Arturo J. Rios-Diaz

Inequalities in Specialist Hand Surgeon Distribution across the United States.

Background: Unequal access to hospital specialists for emergency care is an issue in the United States. The authors sought to describe the geographic distribution of specialist hand surgeons and associated factors in the United States. Methods: Geographic distributions of surgeons holding a Subspecialty Certificate in Surgery of the Hand and hand surgery fellowship positions were identified from the American Board of Medical Specialties Database and the literature (2013), respectively. State-level population and per capita income were ascertained using U.S. Census data. Variations in hand trauma admissions were determined using Healthcare Cost and Utilization Project national/state inpatient databases. Risk-adjusted generalized linear models were used to assess independent association between hand surgeon density and hand trauma admission density, fellowship position density, and per capita income. Results: Among 2019 specialist hand surgeons identified, 72.1 percent were orthopedic surgeons, 18.3 percent were plastic surgeons, and 9.6 percent were general surgeons. There were 157 hand surgery fellowship positions nationwide. There were 149,295 annual hand trauma admissions. The national density of specialist hand surgeons and density of trauma admission were 0.6 and 47.6, respectively. The density of specialist hand surgeons varied significantly between states. State-level variations in density of surgeons were independent and significantly associated with median per capita income (p < 0.001) and with density of fellowships (p = 0.014). Conclusions: Specialist hand surgeons are distributed unevenly across the United States. State-level analyses suggest that states with lower per capita incomes may be particularly underserved, which may contribute to regional disparities in access to emergency hand trauma care.

Are Older Adults With Hip Fractures Disadvantaged in Level 1 Trauma Centers

Background:Large regional hospitals achieve good outcomes for patients with complex conditions. However, recent studies have suggested that some patient groups might not benefit from treatment in higher-level trauma centers. Objective:To test the hypothesis that older adults with isolated hip fractures experience delayed surgical treatment and worse clinical outcomes when treated in higher-level trauma centers. Research Design:Retrospective cohort study using a statewide longitudinal database that captured 98% of inpatients within California (2007–2011). Subjects:All older adults (aged 65 y and above) admitted with an isolated hip fracture who did not require interhospital transfer. Measures:Days to operation, length of stay, inhospital mortality, 30-day risk of unplanned readmission, 30-day venous thromboembolism, decubitus ulcers, and pneumonia. Results:There were 91,401 patients, 6.1% of whom were treated in a level 1 trauma center (L1TC), 17.7% in a level 2 trauma center (L2TC), and 70.2% in a nontrauma center (NTC). Within multivariable logistic and generalized linear regression models, patients treated in L1TCs underwent surgery later (predicted mean difference: 0.30 d; 95% CI, 0.08–0.53), had prolonged inpatient stays (0.99 d, 0.40–1.59), and had higher odds of both 30-day readmission (aOR=1.62; 95% CI, 1.35–1.93) and venous thromboembolism (aOR=1.32, 1.01–1.74) relative to NTCs. There were no differences in mortality, decubitus ulcers, or pneumonias. L2TCs were not different from NTCs across any of the measured outcomes. Conclusions:Older adults with hip fractures may be disadvantaged in L1TCs. Further research should aim to develop our understanding of this disparity to ensure that all patient groups benefit from the resources and expertise available within these hospitals.

Comparing Readmissions and Infectious Complications of Blunt Splenic Injuries Using a Statewide Database

BACKGROUND Although non-operative management of blunt splenic injury (BSI) is increasingly common, the long-term infectious complications after adjunct splenic artery embolization (SAE) are not well described. METHODS Patients aged 18-64 y with BSI were identified in the California State Inpatient Database (2007-2011) and categorized as receiving either non-operative management (NOM) without SAE, NOM with SAE, or operative management (OM). The cumulative incidence of infections (surgical site infections [SSI], pneumonia, urinary tract infections, and sepsis) requiring readmission at different times up to one y after injury were calculated. Patient and treatment factors associated with infectious readmissions were determined using multivariable logistic regression models. RESULTS Of the 4,360 patients with BSI, 61.6% had NOM without SAE, 5.8% had NOM with SAE, and 32.6% had OM. The cumulative incidences of infectious complications after each of the management modes were 1.27%, 1.59%, and 1.76%, respectively, during admission (p = 0.446); 2.16%, 5.18%, and 4.85%, respectively, at 30 d after injury (p < 0.001); and 4.69%, 9.16%, and 8.85%, respectively, at one y after injury (p < 0.001). Risk factors for infection-associated readmissions within one y after injury were Charlson score ≥2 (adjusted odds ratio [AOR] 3.9; 95% confidence interval [CI] 2.61-6.02), length of stay >seven d (AOR 2.47; 95% CI 1.58-3.85), NOM with SAE (AOR 2.00; 95% CI 1.19-3.34), and OM (AOR 1.47; 95% CI 1.05-2.07). CONCLUSIONS The long-term risk of infectious complications in patients with BSI who have NOM with SAE is similar to that in patients who are treated with OM, indicating the need for pro-active strategies to reduce long-term infectious complications after SAE.

