Predictors of excess mortality after fracture: a population-based cohort study

Opportunistically identifiable vertebral fractures on routine radiological imaging predict mortality: observational cohort study

In men and women with opportunistically identifiable vertebral fractures (VFs) on routine CT scans including the chest and/or abdomen, the risk of death is 51% higher than in those with no VF on the CT scan, and 325% higher than an age- and sex-matched general population cohort.

PURPOSE

There is little knowledge about the risk of death in patients with VFs present on routine radiological imaging. We evaluated the risk of death in men and women aged 50 years or older with opportunistically identifiable VFs on routine CT scans and not treated with osteoporosis medications.

METHODS

Thoracic and lumbar VFs were identified through a blinded, two-step approach on CT scans performed as part of normal clinical care in a Danish hospital in 2010 or later. Subjects with VF were matched on age and sex against those with no VF (1:2-ratio) and a general population cohort (1:3-ratio), respectively, and followed for up to 7 years through the national Danish registers. Subjects treated with an osteoporosis medication in the year prior to baseline were excluded.

RESULTS

Subjects with VF had a significantly higher risk of death during follow-up as compared to subjects with no VF on the CT scan (adjusted hazard ratio [HR] 1.51 [95% confidence interval 1.27-1.79; p < 0.001]) and even more so when compared to the general population cohort (HR 4.25 [3.53-5.12; p < 0.001]). In subjects with versus without VF on the CT scan, the risk was higher in those with moderate or severe VF, in those with no malignancy prior to baseline, and in those with a lower Charlson comorbidity index score.

CONCLUSION

Subjects with VF available for identification on routine CT scans face a substantially increased risk of death. Opportunistic identification and reporting of VF is important to identify these patients to allow intervention if indicated.

Major areas of interest of artificial intelligence research applied to health care administrative data: a scoping review

Introduction: The ongoing collection of large medical data has created conditions for application of artificial intelligence (AI) in research. This scoping review aimed to identify major areas of interest of AI applied to health care administrative data.

Methods: The search was performed in seven databases: Medline, Embase, CINAHL, Web of science, IEEE, ICM digital library, and Compendex. We included articles published between January 2001 and March 2021, that described research with AI applied to medical diagnostics, pharmacotherapy, and health outcomes data. We screened the full text content and used natural language processing to automatically extract health areas of interest, principal AI methods, and names of medications.

Results: Out of 14,864 articles, 343 were included. We determined ten areas of interest, the most common being health diagnostic or treatment outcome prediction (32%); representation of medical data, clinical pathways, and data temporality (i.e., transformation of raw medical data into compact and analysis-friendly format) (22%); and adverse drug effects, drug-drug interactions, and medication cascades (15%). Less attention has been devoted to areas such as health effects of polypharmacy (1%); and reinforcement learning (1%). The most common AI methods were decision trees, cluster analysis, random forests, and support vector machines. Most frequently mentioned medications included insulin, metformin, vitamins, acetaminophen, and heparin.

Conclusions: The scoping review revealed the potential of AI application to health-related studies. However, several areas of interest in pharmacoepidemiology are sparsely reported, and the lack of details in studies related to pharmacotherapy suggests that AI could be used more optimally in pharmacoepidemiologic research.

Cost-Effectiveness of Sequential Denosumab/Zoledronic Acid Compared With Zoledronic Acid Monotherapy for Postmenopausal Osteoporotic Women in China

Objective: The primary purpose of this study was to estimate the cost-effectiveness of sequential denosumab/zoledronic acid versus zoledronic acid monotherapy for postmenopausal osteoporotic women in China.

Methods: We updated and utilized a previously validated Markov microsimulation model to obtain the cost-effectiveness of two strategies for treating postmenopausal osteoporotic women. We compared the incremental cost-effectiveness ratios (ICERs) (US dollars [$] per quality-adjusted life year [QALY]) of sequential denosumab/zoledronic acid (i.e., biannual subcutaneous denosumab for 3 years followed by annual intravenous zoledronic acid for 3 years) with zoledronic acid monotherapy for 3 years in Chinese women with postmenopausal osteoporosis at ages 65, 70, 75, and 80 from the health care payer perspective. Our study’s willingness-to-pay (WTP) threshold was set to three times the value of China’s per capita GDP in 2020 ($31,512).

