Article Text

Download PDFPDF

Reducing black/white disparity: changes in injury mortality in the 15–24 year age group, United States, 1999–2005
  1. G Hu1,2,
  2. S P Baker2
  1. 1
    Department of Epidemiology and Health Statistics, School of Public Health, Central South University, Changsha, Hunan, China
  2. 2
    Center for Injury Research and Policy, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
  1. S P Baker, Center for Injury Research and Policy, Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, 624 North Broadway, MD 21205, USA; sbaker{at}


In 1999, the injury death rate for black males aged 15–24 in the USA was 80% greater than for white males: 148.5 vs 82.5/100 000, a difference of 66/100 000. Injury-specific changes between 1999 and 2005 in death rates for the 15–24 age group and in racial disparity were analysed using data from CDC’s WISQARS. The gap between black and white all-injury death rates in males was reduced by 24%, to a difference of 50/100 000, largely because of greater decreases in the rates for motor vehicle crashes and firearm suicide in young black men than young white men, and large increases in suicide by suffocation and unintentional poisoning in the latter. Among females, despite a reduction in the black/white gap in firearm homicide rates, the gap between the races in total injury rates changed from a small black excess to a higher rate in young white women, which was due primarily to greater increases in these white women than black women in unintentional poisoning and suicide by suffocation, and greater decreases in black women than white women in firearm suicide.

Statistics from

Injury has long been the leading cause of death for youths in the USA. In 2004, unintentional injury, homicide, and suicide were the three leading causes of death among people aged 15–24 and accounted for 74% of all deaths in this age group.1 Because rates of fatal injury are much higher in the black than the white population, reduction in racial disparity is one of the Healthy People 2010 Goals.2 Monitoring the progress in black/white disparity in fatal injury among young people is crucial for attaining the Healthy People 2010 Goals and improving the health of youths in the USA. Although some studies have addressed racial disparity for certain causes of injury among youth,39 we found no study that examined the most recent trends in black/white disparity. The objective of this study was to analyze changes in injury mortality and in black/white disparity in fatal injury rates for youths aged 15–24, so as to provide a basis for eliminating black/white inequalities in injuries.



Data came from WISQARS (Web-based Injury Statistics Query and Reporting System) mortality reports, which provide the numbers and rates per 100 000 of injury-related deaths in the USA.10 The mortality data came from annual data files of the National Center for Heath Statistics (NCHS), CDC, and are derived from multiple cause of death data. Population data came from the Bureau of the Census. WISQARS provides data on deaths according to cause (mechanism) and intent of injury by sex, age, race, and state beginning in 1981 and currently available through 2005. Age-adjusted rates are calculated by the direct method standardized to the total US population. We selected 2000 as the standard year.


To analyze recent changes in black/white disparities, we first plotted the 1999–2005 trends in total injuries by sex and race. Then, the 10 most common causes of injury were selected for decomposing the change from 1999 to 2005 in the total injury rate and searching for changes in specific causes of injury. The 10 causes were determined from WISQARS’ Leading Causes of Death Reports of 1999.11 Percentage changes in death rates/100 000 population and Poisson regression were used to examine the change for each race/sex group between 1999 and 2005; p<0.05 was selected as the significant level. Percentage change in mortality was calculated as: ((death rate in 2005) − (death rate in 1999))/(death rate in 1999) × 100%. Black/white disparity was measured using the difference in mortality between black and white people: (death rate of black people) − (death rate of white people). The statistical significance of black/white disparity was tested using Poisson regression. The percentage change in the difference/100 000—((difference in 2005) − (difference in 1999))/(difference in 1999) × 100%—was used to measure the change in black/white disparity.


Between 1999 and 2005, the total injury death rate for white youths aged 15–24 increased by 5%, from 55.7 to 58.6/100 000, and the rate for black youths in the same age group decreased by 7%, from 87.5 to 81.8. The combined effect of the increase in the white rate and the decrease in the black rate was a narrowing of the difference between the two races by 27%—from 31.8/100 000 in 1999 to 23.2/100 000 in 2005.

Examination by sex revealed that the black/white difference in the death rate from all injury causes decreased markedly for young men between 1999 and 2005, because the white rate increased while the black rate decreased (fig 1A and table 1). Similar changes in young women led to a reversal from a slightly higher injury rate in the black population in 1999 to a substantially higher rate in the white population in 2005, resulting in an enlarged difference between the two races (fig 1B and table 2).

Figure 1 Trends in total injury mortality by sex and race (USA; 1999–2005; ages 15–24). (A) Males; (B) females.
Table 1 Changes in black and white fatal injury rates and mortality gaps (males aged 15–24; USA; 1999 vs 2005)
Table 2 Changes in black and white fatal injury rates and mortality gaps (females aged 15–24; USA; 1999 vs 2005)

In 1999, the injury death rate for young black men aged 15–24 in the USA was 80% higher than for young white men: –148.5 vs 82.5/100 000, a difference of 66/100 000 (table 1). Between 1999 and 2005, the all-injury death rate increased by 7% in white youth and decreased by 7% in black youth; as a result, the injury death rate for black males was only 57% higher than for white males: 137.8 vs 87.9/100 000. This represented a 24% reduction in the difference between the two races, to a difference of 50/100 000. The main components of this reduction in racial disparity were a greater decrease in death rates from motor vehicle crashes and firearm suicide in young black men than in young white men, as well as greater increases in white youth than in black youth in unintentional poisoning and suicide by suffocation. The greatest proportional change in injury mortality among males was in the white rate for unintentional poisoning, which more than doubled, whereas the black rate decreased. Suicide by suffocation (typically, hanging) increased by almost half in young white men. Fatal motor vehicle crashes in young black men decreased by 9%. Unintentional and suicidal firearm mortality decreased in both races.

