.1186/1745-6215-12-125

¹⁸ Barazzoni R, Bischoff S, Boirie Y, et al. Sarcopenic Obesity: Time to Meet the Challenge. Obes Facts 2018; 11: 294–305. doi:10.1159/000490361

¹⁹ Abu Hammattah A, Mohd Yunus R, Matthias Müller A, et al. Association between structural social support and quality of life among urban older Malaysians. Australas J Ageing 2021; 40: 390–6. doi:10.1111/ajag.12919

²⁰ Nalathamby N, Morgan K, Mat S, et al. Validation of the CASP-19 Quality of Life Measure in Three Languages in Malaysia. J Trop Psychol 2017; 7: 1–8. doi:10.1017/jtp.2017.4

²¹ Ong JYQ, Mat S, Kioh SH, et al. Cognitive frailty and 5-year adverse health-related outcomes for the Malaysian elders longitudinal research (MELoR) study. Eur Geriatr Med 2022; 13: 1309–16. doi:10.1007/s41999-022-00673-x

²² Malmstrom TK, Morley JE. SARC-F: a simple questionnaire to rapidly diagnose sarcopenia. J Am Med Dir Assoc 2013; 14: 531–2. doi:10.1016/j.jamda.2013.05.018

²³ Wong CJ. Involuntary weight loss. Med Clin North Am 2014; 98: 625–43. doi:10.1016/j.mcna.2014.01.012

²⁴ Gaddey HL, Holder KK. Unintentional Weight Loss in Older Adults. Am Fam Physician 2021; 104: 34–40.

²⁵ Barry E, Galvin R, Keogh C, et al. Is the Timed Up and Go test a useful predictor of risk of falls in community dwelling older adults: a systematic review and meta-analysis. BMC Geriatr 2014; 14: 14. doi:10.1186/1471-2318-14-14

²⁶ Hamer M, O’Donovan G. Sarcopenic obesity, weight loss, and mortality: the English Longitudinal Study of Ageing. Am J Clin Nutr 2017; 106: 125–9. doi:10.3945/ajcn.117.152488

²⁷ Jensen MD, Ryan DH, Apovian CM, et al. 2013 AHA/ACC/TOS guideline for the management of overweight and obesity in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and The Obesity Society. Circulation 2014; 129: S102–38. doi:10.1161/01.cir.0000437739.71477.ee

²⁸ Newman AB, Lee JS, Visser M, et al. Weight change and the conservation of lean mass in old age: the Health, Aging and Body Composition Study. Am J Clin Nutr 2005; 82: 872–8. doi:10.1093/ajcn/82.4.872

²⁹ Kang H. Sample size determination and power analysis using the G*Power software. J Educ Eval Health Prof 2021; 18: 17. doi:10.3352/jeehp.2021.18.17

³⁰ Faul F, Erdfelder E, Buchner A, et al. Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav Res Methods 2009; 41: 1149–60. doi:10.3758/BRM.41.4.1149

³¹ Aravindhan K, Mat S, Hamid TA, et al. Development of Virtual Surveys for the COVID-19 Wave of the AGELESS Longitudinal Study in Malaysia. Gerontology 2022; 68: 551–5. doi:10.1159/000517946

³² Alex D, Khor HM, Chin AV, et al. Cross-sectional analysis of ethnic differences in fall prevalence in urban dwellers aged 55 years and over in the Malaysian Elders Longitudinal Research study. BMJ Open 2018; 8: e019579. doi:10.1136/bmjopen-2017-019579

³³ Villareal DT, Apovian CM, Kushner RF, et al. Obesity in older adults: technical review and position statement of the American Society for Nutrition and NAASO, The Obesity Society. Am J Clin Nutr 2005; 82: 923–34. doi:10.1093/ajcn/82.5.923

³⁴ Batsis JA, Villareal DT. Sarcopenic obesity in older adults: aetiology, epidemiology and treatment strategies. Nat Rev Endocrinol 2018; 14: 513–37. doi:10.1038/s41574-018-0062-9

³⁵ Chan YY, Lim KK, Lim KH, et al. Physical activity and overweight/obesity among Malaysian adults: findings from the 2015 National Health and morbidity survey (NHMS). BMC Public Health 2017; 17: 733. doi:10.1186/s12889-017-4772-z

³⁶ Kyaw TM, Ismail Z, Selamat MI, et al. Obesity and its associated factors among older adults: MyHEBAT (Malaysian HEalth and Well‐Being AssessmenT) study. Heal Sci Rep 2022. doi:10.1002/hsr2.668

³⁷ Oh K-J, Lee DS, Kim WK, et al. Metabolic Adaptation in Obesity and Type II Diabetes: Myokines, Adipokines and Hepatokines. Int J Mol Sci 2016; 18: 8. doi:10.3390/ijms18010008

³⁸ Donini LM, Busetto L, Bischoff SC, et al. Definition and Diagnostic Criteria for Sarcopenic Obesity: ESPEN and EASO Consensus Statement. Obes Facts 2022; 15: 321–35. doi:10.1159/000521241

