DEPARTMENT OF TRANSPORTATION

NATIONAL HIGHWAY TRAFFIC SAFETY ADMINISTRATION

TECHNICAL SUMMARY

On July 1, 2000, Florida repealed the legal requirement that all motorcyclists wear protective helmets. Instead, state law now requires helmet use only by riders under the age of 21, or by older riders who do not have a minimum of $10,000 medical insurance coverage

The Florida Motorcycle Helmet Law Change

Helmet Use -- Observational Surveys

A Florida helmet use observation survey carried out in 1998 before the helmet law change showed that virtually all observed riders were wearing helmets. However, only 59 percent of the observed sample wore compliant helmets (headgear that meets FMVSS No. 218) while 40 percent were wearing noncompliant helmets (headgear that does not meet FMVSS No. 218). These figures compare to 84 percent compliant and 15 percent noncompliant observed in a 1993 survey suggesting that noncompliant helmet use was increasing over time. Following weighting, the 1998 survey results yielded estimated statewide helmet use of 65 percent compliant helmets and 35 percent noncompliant helmets.

A post law change survey, done in 2002, found 47 percent compliant helmet use, 6 percent noncompliant helmet use and 47 percent no helmet use. These results indicate that use of compliant helmets has declined following the law change, while wearing noncompliant helmets has largely been abandoned.

Helmet Use – Crash Reports

Among the 515 motorcyclists killed in Florida in traffic crashes in the three years prior to the helmet law change (1997-1999), 9 percent were recorded in FARS as not wearing a helmet. In the three years following the law change (2001-2003), 61 percent of the 933 fatally injured motorcyclists were reported being unhelmeted. In 1997-1999, there were 35 motorcyclists under the age of 21 killed in Florida. Of these, 26 percent were not helmeted. In 2001-2003, 101 motorcyclists under age 21 were killed, with 45 percent of them being unhelmeted.

Excluding cases where helmet use was not recorded (less than 7 percent of the cases), 27 percent of all motorcyclists involved in crashes in 1999, of all degrees of severity, were recorded in the Florida crash database as being unhelmeted. In 2001, the figure was 51 percent. Among motorcyclists who sustained incapacitating injury, 21 percent of those involved in 1999 crashes were unhelmeted while 50 percent of those involved in year 2001 crashes were not wearing helmets.

Among riders under the age of 21, 40 percent of those involved in crashes in 1999 were unhelmeted. In 2001, the figure was 49 percent. For those sustaining incapacitating injury, 35 percent of those in 1999 crashes were not helmeted, while in 2001 49 percent were not helmeted.

Motorcyclist Fatalities

While Florida's all-rider helmet law was repealed in July 2000, there was a substantial increase in motorcyclists killed in Florida beginning in the first 6 months of the year, before the effective date of the law. Fatalities in the two years following the law change (2001-2002, N=575) were 71 percent greater than the 337 fatalities that occurred in 1998-1999, compared to an increase of 37 percent for the nation as a whole (4,560 to 6,227). Fatalities in Florida per 10,000 registered motorcycles increased 21 percent compared to 13 percent nationally for the two years before and after the law change. There was an annual average of 181 motorcyclists killed in Florida in the 30 months before the law change, and an annual average of 280 in the 30 months after the law change; a 55 percent increase. Registrations increased an average 33.7 percent (219,486 to 293,393) in this time period.

The expected average annual number of motorcycle fatalities as a result of the increase in registrations was 242 (181 x 1.337 = 242). The actual number who died was 301 in 2002, 59 more motorcycle fatalities than expected as a result of increased registrations alone (a 24 percent increase).

Motorcyclist fatalities in Florida have continued to increase. In the three years after the law change (2001-2003), 933 motorcyclists were killed, 81 percent more than the 515 motorcyclists who were killed in 1997-1999. The actual number who died in 2003 was 358.

While the reduction in helmet use observed after the law was repealed would be expected to result in more non-helmeted fatalities, the actual increase was above what would be expected based solely on the decrease in helmet use. Non-helmeted motorcyclists who were killed in Florida increased from 15 (9 percent) in 1998, when observed helmet use was close to 100 percent, to 198 (66 percent) of the total motorcycle fatalities in 2002, an increase of over thirteen times. When the increase in motorcycle registrations after the law was changed is taken into account, the non-helmeted fatality rate per 10,000 registered motorcycles increased from 0.7 fatalities in 1998 to 6.1 in 2002. Helmeted motorcyclist fatalities, on the other hand, fell from 7.6 per 10,000 registered motorcycles in 1998 to 3.2 in 2002.

