Surveillance for Injuries and Violence Among Older Adults

Judy A. Stevens, M.S., M.P.H., Ph.D.¹
La Mar Hasbrouck, M.D., M.P.H.^2,3
Tonji M. Durant, M.A., Ph.D.^2,3
Ann M. Dellinger, M.P.H., Ph.D.¹
Prabhansu K. Batabyal, M.B.A., M.P.H.⁴
Alexander E. Crosby, M.D., M.P.H.²
Balarami R. Valluru, M.S.⁴
Marcie-jo Kresnow, M.S.⁴
Janet L. Guerrero, M.S.⁴

Problem/Condition: Injuries and violence are major causes of disability and death among adults aged greater than or equal to 65 years in the United States. Injuries impair older adults' quality of life and result in billions of dollars in health-care expenditures each year.

Reporting Period: This report reviews 1987-1996 data regarding fall-related deaths, 1988-1996 data on hospitalizations for hip fracture, 1990-1997 data regarding motor vehicle-related injuries, 1990-1996 data on suicides, and 1987-1996 data on homicides.

Description of Systems: Data on fall-related deaths, suicides, and homicides are from the National Center for Health Statistics annual mortality data tapes for 1987-1996. Homicide data are supplemented with information from the Federal Bureau of Investigation's Supplemental Homicide Reports for 1987-1996. Data on hospitalizations for hip fracture are from the 1988-1996 National Hospital Discharge Surveys. Information regarding motor vehicle-related injuries for 1990-1997 is from the National Highway Traffic Safety Administration's Fatality Analysis Reporting System and General Estimates System.

Results: Rates of fall-related deaths for older adults increased sharply with advancing age and were consistently higher among men in all age categories. Men were 22% more likely than women to sustain fatal falls. A trend of increasing rates of fall-related deaths was observed from 1987 through 1996 in the United States, although rates were consistently lower for women throughout this period. Rates of hospitalizations for hip fracture differed by age and were higher for white women than for other groups. Rates increased with advancing age for both sexes but were consistently higher for women in all age categories. U.S. hospitalization rates for hip fracture increased for women from 1988 through 1996 while the rates for men remained stable. Rates of motor vehicle-related injuries increased slightly from 1990 through 1997, and marked variations in state-specific death rates were observed; in most states, older men had death rates approximately twice those for older women. Although suicide rates remain higher among older adults than among any other age group, rates of suicide among adults aged greater than or equal to 65 years decreased 16% during the study period. Suicide rates among older adults varied by sex and age group. Homicide rates declined 36% among older adults. Homicide rates were highest for black men, followed by black women and white men; the homicide risk for blacks relative to whites decreased from 4.8 to 3.9 per 100,000 persons, indicating that the gap between rates for blacks and whites is closing. Half of the older homicide victims were killed by someone they knew.

Interpretation: The increase in rates of fall-related deaths and hip fracture hospitalizations from 1988 through 1996 might reflect a change in the proportion of adults aged greater than or equal to 85 years compared with those aged 65-84 years -- a change that results, in part, from reduced mortality from cardiovascular and other chronic diseases. Fall-related death rates might be higher among older men because they often have a higher prevalence of comorbid conditions than women of similar age. Racial differences in hospitalization rates might have some underlying biologic basis; the prevalence of osteoporosis, a condition that contributes to reduced bone mass and increased bone fragility, is greatest among older white women. Compared with whites aged greater than or equal to 65 years, blacks of comparable ages have greater bone mass and are less likely to sustain fall-related hip fractures. Additional studies are needed to determine why rates of motor vehicle-related injury have increased slightly among older adults and why these rates vary by state. Declining rates of suicide among older adults might be related to changes in the effect or type of risk factors traditionally observed in this age group. Research is needed to identify reasons for variations in suicide rates among older persons. Homicides among older adults declined. Possible explanations for this decrease include a reduction in victimization rates; advances in medical care for potentially fatal injuries; and benefits from increased public awareness of the problem. The black-white gap in homicide rates is closing because rates are decreasing among blacks.

