Abstract
Spine fractures represent a significant public health concern, particularly among aging populations. They are among the most common osteoporotic fractures and are associated with substantial morbidity, mortality, and healthcare costs. Trends in the prevalence of spine fractures have not been well described in subgroups of demographic characteristics, and understanding trends in the prevalence of spine fractures and risk factors for spine fractures is critical to planning public health approaches to prevent and manage the disease in US adults. In this study, we evaluated the time trends in the prevalence of spine fractures and their associated risk factors in the US adult population from 1999 to 2018. The survey study comprised a series of cross-sectional analyses using nationally representative data from 10 cycles of the National Health and Nutrition Examination Survey (NHANES) spanning from 1999 to 2000 to 2017–2018. All study samples were weighted to represent the civilian resident US population. Spine fractures were defined by self-reporting of spine fracture diagnosis by a physician that they had spine fractures. Trends in the prevalence of spine fractures in subgroups of demographic characteristics for spine fractures were estimated using logistic regression analysis. The age-standardized prevalence (95% CI) of spine fractures increased from 2.54% (1.94-3.14%) in the 1999–2002 cycles to 5.04% (3.62-6.46%) in the 2015–2018 cycles (p < 0.05 for trend). The prevalence of spine fractures among Americans aged 50 and above was 3.24%, which was similar between those under 65 and those aged 65 and above (< 65 3.20% vs. aged ≥ 65 3.31%, p = 0.74). However, the prevalence of spine fractures in males is higher than that in females (males 3.81% vs. females 2.76%, p = 0.005). Univariate analysis showed that age, sex, race (mainly non-Hispanic white), marital status, osteoporosis, smoking, alcohol consumption, hypertension, and diabetes were risk factors for spine fractures. In multivariate analysis, sex (male), race (mainly non-Hispanic white), osteoporosis, smoking, and hypertension were independently associated with spine fractures. Based on NHANES surveys of US adults from 1999 to 2000 to 2017–2018, the prevalence of spine fractures showed an overall increasing trend, with variations across sociodemographic subgroups. In addition, age, sex, race (mainly non-Hispanic white), osteoporosis, smoking, alcohol consumption, hypertension, and diabetes may be closely linked to the occurrence of spine fractures.
Supplementary Information
The online version contains supplementary material available at 10.1038/s41598-025-94871-9.
Keywords: Trend, Prevalence, Spine fractures, Risk factor, NHANES(National health and nutrition examination survey)
Subject terms: Health care, Risk factors
Introduction
Spine injuries account for a relevant proportion of trauma patients, with thoracolumbar fractures being one of the most severe injuries in the human skeleton1–3. Spine fractures are among the most common osteoporotic fractures and are significantly associated with an increased risk of subsequent vertebral and non-vertebral fractures, including hip fractures4–6. As the global population ages, the prevalence of osteoporosis continues to rise, making vertebral fractures a significant public health concern affecting tens of millions worldwide7. Impaired quality of life is often reported in patients with thoracolumbar fractures, independent of treatment modality but related to the severity of the injury8,9. Although many patients with vertebral fractures can be treated as outpatients, a considerable number require ongoing care, which is significantly associated with healthcare cost3,10–12.
According to the Global Burden of Disease Study 2019, the prevalence of spine fractures increased by 37.7% between 1990 and 201913. It is predicted that the cost of treating fractures will increase to $50 billion by 2040, significantly increasing the burden on national healthcare systems14. In Finland, the population-based hospitalization rate for spine fractures increased by 57% between 1998 and 2017, from 57 cases per 100,000 person-years to 89 cases per 100,000 person-years15. In Germany, the number of documented vertebral fractures increased by 45.6% between 2009 and 2019, reaching 150.7 per 100,000 inhabitants3. Because spine fractures cause serious personal and societal problems, understanding how the prevalence of spine fractures has changed over time and identifying their risk factors is critical to public health.
However, in the United States, there is relatively little evidence regarding whether the prevalence of spine fractures is increasing. Furthermore, globally, there is little information in the literature on the trend of spine fracture by demographic subgroups (e.g., age, sex, income, race, education) until now. Meanwhile, a theoretical basis for effective spine fracture prevention can be provided by describing national trends in the prevalence of spine fractures and risk factors for spine fractures in the general adult population. The National Health and Nutrition Examination Survey (NHANES) is an ongoing, multi-stage, nationally representative cross-sectional survey targeting the noninstitutionalized civilian population of the United States. Thus, this study aimed to comprehensive analyze the prevalence, trends, and risk factors of spine fractures among the general adult population in the United States from 1999 to 2018, based on NHANES.
Methods
Data source and study population
To prevent and comprehensively address emerging US public health concerns and supply objective data on the health status of the US civilian population, the American Centers for Disease Control and Prevention (CDC) established and has been continuously updating and refining the NHANES database16. Since 1999, NHANES has published data every two years as a research cycle in which the same data can be combined, allowing for better analysis of disease trends and generating estimates with greater precision and less sampling error. All NHANES data have been approved by the CDC and are available for public use globally, furthermore each participant has signed an informed consent form.
In this study, we incorporated data from 10 NHANES cycles from 1999 to 2000 to 2017–2018, with a total of 39,348 participants reporting spine fractures. Fifty-six subjects in the study were excluded due to they refused to answer or did not remember answers. Participants were 20 years of age or older. The prevalence and trend temporal of spine fractures were presented from 1999 to 2018 for adults. All trends were investigated in terms of sex, age, race, income status, marital status, and education level.
