Article Text

Download PDFPDF

Original research
Risk of falls in 4 years of follow-up among Chinese adults with diabetes: findings from the China Health and Retirement Longitudinal Study
  1. Yue Wen1,
  2. Jing Liao1,
  3. Yiqiong Yin2,
  4. Chunjuan Liu3,
  5. Renrong Gong4,
  6. Dongmei Wu5,6
  1. 1Department of Gastrointestinal Surgery, West China Hospital, Sichuan University, Chengdu, China
  2. 2Gastrointestinal Department, West China Hospital, Sichuan University, Chengdu, China
  3. 3West China School of Nursing, Sichuan University/ West China Hospital, Sichuan University, Chengdu, China
  4. 4Department of Surgery, West China Hospital, Sichuan University/ Department of Surgery, West China School of Nursing, Sichuan University /Nursing Key Laboratory of Sichuan Province, Chengdu, China
  5. 5The Clinical Hospital of Chengdu Brain Science Institute, MOE Key Lab for Neuroinformation, University of Electronic Science and Technology of China, Chengdu, China
  6. 6Department of Nursing, The Fourth People's Hospital of Chengdu & Chengdu Mental Health Center, Chengdu, China
  1. Correspondence to Dr Dongmei Wu; wudongmei_2001{at}163.com; Professor Renrong Gong; gongrenrong{at}wchscu.cn

Abstract

Objectives This study was to determine the incidence of falls and identify baseline factors increased risk for incident falls over time among people with diabetes.

Design This study was a secondary analysis using the baseline and 4 years of follow-up data from the China Health and Retirement Longitudinal Study (CHARLS).

Setting A nationally representative survey of 17 500 Chinese residents aged 45 years and older were recruited in the baseline national survey in 2011. These participants were followed up every 2 years.

Participants A total of 1238 middle-aged and older adults with diabetes and no history of falls at baseline were included in the current study.

Primary and secondary outcome measures Information on incidence of falls and medical treatment resulting from falls were determined by self-report.

Results The findings showed that the incidence of falls was 29.4% during 4 years of follow-up. Participants with incident falls were younger, were more likely to be women, had lower education level and were less likely to be current drinkers. In addition, former drinkers were 2.22 times more likely to fall. Socially active individuals were 47% less likely to fall compared with those without social activities. Every 5 kg increase in grip strength was associated with a 13% lower risk of falls. A 10 mg/dL higher total cholesterol and 1 mg/dL higher blood urea nitrogen were associated with a 4% and 6% higher risk of falls. Finally, participants with depressive symptoms were 1.47 times more likely to fall compared with those without depressive symptoms.

Conclusions These findings underscore the importance of developing a fall prevention programme for those with diabetes, and this programme should address potentially modifiable risk factors, including levels of total cholesterol, blood urea nitrogen, social activity, depressive symptoms and grip strength.

  • epidemiology
  • diabetes & endocrinology
  • public health

Data availability statement

Data are available in a public, open access repository. Study data sets can be downloaded at the CHARLS home page at http://charls.pku.edu.cn/en.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Strengths and limitations of this study

  • We used a national representative sample of participants with diabetes. Therefore, findings of our study can be generalised to adults with diabetes in China.

  • Prospective associations between baseline risk factors for falls and fall events were examined in 4 years of follow-up.

  • There might be non-differential recall bias in the current study due to using self-report of falls, which led to lower power to detect potential risk factors.

  • Due to the observational nature of the study design, causal associations between these risk factors and fall events cannot be drawn from the current study.

