INTRODUCTION

For many motor vehicle crash victims, acute hospital care might be only the first stage of a long and costly treatment program. For many crash victims, lost wages from missed work or reduced work opportunities resulting from permanent impairment will outweigh medical costs. This study was designed to increase knowledge about crash costs and consequences, and provide a more comprehensive picture of the full cost of motor vehicle and crash injuries. Some aspects of the these costs are well documented, but data on rehabilitation costs and permanent disabilities resulting from motor vehicle injury and the source of payment for these costs have been lacking.

This study separately analyzes motorcycle and other motor vehicle injuries because motorcycles account for a disproportionately large share of the burden imposed by highway crashes (Miller, Spicer, Lestina, and Levy, 1999), and may have differential rehabilitation costs. The objectives of this study were to: (1) define rehabilitation costs resulting from injuries received in a traffic crashes; (2) analyze these costs to develop a national model for estimating rehabilitation costs resulting from crashes; (3) identify the source of payment for short- and long-term rehabilitation costs; and (4) capture other data that rehabilitation-related data sets included about the long–term outcomes of motor vehicle injury. Comparable data on other injury causes were captured where possible for comparative purposes. Unless otherwise stated, data in this analysis includes only crashes that occurred on public roadways. The costs reported in this study are only for those patients who were admitted to a hospital for rehabilitation and exclude additional charges for outpatient rehabilitation and related professional services.

Background
Little information is available on the rehabilitation costs of motor vehicle and motorcycle injuries. For traumatic brain injury, which is the most common diagnosis for hospitalized motor vehicle injury, average costs rise dramatically for those individuals who undergo rehabilitation. In one study (Brooks et al., 1995) after a four-year follow-up, average costs for medical and long-term care services averaged $196,460 for survivors receiving rehabilitation services, compared to $17,893 for those receiving no rehabilitation.

A study of Chicago motorcycle crash victims in the acute rehabilitation setting identified 77 patients. Of these, the majority were admitted with a primary diagnosis of traumatic brain injury (TBI, n=50), followed by spinal cord injury (SCI, n=18) and orthopedic injuries or amputations (n=9). Most of the patients with TBI were not wearing their helmets at the time of injury (n=39, 78%). People with spinal cord injuries (SCI) incurred the largest rehabilitation hospital charges (mean $94,548) due to significantly longer rehabilitation stays (mean 42.6 days). Helmet use was unrelated to discharge destination. Although there was no difference in total Functional Independence Measures (FIM) instrument, (discussed further in chapter 3) scores for helmeted and unhelmeted patients, people not wearing helmets had significantly lower cognitive FIM scores at admission and discharge. The study authors conclude that the protective effect of helmet use in motorcycle crashes is reflected even in this post-acute care population of rehabilitation patients (Lombard, Kelly, Heinemann, and Kychlik, 2003).

In 1998 a comment in the Journal of the American Medical Association (JAMA) described a study that analyzed costs for 105 motorcyclists hospitalized in a major trauma center over a 12-month period. Here total direct costs were followed up for a mean of 20 months and were more than $2.7 million with an average of $25,764 per patients. Only 60 percent of the direct costs were accounted for by the initial hospital care; 23 percent of costs were for rehabilitation care or preadmission for treatment of acute problems. The majority (63.4%) of care was compensated with public funds, with Medicaid accounting for more than half of all charges (Rivara, Dicker, Bergman, Dacey, and Herman, 1988).

Methodology
In this study, secondary data were collected on the frequency, duration, and costs of motor vehicle and motorcycle injury from five sources: American Medical Rehabilitation Provider’s Association (AMRPA), Uniform Data System for Medical Rehabilitation; Traumatic Brain Injury Model System National Database; pooled 1997 or 1998 Hospital Discharge Survey census data from 21 States; the Health Care Utilization Program 2000 National Inpatient Sample; and the Colorado Traumatic Brain Injury Database. We had intended to use the National Spinal Cord Injury Data Base but new Health Information Portability and Protection Act (HIPPA) regulations precluded getting access to these data within the contract period. When they were available, we collected demographic data, employment data, and outcome data such as FIM scores.

