Trends in femoral neck fracture-related mortality in the United States: A nationwide retrospective study from 1999 to 2023

Abstract

Femur neck fracture is a significant cause of mortality and morbidity in the U.S. This study examines the long-term temporal trends in femur neck fracture-related mortality among individuals aged ≥ 65 in the United States from 1999 to 2023. A retrospective analysis of death records listing International Statistical Classification of Diseases, Tenth Revision code S72.0 (femoral neck fracture) as the underlying cause of death was conducted using the Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research database from 1999 to 2023. We analyzed crude and age-adjusted mortality rates (AAMRs) per 100,000 population. Joinpoint regression analysis was used to calculate annual percent changes and evaluate trends across various demographics and geographics in the U.S. Between 1999 and 2023, femur neck fractures caused 323,534 deaths among U.S. adults aged ≥ 65 years. The AAMR peaked at 39.6 in 2002 and declined to 22.4 in 2023. A significant drop occurred from 2002 to 2018 (annual percent change ‐3.00 P < .05). Males (30.81), Non-Hispanic (NH) Whites (34.35) and rural areas (41.47) showed the highest AAMR among their demographic and geographic counterparts. Further, among states, Montana (58.9) and Minnesota (58.8) had the highest cumulative AAMRs. Despite overall declines in mortality among individuals with femur neck fractures, persistent disparities by sex, race, and geography highlight the need for targeted prevention strategies.

1. Introduction

Femoral neck fracture is a type of hip fracture that represents a significant mortality burden in the world, particularly among the aging population.^[1] In the United States, hip fractures occur in over 250,000 individuals, and the number is likely to increase to 500,000 by the year 2050.^[2] There are also studies that record global estimations of 6.26 million by 2025, which is alarming because femoral neck fracture causes disability, morbidity, and mortality, especially in older populations.^[3] It is also a big financial burden since it requires extended periods of hospitalization and surgery.^[4] According to the Medicare calculations, the treatment of proximal femoral fracture was the 13th most expensive diagnosis in 2011.^[5] Moreover, hip fractures cost the U.S. between $10.3 billion and $15.2 billion every year^[6] and cost around $65,000–$80,000 in nursing care for patients.^[7] These are some staggering statistics that point towards a considerable financial strain on society.

With the aging population, it is likely that the elderly will encounter more healthcare issues. It is worth noting that 13.3% elderly individuals died within the 30-day window period, and 36% of aged males and females succumbed to death within a year of hip fracture, indicating the high risks of death among older populations.^[8,9] Recent studies showed that factors such as reduced bone density and the loss of balance might be the cause of frequent femoral fracture mortality in elderly individuals.^[10] The low-energy trauma among older adults is the most widespread cause, which can lead to an interrupted or impaired blood flow, causing avascular necrosis as a secondary complication.^[1,11] Also, fractures lead to immobility, which consequently causes many complications, including deep vein thrombosis, pulmonary embolism, pneumonia, pressure ulcers, and muscle atrophy.^[12–15] Chronic morbidity can further play a part in frustrating, isolating, and depending on oneself, which can also complicate the recovery process.^[16] Thus, preventive measures are vital to safeguard the physical and mental health of the people who are at risk.

Although the death rate resulting from hip fractures has been reduced over the years through improved management of hip fractures through improved preoperative care and postoperative rehabilitation,^[17] there is still a need to identify vulnerable populations so that they can have timely interventions. In this regard, we have reviewed long-term mortality trends related to the femoral neck fracture using the Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research database from 1999 to 2023 to identify demographic and regional mortality trends among the elderly population.

2. Methods

2.1. Study setting and population

Our analysis included death certificates of older adults (aged ≥ 65 years) that reported the fracture of the femoral neck as the underlying cause of death between 1999 and 2023 in the U.S.^[18] Death records were traced using the International Statistical Classification of Diseases, Tenth Revision code of S72.0 (fracture of neck of femur). Since the database is publicly available and contains de-identified datasets that are collected by a governed source, it obviates the need for approval from the Institutional Review Board.