Assessment of spine fractures
In the NHANES questionnaire data, within the household questionnaire in the program Osteoporosis, all subjects were asked, “Has a doctor ever told you that you had a spine fracture?“. Answers include yes, no, don’t know, refused, and missing. Participants who refused to answer the question and answered don’t know or missing were excluded from the study. Since the NHANES data relies on self-reported diagnoses, undiagnosed vertebral fractures may not be included in the statistics. This limitation may affect the estimation of the prevalence.
Assessment of sociodemographic characteristics and risk factors
Basic information regarding participants’ sex, age, race, income status, marital status, and education level was gathered via the household questionnaire in NHANES. Race was categorized as Mexican American, other Hispanic, non-Hispanic white (white without Hispanic ancestry), non-Hispanic black (black without Hispanic ancestry), and other races (non-Hispanic Asian, Native Alaskans, American Indians, Native Hawaiians, or other Pacific Islanders, and multiracial populations). The study adopted the poverty income ratio (PIR) as an indicator of socioeconomic status, with PIR less than 1.00 denoted those below the official poverty threshold and PIR greater than 1.00 representing those above the poverty threshold17,18. The education level was classified into three groups, including less than high school diploma, high school graduation or equivalent, and more than high school diploma. Based on their marital status, the subjects were divided into married, never married, and other marital statuses (widowed, divorced, separated, and living with a partner).
Risk factors for spine fractures have been previously reported in the literature in the past, mainly including age, prior clinical fractures, bone mineral density (BMD), and alcohol consumption19,20. Regarding osteoporotic fractures, the risk factors are primarily low BMD, osteopenia observed on plain radiographs, glucocorticoid use, increasing age, low calcium intake, and excessive alcohol consumption21,22. The study analyzed the risk factors for spine fracture from basic information in demographic data, smoking, hypertension, diabetes, osteoporosis, serum calcium, alcohol consumption, lumbar bone density, and thoracic bone density. All data were collected by trained interviewers through the Computer-Assisted Personal Interviewing (CAPI) system, with hypertension, serum calcium, alcohol consumption, and bone density carried out at a mobile examination center (MEC), while the remaining variables were obtained through inquiries during home examinations. Smoking status was evaluated in adults only, classified subjects as smokers if they had smoked at least 100 cigarettes in their lifetime. Hypertension was defined as those with a systolic blood pressure of 140 mmHg or more or a diastolic blood pressure of 90 mmHg or more measured on each of the three uses of a mercury sphygmomanometer who had been told by their doctor or other health professional that they had hypertension and who were taking hypertension medications. Individuals who had been told by a doctor that they had diabetes or were now taking diabetic pills were classified as having diabetes. Osteoporosis was ever being told by a doctor that they had osteoporosis, thin or brittle bones. For adults 18 and older, drinking alcohol at least 12 times a year was defined as alcohol consumption. The descriptions of lumbar bone density and thoracic bone density have been presented in other literature23.
Statistical analysis
Sample weights, clustering, and stratification were incorporated into all complex sampling designs of the study, with 4-year weights used in 1999–2002 and 2-year weights used in other years according to the NHANES analysis guidelines to ensure nationally representative estimates24. Descriptive statistics were employed to characterize the study participants (continuous variables were expressed using weighted means ± SE, and categorical variables were described using numbers and percentages). The Chi-square test was employed to evaluate the association between spine fractures (over the entire period) and demographic characteristics (age, sex, race, income status, education, and marital status). To enhance the reliability and accuracy of the weighted estimates and to take into account the relatively low prevalence of spine fractures, two adjacent cycles were merged. The weighted prevalence rates of spine fractures and their 95% CI were calculated from 1999 to 2000 to 2017–2018 for the adult population in the US. During the period 1999–2018, linear logistic regression using survey years as a continuous variable predictive factor was used to evaluate time trends in the prevalence of spine fractures in both the overall and sociodemographic subgroups. In addition, univariate and multivariate analysis were performed for smoking, hypertension, diabetes, osteoporosis, blood calcium, alcohol consumption, lumbar bone density, and thoracic bone density to screen for risk factors for spine fractures in adults by generalized linear model. The covariate with p < 0.05 in univariate analysis was included in the multivariate logistic regression analysis and the relationship between spine fracture and associated risk factors was expressed as an odds ratio (OR) with a 95% CI. All statistical analyses of NHANES data were conducted using the R version 4.3.0 survey command, and a p < 0.05 was considered a statistically significant difference.
Results
The study analyzed data from 39,292 subjects aged 20 years and older to assess demographic characteristics. The weighted mean age was 48.78 (0.18) years, with 52.22% identifying as female and 47.78% as male. Regarding race, 7.45% identified as Mexican American, 11.05% as non-Hispanic black, 70.40% as non-Hispanic white, 5.09% as other Hispanic, and 6.01% as other races (eTable1 in Supplement). Age was categorized into two groups: under 65 years (n = 27894; before retirement) and greater than 65 years (n = 11398; after retirement). Since the questionnaire survey on spine fractures in the NHANES database from 2017 to 2018 only included people aged over 50 years old, we adjusted the age to include only those over 50 when calculating the prevalence of spine fractures. The demographic characteristics of all subjects across the study cycles are presented in Table 1. The proportion of female subjects increased from 51.81% in the 2007–2010 cycle to 53.55% in the 2015–2018 cycle. The proportion of subjects aged 65 and older increased from 16.23 to 42.61% over time during the study period 1999–2002 to 2015–2018. Among participants, the proportion of other races increased from 4.59 to 9.26%. The percentage of those with less than high school education decreased from 21.72 to 11.98%, while the percentage of those with more than high school education increased from 52.46 to 58.78%. In addition, the percentage of married increased from 54.40 to 60.87%.