Introduction

China is now home to the world’s largest number of people with diabetes, with a report of 109.6 million adults having diabetes.1 2 The prevalence of diabetes increases with age, and an estimate of about 20.2% of adults over 60 years of age were diabetic in 2013.2 Recently, emerging evidence suggests that diabetes is associated with an increased risk of falls among older adults, especially for insulin users.3 The underlying mechanisms may lie in the diabetes-related pathological changes, which may include vestibular dysfunction, peripheral neuropathy, diabetic retinopathy, declines in muscle strength and severe hypoglycaemic events associated with insulin use.4 5 Falls are common in older adults with diabetes, with annual incidence rates of 39% among individuals aged 65 years or older and occurring more often in those with poor glycaemic control.6 Since falls are the leading cause of injury in older adults and can lead to decreased functional independence and lower quality of life,7 it is critical to identify predictors that can be easily used to assess fall risks in the clinical settings among older adults with diabetes.

Recently, a few studies have examined risk factors that are associated with risk of falls among individuals with diabetes. For example, using data from the 2010 wave of the Health and Retirement Study, Blackwood8 found that cognitive dysfunction, impairment in executive function and delayed recall, was associated with an increased risk for falls among the entire sample of older diabetes over 65 years of age and among a subsample of diabetess over 75 years of age, respectively. People receiving diabetes treatment, specially insulin users, had a significantly higher risk of falls (relatively risk=1.94) compared with non-insulin users (relative risk=1.27).3 Other risk factors include depressive symptoms,9 reduced physical performance (eg, walking speed,10 repeated chair stands11 and grip strength),12 being overweight/obese10 and problems with instrumental activities of daily living (IADLs).13 Hypoglycaemia14 and vision impairment15 that are associated with the disease also contribute diabetic fall risk. However, these studies were mainly cross-sectional or had limited follow-up time. Large cohort studies with longer follow-up time are needed to study factors associated with falls in older Chinese adults with diabetes. Meanwhile, associations between serum biomarkers and risk of falls are not well studied. Such prospective associations may provide valuable insights into how baseline health conditions may predict risk of falls in the follow-up periods and help develop effective strategies to address these risk factors during primary healthcare service. Therefore, the purpose of this study was to determine the incidence of falls and examine a comprehensive list of potential risk factors associated with falls in 4 years of follow-up among participants with diabetes of the China Health and Retirement Longitudinal Study (CHARLS).

Methods

Study design

This was a secondary data analysis of prospective data from the CHARLS. The CHARLS is an ongoing, biannual national survey, sponsored by the National Natural Science Foundation of China, the National Institute on Aging and the World Bank.16 The CHARLS questionnaire collects a substantial data on an individual’s sociodemographic information, family structure, biomarkers, health status, physical performance, health insurance, employment history, retirement and pension, individual and household assets, and community-level information.16 Participants were selected using a multistage, stratified, cluster probability sampling strategy.17 The sampling strategy has been previously described in detail,18 and the study data sets can be downloaded at the CHARLS home page at http://charls.pku.edu.cn/en. A representative sample of 17 314 community-dwelling individuals of 45 years or older across the country was recruited into the baseline wave of CHARLS, which was fielded in 2011–2012. The current analysis used data from baseline, the first and the second follow-up surveys.

Study sample

CHARLS participants with diabetes who had no history of falls and had complete data on age, gender, education level, falls, cognitive measures, body weight, height, walking speed, standing balance test, depressive symptoms, diabetes treatment and plasma biomarkers at the time of the baseline survey were included in the current study. The final sample included in this study consisted of 1238 Chinese adults with diabetes.

The American Diabetes Association19 criteria was used to define diabetes status. Specifically, participants were diagnosed with diabetes if self-reporting to take hypoglycaemic agents (ie, insulin use, or taking oral hypoglycaemic medications including traditional Chinese, modern western medicine, or other diabetes treatment), fasting blood glucose ≥126 mg/dL, or random blood glucose ≥200 mg/dL, or HbA1c ≥6.5%. Fasting venous blood samples were collected by trained health professionals on the day of physical examination. However, a small proportion (8%) of participants did not fast.20 For those participants, random blood glucose and/or HbA1c were used to define diabetes.

Patient and public involvement

In the current study, we used deidentified data from the CHARLS with no direct involvement of or interaction with participants in the design, recruitment or conduct of the original cohort study.

