Deaths in the United States in which both pulmonary embolism and a fracture were documented on the death certificate declined substantially between 1999 and 2020, according to a population-based analysis of national mortality records published in BMC Cardiovascular Disorders. Researchers identified 22,214 eligible deaths among people aged 25 years and older and calculated a pooled age-adjusted mortality rate of 0.475 per 100,000 person-years of the general US population, standardized to the 2000 US standard population. Joinpoint regression estimated an average annual percent change of −3.164 percent over the study period, a statistically significant decline with a P value below 0.001. The authors stress that these figures describe population-level mortality patterns of deaths listing both conditions, not the risk of dying from pulmonary embolism among individuals who sustain fractures.
The study drew on the Multiple Cause of Death database available through the CDC WONDER platform. Because this registry contains no clinical S-series traumatic fracture codes and provides no denominator of persons with fractures, the team adopted a validated hierarchical coding strategy using M-series fracture disease codes as the definitive diagnostic basis. Fracture confirmation relied on the ICD-10 codes M80 and M84 to verify definite fracture events, while external cause codes W00–W19 and X59 were used only to attribute traumatic injury etiology for subtype classification rather than to define fractures independently. The researchers also constructed and validated this hierarchical fracture case definition against national mortality registry coding specifications and supplemented it with subtype-stratified sensitivity analyses covering osteoporotic pathological fractures, non-osteoporotic pathological fractures, and traumatic fractures.
Subtype sensitivity analysis revealed distinct mortality burdens across fracture categories. Traumatic fracture-associated pulmonary embolism deaths accounted for the largest proportion, followed by deaths associated with osteoporotic pathological fractures and then other pathological fractures, and temporal trends differed across these subtypes. Pulmonary embolism is described in the study as a critical preventable fatal complication among patients with fractures, yet prior work had lacked population-based, time-trended evidence on mortality trends and disparities for deaths with co-occurring fracture and pulmonary embolism documented on death certificates.
Despite the overall downward trajectory, the analysis uncovered pronounced disparities and several nonsignificant upward late-phase trends across subgroups. Female mortality was higher than male mortality throughout, and women experienced a nonsignificant upward trend after 2017, whereas male mortality declined steadily over the period. Older adults aged 65 years and above accounted for most of the deaths and showed a nonsignificant upward trend after 2016. These late-phase reversals, though not statistically significant, prompted the authors to call for continuous monitoring.
Race and ethnicity shaped the geographic and demographic pattern of these deaths. White populations carried the highest observed population mortality burden and displayed a nonsignificant upward trend after 2018, while Hispanic populations had the lowest rates. Black groups showed slowed declines after 2010. The stratified analyses were performed according to age, sex, race and ethnicity, US census region, urban-rural residence, and fracture etiological subtypes, allowing the team to map how the burden was distributed unevenly across the population.
Geography mattered as well. Mortality was higher in rural areas and in the Midwest, and nonsignificant upward late-phase trends appeared in the Northeast, South, West, and metropolitan regions. Remote Midwestern states bore the highest observed population mortality rates, presenting an uneven national geographic distribution. The contrast between declining national figures and stagnating or rising local trends suggests that the improvement was not shared uniformly across regions or residence types.
The methodological choices carry important caveats for interpretation. Because CDC WONDER mortality data cannot supply a denominator of persons with fractures, the calculated rates reflect deaths among the general population rather than risk among fracture patients. The reliance on M80 and M84 codes, necessitated by the absence of S-series traumatic fracture codes in the registry, means the case definition captures definite fracture events recorded as disease codes, with external cause codes serving only to classify etiology. The authors accordingly frame their conclusions narrowly: the findings reflect population-level mortality patterns of deaths listing both fracture and pulmonary embolism on death certificates and do not represent pulmonary embolism mortality risk among individuals with fractures.
Within those limits, the study identifies groups and regions that may warrant further evaluation of fracture-related venous thromboembolism prevention and surveillance strategies. The persistence of higher mortality among women, older adults, White populations, and residents of rural and Midwestern areas, alongside the nonsignificant upward late-phase trends among females after 2017, older adults after 2016, White populations after 2018, metropolitan residents, and populations across multiple Midwestern and Western regions, highlights heterogeneity in mortality burden across sex, age, race and ethnicity, urban-rural status, states, and fracture etiological subtypes. The authors conclude that continuous monitoring of these late-phase upward trends is warranted.
The work, published open access on 4 November 2026, offers one of the most granular long-term portraits to date of deaths in which fracture and pulmonary embolism co-occur on American death certificates. Its hierarchical coding framework, adapted specifically to national mortality registry specifications, provides a template for future surveillance studies that must work within the constraints of vital statistics data. As the national decline continues to slow in certain subgroups, the findings underscore that aggregate progress can mask widening gaps that only stratified, population-level monitoring can detect.
Subject of Research: Population-level mortality trends and disparities for US deaths documenting both pulmonary embolism and fracture, 1999–2020
Article Title: Temporal trends and population disparities in mortality from pulmonary embolism with co-occurring fracture-related injuries in the United States, 1999–2020
Article References: Wang, G., Hou, T., Zhang, W., Zhuo, T., Su, Y., Zhou, J., Lu, C., & Ding, Q. (2026). Temporal trends and population disparities in mortality from pulmonary embolism with co-occurring fracture-related injuries in the United States, 1999–2020. BMC Cardiovascular Disorders. https://doi.org/10.1186/s12872-026-06971-7
Image Credits: AI Generated
DOI: 10.1186/s12872-026-06971-7
Keywords: pulmonary embolism, fracture, mortality trends, CDC WONDER, health disparities, age-adjusted mortality rate, joinpoint regression, ICD-10 coding, rural health, venous thromboembolism, United States, death certificates