Table 1.
Sociodemographic characteristics of U.S. Adults 20 years of age and older via the NHANES survey cycles from 1999 to 2018.
| Variable | No. of participants (Weighted %)a | |||||
|---|---|---|---|---|---|---|
| 1999–2002 | 2003–2006 | 2007–2010 | 2011–2014 | 2015–2018 | P-trend | |
| sex | 0.37 | |||||
| Female | 5477(52.33) | 5206(51.98) | 6232(51.81) | 2005(52.76) | 1547(53.55) | |
| Male | 4793(47.67) | 4799(48.02) | 5910(48.19) | 1806(47.24) | 1517(46.45) | |
| ageb | < 0.0001 | |||||
| <65 | 7427(83.77) | 7327(82.91) | 9066(82.76) | 2508(70.80) | 1566(57.39) | |
| ≥65 | 2843(16.23) | 2678(17.09) | 3076(17.24) | 1303(29.20) | 1498(42.61) | |
| racec | 0.13 | |||||
| Mexican American | 2392(6.61) | 1986(7.86) | 2168(8.48) | 494(7.01) | 348(5.58) | |
| Non-Hispanic Black | 1921(10.78) | 2114(11.36) | 2349(11.34) | 782(10.70) | 726(10.24) | |
| Non-Hispanic White | 5059(71.03) | 5174(71.79) | 5733(68.65) | 1683(70.79) | 1177(69.29) | |
| Other Hispanic | 546(6.98) | 306(3.50) | 1297(4.94) | 337(4.42) | 292(5.63) | |
| Other Races | 352(4.59) | 425(5.49) | 595(6.60) | 515(7.09) | 521(9.26) | |
| Education | < 0.0001 | |||||
| Less than high school | 3556(21.72) | 2876(17.98) | 3629(19.74) | 912(15.89) | 701(11.98) | |
| High school or equivalent | 2360(25.54) | 2447(25.87) | 2886(24.12) | 864(21.88) | 758(29.05) | |
| More than high school | 4317(52.46) | 4659(56.00) | 5606(56.02) | 2031(62.14) | 1595(58.78) | |
| Poverty income ratiod | 0.03 | |||||
| <1 | 1711(14.39) | 1668(11.76) | 2364(14.44) | 691(12.51) | 436(10.15) | |
| ≥1 | 7383(85.61) | 7769(88.24) | 8593(85.56) | 2807(87.49) | 2180(89.85) | |
| Maritale | < 0.0001 | |||||
| Married | 5482(54.40) | 5369(57.04) | 6297(56.14) | 2206(63.18) | 1659(60.87) | |
| Never married | 1497(16.58) | 1639(16.62) | 2089(18.12) | 334( 7.72) | 220( 5.23) | |
| Other marital status | 2743(22.82) | 2987(26.19) | 3748(25.68) | 1269(29.06) | 1180(33.82) | |
a All percentage data were adjusted for NHANES survey weights.
b In the NHAANES database, subjects ranged in age from 20 to 150 years in 1999 to 2010, 40 to 150 years in 2013 to 2014, and 50 to 150 years in 2017 to 2018.
c Other races include non-Hispanic Asian, Native Alaskans, American Indians, Native Hawaiians, or other Pacific Islanders, and multiracial populations.
d Poverty income ratio is the ratio of a household’s self-reported income to the appropriate poverty threshold for the household set by the U.S. Census Bureau, with higher numbers indicating higher incomes.
e Other marital status includes widowed, divorced, separated, and living with a partner.
Prevalence of spine fractures
The age-adjusted prevalence of spine fractures was 3.24%, with a prevalence of 3.81% in men and 2.76% in women (eTable2 in Supplement). For individuals under 65 years old, the prevalence of spine fractures was 3.20% and 3.31% for those aged 65 and above. However, the prevalence of spine fractures in males under the age of 65 was 4.08%, while in males aged 65 and above, it was 3.40% (eTable3 in Supplement). Regarding race, the highest prevalence of spine fractures in other races was 4.06%, while the lowest prevalence in non-Hispanic black was 1.11%. The prevalence of spine fractures was the highest at 3.59%for those with more than high school diploma and the lowest at 2.77% for those with less than high school diploma. The prevalence of spine fractures with the PIR less than 1 was 3.99%, and the opposite was 3.14%.