Variables, definitions and measures

Falls

Information on incidence of falls and medical treatment resulting from falls was collected in the CHARLS. The participant was asked if he or she had fallen down in the past 2 years prior to the survey. If the participant answered ‘yes’ to this question, he or she was then asked to indicate how many times falls resulted in a medical treatment.

Measurement of potential risk factors

Potential risk factors included demographics, lifestyle behaviours, depressive symptoms, physical health and functioning variables, biomarkers, cognitive function and diabetes treatment. Demographics and lifestyle behaviours were measured based on self-report. Information on age, gender, marital status and education levels was collected using face-to-face interviews. Marital status was categorised as married or separated. Education levels included no formal education/illiterate, some primary school, finish primary school, and junior high school or above. Lifestyle behaviours included smoking, drinking and social activities. Smoking and drinking status were categorised as never, former and current users. Social activity was measured as no social activity, some social activity and socially active.

Depressive symptoms were measured using the Center for Epidemiological Studies Depression Scale (CES-D) short form.21 The CES-D short form consists of ten items, and each item is rated on a four-point Likert scale ranging from 0 (rarely or none of the time) to 3 (most or all of the time) with a total possible summary score of 0–30.21 Two positive symptoms (ie, ‘was happy’ and ‘hopeful about the future’) were reversely coded before data analysis. The time frame refers to the week prior to the participants’ interview date. The CES-D short form has been validated among a subsample of 742 CHARLS participants aged 60 years and older and has shown adequate psychometric properties. A CES-D score of 12 or higher was defined as having major depressive symptoms.22

Physical health and functioning variable included vision and hearing function, body mass index (BMI), repeated chair stand test, walking speed, grip strength, systolic blood pressure and IADL. Vision and hearing impairments were self-reported by study participants. BMI was calculated by dividing body weight in kilograms by the square of body height in metres, kg/m2. To conduct repeated chair stand test, participants were asked to sit in the middle of the chair and place their hands on the opposite shoulder. Then they were asked to rise to a full standing position and sit back down again for five times. The examiner recorded the time if the participants finished the test without arms. If the participants must use their arms to stand, the examiner stopped the test and recorded ‘0’ for the number and score. The median time of five tests was used in the analysis. All participants aged 60 years or older without physical limitations that may interfere with walking were eligible for the test of walking speed. Participants were instructed to walk on a straight 2.5 m flat course twice (there and back) at their normal walking speed. The examiner used a stopwatch to record the elapsed time necessary to walk the distance. The median time of the two tests was used as a measure of walking speed.23 24 A handheld dynamometer was used to assess grip strength. Participants were instructed to squeeze the dynamometer with all of their strength for a few seconds, typically twice with each hand, and alternate hands between tests. Consistent with a prior CHARLS publication,25 an average score was calculated using the four measurements from both hands. IADL refers to meal preparation, doing housework, shopping, managing personal finances and managing medications, and these activities were measured with scores ranging from 0 to 5.26 Higher scores indicate having more difficulty in performing IADL, and loss of independence and mobility.

Biomarkers included blood lipids (low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, total cholesterol and triglycerides), blood glucose (fasting glucose and haemoglobin A1c), creatinine, cystatin C, uric acid, blood urea nitrogen (BUN), C reactive protein, haemoglobin and haematocrit.

Subdomains of cognition measured in the CHARLS included visuospatial abilities, episodic memory and orientation/attention. Consistent with prior CHARLS publications,23 24 an overall cognitive score was calculated as a sum of these three cognitive subdomains, which could range from 0 to 21 and was used to indicate overall cognitive functioning of the participant.

Figure drawing

As previously presented by CHARLS,23 24 the examiner first demonstrated a picture of two overlapping pentagons, and the participant was required to draw a similar picture. Those who correctly drew the picture received a score of 1, while those who failed to draw the picture received a score of 0. This test was designed to measure a person’s capacity to identify visual and spatial relationships among objects.