Autoregressive Integrated Moving Average (ARIMA) models were used to examine the relationship between the change in Florida's motorcycle helmet law and motorcyclist fatalities. The date of the law change was used as an intervention point in the time series. Data on motorcyclist fatalities in Georgia were employed as a comparison. Georgia was selected based on proximity to Florida and the fact that it had an all-rider helmet law in effect for the entire time period.

Monthly fatalities in Florida and Georgia, before and after the intervention point (July 2000), were modeled in the presence of the annual number of motorcycle registrations in each State. This was done to control for the extent to which changes in fatalities were associated with changes in motorcycle registrations.

The Florida intervention was statistically significant such that there was an average 9.1 increase in the number of monthly motorcyclist fatalities following the law change ( p <.001). The specified final ARIMA parameters were also significant. There was no statistically significant change in the fatality rate following the same intervention date for Georgia. Change in annual motorcycle registrations was not a statistically significant parameter in the final time series model. At the time of this analysis, motorcycle registration data were available through the year 2002, making December 2002 the last point in the time series examined. In 2003, 358 motorcyclists were killed in Florida suggesting that the upward trend has continued.

Based on the available evidence the increase in motorcycle fatalities that occurred after the Florida motorcycle helmet law was repealed is due in part to the reduced use of helmets. Our analysis shows that is the case despite the pre-existing trend of increasing fatalities, the increase in fatalities associated with increased exposure (measured by registrations), the increase in fatalities that occurred in the first six months of 2000 (before the helmet law repeal became effective), and the likely contribution of a demographic shift in motorcycle ridership.

Motorcyclist Injuries

All Motorcycle Crashes and Non-Fatal Injuries, 1994-2001

            A-Incapacitating Injury, B-Evident Injury, C-Possible Injury

            In the first full year following the law change (2001), there were 1,890 motorcyclists who sustained incapacitating injury and 3,886 who sustained lesser injury. These figures are 32 percent and 28 percent higher, respectively, than the comparable figures in 1999, but less when the increase in registrations is taken into account. Injuries per 10,000 registered motorcycles increased in 2000, but decreased in 2001. Some of the motorcyclists coded “C—Possible Injury” by the police officer may not have sought medical treatment. Although the injury rate per registered motorcycle in 2001 is less than the rate in 1999, the previous downward trend of non-fatal injuries per registered motorcycle appears to have slowed following the law change period.

            Hospital Discharge Data

The Hospital Discharge database maintained by the Florida Agency for Health Care Administration shows that in the 30 months immediately following the helmet law change, there were 4,986 motorcyclists admitted to hospitals for treatment, a figure 40 percent greater than the 3,567 admissions during the 30 months just before the law change. Head injury admissions increased by more than 80 percent.

Total gross costs charged to hospital admitted motorcyclists with head, brain or skull injury more than doubled from $21 million to $50 million; the average case cost rose by almost $10,000; the median patient cost increased by almost $4,000; and the range of costs also increased. Adjusted for inflation, total acute care hospital costs rose from $21 million to $44 million and the average cost per case rose from $34,518 to $39,877 in the 30 months after the law change. In the post law change period, 25 percent of the head, brain, skull injured admitted motorcyclists were charged approximately $12,000 or less, while the remaining 75 percent of patients were charged more than this amount. That is, less than one-quarter of the injured would be covered by the $10,000 medical insurance requirement for those who chose not to use helmets. The hospital discharge data indicate that in the post law change period, approximately 63 percent of admitted motorcyclists were covered by commercial insurance ($31 million), 16 percent were classified as “self pay” because they were under insured or uninsured ($8 million), while the remaining 21 percent had their costs ($10.5 million) billed to charitable and public sources (e.g., Medicaid).

In the 30 months before the helmet law change, 52 motorcyclists with head-brain-skull principal injury died after admission to an acute care hospital. The average treatment cost for these cases was $48,126. In the 30 months after the law change, 115 motorcyclists died following admission. Inflation adjusted costs for these cases averaged $52,450.

In 1998 and 1999, the hospital charges for head-brain-skull principal injury cases per 10,000 registered motorcycles were $311,549 and $428,347 respectively. The comparable figures for 2001 and 2002 were $605,854 and $610,386, adjusted for inflation.

The effect of the motorcycle helmet law repeal on injuries is somewhat less clear than the situation for fatalities. Two sources of data were available: police motor vehicle crash reports, which show an increase in injuries, but a small decline in injury rates per 10,000 registrations; and hospital discharge data that show large increases in hospital admissions and admissions for head injuries. The weight of the evidence indicates that the repeal of the helmet law was associated with a slowing of the existing downward trend in injury rates, with an increase in head injuries. The cost data show that total acute care cost more than doubled. As with fatalities, increased exposure (registrations) cannot account for these changes.

