Epidemiologic patterns of injuries treated in ambulatory care settings

Consequences of inequity in the neurosurgical workforce: Lessons from traumatic brain injury

Women and minorities leave or fail to advance in the neurosurgical workforce more frequently than white men at all levels from residency to academia. The consequences of this inequity are most profound in fields such as traumatic brain injury (TBI), which lacks objective measures. We evaluated published articles on TBI clinical research and found that TBI primary investigators or corresponding authors were 86·5% White and 59·5% male. First authors from the resulting publications were 92.6% white. Most study participants were male (68%). 64·4% of NIH-funded TBI clinical trials did not report or recruit any black subjects and this number was even higher for other races and the Hispanic ethnicity. We propose several measures for mitigation of the consequences of the inequitable workforce in traumatic brain injury that could potentially contribute to more equitable outcomes. The most immediately feasible of these is validation and establishment of objective measures for triage and prognostication that are less susceptible to bias than current protocols. We call for incorporation of gender and race neutral metrics for TBI evaluation to standardize classification of injury. We offer insights into how socioeconomic factors contribute to increased death rates from women and minority groups. We propose the need to study how these disparities are caused by unfair health insurance reimbursement practices. Surgical and clinical research inequities have dire consequences, and until those inequities can be corrected, mitigation of those consequences requires system wide change.

Changing Care Settings for Injuries

Tracking injury rates is important for surveillance purposes but little data exist for injuries outside of emergency department visits. We assess the share and type of injuries reported in urgent care centers (UCCs) compared with other settings. We used FAIR Health claims data from 2016 through the first quarter of 2019 to calculate the percent of claims and most common types of injuries. Of the 197 million injury claims, 62% occurred in office settings and 17% in hospital outpatient departments (HOPDs), 5% in inpatient and in ED settings, and less than 2% in UCCs. Injury claims in UCCs increased 6% from 2016 to 2018, whereas injury claims in EDs declined 24%. Overall, physician offices and HOPDs accounted for the largest share of injury care, but UCCs represented the fastest growing setting to treat injuries.

Improving identification of fall-related injuries in ambulatory care using statistical text mining

OBJECTIVES

We determined whether statistical text mining (STM) can identify fall-related injuries in electronic health record (EHR) documents and the impact on STM models of training on documents from a single or multiple facilities.

METHODS

We obtained fiscal year 2007 records for Veterans Health Administration (VHA) ambulatory care clinics in the southeastern United States and Puerto Rico, resulting in a total of 26 010 documents for 1652 veterans treated for fall-related injury and 1341 matched controls. We used the results of an STM model to predict fall-related injuries at the visit and patient levels and compared them with a reference standard based on chart review.

RESULTS

STM models based on training data from a single facility resulted in accuracy of 87.5% and 87.1%, F-measure of 87.0% and 90.9%, sensitivity of 92.1% and 94.1%, and specificity of 83.6% and 77.8% at the visit and patient levels, respectively. Results from training data from multiple facilities were almost identical.

CONCLUSIONS

STM has the potential to improve identification of fall-related injuries in the VHA, providing a model for wider application in the evolving national EHR system.

A Case Study of Data Quality in Text Mining Clinical Progress Notes

Text analytic methods are often aimed at extracting useful information from the vast array of unstructured, free format text documents that are created by almost all organizational processes. The success of any text mining application rests on the quality of the underlying data being analyzed, including both predictive features and outcome labels. In this case study, some focused experiments regarding data quality are used to assess the robustness of Statistical Text Mining (STM) algorithms when applied to clinical progress notes. In particular, the experiments consider the impacts of task complexity (by removing signals), training set size, and target outcome quality. While this research is conducted using a dataset drawn from the medical domain, the data quality issues explored are of more general interest.

Throw rug-related injuries treated in US EDs: are children the same as adults?

BACKGROUND

Various characteristics of floors and floor coverings are well established as injury hazards. Loose carpeting, such as rugs, is often cited as a hazard leading to injury.

PURPOSE

To describe the epidemiology and patterns of rug, mat, and runner-related injuries in patients seeking emergency treatment.

METHODS

Data from the National Electronic Injury Surveillance System from 1990 through 2009 were investigated. Sample weights were used to calculate national estimates. US Census Bureau data were used to calculate injury rates per 100 000 individuals. Linear regression and computation of relative risks (RRs) with 95% confidence intervals (CIs) were performed.