Public Health Actions: Because older adults are the fastest growing age group in the U.S. population, the burden of injuries and violence will increase unless persons who have frequent contact with older adults are aware of the extent of the problem and prevention strategies are improved for this population. Interventions must be multifaceted, and older adults who are able to take an active role in reducing their risks should do so. To prevent fall-related injuries and hip fractures, interventions should promote behavioral and environmental changes as well as the development of safety products (e.g., protective hip pads and impact-absorbing floor materials). Interventions to prevent motor vehicle-related injuries should focus on improving the design of motor vehicles; changing the traffic and pedestrian environment to improve safety; and changing the behavior of older drivers, passengers, and pedestrians. In the area of suicide prevention, interventions are needed to educate health-care providers and caregivers about the extent of the problem and risk factors for suicide among older adults. Effective interventions to prevent homicide and violence against older adults must integrate a variety of disciplines (e.g., criminal justice, social services, education, community advocacy, and public health). The association between homicide, assault, and elder abuse is not well understood, but the limited information that is available supports the need to target family members and other persons known to older victims, because they are most often the assailants.

Injuries and violence are major causes of disability and death among older adults (i.e., persons aged greater than or equal to 65 years). Although older adults die in greater numbers as a result of chronic conditions (e.g., cardiovascular diseases and cancer), injuries and violence remain important public health issues for this age group. For example, unintentional fall-related injuries are the leading cause of unintentional injury deaths for adults aged greater than or equal to 65 years (1,2), and they can result in debilitating injuries (e.g., hip fractures among persons with osteoporosis). Each year, one third of adults aged greater than or equal to 65 years reports at least one unintentional fall (3,4). Dying in a motor vehicle crash is also a concern among older adults, with drivers aged greater than or equal to 65 years having higher rates of motor vehicle crash deaths than all except the youngest drivers (5). Although suicide research and awareness have often focused on youths, U.S. suicide rates have consistently been highest among older adults. Fear of being a victim of violence is also of concern among older adults, who might perceive themselves to be more vulnerable not only to violence outside the home but also to abuse by those on whom they are dependent. This report describes the magnitude of these types of injury and violence among older adults in the United States.

Unintentional injuries are the seventh leading cause of deaths among adults aged greater than or equal to 65 years in the United States. Of these deaths, the greatest number of fatalities result from unintentional fall-related injuries; nearly 8,500 adults in this age group died as a result of injuries from unintentional falls in 1996 (1). Sixty percent of all fall-related deaths in 1996 involved adults aged greater than or equal to 75 years (1). Throughout the past two decades, surveillance data on unintentional fall-related injury have been limited, but available data indicate that unintentional falls accounted for 87% of fractures among adults aged greater than or equal to 65 years treated in emergency departments during 1977 (6) and were the second leading cause of spinal cord and brain injury during 1981 and 1982 (7). Of adults aged greater than or equal to 65 years, 5% were hospitalized for fall-related injury in 1989 (8). Advanced age substantially increases the likelihood of hospital admission following a fall-related injury (9). In 1994, the total direct cost of fall-related injuries among adults aged greater than or equal to 65 years in the United States was an estimated $20.2 billion (10).

Hip fractures are one of the most serious outcomes associated with falls. Approximately 212,000 fall-related hip fractures occur each year among adults aged greater than or equal to 65 years in the United States (11); 75%-80% of these injuries are sustained by women (12). Half of all older adults who suffer hip fractures never regain their former level of function (13,14). Medicare costs for hip fractures were estimated at $2.9 billion in 1991 (15).

In 1996, adults aged greater than or equal to 65 years represented 13% of the U.S. population, yet they accounted for 17% of all motor vehicle-related deaths (5,16). Each year, approximately 7,000 adults aged greater than or equal to 65 years die in motor vehicle crashes, and an additional 175,000 older adults sustain nonfatal injuries in motor vehicle crashes (17,18). These older adults are at risk for motor vehicle-related injury as vehicle occupants and as pedestrians. Nearly 80% of persons in this age group who died in motor vehicle-related crashes in 1997 were occupants of passenger vehicles, whereas approximately 16% were pedestrians (18). In 1996, 25.5 million of the 33.9 million adults aged greater than or equal to 65 years were licensed drivers, accounting for approximately 1.2 billion vehicle miles traveled that year. The number of older adults who drive is expected to increase as the U.S. population ages.

In 1996, suicide was the ninth leading cause of death among all persons in the United States (19). Adults aged greater than or equal to 65 years accounted for 13% of the U.S. population in 1996, yet nearly one fifth of U.S. suicides.

In 1996, homicide was the fourteenth leading cause of death among all persons in the United States. Approximately 5% of all homicides occurred among persons aged greater than or equal to 65 years. Although homicide is a greater problem among younger adults -- ranking as the sixth leading cause of death for adults aged 25-44 years (1) -- fear of being a victim of violence affects older adults' quality of life (20). This fear can limit the participation of older adults in activities outside their homes as well as their use of and access to health care (21).