2.2. Data stratification and variables

We extracted data on deaths associated with femoral neck fracture, including population sizes, year of death, demographic data, and information regarding geographics. Demographics comprised of gender, race (Non-Hispanic [NH] White, NH African Americans, Hispanic or Latino, NH American Indian or Alaskan Native, and NH Asian), and age groups (65–74, 74–85, 85+). Geographic data included place of death (medical facility, home, hospice, or nursing home/long-term care facility), urbanization level (metro and nonmetro areas), census region, and states. These variables were used to stratify the dataset to calculate subgroup-specific mortality rates and identify disparities across sex, race, age, and geography. The urbanization levels followed the National Center for Health Statistics Urban–Rural Classification Scheme, and geographic regions were categorized according to the U.S. Census Bureau divisions: Northeast, Midwest, South, and West.^[19] These variables were used to stratify the dataset to calculate subgroup-specific mortality rates and identify disparities across sex, race, age, and geography. Additionally, the age threshold of ≥5 years was applied across all the defined stratifications according to the inclusion criteria of older adults.

2.3. Statistical analysis

The trends in mortality were measured by calculating crude and age-adjusted mortality rates (AAMR) per 100,000 of the older adult population. The crude mortality rate was obtained by dividing the number of deaths (per year) by the equivalent population in the U.S. Standardization of AAMRs was done by the direct method with the help of the U.S. 2000 population to make fair comparisons over time across different demographic and geographical subgroups.^[20] For each U.S. state, cumulative AAMRs were computed and were used to compare regional mortality burdens during the study period. Trend analysis was done in the Joinpoint Regression Program (V5.0) to estimate the annual percent change (APC) with a 95% confidence interval that identifies statistically significant changes in the trend segments over time across the entire demographic and geographic subgroups.^[21] Moreover, the maximum number of joint points was set to 4, and the data-driven weighted Bayesian information criterion method was employed to select the model. Further, an increasing or decreasing trend was determined if the slope significantly differed from zero. The 2-tailed t test was used by the program, considering a P-value of ≤.05 statistically significant.

3. Results

3.1. Overall mortality distribution by demographic and geographic variables (1999–2023)

A total of 323,534 femur neck fracture-associated deaths occurred among adults (aged ≥ 65 years) in the U.S. from 1999 to 2023 (Fig. 1). Of these, 63% (n = 205,623) were females and 37% (n = 117,911) were males. By race and ethnicity, NH White adults had the highest number of absolute deaths (93%; n = 301,153), followed by NH African Americans or African Americans (3%; n = 9459), Hispanics (2.8%; n = 8889), NH Asians (0.9%; n = 2925), and NH American Indians (0.3%; n = 1108).

Figure 1.

Flowchart illustrating the selection of the study population from the CDC WONDER database between 1999 and 2023.

In terms of age, 65% of deaths occurred in individuals aged 85 years and older (n = 208,848), followed by 28% in those aged 75 to 84 years (n = 90,891) and 7% in individuals aged 65 to 74 years (n = 23,795). Regarding the place of death, 48% of deaths took place in inpatient medical facilities, 34.5% in nursing homes or long-term care facilities, 10% at home, 3% in outpatient or emergency settings, 4.3% occurred in other locations, and 0.2% were dead on arrival.

By urbanization, 76% of deaths occurred in metropolitan areas (n = 219,572) while 24% occurred in nonmetropolitan areas (n = 68,356). Regionally, the South had the greatest number of deaths (34%; n = 111,135), followed by the Midwest (29%; n = 92,995), West (21%; n = 68,563), and Northeast (16%; n = 50,841) (Table S1 and Figure S1, Supplemental Digital Content, https://links.lww.com/MD/R399).