Trends in prevalence of spine fractures
From 1999 to 2018, the prevalence of spine fractures varied by year in both the total population and demographic subgroups, with an increasing trend observed in these groups (Table 2; Fig. 1). The estimated age-standardized prevalence of spine fractures increased significantly from 2.54% (95% CI, 1.94-3.14%) in 1999–2002 to 5.04% (95% CI, 3.62-6.46%) in 2015–2018, which is almost a twofold increase. There trends in prevalence varied across subgroups with different demographic characteristics. The prevalence of spine fractures was higher in men than in women. Among those younger than 65 years, the prevalence of spine fractures increased significantly from 1.94% (95% CI, 1.22-2.67%) in 1999–2002 to 5.04% (95% CI, 3.17-6.92%) in 2015–2018. When stratified by race, the prevalence of spine fractures among non-Hispanic whites increased significantly from 2.58% (95% CI, 1.91-3.26%) in 1999–2002 to 5.47% (95% CI, 3.72-7.23%) in 2015–2018, while the trend remained stable among other races. A significant increase in the estimated prevalence of spine fractures was observed among those with more than high school education, PIR greater than or equal to 1, and those who were married (all P for trend < 0.05).
Table 2.
Trends in weighted prevalence of spine fractures among NHANES study participants from 1999 to 2018.
| Characters | Prevalence in spine fractures NHANES cycle yearsa | |||||
|---|---|---|---|---|---|---|
| 1999–2002(n = 4965) | 2003–2006(n = 4714) | 2007–2010(n = 6097) | 2011–2014(n = 2777) | 2015–2018(n = 3064) | P-trendb | |
| Total | 2.54(1.94,3.14) | 3.15(2.41,3.89) | 3.24(2.56,3.93) | 2.43(1.41,3.45) | 5.04(3.62,6.46) | 0.01 |
| sex | ||||||
| Female | 2.22(1.56,2.87) | 2.28(1.58,2.98) | 2.92(2.04,3.79) | 1.85(0.97,2.73) | 4.72(2.78,6.65) | 0.02 |
| Male | 2.92(1.84,3.99) | 4.17(3.07,5.26) | 3.62(2.75,4.50) | 3.09(1.11,5.07) | 5.41(3.18,7.64) | 0.14 |
| Age | ||||||
| < 65 | 1.94(1.22,2.67) | 3.09(2.28,3.89) | 3.40(2.39,4.41) | 2.74(0.91,4.57) | 5.04(3.17,6.92) | 0.00 |
| ≥ 65 | 3.30(2.42,4.18) | 3.24(2.31,4.16) | 3.02(2.28,3.75) | 2.00(0.98,3.02) | 5.03(3.39,6.68) | 0.28 |
| Race | ||||||
| Mexican American | 2.84(1.83,3.86) | 2.03(0.81,3.24) | 3.25(1.93,4.56) | 2.07(0.34,3.80) | 2.38(0.74,4.02) | 0.74 |
| Non-hispanic Black | 0.89(0.40,1.38) | 1.59(0.66,2.53) | 0.84(0.36,1.31) | 0.48(0.00,1.01) | 1.69(0.74,2.64) | 0.75 |
| Non-hispanic White | 2.58(1.91,3.26) | 3.39(2.49,4.28) | 3.66(2.83,4.48) | 2.79(1.42,4.15) | 5.47(3.72,7.23) | 0.01 |
| Other hispanic | 3.44(1.24,5.64) | 2.24(0.00,6.60) | 1.82(0.55,3.08) | 3.61(0.06,7.16) | 3.89(0.64,7.15) | 0.77 |
| Other races | 3.72(0.42,7.01) | 3.70(1.91,5.49) | 2.61(0.95,4.27) | 0.88(0.08,1.68) | 7.78(0.04,15.52) | 0.33 |
| Education | ||||||
| Less than high school | 2.69(1.19,4.20) | 2.54(1.23,3.85) | 3.33(2.22,4.45) | 1.67(0.76,2.59) | 3.32(1.66,4.98) | 0.85 |
| High school or equivalent | 2.97(1.75,4.19) | 2.17(1.44,2.90) | 3.09(2.11,4.07) | 2.70(0.65,4.75) | 3.75(1.94,5.57) | 0.33 |
| More than highs chool | 2.23(1.48,2.97) | 3.94(2.90,4.97) | 3.28(2.30,4.26) | 2.54(1.52,3.56) | 6.04(3.74,8.34) | 0.01 |
| Poverty income ratio | ||||||
| < 1 | 3.63(1.74,5.53) | 4.67(2.13,7.21) | 4.15(2.06,6.24) | 3.36(2.14,4.58) | 4.18(0.49,7.88) | 0.99 |
| ≥ 1 | 2.47(1.81,3.14) | 2.92(2.05,3.79) | 2.95(2.17,3.73) | 2.50(1.27,3.72) | 5.28(3.77,6.79) | 0.00 |
| Marital | ||||||
| Married | 2.42(1.69,3.15) | 2.79(2.05,3.53) | 3.28(2.40,4.16) | 2.69(1.55,3.83) | 4.43(2.88,5.98) | 0.02 |
| Never married | 0.85(0.00,2.12) | 3.23(0.00,6.85) | 2.04(0.63,3.44) | 2.75(1.30,4.21) | 1.35(0.00,2.76) | 0.94 |
| Other marital status | 2.94(1.92,3.97) | 3.84(2.59,5.08) | 3.41(2.36,4.46) | 1.87(0.65,3.09) | 6.71(3.63,9.80) | 0.07 |
a Because all subjects in the NHANES database regarding spine fracture questionnaires from 2017 to 2018 were older than 50 years of age, we included trend analyses only for subjects older than 50 years of age from 1999 to mid-2018.
b Time trends from the range of listed years to 1999–2018 were assessed using linear logistic regression.