Word recall

This was a memory task for assessing both immediate and delayed recall. The immediate recall test involved presenting the participant with a list of 10 random words, which were read at a constant rate of one word every 2 s. At the end of the presentation, the participant was given up to 2 min to recall the list of words. Approximately 4–10 min later, a delayed recall test was administered by asking the participant to recall the list of 10 words presented earlier. For each task, the number of correctly recalled words was scored, with higher scores indicating better memory performance. In line with prior CHARLS publications,23 24 an episodic memory score could range from 0 to 10 by averaging number of correctly recalled words from both immediate and delayed word tasks.

The Telephone Interview of Cognitive Status (TICS-10)

The original TICS is a global mental status test that can either be administered over the telephone or face to face.27 As previously presented by CHARLS,23 24 10 questions from the original TICS were used in the CHARLS baseline survey, including date (day, month and year), day of the week, the serial subtraction of 7 beginning with the number 100 up to five times and season of the year. The TICS-10 was used to assess orientation/attention, and it was calculated as the sum of correct answers which could range from 0 to 10.28

Treatment of diabetes was based on self-report. Participants were asked whether they took medications, including traditional Chinese medicine, modern medicine and insulin, to treat diabetes. Participants that took any of the medications were coded as receiving antidiabetic treatment.

Ethical considerations

All participants provided signed written consent forms in the original CHARLS study.20

The current study is a secondary analysis of the deidentified CHARLS public data. The Ethics Review Committee at University of Electronic Science and Technology of China granted an exempt research determination to the current study.

Statistical analysis

Baseline characteristics of the participants were summarised as mean and SD or median and IQR for continuous variables and frequency and percentage for categorical variables. Binary associations between incidence of falls and potential risk factors at baseline were tested using χ2 tests for categorical variables and one-way analyses of variance for continuous variables. Significant variables in the binary analyses were added in a multivariable logistic regression model to identify risk factors for fall. ORs and the corresponding 95% CIs were reported. In a sensitivity analysis, we also include variables with a p value <0.15 into the full model. The SAS V.9.4 (SAS Institute Inc, Cary, North Carolina, USA) was used to analyse the data. All p values were two sided, and p<0.05 was considered significant.

Results

Characteristics of the study participants

A total of 1238 participants with diabetes who reported no falls in the 2011 baseline visit were included in the current analyses, of whom, 364 (29.4%) reported to have fall(s) in 4 years of follow-up. As shown in table 1, compared with participants had no incident fall, those with incident falls were younger (59.4 vs 61.8), were more likely to be women (56.9% vs 49.0%), had lower education level (illiterate rate: 33.2% vs 24.9%) and were less likely to be current drinkers (13.8% vs 21.6%). In addition, participants having incident fall(s) were more likely to be socially inactive (68.2% vs 67.0%) and receive diabetes treatment (42.0% vs 36.7%). However, these differences were not statistically significant (p=0.09 and 0.08, respectively). The two group of participants were similar in living areas, marital status and smoking status.

Table 1

Baseline characteristics of diabetic participants of the China Health and Retirement Longitudinal Study according to incidence of fall in 4 years follow-up (2011–2015)

Significant variables in bivariable analyses

In addition to age, gender, education level and drinking status as described above, depressive symptoms, vision impairment, hearing problem, grip strength, IADL, total cholesterol, TICS-10, figure drawing and total cognitive scores were also significantly associated with incident falls in the bivariable analyses (table 1).

Full-model results

As shown in table 2, when putting all significant variables identified in the bivariable analyses in the full model, only drinking status, grip strength, total cholesterol and depressive symptoms were significant predictors of incident fall(s). Vision impairment was a nominally significant predictor. Compared with never drinkers, former drinkers were 2.22 times (95% CI 1.24 to 3.99) more likely to fall. However, current drinker had a similar risk as never drinkers (OR=1.11, 95% CI 0.69 to 1.81). Every 5 kg increase in grip strength was associated with a 13% (95% CI 2% to 21%) lower risk of falls. A 10 mg/dL higher total cholesterol was associated with a 4% (95% CI 0% to 7%) higher risk of falls. Finally, participants with depressive symptoms were 1.47 (95% CI 1.03 to 2.11) times more likely to fall compared with those without depressive symptoms.