RESULTS

An estimated 245 605 patients were treated in US emergency departments for rug-related injuries during the study period, with an average of 12 280 cases per year. Females (72.3%) and individuals older than 64 years (47.1%) sustained the largest number of injuries. Patients younger than 6 years were more likely to injure the head or neck region (RR, 3.52 [95% CI, 3.26-3.81]) compared with all other groups. Patients older than 18 years were more likely to experience a fracture or dislocation (RR, 2.52 [95% CI, 2.13-2.88]) and sustain an injury as a result of tripping or slipping on a rug (RR, 1.36 [95% CI, 1.26-1.41] compared with other age groups. Increasing age was associated with increased risk of hospitalization in this study. Patients who sustained an injury from a rubber or plastic mat/rug were significantly less likely to be admitted (RR, 0.67 [95% CI, 0.55-0.83]). Injuries occurring in kitchens or bathrooms resulted in significantly higher admission rates (RR, 1.45 [95% CI, 1.34-1.54]).

CONCLUSIONS

Rug-related injuries are an important source of injury for individuals of all ages.

Emergency department and outpatient treatment of acute injuries in older adults in the United States: 2009-2010

OBJECTIVES

We sought to describe epidemiological patterns of acute injuries treated in emergency department (ED) and outpatient primary care settings in the United States.

DESIGN

Retrospective cross-sectional analysis of data from the 2009 and 2010 National Health Care Surveys.

SETTING

Emergency departments and outpatient primary care clinics.

PARTICIPANTS

Older adults (≥ 65) with initial visits for acute injuries.

MEASUREMENTS

Frequencies and incidence rates of medically attended injury according to participant characteristics and care setting.

RESULTS

Of the 19.7 million medically attended acute injuries in older adults in 2009-10, 50% were treated at EDs and 50% at outpatient primary care clinics. The annual incidence rate of medically attended injuries rose with age, from 20.8 (95% confidence interval (CI) = 17.0-24.6) per 100 in those aged 65 to 74 to 41.5 (95% CI = 33.5-49.4) per 100 for those aged 85 and older. Of injury-related ED visits, 60% occurred outside standard business hours, 36% were triaged as low acuity, and 25% resulted in admission. Only 9% of injury-related primary care visits had injury prevention counseling documented.

CONCLUSION

Medically attended injuries area common in older adults, and their incidence increases with advancing age. Half of all initial visits for acute injuries in older adults are to primary care clinics. Most injured individuals are discharged home, and injury prevention counseling is rarely documented. To inform injury prevention efforts appropriately and to avoid underestimating the burden of injury, future injury studies should include a range of outpatient and inpatient care settings.

Finding falls in ambulatory care clinical documents using statistical text mining

OBJECTIVE

To determine how well statistical text mining (STM) models can identify falls within clinical text associated with an ambulatory encounter.

MATERIALS AND METHODS

2241 patients were selected with a fall-related ICD-9-CM E-code or matched injury diagnosis code while being treated as an outpatient at one of four sites within the Veterans Health Administration. All clinical documents within a 48-h window of the recorded E-code or injury diagnosis code for each patient were obtained (n=26 010; 611 distinct document titles) and annotated for falls. Logistic regression, support vector machine, and cost-sensitive support vector machine (SVM-cost) models were trained on a stratified sample of 70% of documents from one location (dataset Atrain) and then applied to the remaining unseen documents (datasets Atest-D).

RESULTS

All three STM models obtained area under the receiver operating characteristic curve (AUC) scores above 0.950 on the four test datasets (Atest-D). The SVM-cost model obtained the highest AUC scores, ranging from 0.953 to 0.978. The SVM-cost model also achieved F-measure values ranging from 0.745 to 0.853, sensitivity from 0.890 to 0.931, and specificity from 0.877 to 0.944.

DISCUSSION

The STM models performed well across a large heterogeneous collection of document titles. In addition, the models also generalized across other sites, including a traditionally bilingual site that had distinctly different grammatical patterns.

CONCLUSIONS

The results of this study suggest STM-based models have the potential to improve surveillance of falls. Furthermore, the encouraging evidence shown here that STM is a robust technique for mining clinical documents bodes well for other surveillance-related topics.

A population-based study of repetitive traumatic brain injury among persons with traumatic brain injury

PRIMARY OBJECTIVE

The objective was to estimate and compare the hazards of repetitive traumatic brain injury (RTBI) events as a function of the index TBI severity, in a cohort of TBI hospital discharges include in the South Carolina Traumatic Brain Injury Follow-up Registry.

RESEARCH DESIGN

Retrospective cohort.

METHODS AND PROCEDURES

There were 4357 persons with TBI who were followed from the index hospital discharge through 31 December 2005 for RTBI events through the statewide hospital discharge (HD) and emergency department (ED) records. Prentice, Williams, Peterson total time/conditional probability model (PWP-CP) for recurrent events survival analysis was used to assess RTBI as a function of index TBI severity.

MAIN OUTCOMES AND RESULTS

Index TBI severity approached significance in its relationship with RTBI, with persons with a severe index TBI experiencing events at a higher rate than those with a mild/moderate index TBI. Among the other covariates evaluated, epilepsy/seizure disorder, race, gender, payer status, cause of injury and having a prior history of TBI were associated with RTBI.