Results: The ICERs of sequential denosumab/zoledronic acid versus zoledronic acid monotherapy were $59,389/QALY, $23,821/QALY, $22,710/QALY, and $14,027/QALY at age 65, 70, 75, and 80 years, respectively. One-way sensitivity analyses showed that the most impactful parameter was the persistence rate of the medications. If the persistence rate of denosumab or zoledronic acid was increased by 10%, sequential denosumab/zoledronic acid would be cost-effective at age 65. In probabilistic sensitivity analyses, the probabilities of sequential denosumab/zoledronic being cost-effective compared to zoledronic acid monotherapy were approximately 29.3%, 68.7%, 86.1%, and 99.4% for ages 65, 70, 75, and 80 years, respectively, at the WTP threshold of $31,512/QALY.

Conclusion: Among Chinese postmenopausal osteoporosis women over 70 years old, sequential denosumab/zoledronic acid was cost-effective compared with zoledronic acid monotherapy at the pre-determined WTP threshold.

Associations of clinically recognized vs unrecognized vertebral fracture with mortality

PURPOSE

Our purpose was to contrast mortality associated with clinically undiagnosed prevalent vertebral fracture recognized on densitometric vertebral fracture assessment (VFA) and prior clinically diagnosed vertebral fracture.

METHODS

Between 2010 and 2016, 9679 men and women (mean age [SD] 76 [6.9] years, 93% women) with central site bone density T-score ≤ -1.5 had VFA images obtained at the time of bone densitometry. Vertebrae between T4 and L4 inclusive were evaluated for prevalent vertebral fracture on these images. Participants were categorized into three mutually exclusive groups; those with no vertebral fracture on VFA and no prior clinically diagnosed vertebral fracture (n = 7983), those with a definite vertebral fracture on VFA but no vertebral fracture clinically diagnosed before the index VFA date (n = 1376), and those with a prior clinically diagnosed vertebral fracture (n = 320). We ascertained mortality after the index VFA date using Manitoba provincial vital statistics files over a mean 2.8 (SD 1.7) years. We used Cox proportional hazards models to estimate the hazard ratios (HR) and 95% confidence intervals (C.I.) of prevalent vertebral fractures with mortality adjusted for multiple covariates.

RESULTS

Adjusted for age and sex, those with clinically undiagnosed vertebral fracture on VFA had an HR of 1.22 (95% C.I. 1.00 to 1.50) for mortality, and those with a prior clinically diagnosed vertebral fracture had an HR of 1.78 (95% C.I. 1.27 to 2.50) for mortality compared to those with no prevalent vertebral fracture. After further adjustment for comorbidity score, current smoking, and other causes of mortality, these associations were slightly attenuated (HRs 1.18 [95% C.I. 0.95 to 1.45] and of 1.70 [95% C.I. 1.21 to 2.40], respectively). Stratified by elapsed years since diagnosis, clinical vertebral fractures were associated with excess mortality for up to 10 years after their occurrence.

CONCLUSION

Clinically undiagnosed prevalent vertebral fracture detected on densitometric VFA images are weakly associated with subsequent mortality adjusted for age and sex, but not after accounting for other causes of mortality. Clinical vertebral fractures are associated with increased mortality for ten years after their diagnosis, even after accounting for other causes of mortality.

Early Ambulation After Hip Fracture Surgery Is Associated With Decreased 30-Day Mortality

INTRODUCTION

Hip fractures frequently present in complicated patients and are fraught with high morbidity and mortality rates. Postoperatively, delayed ambulation has been identified as a factor associated with increased mortality, although its magnitude has yet to be quantified. Therefore, this article aims to evaluate mortality after hip fracture surgery because it relates to early postoperative ambulation, taking into account preexisting comorbidity burden.

METHODS

This is a retrospective review of patients older than age 65 years who underwent surgical fixation for hip fractures because of a low-energy mechanism. Ambulation during the first 3 postoperative days was recorded along with age and preexisting comorbidity burden (Modified 5-Factor Frailty Index), and 30-day and 1-year mortality statuses were examined. Multivariable logistic regression was used to analyze the association between postoperative ambulation and mortality.

