Objective To determine the rate of falls reported by older community dwellers in Shenzhen, China and to identify fall-related risk factors.
Method Participants were community dwellers residing in Shenzhen, China, who were aged 60 years and over and were recruited using multistage random sampling. All participants were surveyed about demographic and health-related information, mood, vision and hearing impairment, self-rated health and retrospective falls, and a test of balance was administered. Univariate and multivariate negative binomial regression was used to identify factors associated with a greater number of falls.
Result Study participants were 1290 people aged 60–98 years (mean 68.2 years, SD ±6.5). One hundred and seventy-seven falls were reported. One hundred and eleven (8.6%) participants reported one fall in the past year, 17 (1.3%) participants reported two falls and 10 (0.8%) participants reported three or more falls. Univariate analysis showed that age, living alone, presence of a medical condition, medication usage, visual impairment, poor subjective body sense perception, low mood, poor self-rated health and poor balance were associated with a greater number of falls in the past year. Multivariate analysis identified presence of a medical condition (incidence rate ratio (IRR)=1.40, 95% CI 1.19 to 1.67), living alone (IRR=2.46, 95% CI 1.12 to 5.41), visual impairment (IRR=1.46, 95% CI 1.03 to 2.08), walking aid use (IRR=2.29, 95% CI 1.12 to 4.69) and impaired balance (IRR=1.05, 95% CI 1.00 to 1.10) to be significantly associated with a greater number of falls in the past year.
Conclusion More falls occurred in older Chinese people with presence of a medical condition, living alone, visual impairment, used a walking aid and impaired balance.
- accidental falls
- risk factors
Statistics from Altmetric.com
Falling is an important public health issue among older people and is becoming more and more important with the increasingly ageing population. Falls, defined as ‘events which result in a person coming to rest inadvertently on the ground or floor or other lower level’,1 are one of the most common causes of injuries among older people and can lead to long-term disability and even death.2 Approximately 30% of American community dwellers aged 65 years and over fall annually, and around half of these people fall multiple times.3
A range of risk factors for falls have previously been identified, including sociodemographic factors, impaired mobility and sensory functioning, psychological factors and medication use.4 Risk factors for falls in older Chinese people appear to be similar to those found among Caucasian seniors.5 Older Chinese people, however, have a relatively low annual rate of falls compared with that found among Caucasians,5–7 and the reasons for this are not yet understood. Further exploration of risk or protective factors associated with falls in Chinese people is warranted.
The factors associated with being a faller among older Chinese people have been determined by a range of previous studies.5 8 However, investigation of the risk factors for the rate of falls experienced by older Chinese people has had little attention, and these factors may vary from the risk factors for being a faller with a single fall.9 It could provide a greater understanding of why older Chinese people appear to have a relatively lower annual rate of falls compared with the rate in other populations. Therefore, this study investigated the risk factors associated with rate of falls in older community dwellers in Shenzhen, China.
We randomly selected two districts from all six districts in Shenzhen, and then randomly selected three subdistricts from 18 subdistricts administered by these two districts. Six communities with similar economic status and population size were selected based on government-level data. We excluded all non-eligible participants by checking the family health records in each community against the study inclusion/exclusion criteria, then randomly selected around 210 participants in each community from the health records. If there was more than one eligible person within a family, we randomly selected one person. Participants were included if they were: aged 60 years or older; had lived in the current address for at least 1 year; had the ability to communicate with the interviewers; and were able to ambulate independently (including with the use of walking aid).
All participants completed a study-specific questionnaire that collected information about falls in the past 12 months, demographic characteristics, medical history and medication use, visual and hearing impairment, mood, body sense perception, health status and fall characteristics. The questionnaire was administered by health workers from community health services who were trained in the study procedures prior to study commencement during May and June 2014.
Demographic information collected included sex, age, education level (categorised as illiteracy, primary school, middle school, high school and college level and above) and living status (alone or with others).
A fall was defined as an event that resulted in a person coming to rest inadvertently on the ground or floor or other lower level.1 We collected information on falls by asking the patient whether he or she had a fall in the past 12 months and how many times they fell.
Medical conditions and medication use
Participants were asked whether they had the following medical conditions: hypertension/postural hypotension, cardiac disease, cerebrovascular disease (cerebral infarction/stroke/hemiplegia), diabetes, bone hyperplasia (cervical spondylosis/lumbar disease/spur), osteoporosis, arthritis (rheumatic/rheumatoid), vertigo, nerve dysfunction (compression/pain), Parkinson’s disease, chronic bronchitis/asthma/tuberculosis, eye diseases (cataract/glaucoma/high myopia), Alzheimer’s disease, lower limb disability and chronic diarrhoea. The total number of medical conditions was calculated for the analyses.