3.2. Overall age-adjusted mortality trends (1999–2023)

The AAMR for femur neck fracture-related deaths increased slightly from 36.18 per 100,000 (95% confidence interval [CI]: 35.54–36.82) in 1999 to a peak of 39.60 (95% CI: 38.95–40.25) in 2002. From 1999 to 2002, this increase corresponded to an APC of 2.77 (95% CI: 0.15–7.39; P = .039192). A significant decline followed between 2002 and 2018, with AAMRs falling to 23.22 (95% CI: 22.80–23.64) and an APC of ‐3.00 (95% CI: ‐4.00 to ‐2.78; P < .000001). Between 2018 and 2023, the AAMR declined slightly to 22.39 (95% CI: 21.98–22.80), though the change was not statistically significant (APC: ‐1.09; 95% CI: ‐2.42 to 2.32; P = .338332) (Fig. 2; Tables S2 and S3, Supplemental Digital Content, https://links.lww.com/MD/R399).

Figure 2.

Overall, trends in femur neck fracture-related age-adjusted mortality rates per 100,000 among adults aged 65 and above in the United States, 1999 to 2023. APC = annual percentage change, CI = confidence interval. *Indicates that the annual percentage change (APC) is significantly different from zero at α = 0.05.

3.3. Gender stratified

Although females showed a higher number of deaths, males exhibited higher AAMRs. The AAMR among males was 38.96 (95% CI: 37.79–40.12) compared to 34.89 (95% CI: 34.12–35.66) in females. Among males, the AAMR increased to 41.39 in 2002 (95% CI: 40.23–42.55) and then declined significantly to 22.19 in 2018 (95% CI: 21.53–22.85), corresponding to an APC of −3.50 (95% CI: −4.52 to −3.28; P < .000001). A further decrease to 21.09 in 2023 (95% CI: 20.46–21.73) occurred with a nonsignificant APC of −1.54 (95% CI: −2.88 to 1.99; P = .228354).

Among females, the AAMR rose from 34.89 in 1999 to 38.66 in 2002 (95% CI: 37.86–39.46), then declined to 23.83 in 2018 (95% CI: 23.28–24.38) with an APC of −2.72 (95% CI: −4.42 to −2.48; P < .000001). A further nonsignificant decline to 23.19 in 2023 (95% CI: 22.65–23.73) was noted with an APC of −0.83 (95% CI: −2.22 to 2.78; P = .452709) (Fig. 3; Tables S2 and S3, Supplemental Digital Content, https://links.lww.com/MD/R399).

Figure 3.

Trends in femur neck fracture-related age-adjusted mortality rates per 100,000, stratified by sex among adults aged 65 and above in the United States, 1999 to 2023. APC = annual percentage change, CI = confidence interval. *Indicates that the annual percentage change (APC) is significantly different from zero at α = 0.05.

3.4. Race/ethnicity stratified

NH White adults had the highest mean AAMR of 34.35 (95% CI: 33.74–34.97) among racial/ethnic groups. Their AAMR decreased from 39.38 in 2002 to 26.58 in 2018 (APC: −2.55; 95% CI: −4.31 to −2.32; P < .000001), and further to 25.54 in 2023 (APC: −0.73; 95% CI: −2.09 to 2.69; P = .496301). NH African Americans experienced a decline from 16.15 in 1999 to 7.63 in 2023 (APC: −3.50; 95% CI: −4.07 to −2.92; P < .000001).

Among Hispanics, the AAMR declined from 17.04 in 2010 to 9.82 in 2023, with a significant APC of −6.69 (95% CI: −7.86 to −3.51; P = .024795). NH American Indian showed the highest proportional decline, with AAMR decreasing from 34.19 in 1999 to 15.42 in 2023 (APC: −6.88; 95% CI: −11.02 to −4.95; P < .000001). NH Asian showed overall declines but without statistical significance in recent years (2015–2023 APC: 0.09; 95% CI: −7.88 to 10.72; P = .74865) (Fig. 4; Tables S3 and S4, Supplemental Digital Content, https://links.lww.com/MD/R399).

