1 Large increases in motorcycle-related head injury deaths

Large increases in motorcycle-related head injury deaths, hospitalizations, and hospital
charges following the repeal of Pennsylvania’s mandatory motorcycle helmet law
Word count abstract: 80
Word count text: 1,225
Tables: 2
1
Abstract
To evaluate the 2003 repeal of Pennsylvania’s motorcycle helmet law, we assessed
changes in helmet use and compared motorcycle-related head injuries with nonhead
injuries from 2001-2002 to 2004-2005. Helmet use among riders in crashes decreased
from 82% to 58%. Head injury deaths increased 66%; nonhead injury deaths increased
25%. Motorcycle head injury hospitalizations increased 78% compared with 28% for
nonhead injury hospitalizations. Helmet law repeals jeopardize motorcycle riders. Until
repeals are reversed, states need voluntary strategies to increase helmet use.
2
Introduction
In 1975, Congress stopped requiring states to enact mandatory universal motorcycle
helmet use laws for receipt of federal highway construction funds. By 1980, 27 states
had repealed mandatory helmet laws. Despite consistent evidence that repeals led to
increases in motorcycle-related deaths and injuries,1 several more states followed suit in
the late 1990s and early 2000s. Most recently, in 2003, Pennsylvania repealed its
universal helmet law, now requiring helmets only for riders < 21 years or those with <2
years of riding experience who have not completed a safety program.
In 2006, the Pennsylvania Legislative Budget and Finance Committee released a report
describing a 33% increase in trauma center admission rates for motorcycle-related head
injuries but only an 11% increase in motorcycle-related death rates from 2001-2002 to
2004-2005.2 Media coverage focused on the small post-repeal change in death rates
rather than the full injury picture. To supplement the report, we examined statewide
crash, hospitalization, and death data. To better control for nonhelmet factors affecting
the number of motorcycle-related injuries, such as changes in general motorcycle use,
miles traveled, and weather, we compared changes in head injury rates with changes in
nonhead injury rates, assuming that nonhelmet factors affect head and nonhead injuries
equally.
Methods
The number of Pennsylvania motorcycle registrations and the number of motorcycle
riders, including drivers and passengers, involved in “reportable” crashes were obtained
from the Department of Transportation (PennDOT) for years 2001-2005. “Reportable”
3
crashes include those involving an injury or those in which at least one vehicle is towed.
Helmet status for riders in crashes was categorized as used, used improperly, none, or
unknown. “Motorcycle” in this paper includes motorized pedacycles and motor driven
cycles.
The number of in-state motorcycle traffic-related deaths among Pennsylvania residents
was obtained from Pennsylvania Department of Health (DOH). These deaths were
defined as those with an ICD-10 code of V20-V28 (.3-.9) or V29 (.4-.9) as underlying
cause of death on the death certificate. Motorcycle-related deaths were categorized as
head injuries if one of the cause-of-death codes included a traumatic brain injury (TBI)
diagnosis code.3 The number of Pennsylvania residents hospitalized at acute care
hospitals in Pennsylvania for motorcycle traffic–related injuries was obtained from DOH
using data compiled by the Pennsylvania Health Care Cost Containment Council.
Inpatient records with ICD-9 codes of E810-819 (.2, .3) were selected. Motorcyclerelated hospitalizations were classified as head injuries if one of the first five ICD-9
diagnosis codes included a TBI code.3 DOH also provided aggregate hospital charges
and summary disposition information for motorcycle-related hospitalizations. Drivers
and riders of all ages were included in the analyses to evaluate the overall effect of the
repeal.
The number of riders in motorcycle crashes per 10,000 motorcycle registrations was
calculated for the two years before (2001-2002) and after (2004-2005) the September
2003 repeal. The number of motorcycle-related fatalities and hospitalizations per 10,000
4
motorcycle registrations was calculated for the same two time periods. Hospital charges
were converted to 2005 dollars using the Consumer Price Index for “hospital and related
services.” 4 The percentage change from 2001-2002 to 2004-2005 was calculated for all
injury indices. Differences in rates, assessed with the z statistic, were considered
statistically significant at the 0.05 level.5 The pre- and post-repeal difference in the
percentage of helmet wearers, defined as those categorized as “used” divided by those
with known helmet status, was assessed using the chi-square test.
Results
Helmet use by motorcycle riders involved in reportable crashes in Pennsylvania declined
from 82% in 2001-2002 to 58% after the repeal (2004-2005) (p<.0001).
Both the number of motorcycle registrations and the number of motorcycle riders in
reportable crashes increased from 2001-2002 to 2004-2005 (Table 1). The number of
riders in reportable crashes per 10,000 registrations, however, did not change
significantly (Table 1).