Fig. 1.
Trends in prevalence of spine fracture among US adults. Graph shows the prevalence of spine fractures in different sex and age groups in the United States. Data presented incorporated sample weights and adjusted for clusters and strata of the complex sample design of NHANES 1999–2018 to represent the US population. After adjusting for age, the subjects were all over 50 years old.
Risk factors for spine fractures
Table 3 presents univariate analysis of demographic characteristics associated with spine fractures among US adults. Associated risk factors included older age, adult males, race in non-Hispanic white, osteoporosis, smoking, alcohol consumption, hypertension, and diabetes.
Table 3.
Univariate and multivariate analysis of associated risk factors in spine fractures included in NHANES from 1999 to 2018.
| Character | Univariate analysis | Multivariate analysis | ||
|---|---|---|---|---|
| OR[95% CI] | p-value | OR[95% CI] | p-value | |
| age | ||||
| <65 | 1 ref | 1 ref | ||
| ≥65 | 1.44(1.20,1.71) | < 0.0001 | 0.86(0.67,1.11) | 0.25 |
| sex | ||||
| Female | 1 ref | 1 ref | ||
| Male | 1.47(1.20,1.79) | < 0.001 | 1.76(1.40,2.22) | < 0.0001 |
| Race | ||||
| Non-hispanic white | 1 ref | 1 ref | ||
| Non-hispanic black | 0.29(0.22,0.38) | < 0.0001 | 0.31(0.22,0.43) | < 0.0001 |
| Mexican American | 0.51(0.41,0.64) | < 0.0001 | 0.59(0.46,0.76) | < 0.0001 |
| Other Hispanic | 0.55(0.35,0.86) | 0.01 | 0.58(0.36,0.93) | 0.02 |
| Other Races | 0.90(0.56,1.44) | 0.66 | 0.91(0.57,1.47) | 0.71 |
| Education | ||||
| High school or equivalent | 1 ref | - | ||
| Less than high school | 0.84(0.66,1.08) | 0.17 | - | |
| More than high school | 0.98(0.81,1.19) | 0.88 | - | |
| Poverty income ratio | ||||
| <1 | 1 ref | - | ||
| ≥1 | 0.85(0.68,1.06) | 0.15 | - | |
| Marital | ||||
| Other marital status | 1 ref | 1 ref | ||
| Married | 0.69(0.57,0.84) | < 0.001 | 0.71(0.58,0.88) | 0.002 |
| Never married | 0.44(0.31,0.62) | < 0.0001 | 0.57(0.40,0.81) | 0.002 |
| Osteoporosis | ||||
| No | 1 ref | 1 ref | ||
| Yes | 2.97(2.34,3.78) | < 0.0001 | 3.37(2.45,4.64) | < 0.0001 |
| Smoke | ||||
| No | 1 ref | 1 ref | ||
| Yes | 1.64(1.37,1.96) | < 0.0001 | 1.34(1.09,1.65) | 0.01 |
| Serum calcium | 0.71(0.25,2.01) | 0.51 | - | |
| Drinking alcohol | ||||
| No | 1 ref | 1 ref | ||
| Yes | 1.48(1.17,1.86) | 0.001 | 1.24(0.95,1.60) | 0.11 |
| Lumbar spine BMD | 0.49(0.21,1.15) | 0.10 | - | |
| Thoracic spine BMD | 0.75(0.32,1.76) | 0.51 | - | |
| Hypertension | ||||
| No | 1 ref | 1 ref | ||
| Yes | 1.54(1.30,1.83) | < 0.0001 | 1.28(1.02,1.60) | 0.03 |
| Diabetes | ||||
| No | 1 ref | 1 ref | ||
| Yes | 1.44(1.09,1.89) | 0.01 | 1.21(0.88,1.66) | 0.25 |
To identify independent factors associated with spine fractures, we conducted multivariate logistic regression analyses of risk factors associated with spine fractures in univariate analysis. The prevalence of spine fractures was significant higher in the male population compared with the female population (OR, 1.76; 95% CI, 1.40–2.22). Compared with Mexican American, non-Hispanic white was more likely (OR, 1.69; 95% CI, 1.32–2.16) and non-Hispanic black was less likely (OR, 0.52; 95% CI, 0.36–0.74). Among the remaining factors, married (OR, 0.71; 95% CI, 0.58–0.88), never married (OR, 0.57; 95% CI, 0.40–0.81), osteoporosis (OR, 3.37; 95% CI, 2.45–4.64), smoking (OR, 1.34; 95% CI, 1.09–1.65), hypertension (OR, 1.28; 95% CI, 1.02–1.60) were associated with the prevalence of spine fractures.
Discussion
In this study, we analyzed the time trends in the prevalence of spine fractures among US adults from 1999 to 2018 in the NHANES database. This nationally representative survey revealed that the overall prevalence of spine fractures has increased over the past 20 years. In 2015–2018, approximately 5.04% of adults experienced spine fractures, with a significantly higher prevalence than in 1999–2014. We also assessed the development trend in spine fractures across various demographic characteristics, including sex, age, race, education level, PIR, and marital status. In addition, we analyzed risk factors for spine fractures using both univariate and multivariate logistic regression.