Figure 4.

Trends in femur neck fractures-related age-adjusted mortality rates per 100,000, stratified by race and ethnicity among adults aged 65 and above in the United States, 1999 to 2023. APC = annual percentage change, CI = confidence interval. *Indicates that the annual percentage change (APC) is significantly different from zero at α = 0.05.

3.5. Age group stratified

Older adults consistently experienced higher mortality rates, with AAMRs declining from 178.33 in 1999 to 117.85 in 2023, among those aged ≥ 85 years. The most notable decrease was observed between 2002 and 2019 (APC: −2.72; 95% CI: −4.18 to −2.50; P < .000001). Adults aged 75 to 84 years had an AAMR of 32.84 in 1999, which declined to 17.91 in 2023. Significant declines were observed during 2008–2018 (APC: −4.02; 95% CI: −5.03 to −3.73; P < .000001) and again from 2021 to 2023 (APC: −5.88; 95% CI: −7.98 to −3.35; P < .000001). In the 65- to 74-year age group, the AAMR declined from 5.15 in 1999 to 3.02 in 2023, with a relatively stable trend from 2013 to 2023 (APC: −1.04; 95% CI: −1.86 to 0.30; P = .085183) (Tables S3 and S5; Figure S2, Supplemental Digital Content, https://links.lww.com/MD/R399).

3.6. Geographic trends

Nonmetropolitan regions exhibited higher AAMRs than metropolitan regions. In metropolitan areas, the AAMR declined from 33.92 in 1999 to 22.39 in 2023, with a statistically significant drop from 2002 to 2020 (APC: −2.86; 95% CI: −3.14 to −2.65; P < .000001). In nonmetropolitan areas, AAMRs declined from 45.54 in 1999 to 30.70 in 2020. A significant decline occurred between 2001 and 2015 (APC: −2.44; 95% CI: −2.62 to −2.24; P < .000001) and from 2015 to 2018 (APC: −7.19; 95% CI: −8.32 to − 5.52; P < .000001), followed by a small, short-lived increase from 2018 to 2020 (APC: 5.09; 95% CI: 2.31 to 7.44; P = .0004). Among the 4 U.S. census regions, the Midwest reported the highest AAMR at 37.72 (95% CI: 36.51–38.94), followed by the West at 31.10 (95% CI: 29.84–32.16), South at 29.89 (95% CI: 29.01–30.78), and Northeast at 23.09 (95% CI: 22.09–24.10). A lot of variation was noticed across U.S. states. The highest AAMRs were found in Montana (58.89; 95% CI: 56.29–61.49), Minnesota (58.82; 95% CI: 57.67–59.97), and Colorado (58.30; 95% CI: 56.93–59.67). In contrast, the lowest AAMRs were recorded in Hawaii (14.46; 95% CI: 13.39–15.54), New York (17.96; 95% CI: 17.63–18.29), and the District of Columbia (19.32; 95% CI: 17.22–21.43). These AAMRs are based on data available from 1999 to 2020 (Fig. 5; Tables S3, S6–S8; Figure S3, Supplemental Digital Content, https://links.lww.com/MD/R399).

Figure 5.

Trends in femur neck fracture-related age-adjusted mortality rates per 100,000, stratified by urbanization among adults aged 65 and above in the United States, 1999 to 2020. APC = annual percentage change, CI = confidence interval. *Indicates that the annual percentage change (APC) is significantly different from zero at α = 0.05.

4. Discussion

This retrospective analysis of U.S. mortality data found an overall decline in deaths due to femoral neck fractures from 1999 to 2023. AAMRs fell from the peak of 39.02 in 2002 to almost half their value in 2023 (22.39). This is indicative of a national improvement in not only the management of femoral neck fractures but also associated diseases that can cause them.