Sarcopenia increases risk of long-term mortality in elderly patients undergoing emergency abdominal surgery

BACKGROUND Frailty is associated with poor surgical outcomes in elderly patients but is difficult to measure in the emergency setting. Sarcopenia, or the loss of lean muscle mass, is a surrogate for frailty and can be measured using cross-sectional imaging. We sought to determine the impact of sarcopenia on 1-year mortality after emergency abdominal surgery in elderly patients. METHODS Sarcopenia was assessed in patients 70 years or older who underwent emergency abdominal surgery at a single hospital from 2006 to 2011. Average bilateral psoas muscle cross-sectional area at L3, normalized for height (Total Psoas Index [TPI]), was calculated using computed tomography. Sarcopenia was defined as TPI in the lowest sex-specific quartile. Primary outcome was mortality at 1 year. Secondary outcomes were in-hospital mortality and mortality at 30, 90, and 180 days. The association of sarcopenia with mortality was assessed using Cox proportional hazards regression and model performance judged using Harrells C-statistic. RESULTS Two hundred ninety-seven of 390 emergency abdominal surgery patients had preoperative imaging and height. The median age was 79 years, and 1-year mortality was 32%. Sarcopenic and nonsarcopenic patients were comparable in age, sex, race, comorbidities, American Society of Anesthesiologists classification, procedure urgency and type, operative severity, and need for discharge to a nursing facility. Sarcopenic patients had lower body mass index, greater need for intensive care, and longer hospital length of stay (p < 0.05). Sarcopenia was independently associated with increased in-hospital mortality (risk ratio, 2.6; 95% confidence interval [CI], 1.6–3.7) and mortality at 30 days (hazard ratio [HR], 3.7; 95% CI, 1.9–7.4), 90 days (HR, 3.3; 95% CI, 1.8–6.0), 180 days (HR, 2.5; 95% CI, 1.4–4.4), and 1 year (HR, 2.4; 95% CI, 1.4–3.9). CONCLUSION Sarcopenia is associated with increased risk of mortality over 1 year in elderly patients undergoing emergency abdominal surgery. Sarcopenia defined by TPI is a simple and objective measure of frailty that identifies vulnerable patients for improved preoperative counseling, setting realistic goals of care, and consideration of less invasive approaches. LEVEL OF EVIDENCE Prognostic study, level III.

Routine inclusion of long-term functional and patient-reported outcomes into trauma registries: The FORTE project

BACKGROUND The National Academies of Sciences, Engineering, and Medicine (formerly the Institute of Medicine) recently recommended inclusion of postdischarge health-related quality of life (HRQoL) and patient-reported outcomes (PROs) metrics to benchmark the quality of trauma care. Currently, these measures are not routinely collected at most trauma centers. We sought to determine the feasibility and value of adding such long-term outcome measures to trauma registries. METHODS As part of the FORTE (Functional Outcomes and Recovery after Trauma Emergencies) project, we included patients with an Injury Severity Score of 9 or greater, admitted to the Brigham and Women’s Hospital in Boston, MA, who were identified retrospectively using the institutional trauma registry and contacted 6 or 12 months after injury to participate in a telephone survey evaluating HRQoL (Short Form 12 [SF-12]), PROs (Trauma Quality of Life), posttraumatic stress disorder, return to work, residential status, and health care utilization. RESULTS Data were collected for 171 of 394 eligible patients: 85/189 (45%) at 6 months and 86/205 (42%) at 12 months; 25%/29% (6/12 months) patients could not be contacted, 15%/16% (6/12 months) declined to participate, and 15%/13% (6/12 months) were interested in participating at another time but were not reached again. Approximately 20% patients screened positive for posttraumatic stress disorder, and half had not yet returned to work. There were significant reductions in SF-12 physical composite scores relative to population norms (mean, 50 [SD, 10]) at 6 months (mean, 44; 95% confidence interval [CI], 41–47) and 12 months (45; 95% CI, 42–47); no difference was noted in the SF-12 mental composite scores (6 months: 51 [95% CI, 48–54]; 12 months: 50 [95% CI, 46–53]). CONCLUSIONS Trauma patients reported considerable impairment 6 and 12 months after injury. Routine collection of PROs and HRQoL provides important data regarding trauma outcomes beyond mortality and will enable the development of quality improvement metrics that better reflect patients’ postinjury experiences. Improved and alternate methods for collection of these data need to be developed to enhance response rates before widespread adoption across trauma centers in the United States. LEVEL OF EVIDENCE Prognostic/epidemiologic, level II; Therapeutic, level III.