Data on fall-related deaths, suicides, and homicides are from the mortality data tapes of CDC's National Center for Health Statistics (NCHS) (22). Cause-of-death data are based on the "medical certification of death" portion of the death certificate, which should be completed by the physician with the most complete knowledge of the patient's immediate antemortem medical course and past medical history or by a coroner or medical examiner (23,24). Original copies of the certificates are filed in state and other registration areas' vital statistics offices, where they are processed according to NCHS coding specifications or are sent to NCHS for processing (25,26). Death data from all registration areas are then entered into a national mortality database.

From the medical conditions reported on the death certificate, a single condition -- the underlying cause of death -- is selected for tabulation and analysis according to coding rules specified in the International Classification of Diseases (ICD) (27,28). The underlying cause is the disease or injury that set in motion the chain of morbid events leading to death (e.g., chronic ischemic heart disease leading to an acute myocardial infarction) or the circumstances of the accident or violence that produced the fatal injury (e.g., gunshot wound to the thorax leading to a perforation of the heart) (27). The ICD also specifies how physicians are to report causes of death on death certificates; this specification is reflected in the design of the U.S. Standard Certificate of Death, which serves as a model for death certificates used by states and other registration areas (24).

NCHS data on unintentional fall-related deaths include cases in which the underlying cause of death was coded as E880-E886 or E888, according to the International Classification of Diseases, Ninth Revision (ICD-9) (27). National estimates of hospital admissions for hip fracture are from the 1988-1996 National Hospital Discharge Surveys (NHDS), ongoing surveys conducted by NCHS (29). NHDS collects data from a sample of inpatient records acquired from a national probability sample of nonfederal, short-stay hospitals; data represent a sample of hospital discharges, not a sample of persons (i.e., one person with multiple discharges during the year might be counted more than once). In 1996, data were collected for approximately 282,000 patient discharges from 480 participating hospitals. Hospitalizations for hip fracture include cases in which the discharge diagnosis was coded as 820, according to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) (30).

Data for motor vehicle-related injury are from two sources -- the Fatality Analysis Reporting System (FARS) and the General Estimates System (GES), which are maintained by the National Highway Traffic Safety Administration. FARS is a census of all fatal traffic crashes (within the 50 states, Puerto Rico, and the District of Columbia) that occur on a roadway customarily open to the public and that result in at least one death within 30 days of the crash. The fatality can affect a motor vehicle occupant (e.g., driver or passenger) or nonoccupant (e.g., pedestrian or cyclist) involved in the crash. The system compiles data on these crashes from police accident reports, state vehicle registration files, state driver's licensing files, state highway department data, vital statistics, death certificates, coroner and medical examiner reports, hospital medical reports, and emergency medical service reports (17). Most data on crash events come from police accident reports.

Data for estimates of nonfatal motor vehicle-related injury are from GES, a nationally representative sample of police-reported motor vehicle crashes of all severities, including death, nonfatal injury, and property damage. Data collectors make weekly visits to approximately 400 police jurisdictions at 60 U.S. sites, where they randomly obtain a sample of approximately 48,000 crashes annually (17).

Suicide data are from 1990-1996 NCHS annual mortality data tapes; 1990 was selected as the starting point because that was the year most states reported ethnicity. Suicides include cases in which underlying cause-of-death codes E950-E959 were designated.

Homicides are those deaths in which the underlying cause of death was coded as E960-E969 (these codes exclude legal intervention). The exact number of deaths will not match NCHS published reports, which include legal intervention. Firearm-related homicides are those deaths in which the underlying cause of death was coded as E965.0-E965.4. NCHS homicide data for 1987-1996 are supplemented with 1987-1996 information from the Federal Bureau of Investigation's Supplemental Homicide Reports (FBI-SHR), which provide information on the demographic characteristics of the victim and assailant, the relationship between the victim and assailant, the circumstances of the death, and the type of weapon used. FBI-SHR data are based solely on the reports of investigating law enforcement officials. These data are not linked with NCHS mortality data; however, the proportion of homicide victims in corresponding demographic categories have been shown to be highly congruent between these two sources (31).