CONCLUSIONS

While TBI severity approached significance with RTBI, other variables, such as epilepsy/seizure disorder, seem to have a more significant relationship with RTBI.

The prevalence of injury of any type in an urban emergency department population

BACKGROUND

National estimates of injury prevalence in the Emergency Department (ED) are based on medical record review and vary considerably. By using a more robust approach to surveillance, we (1) determine the prevalence of injury of any type in an urban ED population and (2) explore the association between violence-related injury and personal characteristics of injury victims.

METHODS

This cross-sectional study was performed at an urban level I trauma center from June to August, 2005. We prospectively screened 4,246 consecutive ED patients for injury during a randomized schedule of shifts totaling 336 hours. The ED record of each injured patient was reviewed to catalogue injury type and intent (International Classification of External Causes of Injury, Short Form) as well as to estimate injury severity (New Injury Severity Score). We interviewed noncritically injured, adult patients who provided consent to collect demographic (race, income, and education) and personal information (substance abuse, domestic violence, handgun ownership, and homelessness). We sought independent associations between these variables and violence-related injury in an exploratory analysis using multivariate logistic regression.

RESULTS

Injury contributed to 1,036 of 4,246 ED visits (24.4%, 95% confidence interval [CI], 23.1-25.7%). Eleven percent of injured patients were admitted to the hospital and two patients died in the ED. The majority of patients (75%) suffered minor injury. Among the 434 injured patients consenting to interview, the prevalence of established injury risk factors, such as substance use or handgun ownership, varied by gender. The adjusted odds of violence-related injury among this subset of patients were increased for males (odds ratio [OR], 2.22; 95% CI, 1.17-4.23), patients with an annual income less than $5,000 (OR, 2.85; 95% CI, 1.64-4.97), those reporting a history of domestic violence (OR, 2.69; 95% CI, 1.43-5.07), and heavy alcohol users (OR, 1.79; 95% CI, 1.01-3.19).

CONCLUSION

One in four ED visits to this urban, county hospital is due, at least in part, to injury. Patient characteristics associated with violence-related injury may generate hypotheses for further study.

Utilization of and direct expenditure for emergency medical care in Taiwan: a population-based descriptive study

Background

We surveyed the emergency medical system (EMS) in Taiwan to provide information to policymakers responsible for decisions regarding the redistribution of national medical resources.

Methods

A systematic sampling method was used to randomly sample a representative database from the National Health Insurance (NHI) database in Taiwan, during the period from 2000 to 2004.

Results

We identified 10,124, 10,408, 11,209, 10,686, and 11,914 emergency room visits in 2000, 2001, 2002, 2003, and 2004, respectively. There were more males than females, and the majority of adults were younger than 50 years. Diagnose of injury/poisoning was the most frequently noted diagnostic category in emergency departments (EDs) in Taiwan. There were 13,196 (24.3%) and 2,952 (5.4%) patients with 2 and 3 concomitant diagnoses, respectively. There was a significant association between advanced age and the existence of multiple diagnoses (P < 0.001). With the exception of the ill-defined symptoms/signs/conditions, the two most frequent diagnoses were diseases of the circulatory system and diseases of the respiratory system in patients aged 65 years or older. On average, treatment-associated expenditure and drug-associated expenditure in Taiwan EDs averaged NT$1,155 ($35.0) and NT$190 ($5.8), respectively, which was equal to 64.5% and 10.6% of the total ED-associated cost. General ED medical expenditure increased with patient age; the increased cost ratio due to age was estimated at 8% per year (P < 0.001).

Conclusions

The frequency of major health problems diagnosed at ED visits varied by age: more complicated complaints and multiple diagnoses were more frequent in older patients. In Taiwan, the ED system remains overloaded, possibly because of the low cost of an ED visit.

Paediatric trauma and trauma care in Flanders (Belgium). Methodology and first descriptive results of the PENTA registry

Paediatric injury surveillance and prevention are definite priorities for the European, Belgian, and Flemish authorities. Current available data for Flanders (Belgium) are fragmentary and out-of-date. The PENTA registry (PaEdiatric Network around TraumA) was therefore set up to obtain recent population-based data on trauma and trauma care in children and youngsters in Flanders. Data were collected prospectively in a representative sample (n = 18) of Flemish emergency departments (ED). All children (age 0-17 years) who presented at the ED in 2005 or died prehospital due to trauma were included. The registry was split into two levels. The basic A registry ('all' trauma) consisted of 30 variables, and the more exhaustive B registry ('severe trauma', defined as length of hospitalisation >48 hours, including all nonsurvivors) collected data on 291 variables. The incidence for paediatric trauma presenting at Flemish ED was approximately 119/1000/year. Further data were collected in a random sample of 7,879 cases (21.9% of 35,900 eligible patients). Of all cases, 0.8% were considered 'severe' and included in the B registry. In conclusion, the 'burden' of injury in Flanders is still enormous. PENTA provides the first population-based data about the circumstances and the extent of injury in children and youngsters for the Flemish region. In this article we present in detail the surplus value of the methods used, the difficulties encountered, and the most relevant epidemiological findings from the registry.