RESULTS

Of 485 patients initially identified, 218 met the inclusion criteria. Overall mortality rates were 6.4% at 30 days and 18.8% at 1 year. Two-thirds of patients ambulated in the first 3 postoperative days versus one-third who did not. Patients who did not ambulate had both significantly increased 30-day mortality (odds ratio [OR] 4.42, P = 0.010, 95% confidence intervals [CIs] 1.42 to 13.75) and 1-year mortality (OR 2.26, P = 0.022, 95% CI 1.12 to 4.53). After multivariable logistic regression accounting for age and comorbidity status, ambulation remained strongly associated with 30-day (OR 3.87, P = 0.024, 95% CI 1.20 to 12.50) but not 1-year mortality (OR 1.66, P = 0.176, 95% CI 0.80 to 3.48). Although neither were significant at 30 days, both increasing age (OR 1.05, P = 0.020, 95% CI 1.01 to 1.10) and Modified 5-Factor Frailty Index (OR 1.62, P = 0.005, 95% CI 1.16 to 2.26) correlated with increased mortality at 1 year.

CONCLUSION

Early ambulation after hip fracture surgery bears a notable, almost four-fold, association with early postoperative mortality independent of age and medical comorbidities. Our results support a growing body of evidence that ambulation is a powerful tool that should continue to be emphasized to optimize mortality in hip fracture patients.

The Role of Comorbidity in Mortality After Hip Fracture: A Nationwide Norwegian Study of 38,126 Women With Hip Fracture Matched to a General-Population Comparison Cohort

Hip fracture patients often have comorbid conditions. We investigated whether the combination of comorbidity and hip fracture could explain the previously observed excess mortality among hip fracture patients as compared with the general population. Using a population-based matched study design with 38,126 Norwegian women who suffered a hip fracture during the period 2009–2015 and the same number of women in a matched comparison cohort, we matched participants on prefracture comorbidity, age, and education. We estimated relative survival and additive and multiplicative comorbidity–hip fracture interactions. An additive comorbidity–hip fracture interaction of 4 or 9 additional deaths per 100 patients, depending on Charlson Comorbidity Index (CCI) score, was observed 1 year after hip fracture. Among women with a CCI score of ≥3, 15 additional deaths per 100 patients were observed; of these, 9 deaths could be attributed to the interaction and 6 to the hip fracture per se. On the relative scale, we observed increasing heterogeneity in survival by comorbidity over time; survival was reduced by 39% after 6 years among patients with a CCI score of ≥3, while among women with no comorbidity, survival was reduced by 17% (hip fracture vs. no hip fracture). In summary, prefracture comorbidity was associated with short-term absolute excess mortality and long-term relative excess mortality.

Comorbidities Only Account for a Small Proportion of Excess Mortality After Fracture: A Record Linkage Study of Individual Fracture Types

Nonhip, nonvertebral (NHNV) fractures constitute the majority of osteoporotic fractures, but few studies have examined the association between these fractures, comorbidity, and mortality. Our objective was to examine the relationship between individual nonhip, nonvertebral fractures, comorbidities, and mortality. The prospective population-based cohort of 267,043 subjects (45 and Up Study, Australia) had baseline questionnaires linked to hospital administrative and all-cause mortality data from 2006 to 2013. Associations between fracture and mortality were examined using multivariate, time-dependent Cox models, adjusted for age, prior fracture, body mass index, smoking, and comorbidities (cardiovascular disease, diabetes, stroke, thrombosis, and cancer), and survival function curves. Population attributable fraction was calculated for each level of risk exposure. During 1,490,651 person-years, women and men experienced 7571 and 4571 fractures and 7064 deaths and 11,078 deaths, respectively. In addition to hip and vertebral fractures, pelvis, humerus, clavicle, rib, proximal tibia/fibula, elbow and distal forearm fractures in both sexes, and ankle fractures in men were associated with increased multivariable-adjusted mortality hazard ratios ranging from 1.3 to 3.4. Comorbidity independently added to mortality such that a woman with a humeral fracture and 1 comorbidity had a similarly reduced 5-year survival as that of a woman with a hip fracture and no comorbidities. Population mortality attributable to any fracture without comorbidity was 9.2% in women and 5.3% in men. All proximal nonhip, nonvertebral fractures in women and men were associated with increased mortality risk. Coexistent comorbidities independently further increased mortality. Population attributable risk for mortality for fractures was similar to cardiovascular disease and diabetes, highlighting their importance and potential benefit for early intervention and treatment. © 2018 American Society for Bone and Mineral Research.