Previous studies had shown that in 2019, the estimated number of new vertebral fractures globally was 8.58 million, representing a 37.7% increase from 199013. The number of new spine fractures was 4.90 million (3.87–6.28) in males, compared to 3.68 million (2.71–5.04) in females. Similarly, the age-standardized spine fracture prevalence rate in 2019 was 109 cases per 100,000 population, with the age-standardized rate being 125 cases per 100,000 population for males, which was higher than 92 cases per 100,000 population for females13, which was similar to our findings. In our study, the prevalence of spine fractures was higher in males than females from 1999 to 2018, but the trend of increase was more significant in the subgroup of females. The prevalence of spine fractures in 2015–2018 was nearly double that of 1999–2002. Previous studies have reported an increasing prevalence of hospitalization for cervical, thoracic, and lumbar fractures in Finland during 1998-201715. These studies also documented a gradual increase in the prevalence of fall-related cervical spine trauma among the elderly from 1970 to 2000, following by a rapid rise until 201126–28. Similar trends have been observed in the United States, Iceland, and Sweden28–30.
Between 2003 and 2018, we observed that the prevalence of spine fractures was higher among male aged 50 to 65 years compared to those aged 65 years and older, which was not found in the female population, but there was a significant time trend in the prevalence of spine fractures in the female population aged 50 to 65 years (eTable 4 in Supplement). This may be due to the fact that males are more occupational hazards in more fracture-risk jobs than females before retirement31–34. Moreover, males are more likely to engage in high-risk jobs and physically demanding activities such as construction, working at heights, and operating machinery, which may contribute to the higher prevalence of spine fractures in males compared to females. Additionally, the prevalence of spine fractures is increasing among non-Hispanic white, more than high school diploma, PIR greater than or equal to 1, and those married. However, little is known about the effects of race, education, PIR, and marital status on spine fractures. These factors may be influenced to racial beliefs, the economic development of the region, and the lifestyle habits of the race. Therefore, it may be necessary to investigate how changes in demographic characteristics impact the prevalence of spine fractures in specific community or region. During the study period, the proportion of people aged over 65 increased significantly, from 16.23% in 1999–2002 to 42.61% in 2015–2018. This change may reflect the aging trend of the US population and may also have an impact on the increase in the prevalence of spine fractures. As people age, the risk of osteoporosis and fractures increases significantly, which may be an important factor leading to the rise in the prevalence of spine fractures.
This finding of this study underscore the urgent need to recognize the overall prevalence of spine fractures, particularly within certain demographic subgroups. Additionally, some studies have reported a significant increase in the prevalence of hospitalization and surgery for spine fractures, especially cervical fractures, during the period 1998-201715. At the same time, the most substantial increase was observed in surgical treatment for patients over 60 years old, with a 400% increase in surgery15. Therefore, there is a compelling need for targeted interventions and greater attention to subgroup classification and demographic subgroups of spine fractures.
Prediction risk factors for spine fractures may inform healthy lifestyle choices and preventive strategies for orthopedic surgeons and potentially high-risk populations. In our study, a univariate analysis of the spine fracture population in the NHANES database revealed that age, sex, race (mainly non-Hispanic white), marital status, osteoporosis, smoking, alcohol consumption, hypertension, and diabetes may be significant risk factors for spine fractures. Notably, age and sex are critical risk factors, especially among men and individuals aged 65 years and older. Non-Hispanic white older adults exhibit a higher prevalence of spine fractures compared to other races included in the NHANES database, which may be attributed to factors such as specific racial genetics, dietary characteristics, and lifestyle habits. According to WHO diagnostic criteria, 54% of postmenopausal white US women have osteopenia, while another 30% have osteoporosis. Individuals with osteoporosis are more likely to.
experience vertebral fractures compared to those with osteopenia or normal BMD. Our findings align with those of the National Osteoporosis Risk Assessment study, highlighting the importance of early screening and appropriate management for patients with osteopenia and those with normal BMD who have risk factors35. Previous research has shown that a history of smoking is associated with poorer bone health in old age, with continue smokers experiencing more rapid bone loss36. Similarly, alcohol consumption may contribute to spine fractures in older adults by affecting BMD. This study further demonstrates the importance of smoking cessation and limiting alcohol consumption in preventing spine fractures in older adults. Although our analysis shows an association between alcohol consumption, smoking, and the prevalence of spine fractures, the biological mechanism underlying this association remains unclear. The definitions of alcohol consumption and smoking may be too broad, causing the results to be affected by confounding factors. Alcohol consumption was defined as having consumed alcohol at least 12 times in the past year, and this criterion may cover a wide range of situations from light monthly drinking to alcoholism. Similarly, smoking was defined as having smoked at least 100 cigarettes in one’s lifetime, and this criterion may not accurately reflect the impact of long-term smoking. Future studies could consider more detailed data on alcohol consumption and smoking to further explore their effects on spine fractures. Additionally, in our study, hypertension and diabetes emerged as risk factors for spine fracture, which may be related to complications caused by hypertension and diabetes (leading to dizziness), and further analysis is needed.