National and state-specific rates of hospital admissions for hip fracture were computed by using civilian population estimates from the U.S. Bureau of the Census. Regional rates were calculated for U.S. regions designated by the U.S. Bureau of the Census.*

National and state-specific death rates were computed by using U.S. decennial census counts and mid-year intercensal and postcensal population estimates for adults aged greater than or equal to 65 years from the U.S. Bureau of the Census (16). Because of the limited age range, rates were not age-adjusted. Death rates were calculated per 100,000 U.S. population. Because of the minimal number of motor vehicle-related deaths in some states, a different methodology was used to calculate state-specific rates. To increase the stability of state estimates, the average annual rate for 1990-1997 for each state was calculated.

For rates of fall-related death and hip fracture hospitalization, race was categorized as white, black, other, or unspecified. Three states were excluded from the analysis of suicide rates by ethnicity during the years in which they did not record ethnicity data: Louisiana (1990), New Hampshire (1990-1992), and Oklahoma (1990-1996).

With the exception of the FBI-SHR data, findings for any subgroup that generated estimates with relative standard errors greater than or equal to 30% were not reported. With the same exception, reported differences and trends were significant (p less than or equal to 0.05) unless otherwise noted. No statistical testing was done for analyses of the FBI-SHR data, and these findings should therefore be considered descriptive in nature.

Rates of fall-related deaths for men and women increased substantially with increasing age and were consistently higher among men in all age categories (Table 1). Overall, men were 22% more likely than women to sustain fatal falls. Men aged greater than or equal to 85 years were 43% more likely to die as a result of falls than were women in the same age category. For both men and women, rates were higher for whites than for blacks. In 1996, the pattern of state-specific fall-related death rates was similar for men and women. The Midwest region had the highest rates of fall-related deaths for both men and women (Table 1). Rates of fall-related death increased from 1987 through 1996 in the United States, although rates were consistently lower for women throughout this period (Figure 1).

In 1996, rates of hospitalization for hip fracture differed by age and race for both men and women (Table 2). Rates increased with age for both sexes but were consistently higher for women in all age categories. Women aged greater than or equal to 85 years were almost eight times more likely to be hospitalized for hip fractures than women aged 65-74 years. Among adults aged greater than or equal to 85 years, the rate of hospitalization for hip fracture was 1.8 times higher among women than among men. Within racial categories, hip fracture hospitalization rates were highest among white women and were approximately three times higher for women than men. Sex-specific rates were similar for all geographic regions.

Rates of hospitalization for hip fracture increased from 1988 through 1996 (Figure 2). Rates for men remained stable.

National rates of motor vehicle-related death indicated a small increase (4.5%) for adults aged greater than or equal to 65 years, from 20.6 per 100,000 persons in 1990 to 21.5 in 1997. These rates included deaths to vehicle occupants and nonoccupants (e.g., pedestrians). For adults aged 65-74 years, rates were as low as 7.9 per 100,000 persons in the District of Columbia and as high as 30.1 deaths per 100,000 persons in Mississippi; the rate for all states combined was 17.0. For the oldest group (adults aged greater than or equal to 75 years), rates were as low as 8.9 in Rhode Island and as high as 39.3 deaths per 100,000 persons in Georgia; the rate for all states combined was 25.8. For any given state, rates appear to be higher in the greater than or equal to 75-year age group (Table 3). In most states, men had rates of motor vehicle-related death that were nearly twice the rates for women (Table 4).

The number of older adults dying in traffic crashes increased 14% -- from 6,427 in 1990 to 7,326 in 1997. For drivers only, the number of deaths increased 30%. Information about the number of licensed drivers during this period was available for the years 1990 through 1995. During this time, the number of licensed drivers aged greater than or equal to 65 years increased 13%, and the proportion of licensed drivers aged greater than or equal to 65 years increased from 13% to 14% (32). Thus, some of the increase in deaths among older drivers during the study period would be expected because of the increased number of drivers.

Patterns of deaths for pedestrians differed from those for drivers. In a comparison of 1990 and 1997 data, the number of older pedestrians dying decreased 23%.

In the United States, an estimated 1,869,308 nonfatal motor vehicle-related injuries were reported for older adults from 1990 through 1997. Although both the rate and number of nonfatal motor vehicle-related injuries increased during the study period, these increases did not reach statistical significance. The rate of nonfatal injury for adults aged greater than or equal to 65 years increased 9%, from 684 per 100,000 persons in 1990 to 748 in 1997. The number of nonfatal injuries was estimated from a national sample of cases that were designed to be representative of the nation as a whole; therefore, reporting estimates by state was not possible. The number of nonfatal injuries increased 19% overall, from 213,463 in 1990 to 254,799 in 1997. For drivers aged greater than or equal to 65 years, the number of nonfatal injuries increased 21%, from 143,997 in 1990 to 174,609 in 1997. For pedestrians aged greater than or equal to 65 years, the number of nonfatal injuries declined 24%, from 9,092 in 1990 to 6,897 in 1997.