Among females (table 2), between 1999 and 2005, the total injury death rate for young white women did not change significantly, whereas there was a 11% decrease for their black counterparts; as a result, the difference between the two races in the overall injury rate changed from a black excess of 0.8/100 000 to a white excess of 2.6/100 000, creating a net increase of 3.4/100 000. This shift was caused by greater increases in unintentional poisoning and suicidal suffocation in white girls than black girls, and greater decreases in firearm suicide in black girls than white girls. The largest increases were in rates of unintentional poisoning (which increased by 100% in this black population and by 150% in the white population) and a 100% increase in the rate of suicide by hanging/suffocation in young white women.


“…And then quite suddenly (just like us),

one got better and the other got wuss.”

A. A. Milne, Twice Times, 1927

The above quote described the change in behavior of two bears, whereas the results of our analyses reflect behavioral changes among adolescents and young adults, with some death rates for the two races moving in opposite directions during the 7 years studied.

Our primary objective was to examine the trend in black/white disparity in injury fatality rates in the 15–24-year age group. The best news to be gleaned from this analysis is that the all-injury death rate for black people aged 15–24 decreased between 1999 and 2005, by a significant 7%. This progress is overshadowed by the bad news of a 5% increase in the injury death rate for white people in the same age group. Thus, the reduction in the disparity between young black and white people is the combined effect of both trends—although the goal of reduced racial disparity no doubt was based on the hope for reductions of injury rates in all races, with the greatest reductions where most needed. The analysis illustrates that a major component of reduced racial disparity can be an undesired trend—that is, an increase in death rates for the white population. Similarly, Harper and colleagues12 point to increasing mortality among middle-aged white people in 1993–2003 as an important contributor to the decline in the black/white disparity in life expectancy.

The reductions in firearm-related deaths (from homicide, suicide, and unintentional injury) for both races are encouraging, as is the decrease in road traffic deaths for black youth. This progress probably reflects a variety of preventive efforts as well as demographic and economic changes.13 Among effective preventive efforts, graduated driver licensure has been shown to cause substantial reductions in the crash rates of novice drivers.14 Societal changes have included some improvements in economic status (rates of virtually all categories of fatal injury are lower where per capita income is higher).15

Firearm suicide has been declining in the 15–24 age group for more than 25 years.16 Firearm homicide, on the other hand, has followed an erratic course: Fingerhut and colleagues17 analyzed trends in firearm homicide rates for the 15–24 age group during 1987–1995 and reported an increase in rates followed by a decrease, changes that occurred across all urban/rural levels. The absence of substantial change in the firearm homicide rate for both black and white males aged 15–24 during 1999–2005 and the decline in females is encouraging, and especially noteworthy in contrast with the increase in firearm homicide among men aged 25–39.18

Key points

  • Between 1999 and 2005, the difference in total injury death rate between black and white males aged 15–24 decreased by 24%; in females, the difference was reversed from a slightly higher rate in black females to a higher rate in white females.

  • Black/white disparity in the total injury rate was reduced by a combination of decreases in injury rates for black people and increases in rates for white people.

  • Death rates from intentional poisoning more than doubled in white males and females and doubled in black females between 1999 and 2005.

The greatest increases seen among both male and female youths occurred in fatal unintentional poisoning and suicide by suffocation/hanging. The reasons underlying these increases are not known, but a CDC report suggests that the similarity between the age groups 20–29 and 45–54 in trends related to these mechanisms of fatal injury points to a possible increase in shared risk factors such as drug abuse; the report further suggests the potential value of prevention programs based on shared risk factors.19

Unintentional fatal poisoning increased most in people aged 15–2420 and is a special cause for concern because it doubled in both white males and white females. In white males, doubling occurred only at the ages 15–24, whereas in white females the doubling occurred in each 5-year age group between 15 and 69.10 A remarkable finding is the absence of an increase in black males aged 15–24. Research exploring the reasons for this racial difference might contribute to both our understanding of the increase in poisoning and lowering the rate in white youths. Similarly, our finding that unintentional and suicidal poisoning of young black females did not increase, despite increases among their white counterparts, may be worth exploring for possible clues to prevention. Recent increases in specific substances involved in fatal poisoning clearly deserve attention; for example, Fingerhut21 reports that the largest increase in fatal methadone poisoning between 1999 and 2004 was in young persons aged 15–24; the rate in 2004 was 11 times that in 1999; the increase was primarily in white people, whose rate for 2003–2004 was 6 times the black rate (1.57 vs 0.24/100 000) (personal communication from L Fingerhut, NCHS; 4 January 2008). Use of narrative text from death certificates may shed further light on the characteristics of poisoning deaths, both unintentional and of unknown or suicidal intent, and contribute to preventing these fatalities.22

The findings of this study are limited by the changes in E-codes in 1999, from Version 9 of the International Classification of Diseases (ICD) to Version 10. Because ICD-10 differs greatly from the previous version, we did not analyze long-term trends among adolescents, thereby possibly excluding information relevant to injury control in adolescents. The second limitation is that the study did not include non-fatal injury. Unlike rates for fatal injuries, rates for non-fatal injuries provided by WISQARS are estimated on the basis of a sample, and for some specific causes are not stable and robust. It would nevertheless be informative to perform comparable analyses on non-fatal injuries.

In conclusion, the discrepancy between black and white people aged 15–24 in overall injury rates has been reduced in recent years, primarily by a combination of decreases in injury rates in black people and increases in rates in white people, especially from unintentional poisoning. Exploration of the differences between the two races in recent trends may offer clues to prevention.


This research was supported by the Center for Injury Research and Prevention, Centers for Disease Control and Prevention (Grant CCR302486).


Supplementary materials

Related Data


  • Competing interests: None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.