The Arkansas, Kentucky, Louisiana, and Texas Law Changes

Arkansas and Texas repealed all-rider helmet laws in 1997 while Kentucky did so in 1998 and Louisiana in 1999. Comparing motorcyclist fatalities in the three full years after the law changes indicates that fatalities increased by 130 percent in Louisiana, by 99 percent in Kentucky, by 52 percent in Texas, and by 23 percent in Arkansas.

Time Series Analyses

ARIMA modeling was conducted for these States. As with Florida, analyses explored monthly fatalities over a nine- year period (1994-2002), controlling for the annual number of motorcycles registered in each State. The intervention effects were statistically significant for Kentucky, Louisiana, and Texas, but not Arkansas.

For Kentucky, there was a statistically significant effect of the intervention on fatalities such that there was an average increase of 1.3 motorcycle fatalities per month ( p = .001). Registrations did not enter the model as a statistically significant parameter.

Louisiana also showed a significant effect of intervention. For this State, the repeal of the law raised motorcycle fatalities by an average of 2.6 per month ( p < .001). There was no statistically significant effect of registrations on fatalities.

The results in Texas demonstrated statistically significant effects of both intervention (fatalities increasing an average of 3.7 per month; p = .001) and registrations (fatalities increasing by 1 as registrations increase by 10,000, p < .001) on the number of motorcycle fatalities. That is, while there was an effect of registrations on a change in the rate of fatalities, there was also a separate effect of the law repeal.

Limitations of the Study

National data suggest that as motorcycle registrations increase, motorcyclists' deaths and injuries increase. Conversely, when registrations decline, fatalities and injuries decline. In Florida, motorcycle registrations increased substantially following repeal of the all-rider helmet law, an outcome similar to that seen in the other States that repealed helmet laws in recent years. It is likely that some of the increases in motorcyclist fatalities were due to increased ridership. However, the analyses just described show that increases in motorcycle registrations alone do not account for the magnitude of the increases in fatalities. Other factors than the decrease in helmet use that may have contributed to the fatality increase are alcohol use, speed, and a demographic shift in motorcycle ridership. Unfortunately, the available data do not allow for a precise determination of the extent to which these various factors contributed to the increase in fatalities.

Nationally, motorcycle vehicle miles of travel (VMT) increased gradually throughout the 1990s, but decreased in 2001 and 2002. The VMT measure, provided by the Federal Highway Administration, is regarded as a good indicator of trends year to year, but cannot be broken down reliably to the individual State level for motorcycles. Nationally between 1998 and 2002, motorcycle registrations increased by approximately 29 percent. In 1998, the average motorcycle traveled 2,645 miles, while in 2002 this figure had declined to 1,909 miles. The extent to which this effect occurs at the individual State level is not knowable from existing data sources.

Summary

The effects of Florida's repeal of its all-rider motorcycle helmet use law are similar to those seen in the other States that have repealed such laws in recent years. Based on these findings, it is reasonable to conclude that the following are likely outcomes in a State considering elimination of an all-rider helmet law:

•  Helmet use will decline markedly, from virtually full daytime compliance to voluntary use by about 50 percent of riders;

•  Helmet use likely will decline among all riders regardless of restrictions remaining in the law (use required by young riders, those without insurance) because of enforceability factors.

•  Motorcycle registrations will increase. This, in turn, will contribute to an increase in motorcycle crashes of all degrees of severity.

•  Motorcyclist fatalities will increase significantly, typically by 50 to 100 percent comparing the years following the law change with the years immediately before repeal. The fatality rate per registered motorcycle will also increase.

The Florida results also showed that non-fatal serious injuries increase more than lesser injuries following law repeal. Injured motorcyclists' hospital admissions increased by 40 percent following the law change. Admissions for head-brain-skull injuries increased by more than 80 percent following the law change. Total gross treatment costs for these cases more than doubled and the cost per case also increased substantially. Fewer than 25 percent of hospital admitted motorcyclists for head-brain-skull injuries had treatment costs under $10,000, indicating that the law's medical insurance provision is largely inadequate to cover the costs incurred. Only about two-thirds of admitted motorcyclists have medical insurance.

The Florida law continues to require helmet use by riders under the age of 21. The data indicate that this provision is not being observed. The number of under age 21 motorcyclists killed in Florida in the two years after the law change nearly tripled, compared to the two years before the change. Almost one-half of the post law change victims were not helmeted compared to about 26 percent before the law change, an increase of 188 percent. The number of young motorcyclists involved in crashes of lesser severity increased by about 47 percent.

HS Form 321

July 1974