³⁹ Ponti F, Santoro A, Mercatelli D, et al. Aging and Imaging Assessment of Body Composition: From Fat to Facts. Front Endocrinol (Lausanne) 2020; 10: 488049. doi:10.3389/FENDO.2019.00861/BIBTEX

⁴⁰ Willoughby D, Hewlings S, Kalman D. Body Composition Changes in Weight Loss: Strategies and Supplementation for Maintaining Lean Body Mass, a Brief Review. Nutrients 2018; 10: 1876. doi:10.3390/nu10121876

⁴¹ Greendale GA, Sternfeld B, Huang M, et al. Changes in body composition and weight during the menopause transition. JCI Insight 2019; 4: e124865. doi:10.1172/jci.insight.124865

⁴² Sherlock M, Toogood AA. Aging and the growth hormone/insulin like growth factor-I axis. Pituitary 2007; 10: 189–203. doi:10.1007/s11102-007-0039-5

⁴³ Bartke A. Growth hormone and aging. Rev Endocr Metab Disord 2021; 22: 71–80. doi:10.1007/S11154-020-09593-2/METRICS

⁴⁴ Juul A, Skakkebaek NE. Androgens and the ageing male. Hum Reprod Update 2002; 8: 423–33. doi:10.1093/humupd/8.5.423

⁴⁵ Kirchengast S, Huber J. Gender and age differences in lean soft tissue mass and sarcopenia among healthy elderly. anthranz 2009; 67: 139–51. doi:10.1127/0003-5548/2009/0018

⁴⁶ Li X, Liu J, Zhou B, et al. Sex Differences in the Effect of Testosterone on Adipose Tissue Insulin Resistance From Overweight to Obese Adults. J Clin Endocrinol Metab 2021; 106: 2252–63. doi:10.1210/clinem/dgab325

⁴⁷ Newman AB, Kupelian V, on Behalf of the Health A and BCSI. Strength, But Not Muscle Mass, Is Associated With Mortality in the Health. Aging Body Comp Study Cohort Journals Gerontol Ser A 2006; 61: 72–7. doi:10.1093/gerona/61.1.72

⁴⁸ Goodpaster BH, Park SW, Harris TB, et al. The loss of skeletal muscle strength, mass, and quality in older adults: the health, aging and body composition study. J Gerontol A Biol Sci Med Sci 2006; 61: 1059–64. doi:10.1093/gerona/61.10.1059

⁴⁹ Kioh SH, Mat S, Kamaruzzaman SB, et al. Gender Differences in Association Between Measures of Central Obesity and Falls in Community-Dwelling Middle-Aged and Older Adults in Malaysia. Aging Med Healthc 2021; 12: 95–104. doi:10.33879/AMH.123.2020.07024

⁵⁰ Lim U, Ernst T, Buchthal SD, et al. Asian women have greater abdominal and visceral adiposity than Caucasian women with similar body mass index. Nutr Diabetes 2011; 1: e6. doi:10.1038/nutd.2011.2

⁵¹ Jeong S-M, Lee DH, Rezende LFM, et al. Different correlation of body mass index with body fatness and obesity-related biomarker according to age, sex and race-ethnicity. Sci Rep 2023; 13: 1–11. doi:10.1038/s41598-023-30527-w

⁵² Lynch GM, Murphy CH, Castro E de M, et al. Inflammation and metabolism: the role of adiposity in sarcopenic obesity. Proc Nutr Soc 2020; 79: 1–13. doi:10.1017/S0029665120007119

⁵³ Romli MH, Mackenzie L, Tan PJ, et al. Comparison of Retrospective and Prospective Falls Reporting Among Community-Dwelling Older People: Findings From Two Cohort Studies. Front Public Health 2021; 9: 612663. doi:10.3389/fpubh.2021.612663

⁵⁴ Ariaratnam S, Rodzlan Hasani WS, Krishnapillai AD, et al. Prevalence of obesity and its associated risk factors among the elderly in Malaysia: Findings from The National Health and Morbidity Survey (NHMS) 2015. PLoS One 2020; 15: e0238566. doi:10.1371/journal.pone.0238566

⁵⁵ Baharudin Shaharuddin A, Abdul Aziz NS, Ahmad MH, et al. Abdominal obesity and its associated factors among older adults in Malaysia. Geriatr Gerontol Int 2020; 20 Suppl 2: 68–72. doi:10.1111/ggi.13962

⁵⁶ Goisser S, Kiesswetter E, Schoene D, et al. Dietary weight-loss interventions for the management of obesity in older adults. Rev Endocr Metab Disord 2020; 21: 355–68. doi:10.1007/s11154-020-09577-2

⁵⁷ Redman LM, Ravussin E. Caloric restriction in humans: impact on physiological, psychological, and behavioral outcomes. Antioxid Redox Signal 2011; 14: 275–87. doi:10.1089/ars.2010.3253