Limitations
This study also has several limitations. Firstly, some variables were based on self-reporting, such as history and factors of spine fracture and osteoporosis, which are susceptible to randomization and systematic errors, as well as reporting and recall biases, which cannot be entirely avoided. NHANES data is only based on self-reported diagnoses and may not fully reflect the actual prevalence of vertebral fractures. Future studies could consider combining imaging examinations or other diagnostic tools to more accurately assess the prevalence of vertebral fractures. However, self-reporting history and factors of fractures have been used in epidemiologic studies, and measurement error is unlikely to affect the results of studies with long-term trends16,37. Secondly, due to the limited variables included in the NHANES database, certain covariates related to spine fracture, such as previous clinical fracture and spine fracture, parental history of spine fracture, and specific fracture causes, were not included in this study. Thirdly, the study did not analyze the specific distribution of spine fractures. Fourthly, this study contains possible unrecognized confounding factors. For example, with the advancement of medical technology, the increased use of X-ray and DXA examinations may lead to more vertebral fractures being diagnosed. In addition, socioeconomic status (such as the PIR) may affect an individual’s willingness to seek medical care, thus influencing the prevalence of self-reported vertebral fractures. Future studies could consider further exploring the impact of these confounding factors on the prevalence of vertebral fractures. Finally, although two adjacent NHANES cycles were merged, there may not be sufficient ability to detect trends over time, especially in some subgroups with a limited sample size.
Conclusions
By analyzing data from US adults in NHANES from 1999 to 2018, our study demonstrated an overall increasing trend in the prevalence of spine fractures. This trend varied across different demographic characteristics subgroups. Additionally, older age, sex (male), race (mainly non-Hispanic white), osteoporosis, smoking, drinking, hypertension, and diabetes were risk factors for spine fracture.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Author contributions
N.-H.G.: Data curation, Conceptualization, Formal analysis, Methodology, Supervision, Visualization, Writing - original draft. H.-Y.:Conceptualization, Supervision, Writing − review & editing. G.-Y.K. and H.-G.K.: Conceptualization; Writing − review & editing. Y.-G.Q.: Conceptualization, Writing − review & editing. G.-W.S.: Conceptualization, Supervision, Methodology, Project administration, Writing − review & editing.
Data availability
The datasets generated and/or analyzed during the current study are available in the NHANES database, https://wwwn.cdc.gov/nchs/nhanes/.
Declarations
Competing interests
The authors declare no competing interests.
Ethics approval and consent to participate
All NHANES data have been approved by the Centers for Disease Control and Prevention and are available for public use worldwide, and each participant has signed an informed consent form.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Katsuura, Y., Osborn, J. M. & Cason, G. W. The epidemiology of thoracolumbar trauma: A meta-analysis. J. Orthop.13 (4), 383–388 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Smits, A. J. et al. Incidence of traumatic spinal fractures in the Netherlands: analysis of a nationwide database. Spine45 (23), 1639–1648 (2020). [DOI] [PubMed] [Google Scholar]
- 3.Lang, S. et al. Increased incidence of vertebral fractures in German adults from 2009 to 2019 and the analysis of secondary diagnoses, treatment, costs, and in-hospital mortality. Sci. Rep-Uk. 13 (1), 6984 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Wang, Y. et al. Elderly males with or without existing osteoporotic vertebral fracture have much lower future vertebral fracture risk than elderly females: the MrOS (Hong Kong) year-4 follow-up spine radiograph study. Osteoporos. Int.30 (12), 2505–2514 (2019). [DOI] [PubMed] [Google Scholar]
- 5.Ferrar, L. et al. Association between incident and baseline vertebral fractures in European women: vertebral fracture assessment in the osteoporosis and ultrasound study (OPUS). Osteoporos. Int.23 (1), 59–65 (2012). [DOI] [PubMed] [Google Scholar]
- 6.Mccloskey, E. V. et al. Vertebral fracture assessment (VFA) with a densitometer predicts future fractures in elderly women unselected for osteoporosis. J. Bone Min. Res.23 (10), 1561–1568 (2008). [DOI] [PubMed] [Google Scholar]
- 7.Salari, N. et al. Global prevalence of osteoporosis among the world older adults: a comprehensive systematic review and meta-analysis. J. Orthop. Surg. Res.16 (1), 669 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Cooper, C., O’Neill, T., Silman, A. & European Vertebral Osteoporosis Study Group. The epidemiology of vertebral fractures. Bone14, S89–S97 (1993). [DOI] [PubMed] [Google Scholar]
- 9.Kitzen, J. et al. Surgeon reported treatment choices for AO type B and C thoracolumbar fractures without neurological deficits: an expert survey. Injury55 (3), 111389 (2024). [DOI] [PubMed] [Google Scholar]
- 10.Briem, D. et al. Pain regulation and health-related quality of life after thoracolumbar fractures of the spine. Eur. Spine J.16 (11), 1925–1933 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Schnake, K. J., Stavridis, S. I. & Kandziora, F. Five-year clinical and radiological results of combined anteroposterior stabilization of thoracolumbar fractures. J. Neurosurg-Spine. 20 (5), 497–504 (2014). [DOI] [PubMed] [Google Scholar]