Table 2

Associations between baseline potential risk factors and incidence of fall in 4-year follow-up (2011–2015) among diabetes participants of the China Health and Retirement Longitudinal Study

Furthermore, when adding all variables with a p<0.15 in the bivariable analyses, p values for depressive symptoms (p=0.09) and total cholesterol (p=0.08) became nominally significant; however, magnitudes of the associations did not change much. Furthermore, two new variables, social activity and BUN were significantly associated with risk of fall. Compared with people with no social activity, socially active individuals were 0.53 (95% CI 0.31 to 0.91) times less likely to have fall. On contrary, 1 mg/dL higher BUN was associated with 1.06 (95% CI 1.01 to 1.11) times higher risk for fall.

Discussion

In this prospective analysis among this nationally representative sample of middle-aged and older Chinese diabetes participants without any history of fall(s), 29.4% reported to have incidence of falls in 4 years of follow-up. Compared with participants had no incident fall, those with incident falls were younger, were more likely to be women, had lower education attainment and were less likely to be current drinkers. Furthermore, we identified four factors predicting risk for falls, including drinking status, grip strength, total cholesterol and depressive symptoms. These findings will not benefit allocation of healthcare resources to address health conditions and provide evidence for prevention strategies of falls in this population.

Alcohol consumption has been considered as a risk factor for falls by the WHO, as physiological changes related to ageing may increase sensitivity to alcohol use in older adults.29 However, few studies have examined alcohol use as a risk for falls in community-dwelling older adults with diabetes. In this study, we found that former drinkers had a more than twice the risk of falls compared with never drinkers; however, current drinkers had a similar risk as never drinkers. The findings are consistent with a previous study; in a large-scale study among 289 187 adults in the 2004–2013 US National Health Interview Surveys, former drinker had a similar higher risk of falls as at-risk drinkers, compared with lifetime abstainers.30 A possible reason could be that former drinkers might have stopped drinking due to poor health status which predisposed them to risk of fall. In the current study, compared with never drinkers, former drinkers were more likely to have vision impairment and take diabetes medications. It is possible that the vision impairment was due to worse glucose control, and former drinkers were highly suggested by doctors to take diabetes medications. Future studies with larger sample sizes, particularly for former drinkers, and more detailed measures of health status are warranted to further validate our findings. Nevertheless, studies have shown that low to moderate drinking was associated with a reduced risk of fall.31 In the CHARLS, about two-thirds of the current drinkers were moderate drinkers (defined as having ≤14 drinks per week), and the remaining were at-risk drinkers (defined as having >14 drinks per week).32 In the current study, current drinkers had a similar risk of fall as never drinkers. This may be a result of mixture of moderate and at-risk drinking. However, these are preliminary findings that need to be better investigated in future studies.

The hand grip strength is an indicator of muscle strength, which is important to fall prevention.33 Grip strength was negatively associated with risk of falls in the current study, and this finding is intuitive and consistent with previous studies, which have repeatedly demonstrated a negative association between stronger grip strength and reduced fall events.12 34–36 Our study provided further evidence that stronger grip strength at baseline was longitudinally associated with a less risk of falls in 4-year follow-up among an older population with diabetes in China. In addition, hand grip strength has also been linked to other important diabetes outcomes. Using longitudinal data from the UK Biobank, Celis-Morales et al37 found that patients with diabetes who had stronger grips had a reduced risk of all-cause mortality, lower risk of death from cardiovascular disease mortality and less risk of developing cardiovascular disease. However, a decline in hand grip strength has been reported among individuals with diabetes compared with healthy individuals.38 Taken together, these studies suggest that reduced grip strength may be used to identify a subgroup of patients with diabetes who are at a higher risk of falls and other important diabetes outcomes, and it is important to develop strategies to increase hand grip strength in this population.

