Current study

Low-energy fracture was defined as a fracture that was caused by slip, trip or fall from standing height. Fractures caused by high-energy injuries (traffic trauma, fall from height, crushing injury, sharp trauma and others) were excluded. A total of 154099 women and men aged 50 years and older participated in this study and 937 participants had at least one fracture of any site caused by either low-or high-energy in 2014. 215 patients reporting 217 cases of low-energy fractures of femur and tibia (fibula) were identified as case group. The remaining 153162 participants without any fracture were defined as control group. The detailed flow chart was presented in Fig 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. The flow chart for inclusion of the participants.

https://doi.org/10.1371/journal.pone.0209203.g001

Variables of interest included age, height, weight and body mass index (BMI), residency category (rural or urban), regions, latitude zone (20°-29.9°, 30°-39.9° or 40°-49.9°), ethnic origins, occupation, educational level, frequency of drinking intake (tea, carbonate beverages, coffee), smoking status, alcohol consumption, dwelling place (ground floor, > 2^nd floor with or without elevator), average sleep duration per day, history of past fracture, living situation (alone or with others), supplementation of calcium or Vitamin D or both for women and men; and age of menopause and number of births extra for women.

BMI was grouped on basis of the criteria suited to Chinese populations: underweight, <18.5 kg/m²; normal, 18.5–23.9 kg/m²; overweight, 24–27.9 kg/m²; obesity, > = 28 kg/m² [5, 13]. Current smoking and alcohol consumption was defined as positive (yes), if participants smoked >1 time/week or drank > 1time/month during the 2014 year or the past year before fracture occurrence. Similarly within the time window, the average intake frequency of certain type of drinking was reported by participants; and supplementation of calcium or Vitamin D or both was defined as positive (yes) if participants acknowledged they take these medications for at least 1 month; otherwise, as negative (no).

Statistical analysis

Incidence rates of low-energy lower extremity fractures were estimated by the number of fractures divided by the total number of overall population and population in each subgroup by age (5-year interval), as well as by demographic factors such as ethnic origin, geographical region, education, occupation, and residency category (rural or urban), stratified by gender. Differences in incidence rate between categories of nominal variables, such as ethnic origin, regions, occupation, and residency category, were tested using the Chi-square test. Age and education as ordered categorical variables and trends in incidence rates by them were tested by univariate logistic regression models. We also assessed incidence rates of low-energy femur or tibia (fibula) fractures based on sites (proximal, shaft and distal), in men or women or combined.

Two separate gender-based multiple logistic regression models were constructed to explore the independent risk factors associated with low-energy lower extremity fractures in women and men. Given the limited cases of fracture for each site (proximal, shaft and distal), we did not perform the subgroup analysis based on fracture site. The above-mentioned variables that were reported in previous literature to have positive or negative effects on bone mass density, or be associated with falls or low-energy fracture were all entered into the multivariate model. A stepwise backward elimination approach was used to exclude confounding covariates from the final model. Covariates were retained in the final model if the p value was ≤ 0.10. Odd ratio (OR) and 95% confidence interval (95% CI) were used to indicate the correlation magnitude between variables and the risk of low-energy fracture. The significance threshold was set at P < 0.05. The Hosmer–Lemeshow test was used to examine the goodness-of-ft of the final model, and a p value >0.05 indicated an acceptable fitness. All analyses were performed using SPSS 19.0 software (IBM Corporation, Armonk, NY)

Multivariate analysis

After adjustment for confounding variables, alcohol consumption, sleep duration<7h/d and history of past fracture were identified as significant risk factors associated with low-energy lower extremity fractures in men (Table 3). The Hosmer–Lemeshow test demonstrated the adequate fitness (X² = 7.105, p = 0.130).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Results of multivariate logistic regression of risk factors for low-energy lower extremity fractures in men and women.

https://doi.org/10.1371/journal.pone.0209203.t003

In women, advanced age (80+ years), alcohol consumption, sleep duration <7h/d, and history of past fracture were identified to be associated with increased risk of lower extremity fractures. Living in west region (compared to eastern region) and weight of 50–59.9kg (compared to <40kg) were identified as protective factors for low-energy fracture. Supplementation of calcium or Vitamin D or both seemed to have positive effect on prevention of low-energy lower extremity fractures, but the significance did not approach to the statistical level (p = 0.074). The detailed statistical results were presented in Table 3. The Hosmer–Lemeshow test demonstrated the adequate fitness (X² = 6.087, p = 0.638).