Approximately 43,000 (20%) of the 216,631 suicides that occurred in the United States from 1990 through 1996 involved adults aged greater than or equal to 65 years. The annual number of suicides among adults in this age group decreased 8.4%, from 6,394 in 1990 to 5,855 in 1996 (Figure 3). In comparison, rates for this age group decreased 16%, from 20.6 to 17.3 per 100,000 persons. Men accounted for 82% of suicides among adults aged greater than or equal to 65 years; from 1990 through 1996, the rate for men decreased 15%, from 41.6 to 35.2 per 100,000 persons. For women, the rate decreased 25%, from 6.4 to 4.8 per 100,000 persons.

For the period 1990 through 1996, the rate of suicide among adults aged 65-74 years was greater than the rate for those aged 75-84 years; however, the rate remained unchanged between the 75-84-year age group and the greater than or equal to 85-year age group (Table 5). Suicide rates among men aged greater than or equal to 85 years were higher (65 per 100,000 persons) than rates among those aged 65-74 years and 75-84 years. Among women, suicide rates were lowest among those aged greater than or equal to 85 years, but the difference in rates between age groups was not significant. Men had a higher overall suicide rate (38 per 100,000 persons) than women (5.7 per 100,000 persons).

Firearm use was the predominant suicide method for both men and women. Firearm-related suicides accounted for 70% of all suicides among adults aged greater than or equal to 65 years during the years 1990 through 1996 (77.3% of suicides among men compared with 34.4% among women). Poisoning was the second most common method among men (12%) and women (29%).

Suicide rates among adults aged greater than or equal to 65 years also varied by race and ethnicity (Figure 4). Rates increased steadily for persons of Hispanic ethnicity for each 10-year age interval; however, the difference between the 75-84-year age group and the greater than or equal to 85-year age group was not significant. For black persons not of Hispanic ethnicity, the rate was similar in each 10-year age group. For white persons not of Hispanic ethnicity, the rate for those aged 65-74 years was lower than rates for the other age groups.

The rate of homicides among adults aged greater than or equal to 65 years decreased 36% from 1987 through 1996. A downward trend was observed in homicide rates, from 4.6 per 100,000 persons in 1987 to 2.9 per 100,000 persons in 1996. The overall rate for homicides related to firearms use decreased 30%, from 1.6 per 100,000 persons in 1987 to 1.1 per 100,000 persons in 1996. A slight increase in firearm-related homicides was observed in 1990.

In 1996, the rate of homicide was 4.1 per 100,000 among men and 2.1 per 100,000 among women aged greater than or equal to 65 years. Throughout the 10-year period, homicide rates were highest for black men, followed by black women and white men (Figure 5). Homicide rates decreased 49% for black men and 38% for black women. The homicide risk for blacks relative to whites decreased from 4.8 in 1987 to 3.9 in 1996.

From 1987 through 1990, the highest overall homicide rates for adults aged greater than or equal to 65 occurred among those aged greater than or equal to 85 years (Figure 6). Beginning in 1994, however, rates were similar for all age groups. The homicide rate decreased for each age group from 1987 through 1996. For the oldest age group (greater than or equal to 85 years), rates began decreasing from the 10-year high of 5.2 per 100,000 persons in 1987 to 2.9 per 100,000 persons in 1996. This corresponds with a 44% decrease in rates.

For the period 1987 through 1996, rates of firearm-related homicide were 25% higher for persons aged 65-74 years than for those aged 75-84 years. The rate was 32% higher for persons aged 75-84 years than for those aged greater than or equal to 85 years.

Because the number of homicides caused by firearm use did not decline as much as the number of all homicides during the study period, the proportion of homicides attributed to firearm use increased: homicides caused by firearm use accounted for 35% (478/1,355) of homicides among adults aged greater than or equal to 65 years in 1987 and 39% (382/989) in 1996.

According to FBI-SHR data for 1996, a total of 35% of homicides among older adults involved firearms (Figure 7), with 72% of these firearm-related deaths involving a handgun. The other most commonly reported methods of homicide were cutting (23%), use of blunt objects (14%), bodily force (11%), and strangulation (4%). Other, or unspecified, means accounted for 13% of all homicides. These relative proportions changed only slightly during 1987-1996. The proportions did not differ by race.