Higher total cholesterol and BUN levels at baseline increased the risk of falls at follow-up in the current study. Previous studies that examined the relationship between biomarkers and fall risks have primarily focused on levels of high-density lipoprotein cholesterol,39 which were not associated with fall events in the current study. Levels of total cholesterol are inversely associated with plasma 25-hydroxyvitamin D levels,40 which are an important marker for frailty among Chinese community-dwelling older adults.41 Levels of total cholesterol are also inversely associated with lean body mass,42 which is important to determine risks of falls in older Asian adults.43 However, plasma 25-hydroxyvitamin D and lean body mass were not measured in the CHARLS, so we were not able to examine whether these two biomarkers may explain the association between levels of total cholesterol and fall events in this study. Since grip strength is correlated with lean body mass, we tried to evaluate total cholesterol and fall associations before and after controlling for grip strength; however, adding grip strength only slightly changed the effect of total cholesterol on fall. BUN reflects fluid depletion and is also influenced by protein intake, catabolism and tubular reabsorption.44 45 Dehydration is an important risk factor for fall and water intake could significantly reduce risk for fall in nursing homes.46 However, in a case–control study among hospitalised patients, BUN was not associated fall.45 Our study provided longitudinal evidence among a community-dwelling older adults that higher BUN is associated with high risk for fall. Future studies of more biomarkers, such as metabolomics studies, may identify additional biomarkers and reveal underlying mechanisms of falls in this population.

The current study found that higher depressive symptoms may predict falls over time among individuals with diabetes. Such association has been widely reported in case–control, cross-sectional and short-term prospective studies in the general populations.47–50 Our study added further evidence with a larger sample size and a longer follow-up time among individuals with diabetes. Evidence suggested that baseline depressive symptoms increased fall risks through at least three different mechanisms. First, overall antidepressant use, particularly tricyclic antidepressants, are considered to contribute to falls because they increase sedation and risk of orthostatic hypotension.51 Second, compared with the general population, patients with diabetes are more likely to experience depression.52 A meta-analysis53 demonstrated a significant association between depression and treatment non-adherence, including failing to engage in regular exercise among individuals with diabetes. As regular exercise contributes to muscle strength,54 those who do not exercise regularly may have weaker muscle strength, which is associated with an increased risk of falls at follow-up. Third, depression and falls are also linked with each other through several common risk factors. Fear of falling, functional decline, history of falls and cognitive dysfunction have been separately linked to both depression and falls.48 However, due to social discrimination, depression was not routinely screened in clinical practice in China. Only a small proportion of patients with depressive symptoms (5%–8%) were diagnosed with depression, and less than half of patients who were diagnosed with depression sought care over time.55 56 Our study highlighted the importance of screening depression to prevent fall among patients with diabetes in China.

Socially active individuals also had lower risk of fall in the current study. This finding is consistent with previous studies, in which social engagement could improve physical and mental health and reduce risk of fall.57–60 In the current study, even after adjusting for physical and cognitive function measures, social activity was still associated with lower risk of fall. As hypothesised in previous studies, for socially active individuals, attention to environmental hazards and adaptation to changes in physical function may be triggered and encouraged by peers in social activities. Furthermore, socially active individuals may be more likely to get assistance from peers in routine or occasional tasks, such as reaching to an object on a high shelf during shopping.57