Participants were asked whether they had taken the following drugs for at least 3 months in the previous 12 months: hypotensors, hypoglycaemic agents, sedatives, antidepressants, neuroleptics, central analgesics, hormones such as cortisone and prednisone, diuretics, antiarrhythmics and vestibular inhibitors. The medication variable used in the analyses was the total number of medications taken stratified into four categories: 0, 1, 2 and 3+ medications.
Participants who reported that they needed to go to the bathroom at least three times per night on average in the previous 12 months were classified as having nocturia.10
Participants who reported a visual impairment that made them unable to recognise a person at a distance of 4 m in the previous 12 months were classified as being visually impaired.11
Participants who reported that they could not hear clearly during daily conversations with others in the community in the previous 12 months were classified as having hearing impairment.12
Subjective body sense perception
Participants who reported lower limb dysfunction, pain or impaired sensation of warmth or touch in the previous 12 months were classified as having a poor subjective body sense perception.
Participants were asked to rate their mood on average in the previous 12 months in answer to the following question: ‘Did you feel happy for the most of time in the previous 12 months?’, and participants who answered other than yes were classified as having low mood.
Walking aid use
Participants were asked about walking aid use in the past 12 months. The responses from our participants were categorised into four options: (1) never used walking aid; (2) was recommended to use walking aid but never used; (3) used walking aid occasionally and (4) used walking aid usually. Then we combined options 1 and 2 as ‘did not use walking aid’ and combined 3 and 4 as ‘used walking aid’.
Participants were asked to rate their health over the past week and those who rated their health as other than good were classified as having poor self-rated health.
Participants underwent an assessment of balance ability. The test was based on the balance ability test in the fall prevention guideline published by the Ministry of Health of the People’s Republic of China and involved a static balance test, postural control ability test and a dynamic balance test.13 A total score out of 24 was allocated with higher scores indicating poorer balance (online supplementary appendix A).
Supplementary file 1
This study was conducted as a part of a national survey. Falls incidence was estimated as 14% from previous literature, allowable deviation was 15%, uα was 1.96 and design effect was 1.25; hence, the required sample size was estimated as 1310.
Descriptive statistics were used to summarise the characteristics of participants. Negative binomial regression was performed to analyse the association between each risk factor and the rate of falls reported in the past 12 months. Variables that were significantly associated with falls in the univariate analyses were entered into a multivariate negative binomial regression model. Final variables were selected using backwards stepwise regression (probability for removal=0.1, probability for addition=0.05). Data were analysed with Stata V.14, and a P value of less than 0.05 was considered statistically significant.
A total of 1310 potential participants were initially interviewed; 1290 people were recruited and completed the study measures, representing a response rate of 98.5%. The participants were aged 60–98 years (mean 68.2 years, SD ±6.5) and 740 (57.4%) were female. There were 138 (10.7%) people who had at least one fall in the past year, and a total of 177 fall events were reported. The rate of falls was 13.7/100 person-years. One hundred and eleven (8.6%) participants fell once in the past 12 months, 17 (1.3%) participants fell twice and 10 (0.8%) participants fell at least three times (9 fell three times and 1 fell five times). The characteristics of the sample are shown in table 1.
Factors associated with falls
The results of the univariate analyses are shown in table 2.
The factors associated with rate of falls were age (IRR=1.55, 95% CI 1.19 to 2.03), living alone (IRR=2.64, 95% CI 1.12 to 6.22), presence of a medical condition (IRR=1.57, 95% CI 1.34 to 1.85), medication usage (IRR=1.38, 95% CI 1.07 to 1.78), visual impairment (IRR=1.65, 95% CI 1.16 to 2.36), poor subjective body sense perception (IRR=2.42, 95% CI 1.70 to 3.43), low mood (IRR=1.65, 95% CI 1.13 to 2.41), walking aid use (IRR=5.15, 95% CI 2.84 to 9.36), poor self-rated health (IRR=1.62, 95% CI 1.10 to 2.37) and impaired balance (IRR=1.11, 95% CI 1.07 to 1.15). No associations were detected for sex, education level, hearing impairment and nocturia.
The variables retained in the multivariate model using backwards stepwise regression to determine those factors independently associated with falls are shown in table 3.
The factors associated with rate of falls in the past year were presence of a medical condition (IRR=1.40, 95% CI 1.19 to 1.66), living alone (IRR=2.46, 95% CI 1.12 to 5.41), visual impairment (IRR=1.46, 95% CI 1.03 to 2.08), walking aid use (IRR=2.29, 95% CI 1.12 to 4.69) and impaired balance (IRR=1.05, 95% CI 1.00 to 1.10).
This study found that 138 (10.7%) people among 1290 participants in Shenzhen experienced at least one fall in the previous 12 months and a rate of falls of 13.7 per 100 person years, which is a lower fall rate than in studies conducted in other countries. This result might be attributed to different lifestyle and cultural factors.7 Planned outdoor activities are popular in China, which might be associated with a lower risk of falling.