Global Patterns of QALY and DALY Use in Surgical Cost-Utility Analyses: A Systematic Review

Background Surgical interventions are being increasingly recognized as cost-effective global priorities, the utility of which are frequently measured using either quality-adjusted (QALY) or disability-adjusted (DALY) life years. The objectives of this study were to: (1) identify surgical cost-effectiveness studies that utilized a formulation of the QALY or DALY as a summary measure, (2) report on global patterns of QALY and DALY use in surgery and the income characteristics of the countries and/or regions involved, and (3) assess for possible associations between national/regional-income levels and the relative prominence of either measure. Study Design PRISMA-guided systematic review of surgical cost-effectiveness studies indexed in PubMed or EMBASE prior to December 15, 2014, that used the DALY and/or QALY as a summary measure. National locations were used to classify publications based on the 2014 World Bank income stratification scheme into: low-, lower-middle-, upper-middle-, or high-income countries. Differences in QALY/DALY use were considered by income level as well as for differences in geographic location and year using descriptive statistics (two-sided Chi-squared tests, Fischer’s exact tests in cell counts <5). Results A total of 540 publications from 128 countries met inclusion criteria, representing 825 “national studies” (regional publications included data from multiple countries). Data for 69.0% (569/825) were reported using QALYs (2.1% low-, 1.2% lower-middle-, 4.4% upper-middle-, and 92.3% high-income countries), compared to 31.0% (256/825) reported using DALYs (46.9% low-, 31.6% lower-middle-, 16.8% upper-middle-, and 4.7% high-income countries) (p<0.001). Studies from the US and the UK dominated the total number of QALY studies (49.9%) and were themselves almost exclusively QALY-based. DALY use, in contrast, was the most common in Africa and Asia. While prominent published use of QALYs (1990s) in surgical cost-effectiveness studies began approximately 10 years earlier than DALYs (2000s), the use of both measures continues to increase. Conclusion As global prioritization of surgical interventions gains prominence, it will be important to consider the comparative implications of summary measure use. The results of this study demonstrate significant income- and geographic-based differences in the preferential utilization of the QALY and DALY for surgical cost-effectiveness studies. Such regional variation holds important implications for efforts to interpret and utilize global health policy research. PROSPERO registration number: CRD42015015991

Geographic Distribution of Trauma Burden, Mortality, and Services in the United States: Does Availability Correspond to Patient Need?

BACKGROUND The association between the need for trauma care and trauma services has not been characterized previously. We compared the distribution of trauma admissions with state-level availability of trauma centers (TCs), surgical critical care (SCC) providers, and SCC fellowships, and assessed the association between trauma care provision and state-level trauma mortality. STUDY DESIGN We obtained 2013 state-level data on trauma admissions, TCs, SCC providers, SCC fellowship positions, per-capita income, population size, and age-adjusted mortality rates. Normalized densities (per million population [PMP]) were calculated and generalized linear models were used to test associations between provision of trauma services (higher-level TCs, SCC providers, and SCC fellowship positions) and trauma burden, per-capita income, and age-adjusted mortality rates. RESULTS There were 1,345,024 trauma admissions (4,250 PMP), 2,496 SCC providers (7.89 PMP), and 1,987 TCs across the country, of which 521 were Level I or II (1.65 PMP). There was considerable variation between the top 5 and bottom 5 states in terms of Level I/Level II TCs and SCC surgeon availability (approximately 8.0/1.0), despite showing less variation in trauma admission density (1.5/1.0). Distribution of trauma admissions was positively associated with SCC provider density and age-adjusted trauma mortality (p ≤ 0.001), and inversely associated with per-capita income (p < 0.001). Age-adjusted mortality was inversely associated with the number of SCC providers PMP. For every additional SCC provider PMP, there was a decrease of 618 deaths per year. CONCLUSIONS There is an inequitable distribution of trauma services across the US. Increases in the density of SCC providers are associated with decreases in mortality. There was no association between density of trauma admissions and location of Level I/Level II TCs. In the wake of efforts to regionalize TCs, additional efforts are needed to address disparities in the provision of quality care to trauma patients.

How long should we fear? Long-term risk of venous thromboembolism in patients with traumatic brain injury.