The findings of the current study have important implications for clinical practice. Given the high incidence of fall events (29.4%) among middle-aged and older adults with diabetes, it is important to develop a screening programme with the goal of identifying at risk and ensuring these individuals receive fall prevention programmes. The identified risk factors, including drinking status, hand grips strength, blood lipids and depressive symptoms, should be integrated into the screening programme. In addition, as symptoms of depression are a potentially modifiable risk factor for falls, they should be addressed in fall prevention programmes. Treatment of depressive symptoms by non-pharmacological approaches, such as physical exercise or psychosocial therapies such as mindfulness, should be considered as part of fall prevention programmes in this high-risk population for falls.61 Interventions to reduce the likelihood of falls among individuals with diabetes who have weak grip strength should focus on a combination of physical exercise, including resistance54 and aerobic training exercises,62 which has been shown to improve grip strength in patients with diabetes. Fall prevention programmes should also target reducing levels of total cholesterol, as these individuals were at a higher risk of falls. Participants should be advised to first make lifestyle changes to improve cholesterol, including diet changes and increasing physical activity.

This study has several important strengths. A major strength of our study is that we used a national representative sample of participants with diabetes. Therefore, findings of our study can be generalised to persons with diabetes in China. In addition, we have also evaluated a comprehensive list of risk factors to predict fall events in 4 years of follow-up. Previous studies of falls have focused on limited number of factors.3 8–15 Furthermore, a prospective cohort design was used, which avoided temporal ambiguity of potential factors and falls, and reduced survival bias, in which an identified factor may be a result of fall rather than a risk factor. Finally, to ensure data integrity and validity, quality assurance measures have been implemented throughout the process of data collection in the CHARLS. These measures include reviewing data collection forms for completeness and accuracy of the data, verifying accuracy of data in electronic databases, as well as calling back participants.17

Our study also has limitations that need to be acknowledged. As in all population-based studies, fall events were determined by self-report. Previous research consistently showed that older adults tended to under-report falls because they did not recognise the severity of a fall or did not remember a fall with less severe consequences.63 Therefore, there might be recall bias, and we should expect that the fall events may be under-reported by participants in the current study. A prospective design with ‘daily fall calendar’ is considered the golden standard to measure fall events,63 and this method should be incorporated into future studies to more accurately capture fall events among individuals with diabetes. Another limitation of the study is that information on number of glucose-lowering medications was not collected. Therefore, we could not assess whether number of medications may be a risk factor for fall risks in this population. Furthermore, some important and widely acknowledged risk factors, such as diabetic neuropathy and autonomic neuropathy, were not measured in the CHARLS. We were not able to assess their contributions to fall in this population. Finally, this study examined the prospective relationships of baseline risk factors for falls with fall events in 4-year follow-up, and therefore, no causal associations between these risk factors and fall events can be drawn from the current study.

Conclusions

To conclude, through a longitudinal study among a nationally representative sample of middle-aged and older participants with diabetes, we estimated that the incidence of fall events in 4 years of follow-up was as high as 29.4%. We also identified four factors predicting fall events. Besides the five risk factors (ie, alcohol drinking, grip strength, social activity, depressive symptoms and BUN) that have been reported in previous case–control, cross-sectional and/or short-term prospective studies, our study identified a novel risk factor (ie, levels of total cholesterol) to predict fall events among individuals with diabetes. Future studies of more biomarkers, such as a metabolomics study, are warranted to identify additional biomarkers for and reveal underlying mechanisms of fall events among this vulnerable population.

Data availability statement

Data are available in a public, open access repository. Study data sets can be downloaded at the CHARLS home page at http://charls.pku.edu.cn/en.

Ethics statements

Acknowledgments

The authors would gratefully acknowledge the assistance Dr Tingting Liu, an assistant professor at University of Arkansas Eleanor Mann School of Nursing for editing the manuscript. The China Health and Retirement Longitudinal Study (CHARLS) is funded by Peking University, the National Natural Science Foundation of China, the Behavioural and Social Research Division of the National Institute on Ageing and the World Bank.

References

Footnotes

  • Contributors All authors contribute to the conception and design of this study. YW, JL, CL and DW were responsible for the design, analysis, drafting and revision of this manuscript. RG and YY were responsible for interpretation of data and preparation of the manuscript.

  • Funding This work was supported by the Sichuan Science and Technology Program grant number 2018JY0306 to DW.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.