Fracture-related mortality in southern Sweden: A multiple cause of death analysis, 1998-2014

PURPOSE

To assess fracture-related mortality among adults (aged ≥20years) in southern Sweden using multiple causes of death approach.

METHODS

All death certificates (n=201 488) in adults recorded in the region of Skåne from 1998 to 2014 were examined. We identified fracture-related deaths and computed mortality rates by sex, age group, and fracture site. Temporal trends were evaluated using joinpoint regression and associated causes were identified by age- and sex-adjusted observed/expected ratios.

RESULTS

Fractures were mentioned on 6 226 (3.1%) death certificates, with majority of these occurred among women (60%) and those aged≥80years (77%). While hip was the most common site overall (61% of all fracture-related deaths), skull was the most common site in people <60years (60% of all fracture-related deaths). Proportion of death certificates mentioning fracture was stable in women but increased by 0.4% (95% CI: 0.1 to 0.6) in men between 1998-2002 and 2010-2014. The mean age at death was higher in death certificates mentioning fracture than those without and this gap widened over time. The mean age-standardized fracture-related mortality rate was 18.8 (14.0) per 100 000 person-year in men (women) and declined by 1.5% (1.3%) per year during 1998-2014. Injuries (84.6%) and cardiovascular disorders (64.6%) were the most common comorbidities on death certificates mentioning fracture.

CONCLUSIONS

Fracture is a contributing cause of death in more than 3% of all deaths in southern Sweden with hip in lead among older and skull fracture among younger people. There was a slight increase in proportion of deaths associated with fracture in men but not women during the study period.

Fractures in Kidney Transplant Recipients: A Comparative Study Between England and New York State

OBJECTIVES

Fractures are associated with high morbidity and are a major concern for kidney transplant recipients. No comparative analysis has yet been conducted between countries in the contemporary era to inform future international prevention trials.

MATERIALS AND METHODS

Data were obtained from the Hospital Episode Statistics and the Statewide Planning and Research Cooperative databases on all adult kidney transplants performed in England and New York State from 2003 to 2013, respectively, and on posttransplant fracture-related hospitalization from 2003 to 2014.

RESULTS

Our analysis included 18 493 English and 11 602 New York State kidney transplant recipients. Overall, 637 English recipients (3.4%) and 398 New York State recipients (3.4%) sustained a fracture, giving an unadjusted event rate of 7.0 and 5.9 per 1000 years, respectively (P = .948). Of these, 147 English (0.8%) and 101 New York State recipients (0.9%) sustained a hip fracture, giving an unadjusted event rate of 1.6 and 1.5 per 1000 years, respectively (P = .480). There were no differences in the cumulative incidence of all fractures or hip fractures. One-year mortality rates after any fracture (9% and 11%) or after a hip fracture (15% and 17%) were not different between cohorts.

CONCLUSIONS

Contemporaneous English and New York State kidney transplant recipients have similar fracture rates and mortality rates postfracture.

The impact of comorbidities on hip fracture mortality: a retrospective population-based cohort study

The impact of comorbidities on hip fracture-related excess mortality was assessed in a population-based age- and sex-matched cohort over 10 years. On average, only 1 out of 12 excess deaths over 10 years was related to pre-fracture life-threatening comorbidities. The presence of life-threatening comorbidities increased the excess risk of death after hip fracture.

PURPOSE

This work aimed to estimate the impact of pre-fracture comorbidities on the 10-year excess risk of all-cause death after hip fracture among Estonian men and women ≥ 50 years of age.

METHODS

Retrospective, population-based 10-year study of people aged ≥ 50 in two cohorts: those with a hip fracture and an age- and sex-matched random sample from the national health insurance fund for comparison.