- 12.Lang, S. et al. Radiological and mid- to long-term patient-reported outcome after stabilization of traumatic thoraco-lumbar spinal fractures using an expandable vertebral body replacement implant. Bmc Musculoskel Dis.22 (1), 744 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.GBD 2019 Fracture Collaborators. Global, regional, and National burden of bone fractures in 204 countries and territories, 1990–2019: a systematic analysis from the global burden of disease study 2019. Lancet Healthy Longev.2 (9), e580–e592 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lane, N. E. Epidemiology, etiology, and diagnosis of osteoporosis. Am. J. Obstet. Gynecol.194 (2 Suppl), S3–11 (2006). [DOI] [PubMed] [Google Scholar]
- 15.Ponkilainen, V. T. et al. Incidence of spine fracture hospitalization and surgery in Finland in 1998–2017. Spine45 (7), 459–464 (2020). [DOI] [PubMed] [Google Scholar]
- 16.Zhang, Y. W. et al. Prevalence, characteristics, and associated risk factors of the elderly with hip fractures: A Cross-Sectional analysis of NHANES 2005–2010. Clin. Interv Aging. 16, 177–185 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sabanayagam, C. & Shankar, A. Income is a stronger predictor of mortality than education in a National sample of US adults. J. Health Popul. Nutr.30 (1), 82–86 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Singh, G. K. et al. Social determinants of health in the united States: addressing major health inequality trends for the Nation, 1935–2016. Int. J. Mch Aids. 6 (2), 139–164 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Fujiwara, S. et al. Vertebral fracture status and the world health organization risk factors for predicting osteoporotic fracture risk in Japan. Bone49 (3), 520–525 (2011). [DOI] [PubMed] [Google Scholar]
- 20.Samelson, E. J. et al. Incidence and risk factors for vertebral fracture in women and men: 25-year follow-up results from the population-based Framingham study. J. Bone Min. Res.21 (8), 1207–1214 (2006). [DOI] [PubMed] [Google Scholar]
- 21.Campion, J. M. & Maricic, M. J. Osteoporosis in men. Am. Fam Physician. 67 (7), 1521–1526 (2003). [PubMed] [Google Scholar]
- 22.Olszynski, W. P. et al. Osteoporosis in men: epidemiology, diagnosis, prevention, and treatment. Clin. Ther.26 (1), 15–28 (2004). [DOI] [PubMed] [Google Scholar]
- 23.Cosman, F. et al. Spine fracture prevalence in a nationally representative sample of US women and men aged ≥ 40 years: results from the National health and nutrition examination survey (NHANES) 2013–2014. Osteoporos. Int.28 (6), 1857–1866 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Curtin, L. R. et al. The National health and nutrition examination survey: sample design, 1999–2006. Vital Health Stat. 2 (155)1–39. (2012). [PubMed]
- 25.Korhonen, N. et al. Rapid increase in fall-induced cervical spine injuries among older Finnish adults between 1970 and 2011. Age Ageing. 43 (4), 567–571 (2014). [DOI] [PubMed] [Google Scholar]
- 26.Kannus, P. et al. Alarming rise in the number and incidence of fall-induced cervical spine injuries among older adults. J. Gerontol. a-Biol. 62 (2), 180–183 (2007). [DOI] [PubMed] [Google Scholar]
- 27.Kannus, P. et al. Continuously increasing number and incidence of fall-induced, fracture-associated, spinal cord injuries in elderly persons. Arch. Intern. Med.160 (14), 2145–2149 (2000). [DOI] [PubMed] [Google Scholar]
- 28.Devivo, M. J. Epidemiology of traumatic spinal cord injury: trends and future implications. Spinal Cord. 50 (5), 365–372 (2012). [DOI] [PubMed] [Google Scholar]
- 29.Knutsdottir, S. et al. Epidemiology of traumatic spinal cord injuries in Iceland from 1975 to 2009. Spinal Cord. 50 (2), 123–126 (2012). [DOI] [PubMed] [Google Scholar]
- 30.Brolin, K. Neck injuries among the elderly in Sweden. Inj Control Saf. Promot. 10 (3), 155–164 (2003). [DOI] [PubMed] [Google Scholar]
- 31.GBD 2019 Diseases and Injuries Collaborators. Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the global burden of disease study 2019. Lancet396 (10258), 1204–1222 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.JamesSL et al. Estimating global injuries morbidity and mortality: methods and data used in the global burden of disease 2017 study. Injury Prev.26 (Supp 1), i125–i153 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kica, J. & Rosenman, K. D. Surveillance for work-related skull fractures in Michigan. J. Saf. Res.51, 49–56 (2014). [DOI] [PubMed] [Google Scholar]
- 34.Barreto, S. M. et al. Predictors of first nonfatal occupational injury following employment in a Brazilian steelworks. Scand. J. Work. Environ. Health.26 (6), 523–528 (2000). [DOI] [PubMed] [Google Scholar]
- 35.Siris, E. S. et al. Identification and fracture outcomes of undiagnosed low bone mineral density in postmenopausal women: results from the National osteoporosis risk assessment. Jama-J Am. Med. Assoc.286 (22), 2815–2822 (2001). [DOI] [PubMed] [Google Scholar]
- 36.Marques, E. A. et al. Cigarette smoking and hip volumetric bone mineral density and cortical volume loss in older adults: the AGES-Reykjavik study. Bone108, 186–192 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Ye, J., Li, Q. & Nie, J. Prevalence, characteristics, and associated risk factors of wrist fractures in Americans above 50: the Cross-Sectional NHANES study. Front. Endocrinol. (Lausanne) 13 800129. (2022). [DOI] [PMC free article] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated and/or analyzed during the current study are available in the NHANES database, https://wwwn.cdc.gov/nchs/nhanes/.