The specific description of each database and the methodology used in the analysis can be found in the chapters of this report, which each provide an introduction, methods, results, and a discussion describing strengths and limitations of the analysis and salient implications.

As an overview, AMRPA provided data on 83,000 non-Medicare patients from rehab hospitals, 1997–2001, and contained ICD–9 data. The dataset overlaps heavily with the UDSMR data, but has additional cost and charge data; and is considered the most reliable source of cost data. These data allowed us to examine rehabilitation costs for selected diagnoses (e.g., fractures, sprain/strain, head injury, amputation, spinal cord injury, and other/unspecified).

The Uniform Data System for Medical Rehabilitation collects and redistributes data from rehabilitation hospitals nationwide for use in evaluating the effectiveness and efficiency of their rehabilitation programs. It provides the most comprehensive data available on rehabilitation patients across diagnostic categories. This report presents data on people injured in motor vehicle crashes, including demographics, type of injury, length of stay, primary payers, and post-injury rehabilitation circumstances such as employment status, living situation, and the Functional Independence Measure (Guide for the Uniform Data Set for Medical Rehabilitation, 1996) the most widely accepted functional assessment measure in use in the rehabilitation community. The FIM is an 18-item ordinal scale used with all diagnoses within a rehabilitation population.

Traumatic brain injuries are one of the most common injuries of admitted motor vehicle and motorcycle crash victims. The Traumatic Brain Injury Model System National Database collects data from 17 sites across the United States and provided one of the largest available samples of rehabilitation patients who incurred TBI: 2,071 cases in motor vehicle crashes and 227 in motorcycle crashes. These data were collected from 1990 to 2002. This data set provides important pre- and post-admission measures on selected TBI impacts on patient residence, marital status, and employment status.

The Hospital Discharge Survey census data were leased or purchased from the individual States, then “cleaned” and pooled by PIRE for other purposes. These data come from all hospitals in 19 States in 1997 and 2 additional States that lacked 1997 data but had 1998 data. Rehabilitation hospital discharges only were identifiable in a few of the States.

The HCUP National Inpatient Sample is a large, statistically representative sample of U.S. hospital discharges compiled by the Agency for Health Quality and Research (AHQR) of the U.S. Department of Health and Human Services (DHHS). Data on the file identified injury cause for 87 percent of injury admissions. The external cause codes used pinpoint motor vehicle and motorcycling injuries.

The Colorado Traumatic Brain Injury Registry and Follow-Up System provided data on incidence of new cases of TBI among Colorado residents. Although it does not contain rehabilitation data, it does provide information on the prevalence and severity of crash-related TBI for specific demographic groups, as well as different outcome variables.

In collecting rehabilitation costs we found information on longer-term outcomes that had not previously been reported and have reported it in this report.

New Prospective Payment System in 2002
As of January 2002, the Health Care Financing Administration (HCFA) (now the Center for Medicare and Medicaid Services, called CMS) adopted a new system of payment for Medicare using Prospective Payment System rates. Medicare has paid acute-care hospitals under a prospective payment system since 1983. Rehabilitation facilities, which provide extensive occupational, physical, and speech therapy services, were formerly exempt from that system. Starting in 2001, rehabilitation facilities are paid on a per-discharge basis based on the patient diagnoses, with hospitals paid more to care for patients with greater needs. For the most part, private insurers appear to have adopted the Medicare reimbursement rates for rehabilitation.

Chapter 3 provides estimates of the national costs of inpatient rehabilitation for motor vehicle and motorcycle injuries and who pays those costs. The methods used here are a model that can readily be applied to HCUP-NIS data to update the national cost estimate in the future.