We observed a strong decrease in AAMR in the overall population, which can be explained by a number of factors. Studies have shown that the number of femoral neck fractures among the U.S. population is currently on the decrease, which can be attributed to the fact that the clinical management of osteoporosis has improved. Newer formulations such as teriparatide and bisphosphonates, and traditional vitamin D and calcium supplementation can also be used to preserve the bone density and reduce the rate of bone resorption. This lessens the chances of having femoral neck fractures.^[22,23] This can be further explained by the decline in smoking and alcohol consumption and rising levels of physical activity, which enhances bone mass. This lessens the chances of having femoral neck fractures.^[24–26] Moreover, the developments in the surgical procedures and fixation devices, including the femoral neck system, have enhanced patient outcomes and decreased the mortality due to surgical and postsurgical complications. Physical rehabilitation, including weight-bearing and mobility exercises, has also led to reduced healing times and faster recovery.^[27,28]

Our results also showed that deaths in women (63%) were double that of men (37%). This can be due to the female sex being at a higher risk of fracturing the neck of the femur due to a more rapid decline in bone mineral density with increasing age. Osteoporosis, a leading cause of femoral neck fractures, is also more likely to develop in women in their postmenopausal years as their estrogen levels decline.^[29] Lifestyle differences, such as women being less inclined towards participation in weight-bearing exercises that build bone density and muscle mass, also add to the higher risk of mortality in women.^[30] Conversely, men (30.81) had slightly higher AAMR than women (30.47). Studies have found that men have a higher rate of mortality 1 year after total/hemiarthroplasty.^[31] A possible cause is a reluctance to seek medical care after accidents or falls. This leads to higher rates of delayed presentation, which also increases mortality.^[32,33] Additionally, men have more complicated recovery courses following surgeries due to lower levels of independence in daily activities.^[34] Reduced functional status is an independent contributor to femoral neck fracture survival and may explain higher mortality in men.

In addition, we found that the older population, especially those who were aged above 85 years, had greater AAMRs (155.13). A comparable finding was made by Moon et al, who reported that high age levels, cognitive impairment, and comorbid conditions were associated with increased postoperative complications and mortality.^[35] It is of value mentioning that elderly people are at increased risk of falls, which is the most prevalent cause of femoral neck fracture.^[36] It is explained by degeneration of bone and muscle tissue with increasing age, impaired visual functions, deterioration of cerebellar activity, and taking several medications (polypharmacy). The elderly population also has a slow recovery after the fixation of the femoral neck fracture because they have low levels of mobility and high levels of comorbidities, which include end-stage renal disease. Also, the U.S. aging population, which is attributed to the increased medical care and health care inequalities, could be the reason behind the higher mortality in the elderly.^[37] This is because a demographic shift results in a higher number of at-risk individuals with chronic conditions and increased frailty.^[38]

Racial and ethnic disparities were also evident in our analysis, with the highest number of deaths occurring in NH Whites (93%). This trend is associated with a higher incidence of femoral neck fractures in these populations, likely due to lower bone mineral density and accelerated bone loss. A longer average life expectancy in NH whites also contributes to the higher mortality, as increasing age is an independent risk factor for femoral neck fractures.^[39,40] NH American Indians and Alaska Natives experienced the fastest decline, largely due to targeted campaigns by the U.S. Public Health Service, as well as raising awareness of infrastructural modifications to reduce the risk of falls and increased health-seeking behaviors.^[41] Interestingly, no significant decline was found in the NH Asian population, which may be a result of a lower incidence of femoral neck fractures, explained by higher composite hip strength indices, which offer greater stability as opposed to bone mineral density.^[42]