The number of Pennsylvania residents dying in Pennsylvania from motorcycle-related
injuries increased 40% after repeal (Table 1). The number of head injury deaths
increased 66%, whereas the number of nonhead injury deaths increased 25% (Table 2).
The rate of motorcycle-related head injury deaths per 10,000 registrations increased 32%
(borderline significance; p=0.045), but there was no significant change (-0.1%) in
nonhead injury deaths per 10,000 registrations (Table 2).
5
The number of acute-care motorcycle-related hospitalizations for Pennsylvania residents
increased 43% from 2001-2002 to 2004-2005 (Table 1). Motorcycle-related head injury
hospitalizations increased 78%, whereas nonhead injury hospitalizations increased 28%
(Table 2). The hospitalization rate per 10,000 registrations increased significantly (42%)
for head injuries but not for nonhead injuries (2%) (Table 2).
Total acute care hospital charges (in 2005 dollars) for motorcycle-related head injuries
increased 132%, from $53,501,923 in 2001-2002 to $124,236,056 in 2004-2005,
compared with a 69% increase for nonhead injuries. The percentage increase in the mean
charge per motorcycle-related hospitalization was nearly identical for head and nonhead
injuries (32% and 31%, respectively). The larger percentage increase in total charges for
head injuries compared with nonhead injuries reflects the larger percentage increase in
the number of head-injury hospitalizations. The number of head-injured hospitalized
motorcyclists requiring further care at other facilities, largely for rehabilitation and longterm care, increased 87% (165 patients in 2001-02 and 309 in 2004-05), compared with a
16% increase for nonhead-injured hospitalized motorcyclists. Charges for rehabilitation
and long-term care were not available.
Discussion
As in many states, the number of motorcycle registrations and the number of reported
crashes increased in Pennsylvania following repeal of the mandatory helmet law. The
rate of crashes per 10,000 motorcycle registrations, however, did not change
6
significantly, indicating little change in driving habits, road conditions, or risk behavior
of motorcyclists. Consistent with other studies, our analysis shows increases in the
number and rate of deaths and serious injuries following the repeal of a mandatory
universal motorcycle helmet law. 6-14
Our study is important for two main reasons. First, we use population-based hospital
discharge data compiled from all acute care hospitals in the state, whereas most previous
studies of post-repeal changes in motorcycle-related hospitalizations include data only
from selected trauma centers. Second, we attempt to control for nonhelmet factors by
comparing changes in head injuries to nonhead injuries, assuming that nonhelmet factors,
such as miles traveled, weather, and driving practices, generally affect both head and
nonhead injuries equally. The large post-repeal increases in head injuries relative to
nonhead injuries, both for statewide deaths and hospital discharges, indicate that lower
helmet use was most likely responsible.
Limitations
To identify head injuries, we used ICD codes on death certificates and hospital discharge
data, but these are not routinely assessed for validity and reliability. We assumed that
coding practices did not change during the study period and that trends were not affected.
Also, vital statistics data have been shown to underestimate the number of motorcycle
deaths;15again, we assumed trends were not affected.
7
The hospital charges presented here have two shortcomings: one, they are not equivalent
to costs; two, they do not include physician costs, rehabilitation costs, or nonmedical
costs including loss of productivity, and as such they greatly underestimate the financial
burden of motorcycle-related injuries.
Conclusion
These data strongly suggest that Pennsylvania’s mandatory helmet law was effective in
preventing traumatic brain injury, given that its repeal led to disproportionate increases in
head injuries. Data alone, however, are not sufficient to reverse helmet law repeals;
many states maintain repeals despite multiple studies showing increases in deaths,
injuries, and costs. Until re-instatement of life-saving mandatory helmet policies,
voluntary helmet use programs should be developed and evaluated.
8
References
1.
General Accounting Office. Highway safety: motorcycle helmet laws save lives
and reduce costs to society. Washington, DC: US General Accounting Office; 1991.
Report No GAO/RCED-91-170.
2.
Legislative Budget and Finance Committee. Motorcyclist injuries and fatalities
since the 2003 repeal of the mandatory helmet law: Harrisburg, PA: Pennsylvania
General Assembly; 2006.
3.
CDC. Rates of hospitalization related to traumatic brain injury -- nine states,
2003. MMWR 2007;56:167-70.
4.
US Department of Labor, Bureau of Labor Statistics, Consumer Price Index.
Archived consumer price index detailed report information. Available at: http://
www.bls.gov/cpi/cpi_dr.htm, Accessed November 26, 2007.
5.
Hoyert DL, Heron MP, Murphy SL, Kung HC. Deaths: final data for 2003. Natl
Vital Stat Rep 2006;54:1-120.