Comparison of incidences of low-energy fracture with studies

Knowledge of population-based incidence of low-energy fractures is fundamental to develop public health programs, but national epidemiologic surveys of such were very scarce. Currently, most researches selected a single hospital, a certain region or a subgroup of population as study object to obtain knowledge of epidemiologic situation [14–17]. In addition, geographical location, socioeconomic development, and individual lifestyles differ from each other among countries and or regions in the same country [18, 19]. Therefore, it might be explainable that the incidence rates of fractures in literature were greatly varied. In UK, Kaye et al [20] used the data from General Practice Research Database to report the incidence was 472 per 100000 person years for total lower extremity fracture, and 330/100000 person-years for femur and tibia (fibula) fractures, in individuals aged 50 years. In Finland, incidence of hip fracture was reported as 170 to 210 and 364 to 450 per 100000 person-years in men and women aged 50+ years [21, 22] and the incidence of low-energy ankle fracture was 114 and 174 per 100000 person-years for elderly men and women [23]. In USA, the adjusted annual incidence of hip fracture was 129 and 224 per 100000 person-years for men and women above 45 years [24]. Author also observed the declined trend of hip fracture incidence during the past 10 years before 2006, and suggested this result partially benefit from the active bone health intervention [24]. Compared to those western developed countries, the figures reported in this study was substantially lower, as 1/5to 1/3 of them. Even in Asian, the incidence of lower extremity fracture reported in this study was much lower than that of neighboring countries, as Japna [25] and South Korea [26]. The underlying mechanism might be related to relatively low prevalence of osteoporosis [27, 28]. Despite, low-energy lower extremity fracture caused undoubtedly constitutes a major public health issue in modern China, due to the huge base of mid-aged and elderly individuals (http://www.stats.gov.cn/tjsj/).

Risk factors associated with the low-energy fractures

Alcohol consumption, insufficient daily sleep duration (<7h/d) and history of past fracture were identified as significant risk factors for low-energy lower extremity fracture, both in men and women. Clark et al [29] suggested excess consumption of alcohol increase the risk of fractures through metabolic effects and alcohol-related falls. As was previously reported [30], men consuming >8 units and women consuming >6 units of alcohol on the heaviest drinking day in the past week had the 1.65 to 2.07-time increased risk of lifetime fractures. Stone et al [31] and Holmberg [32] reported the adverse effects of insufficient sleep on the risk of fracture in men or women and suggested frequent fall caused by insufficient sleep was the direct reason for fracture occurrence. History of past fracture as a risk factor for low-energy fracture was well-established in literature, and accelerated bone loss and the increased tendency to fall after the first fracture were suggested to contribute to the secondary fracture in a near future [33, 34]. Therefore, awareness on the importance of sleep and alcohol consumption on fragility fracture should be improved, and health policies that focus on decreasing alcohol consumption and encouraging individuals to improve their sleep quality and duration should be considered to reduce and prevent the low-energy fractures.

In women, another modifiable factor was the weight, and in this study weight of 50 to 59.9kg was shown to have significantly positive effect on the lower extremity fracture, compared to the weight of <40kg. And this result was consistent with that of a previous meta-analysis [35], which showed being underweight increased the risk of lower extremity fractures, while being overweight increased the risk of upper arm fractures, in elderly women. Therefore, maintaining a healthy bodyweight with a normal BMI is of clear significance on prevention of low-energy fractures, for both upper and lower extremity fractures. Other several non-modifiable risk factors have also been identified to be associated with lower extremity fracture in women, including advanced age (80+ years) and geographic position. And the latter this might be related to intense pace of life and more frequent exposure to the hazardous lifestyles for individuals living in east region. However, this result this result should be treated cautiously and dialectically. On the hand, individuals in west region had a relatively shortened life span (72.3 years), (http://bbs1.people.com.cn/post/2/0/0/166947095.html), and it was a possibility that death came prior to the incident fracture, especially hip fracture. On the other hand, western region like Xinjiang province had a highest amount of solar radiation annually, which objectively facilitated the synthesis of vitamin D and calcium absorption to benefit bone health, especially in elderly individuals [36–38]. In addition, inhibition of alcohol due to religious reason among Muslim population in western part of China (Xinjiang area) and economy of that region being still centered on agriculture (higher physical activities) might be also the contributing factors [39].

Strengths and limitations

The main strengths of the current study included the stratified multistage cluster random sampling method used for recruiting subjects, largest sample size, face-to-face interview for data collection and adjustment for numerous important covariates. “Double” identification of fracture case via patients' self-reports and clinical or radiographic data increased the accuracy and precision of diagnosis.

Despite, there are some potential limitations that should be mentioned. Firstly, there was some recall bias due to its nature of retrospective design. Secondly, there remain residual confounding effects, because data on clinical chronic disease status, medication use and other socioeconomic factors that may have affected BMD or propensity to fall could not be available for adjustment in the multivariate analysis model. Thirdly, the study could not capture information about fracture cases in which the patients died. Overall, the incidence rate of low-energy lower extremity fracture was underestimated.