Somnolence effects of antipsychotic medications and the risk of unintentional injury

PURPOSE

This study examined the relationship between antipsychotic medications, categorized by published somnolence effects, and unintentional injury (UI).

METHODS

The study population included patients of 18-64 years of age in a healthcare insurance database with claims from 2001 to 2004 and diagnoses of schizophrenia or affective disorder. A nested case-control design was used with cases defined by an E-code claim (a specified external cause of injury) for selected UIs. For cases, the index date referred to the first injury. For controls, the "control index date" was the date of claim if there was only a single medical claim; for patients with > or =2 claims, one was selected at random as the "control index date." Both groups had a prescription for a first-generation antipsychotic (FGA) or second-generation antipsychotic (SGA) overlapping the index date. Potential somnolence effects were defined as: low (referent)--aripiprazole/ziprasidone; medium--risperidone; high--olanzapine/quetiapine: or any single FGA. Logistic regression models were used to estimate odds ratio (OR) and 95% confidence interval (CI) for UI, adjusted for gender, age, concomitant drug, and psychiatric diagnosis.

RESULTS

Among 648 cases and 5214 controls, high-somnolence SGAs were associated with an OR of 1.41 95%CI (1.03-1.93) for risk of UI, while medium-somnolence SGAs, and FGAs had ORs of 1.17 95%CI (0.83-1.64) and 1.17 95%CI (0.79-1.74), respectively. When quetiapine and olanzapine were disaggregated, ORs were 1.61 95%CI (1.15-2.25) and 1.25 95%CI (0.89-1.74), respectively.

CONCLUSIONS

High-somnolence SGAs may lead to UI among patients. When prescribing antipsychotics, clinicians should consider potential somnolence.

A comparison of self-reported motor vehicle collision injuries compared with official collision data: an analysis of age and sex trends using the Canadian National Population Health Survey and Transport Canada data

OBJECTIVE

The purpose of this study was to compare the age and sex trends of motor vehicle collision injuries between a nationally representative self-report survey and official police motor vehicle collision report data. To do this, population-based estimates of motor vehicle collision injuries were established using data from the National Population Health Survey (NPHS), a prospective, population-based, longitudinal survey that was compared to Transport Canada's official motor vehicle collision report statistics.

METHODS

Aggregated mean data (1994-1996) were compared for seven age groups (15-19, 20-24, 25-34, 35-44, 45-54, 55-64, and 65+) from both data sets.

RESULTS

No significant differences were found between males' and females' MVC injuries for any age category in the NPHS. A comparison of the NPHS and Transport Canada data found two small (significant) within-sex differences between the data sets, but overall, the analysis largely revealed similar trends for self-reported injury for all age categories and sex.

CONCLUSIONS

The results indicate that the incidence of injuries based on self-report data in a nationally representative sample is similar to official sources of reporting and are thus a valid indicator or motor vehicle collision injury incidence. The results also confirm that injury trends differ from fatality trends.

Recruitment Strategies for a Fall Prevention Program: If We Build It, Will They Really Come?

BACKGROUND

More than one third of adults over the age of 65 suffer a fall each year, facing morbidity and mortality. Modifiable risk factors for falls have been identified, but specific recruitment strategies for prevention programs have not been evaluated. The purpose of this observational study was to evaluate recruitment strategies for a fall prevention program.

METHODS

Participants were recruited during an 11-month period at a Level I trauma center. Participants were eligible if >65 years old, living independently, and had a fall. Recruitment modalities included (1) emergency medical services, (2) emergency department (ED), (3) primary care providers, and (4) media exposure leading to self-referral. Data were collected on baseline rate of fall victims seen in the ED, demographics, medical history, and source of referral.

RESULTS

There were 91 individuals referred, with 61 (67%) enrolled. Enrollment rates were higher among patients referred by self or primary care providers than among those referred by emergency medical services or the ED. There were no significant differences in demographics or medical history among the eligible but not referred ED population, the referred population, and the enrolled population. Reasons for not enrolling included inappropriate referral (33%), no response (17%), other illness (13%), and patients thinking that they do not need the services (37%).

CONCLUSIONS

These recruitment strategies were successful in enrolling a representative population of patients at risk for recurrent falls, but could be improved to capture more potential participants. Source of referral has a significant effect on rate of enrollment. We outline challenges and solutions to recruitment.