Original ResearchEpidemiologic Patterns of Injuries Treated in Ambulatory Care Settings
Introduction
Injuries remain a major source of morbidity and mortality in the United States. Unintentional injuries are the leading cause of death among US individuals aged 1 to 34 years and the fifth leading cause of death for all age groups combined.1 These deaths represent few injuries, with estimates of approximately 200 emergency department (ED) visits for every injury death.2, 3 The societal cost associated with injuries is also enormous; the Centers for Disease Control and Prevention (CDC) estimated that injuries accounted for approximately 10% of total US medical expenditures in 2000, with costs of more than $224 billion annually if rehabilitation and lost wages and productivity are included.1, 4 Almost one third of all productive years of life that are lost before age 65 result from injuries, a proportion greater than the combined years lost to heart disease, stroke, and cancer.2
The lack of comprehensive national data on injuries remains an obstacle in injury research, and statistics on nonfatal injuries vary widely. The 2002 National Health Interview Survey and a 2003 National Center for Health Statistics report estimated 23.7 million to 34.4 million medically attended injuries and poisonings,5, 6 whereas a 1999 Institute of Medicine report estimated 37 million injury-related visits to EDs alone.2 These discrepancies are due in large part to differences in data sources. A recent study from Greece suggested that national injury estimates are less accurate when based on population surveys than on medical records.7
Most epidemiologic studies of injury morbidity have been restricted to data from ED or inpatient records, only partially revealing the full burden of injuries in the United States. The Institute of Medicine report included little discussion of injuries treated in clinics or physician offices, and a recent review of available metrics for measuring the burden of injuries similarly neglected non-ED ambulatory settings.2, 8 CDC data suggest that there could be almost 3 times as many injury-related visits to clinics or physician offices compared with EDs,9 but these figures include visits for initial injuries and their long-term sequelae. To our knowledge, no study has examined the epidemiologic patterns of acute injuries by combining data from different ambulatory care settings.
The primary objectives of this study were to use medical record data to estimate the annual incidence of nonfatal medically attended injuries in the United States and to compare the characteristics of medically attended injuries treated in EDs, physician offices, and hospital outpatient departments in the United States. Our results should show the magnitude of injury as a public health problem and provide useful information for understanding variations in the epidemiologic patterns of injuries treated in different ambulatory care settings, knowledge important for future injury research.
Section snippets
Study Design and Setting
We conducted a cross-sectional analysis of initial visits for acute injuries to EDs, physician offices, and hospital outpatient departments in the United States using 2002 data from the National Hospital Ambulatory Medical Care Survey (NHAMCS) and National Ambulatory Medical Care Survey (NAMCS). These surveys are conducted annually by the CDC's National Center for Health Statistics and use multistage probability sampling to derive national estimates of patient visits to care sites in the Unites
Results
In 2002, in the United States, an estimated 76.0 million injuries were treated at EDs (46.2%), physician offices (47.8%), and hospital outpatient departments (6.0%), yielding an overall incidence rate of 26.8 per 100 population (95% CI 24.4 to 29.7; Table 1). Injuries accounted for 35.7% of all initial visits to EDs for acute problems in 2002, compared to 17.1% to physician offices and 21.9% to hospital outpatient departments. Age-specific annual rates of medically attended injuries ranged from
Limitations
An important caveat to the implications of our results is that they represent only injury patients who actually visited an ED, physician office, or hospital outpatient department. Some patients do not seek medical care, especially for mild injuries, so our study cannot estimate the true incidence of injuries. The NAMCS and NHAMCS also do not allow for identification of risk factors, because they provide no information about uninjured individuals. Perhaps more important to the interpretation of
Discussion
Research of nonfatal injuries has focused primarily on ED visits and hospital admissions, resulting in considerable underestimates of the magnitude of injury morbidity that exclude the large proportion of injuries treated in non-ED settings.2, 4, 16 The primary objective of this study was to use medical record data to generate a more accurate estimate of the incidence of acute injuries treated in the United States. By combining national data from the main types of ambulatory care settings, we
References (24)
- et al.
Acute healthcare utilization by children after motor vehicle crashes
Accid Anal Prev
(2004) Injury Fact Book 2001-2002
(2002)- et al.
Reducing the Burden of Injury: Advancing Prevention and Treatment
(1999) - et al.
Surveillance summaries: surveillance for fatal and nonfatal injuries: United States, 2001
Morb Mortal Wkly Rep
(2004) - et al.
Medical expenditures attributable to injuries: United States, 2001
Morb Mortal Wkly Rep
(2004) - et al.
Summary health statistics for the US population: National Health Interview Survey, 2002
Vital Health Stat
(2004) - et al.
Health Care in America: Trends in Utilization
(2003) - et al.
Estimating the population burden of injuries: a comparison of household surveys and emergency department surveillance
Epidemiology
(2004) - et al.
Measuring the public health impact of injuries
Epidemiol Rev
(2003) - National Center for Health Statistics. Fastats [National Center for Health Statistics Web site]. Available at:...
Cited by (21)
Throw rug-related injuries treated in US EDs: Are children the same as adults?
2014, American Journal of Emergency MedicineCitation Excerpt :Unintentional injuries are the leading cause of death among individuals aged 1 to 44 years in the United States and the fifth leading cause of death for all age groups combined [1,2]. Injuries account for nearly 10% of US medical expenditures [3]. Despite significant efforts to prevent trauma, injuries continue to represent approximately 30% of emergency department (ED) visits in the United States [4,5].
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
2008, Accident Analysis and PreventionCitation Excerpt :These MVC injuries may include those injuries that are of a less serious magnitude, or those injuries that presented at outpatient medical services, and thus are detected in the NPHS data but not captured in the hospital or police data. As Bishai and Gielen (2001) and Betz and Li (2005) found in their studies, rates of injuries based on hospital or E code data could be missing between 50 and 90% of medically treated MVC injuries. In the case of these national samples (self-report and official police data), the data were very similar in terms of their rates of incidence, but as the degree of injury was not assessed in the initial reports, it is unknown if the severity of the injury is represented differently by the two samples—police reporting the more severe injuries and the NPHS reporting delayed injuries that developed after the initial time of the MVC.
Changing Care Settings for Injuries
2022, Medical Care Research and ReviewConsequences of inequity in the neurosurgical workforce: Lessons from traumatic brain injury
2022, Frontiers in SurgeryImproving identification of fall-related injuries in ambulatory care using statistical text mining
2015, American Journal of Public HealthA case study of data quality in text mining clinical progress notes
2015, ACM Transactions on Management Information Systems
Supervising editor: Arthur L. Kellermann, MD, MPH
Author contributions: GL conceived the study and obtained research funding. MEB obtained, managed, and analyzed the data. GL provided statistical advice. MEB drafted the manuscript, and both MEB and GL contributed substantially to its revision. MEB takes responsibility for the paper as a whole.
Funding and support: This research was supported in part by grants R01 AA09963 and R01 AG013642 from the National Institutes of Health and by grant CCR302486 from the Centers for Disease Control and Prevention.