6.
Bledsoe GH, Li G. Trends in Arkansas motorcycle trauma after helmet law
repeal. South Med J 2005;98:436-40.
7.
Bledsoe GH, Schexnayder SM, Carey MJ, et al. The negative impact of the repeal
of the Arkansas motorcycle helmet law. J Trauma 2002;53:1078-86
8.
Ho EL, Haydel MJ. Louisiana motorcycle fatalities linked to statewide helmet law
repeal. J La State Med Soc;156 :151-2.
9.
Hotz GA, Cohn SM, Popkin C, et al. The impact of a repealed motorcycle helmet
law in Miami-Dade County. J Trauma 2002;52:469-74.
9
10.
Muller A. Florida's motorcycle helmet law repeal and fatality rates. Am J Public
Health 2004;94:556-8.
11.
Preusser DF, Hedlund JH, Ulmer RG. Evaluation of motorcycle helmet law repeal
in Arkansas and Texas. Washington, DC: US Department of Transportation; 2000.
Report No. DOT HS 809 131.
12.
Ulmer RG, Preusser DF. Evaluation of the repeal of motorcycle helmet laws in
Kentucky and Louisiana. Washington, DC: US Department of Transportation; 2003.
Report No. DOT HS 809 530.
13.
Ulmer RG Shabanova Northrup V. Evaluation of the repeal of the all-rider
motorcycle helmet law in Florida. Washington, DC: US Department of Transportation;
2005. Report No. DOT HS 509 849.
14.
National Highway Traffic Safety Administration. Evaluation of the repeal of
motorcycle helmet laws. Ann Emerg Med 2001;37:229-30.
15.
Sosin DM, Sacks JJ, Holmgreen P. Head injury-associated deaths from
motorcycle crashes: relationship to helmet-use laws. JAMA 1990;264:2395-9.
10
Table 1. Number of motorcycle registrations, motorcycle riders in reportable crashes, motorcycle related deaths and hospitalizations,
and corresponding rates per 10,000 motorcycle registrations, Pennsylvania, 2001-2005.
Year
2001
2002
2003
2004
2005
Study
periods:
2001-02
2004-05
% change
p-value*
Number of
registrations
Number
of riders
in
crashes
Riders in
crashes per
10,000
registrations
Number
of
deaths
Deaths per
10,000
registrations
Number of
motorcycle
injury
hospitalizations
Hospitalizations
per 10,000
registrations
237,276
248,775
267,826
291,015
318,283
3,364
3,455
3,513
4,133
4,625
141.8
138.9
131.2
142.0
145.3
106
116
138
133
177
4.5
4.7
5.2
4.6
5.6
1,226
1,319
1,354
1,645
1,986
51.7
53.0
50.6
56.5
62.4
486,051
609,298
25.4%
6,819
8,758
28.4%
140.3
143.7
2.5%
0.133
222
310
39.6%
4.6
5.1
11.4%
0.217
2,545
3,631
42.7%
52.4
59.6
13.8%
<0.001
*2-sided p-value corresponds to z statistic used to test for a difference in rates in 2004-2005 compared with 2001-2002
11
Table 2. Number of head injury and nonhead injury motorcycle-related deaths, hospitalizations, and corresponding rates per 10,000
motorcycle registrations, Pennsylvania, 2001-2005.
Year
Number of
motorcycle
registrations
2001
2002
2003
2004
2005
Study
periods:
2001-02
2004-05
% change
p-value*
Number of
motorcycle
head injury
deaths
Motorcycle
head injury
deaths per
10,000
registrations
Number of
motorcycle
nonhead
injury
deaths
Motorcycle
nonhead
injury deaths
per 10,000
registrations
Number of
motorcycle
head injury
hospitalizations
Motorcycle
head injury
hospitalizations
per 10,000
registrations
Number of
nonhead injury
hospitalizations
Nonhead injury
hospitalizations
per 10,000
registrations
237,276
248,775
267,826
291,015
318,283
31
48
44
51
80
1.3
1.9
1.6
1.8
2.5
75
68
94
82
97
3.2
2.7
3.5
2.8
3.0
354
393
439
602
730
14.9
15.8
16.4
20.7
22.9
872
926
915
1,043
1,256
36.8
37.2
34.2
35.8
39.5
486,051
609,298
25.4%
79
131
65.8%
1.6
2.2
32.3%
0.045
143
179
25.2%
2.9
2.9
-0.1%
0.990
747
1,332
78.3%
15.4
21.9
42.2%
<0.001
1,798
2,299
27.9%
37.0
37.7
2.0%
0.529
*2-sided p-value corresponds to the z statistic used to test for a difference in rates in 2004-2005 compared with 2001-2002
12