Geographical analysis showed that the AAMR of residents in the rural area (41.47) was higher than the AAMR of urban areas (29.33). These disparities are caused by the differences in accessibility to healthcare, such as surgical expertise and multidisciplinary care plans.^[43] In addition, Hsia et al discovered that patients in the rural locations had a greater time period between injury and surgery since they required more time to travel; a factor associated with adverse outcome.^[44] Rural residents are also not well equipped with postoperative physiotherapy and rehabilitation services, and so, have to face more cases of delayed recovery and complications.^[45] Further, the financial stability of the patients in the rural areas is also more fluctuating than in the urban areas, with more patients being underinsured or uninsured.^[46] This minimizes the chances of patients not receiving medical treatment and postponement in arthroplasty. All of these contribute to the increased patient mortality rates. Among regions, the Midwest (37.72) was observed to have the highest AAMRs related to femoral neck fractures compared to other regions in the census, which can be explained by the fact that a higher fraction of the aged population and that most of the cities are classified as rural.^[47,48] Reduced average incomes and socioeconomic status of patients are also associated with delayed presentation to medical centers, and this reduces survival.

State-wise analysis revealed that outliers, such as Montana, Minnesota, and Colorado, had AAMRs that were approximately 3 times higher than those of New York, Hawaii, and the District of Columbia. These differences, similar to other geographical variabilities, are a combination of population demographics, healthcare facilities, educational levels, and socioeconomic standings of susceptible individuals.

4.1. Limitations

Our study has some limitations that should be noted in order to put the findings into perspective. Firstly, the application of the Joinpoint Regression Model adds the constraint of the user, which is the maximum or minimum number of joint points; this influences the trend detection. This paper was based on the accuracy of the death certificates that are accessible in the Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research database, which might be influenced by varying coding practices and standards across regions that have changed within the past 25 years. In addition, the completeness of the provided death certificates can be questioned, possibly due to the potential omission or misclassification of the underlying cause of death. Further clinical specifics, such as thromboembolisms or sepsis, might be regarded as the cause of death, which actually occurred as a consequence of femoral neck fracture. This may result in inaccuracies in the data available. Additionally, the database lacks information on the surgical status and financial information, like insurance status, which may result in bias during comparison between subgroups. Besides, when working with AAMRs, confounding factors, like modifications in management guidelines, cannot be identified and compared. These changes should be factored into future research to provide a more credible interpretation of the mortality patterns.

5. Conclusion

Our analysis of demographic and regional trends in femoral neck fracture mortality rates during the period of 1999 to 2023 has revealed significant disparities, which underline the necessity to work on the development of special prevention and intervention programs to address the differences based on sex, race, and geography. Future comprehensive studies that account for the genetic and socioeconomic issues that could affect mortality rates in various strata could help to achieve this task and further mitigate this mortality burden.

Author contributions

Conceptualization: Muhammad Shaheer Bin Faheem.

Data curation: Muhammad Usama Saeed, Nafila Zeeshan.

Formal analysis: Muhammad Shaheer Bin Faheem, Muhammad Usama Saeed.

Investigation: Nafila Zeeshan, Huda Ahmed.

Methodology: Nafila Zeeshan, Huda Ahmed, Syed Tawassul Hassan, Muhammad Idrees Khan.

Project administration: Muhammad Shaheer Bin Faheem, Syed Tawassul Hassan, Muhammad Idrees Khan.

Resources: Faheem Feroze.

Software: Muhammad Usama Saeed.

Supervision: Muhammad Shaheer Bin Faheem, Faheem Feroze.

Validation: Muhammad Shaheer Bin Faheem, Nafila Zeeshan, Faheem Feroze.

Visualization: Muhammad Shaheer Bin Faheem, Muhammad Usama Saeed, Huda Ahmed.

Writing – original draft: Muhammad Shaheer Bin Faheem, Muhammad Usama Saeed, Huda Ahmed.

Writing – review & editing: Muhammad Shaheer Bin Faheem, Faheem Feroze, Syed Tawassul Hassan, Muhammad Idrees Khan.

Supplementary Material

medi-105-e47754-s001.docx ^{(317.9KB, docx)}