We found that people with at least one medical condition were more likely to experience additional falls. The presence of at least one medical condition increased the rate of falls by 40%. A point increase in the balance impairment scale (min: 0–max: 24) was associated with a 5% greater rate of falls. For the other risk factors, people who used walking aids, who had visual impairment and people who lived alone had 2.29, 1.46 and 2.46 times, the rate of falls, respectively, compared with people without those risk factors. These variables have been identified as fall risk factors in previous research.5
Both walking aid use and objectively measured balance ability were associated with a higher rate of falls. Walking aid use is likely to be a marker of impaired balance but as both variables are retained in the model, the variables appear to be capturing different aspects of balance/mobility performance. Access to walking aids and an individual’s perceived risk of falling may also be important in the use of walking aids. Inappropriate use of walking aids may also contribute to falls.
While most previous studies examined the relationship between risk factors and fall characteristics, very few studies focused on the risk factors associated with rate of falls in older Chinese people. The study conducted by Morris et al 14 indicated that age, back pain, medical conditions, visual impairment and mobility were associated with multiple falls in older Australians. The findings from our study are similar in that we found the presence of a medical condition, visual impairment and impaired balance/walking aid use (a marker of poor mobility) were associated with additional falls. In contrast, however, age was significant in univariate analysis but no longer significant in the multivariate model.
Shi et al 8 analysed the association between risk factors and recurrent falls in older Chinese and found low monthly family income, visual impairment, impaired physical ability, impaired static balance and fear of falling to be associated with recurrent falls. The results of our study are consistent with Shi’s study, as we found visual impairment and impaired static balance to be associated with rate of falls.
The strengths of this study are the large sample size and high response rate and the random selection of participants, increasing the applicability of the findings to older Chinese people more broadly. There were however some limitations. As this is a multistage cluster sampling study, the SE may be underestimated. Since all participants were recruited in Shenzhen, the findings of our study could be generalised to Shenzhen only. We do not know the characteristics of those people who did not participate in the survey. The reporting of falls relied on retrospective recall, which is known to be of limited accuracy.15 Due to recall bias, those participants who had fallen in the past 12 months may have been more likely to report that they had poor health and poor body sense perception. Hence, the association between falls and subjective body sense perception and self-reported health may have been overestimated. Moreover, some potential risk factors were not included in the study such as physical activity level and vitamin D intake, as well as environmental hazards, which may have been associated with falls. Furthermore, people with cognitive impairment were not included in the study, and people who were immobile were also excluded from participation, so the results may not apply to those groups.
Among older community dwellers in Shenzhen, China, the presence of a medical condition, living alone, visual impairment, using a walking aid and impaired balance were significantly associated with a higher rate of falls in the past year. These results may assist with the development of appropriate strategies to reduce the rate of falls among older Chinese people.
What is already known on the subject
Risk factors for falls in older Chinese people are reported to be similar to risk factors in older Caucasians.
Older Chinese people appear to have lower fall rates than their Caucasian peers.
As the reasons for this are not yet understood, further exploration of risk or protective factors of rate of falls in Chinese people is warranted.
What this study adds
This study investigated the risk factors associated with rate of falls in older Chinese people by negative binomial regression model. Important risk factors were presence of a medical condition, living alone, visual impairment, walking aid use and impaired balance.
This study used an objectively measured balance variable that is included in the fall prevention guideline published by the Ministry of Health of the People’s Republic of China.
This study is a part of a national survey study with a relatively large sample size.
The project was organized by the National Centre for Chronic and Non-communicable Disease Control and Prevention, China CDC and conducted by the Shenzhen Centre for Chronic Disease Control. The interviews and data collection were undertaken by trained healthworkers from the community healthservice. Authors Sherrington and Tiedemann are supported by Research Fellowships from the National Health and Medical Research Council of Australia. I would like to thank Dr.Duan Leilei, Dr.Er Yuliang and Dr.Deng Xiao from Division of Injury Prevention, National Centrefor Chronic and Non-communicable Disease Control and Prevention, China CDC for the outstanding leadership and technical guidance in this program, and intellectual discussion regarding fall related issues in China. I would also like to thank Dr.Feng Nongping, Dr.Zeng Nianbin and Dr.Li Heng for the excellent field work coordination and support.
HZ and KP contributed equally.
Contributors KP, HZ and JP participated in field investigation and acquired the data. KP, AT and CS analysed data and wrote the first draft. All authors reviewed the manuscript and approved the submission.
Funding This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Ethics approval Ethics approval was obtained from the Shenzhen Centre for Chronic Disease Control. (reference number 201309).
Provenance and peer review Commissioned; externally peer reviewed.
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.