RESULTS

We found that hip fracture was a strong independent risk factor for death. Upon adjustment for Charlson Comorbidities Index (CCI) score, the impact of life-threatening comorbidities on average hip fracture-related excess mortality was modest: only 8% of excess deaths over 10 years were related to comorbidities. Upon stratification by CCI groups, the excess risk of patients in CCI groups ≥ 3 and 1-2 exceeded that in the CCI 0 group over 5-7 years, indicating that in patients with life-threatening comorbidities, a hip fracture accelerates the chain of lethal events and brings deaths from other conditions forward. The impact of comorbidities was age- and time-dependent: in younger hip fracture patients, the comorbidities almost doubled the excess risk from a fracture in 10 years; in older patients, the effect was shorter and modest.

CONCLUSIONS

The presence of pre-fracture comorbidities increases the risk of excess death in hip fracture patients, but the comorbidity impact on aggregated excess mortality is modest.

The impact of hip fracture on mortality in Estonia: a retrospective population-based cohort study

Background

Estimates of hip fracture mortality in Eastern Europe are scarce. We aimed to estimate the magnitude and duration of excess mortality after hip fracture in Estonia.

Methods

Retrospective, population-based 10-year study of persons aged ≥50 in two cohorts: with hip fracture and an age- and sex-matched (in a 1:4 ratio) random sample from the national health insurance fund database for comparison. Cumulative risks, excess risks and relative risks of death were estimated using Poisson regression with 95% bootstrap confidence intervals (CI). Risks were adjusted for age and Charlson comorbidity index.

Results

We identified 8298 (2383 men, 5915 women) incident hip fracture patients from 2005 to 2013 and 33,191 (9531 men, 23,660 women) individuals for the reference group. 5552 (1564 men, 3988 women) cases and 14,037 (3514 men, 10,523 women) reference individuals died during the 10-year follow-up period. Among hip fracture patients we observed a pronounced and durable excess risk of death that was highest within 3–6 months after fracture and persisted for the full 10-year follow-up period. After adjustment for age and Charlson index, hip fracture was associated with a 21.1% (95% CI 20.0–22.5%) 10-year cumulative excess risk of death (RR 1.37, 95% CI 1.35–1.40). We found a high immediate excess risk of death in older age groups (≥80 years) and gradually accumulating excess risk in younger age groups (50–79 years). The excess risk was more pronounced among men than women.

Conclusions

By the end of the 10-year follow-up, 1 in 4 deaths in the hip fracture group was attributable to the hip fracture. The results indicate a high attributable impact of hip fracture as an independent risk factor for death.

Increased mortality after lower extremity fractures in patients <65 years of age

Background and purpose - The association between mortality and lower extremity fractures (other than hip fractures in older individuals) is unclear. We therefore investigated mortality in adults of all ages after lower extremity fractures that required inpatient care. Patients and methods - Diagnosis code (ICD10), procedure code (NOMESCO), and 7 additional characteristics of patients admitted to the trauma ward at Central Finland Hospital were collected between 2002 and 2008 (n = 3,567). Patients were followed up until the end of 2012. Mortality rates were calculated for patients with all types of lower extremity fractures using data from the population at risk. Results - During the study, 2,081 women and 1,486 men sustained a lower extremity fracture. By the end of follow-up (mean duration 5 years), 42% of the women and 32% of the men had died. For all lower extremity fractures, the standardized mortality ratio (SMR) was 1.9 (95% CI: 1.8-2.0) for women and 2.6 (CI: 2.4-2.9) for men. In patients aged ≥65 years, mortality was increased and of similar magnitude after fractures of the hip, femoral diaphysis, and knee (distal femur, patella, and proximal tibia). In patients aged <65 years, mortality was increased after fractures at all sites. The SMR after fractures at different sites ranged between 2.1 (CI: 1.4-3.2) (ankle) and 6.7 (CI: 5.0-9.0) (hip) in patients aged <65 years and between 0.6 (CI: 0.30-1.1) (leg) and 2.2 (CI: 2.0-2.3) (hip) in patients aged ≥65 years. Interpretation - The post-fracture SMR of patients aged <65 years was at least double that of older patients. Furthermore, the higher mortality observed after proximal fractures of the lower extremity was greater in younger patients. The reasons behind these findings remain unclear.