BACKGROUND Although patients with traumatic brain injury (TBI) are known to be at high risk for venous thromboembolism (VTE), it is not clear how long this risk persists after injury. We aimed to determine the risk of VTE in patients with TBI during one year after injury and to identify associated factors. METHODS Patients 18 years and older with International Classification of Diseases, Ninth Revision, Clinical Modification diagnoses of isolated TBI (head Abbreviated Injury Scale [AIS] ≥3 and AIS <3 for all other body regions) were identified in the California State Inpatient Database (2007–2011). Patient and admission (injury severity score, length of stay, and discharge disposition) characteristics were assessed. Hospital factors (teaching status, trauma center verification, and bed size) were extracted from the American Hospital Association database. Patients who developed VTE during the index admission and at different time points after discharge were determined. Multivariate logistic regression models were used to assess the associated risk factors for VTE after discharge. RESULTS There were 38,984 patients with isolated TBI identified. The incidence of VTE was 1.31% during the index admission and the cumulative incidence of VTE involving hospitalization within one year of injury was 2.83%. The major risk factors for VTE one year after injury (not including the index admission) were discharge to extended care facilities versus home [adjusted odds ratio, 2.69 (95% confidence interval, 2.14–3.37)], age older than 64 years versus 18 to 44 years [2.62 (1.80–3.81)], having an operation during the index admission [1.65 (1.36–2.01)], and hospital length of stay of more than 7 days versus 3 days or less [1.64 (1.27–2.11)]. CONCLUSION The risk of VTE persists long after discharge in a significant proportion of patients with TBI. Demographic and admission characteristics of patients play significant roles in the risk of VTE after discharge. These results highlight the need for sustained surveillance and preventive measures among patients with TBI at increased risk for long-term VTE. LEVEL OF EVIDENCE Epidemiologic study, level III.

Access to post-discharge inpatient care after lower limb trauma

BACKGROUND Most hospitals in the United States are required to provide emergency care to all patients, regardless of insurance status. However, uninsured patients might be unable to access non-acute services, such as post-discharge inpatient care (PDIC). This could result in prolonged acute hospitalization. We tested the hypothesis that insurance status would be independently associated with both PDIC and length of stay (LOS). METHODS An observational study was undertaken using the California State Inpatient Database (2007-2011), which captures 98% of patients admitted to hospital in California. All patients with a diagnosis of orthopedic lower limb trauma were identified using International Classification of Diseases, 9th Revision, Clinical Modification codes 820-828. Multivariable logistic and generalized linear regression models were used to adjust odds of PDIC and LOS for patient and hospital characteristics. RESULTS There were 278,573 patients with orthopedic lower limb injuries, 160,828 (57.7%) of which received PDIC. Uninsured patients had lower odds of PDIC (adjusted odds ratio 0.20, 95% confidence interval 0.17-0.24) and significantly longer hospital LOS (predicted mean difference 1.06 [95% confidence interval 0.78-1.34] d) than those with private insurance. CONCLUSIONS Lack of health insurance is associated with reduced access to PDIC and prolonged hospital LOS. This potential barrier to hospital discharge could reduce the number of trauma beds available for acutely injured patients.

Failure to rescue and disparities in emergency general surgery

BACKGROUND Racial and socioeconomic disparities are well documented in emergency general surgery (EGS) and have been highlighted as a national priority for surgical research. The aim of this study was to identify whether disparities in the EGS setting are more likely to be caused by major adverse events (MAEs) (e.g., venous thromboembolism) or failure to respond appropriately to such events. METHODS A retrospective cohort study was undertaken using administrative data. EGS cases were defined using International Classification of Diseases, Ninth Revision, Clinical Modification diagnostic codes recommended by the American Association for the Surgery of Trauma. The data source was the National Inpatient Sample 2012-2013, which captured a 20%-stratified sample of discharges from all hospitals participating in the Healthcare Cost and Utilization Project. The outcomes were MAEs, in-hospital mortality, and failure to rescue (FTR). RESULTS There were 1,345,199 individual patient records available within the National Inpatient Sample. There were 201,574 admissions (15.0%) complicated by an MAE, and 12,006 of these (6.0%) resulted in death. The FTR rate was therefore 6.0%. Uninsured patients had significantly higher odds of MAEs (adjusted odds ratio, 1.16; 95% confidence interval, 1.13-1.19), mortality (1.28, 1.16-1.41), and FTR (1.20, 1.06-1.36) than those with private insurance. Although black patients had significantly higher odds of MAEs (adjusted odds ratio, 1.14; 95% confidence interval, 1.13-1.16), they had lower mortality (0.95, 0.90-0.99) and FTR (0.86, 0.80-0.91) than white patients. CONCLUSIONS Uninsured EGS patients are at increased risk of MAEs but also the failure of health care providers to respond effectively when such events occur. This suggests that MAEs and FTR are both potential targets for mitigating socioeconomic disparities in the setting of EGS.