From 1987 through 1996, a total of 50% of older adult homicide victims were killed by a person they knew: 25% by a family member and 25% by an acquaintance. This proportion was similar to the proportion for all homicides in the United States (33). In most homicides, the victim and assailant were of the same race.

Older adults are at higher risk for fatal falls compared with persons aged less than or equal to 64 years, and this risk increases sharply with increasing age. In this analysis, the rate of death from unintentional falls was higher for men than women in all age categories. This disparity is well known (2), but the reasons remain unclear. The circumstances of the falls were listed as "other or unspecified" for 71.3% of unintentional fall-related fatalities in 1996. The rate among older men might be higher because men have a higher prevalence of comorbid conditions than women of similar ages. Frail adults (i.e., those with impaired strength, mobility, balance, and endurance) are twice as likely to fall as healthier persons (34) and might sustain more severe injuries. Rates of unintentional fall-related death were lower for blacks, an observation reported by others (35). The likely reason is that blacks aged greater than or equal to 65 years in the United States have greater bone mass (36) and are less likely to sustain fall-related hip fractures than whites of comparable ages (37).

Unintentional fall-related death rates gradually increased from 1987 through 1996. This trend might be related, in part, to a reduction in mortality from cardiovascular and other chronic diseases. Adjusting the rates to account for the changing age distribution in the U.S. population did not alter these findings.

Since 1988, rates of hospitalization for hip fracture among women have steadily increased. One of the Healthy People 2000 objectives (objective 9.7) is to reduce the overall rate of hospitalization for hip fracture among adults aged greater than or equal to 65 years to no more than 607 per 100,000 persons (38). The 1996 rate was 1,003 per 100,000 persons -- 32% higher than the 1988 rate of 761 per 100,000.

Hospitalization rates were consistently higher for women than men. In 1996, white women aged greater than or equal to 65 years had 5.1 times the risk for being hospitalized for hip fracture as black women. Another Healthy People 2000 objective (objective 9.7a) is to reduce rates of hospitalization for hip fracture among white women aged greater than or equal to 85 years to no more than 2,177 per 100,000 (38). The 1996 rate was 3,271 per 100,000, which is almost 18% higher than the 1988 baseline rate.

This analysis has several limitations. First, conclusions about the magnitude of the injury problem among older adults rely on injury information being recorded accurately on death certificates. If fall injury-related deaths were undercounted because injury was not recorded, the fall death rates in this report would underestimate the impact of this injury. The underlying cause of death is determined from the listed diagnoses in Part 1 of the death certificate. When a person dies of injuries, the external cause of the injury (e.g., fall) should be listed in Part 1 of the death certificate and will be recorded as the underlying cause of death; however, external cause-of-injury data might not be as reliably recorded for older adults as for younger persons. Using Los Angeles data, researchers reviewed a section on the 1980 California death certificate designated "Injury Information, Coroner's Use Only" to identify, by age group, injury deaths that had not been coded as injuries and thus determine what proportion these deaths represented among those deaths coded as injuries (39). Of the deaths coded as injury for persons aged greater than or equal to 65 years, an additional 46% of deaths were identified that appeared to be injury deaths that had been inaccurately coded as noninjury deaths; in comparison, an additional 5% of such deaths were identified for persons aged less than or equal to 55 years. Thus, estimates of fall-related death rates in this report might be conservative for persons aged greater than or equal to 65 years.

Second, NHDS identifies hospital admissions, not actual persons. Therefore, a person who is treated for a hip fracture, discharged, and readmitted for additional treatment could receive a second discharge diagnosis of hip fracture. Although such duplications probably represent a small proportion of admissions, they could not be omitted from the analysis because NHDS does not include personal identifiers. Thus, NHDS slightly overestimates rates of hospitalizations for hip fracture. Moreover, because race was unspecified for 19.5% of the 1996 hospital admissions, race-specific rates were underestimated.

Because adults aged greater than or equal to 65 years are the fastest growing age group in the U.S. population, the burden of falls and hip fractures will likely increase unless fall prevention strategies are improved for this group. Risk factors for falls include both personal factors (e.g., neurologic and musculoskeletal disabilities, difficulties with gait and balance, use of psychoactive medications, and visual impairments) and environmental hazards (e.g., poor lighting, slippery surfaces, loose rugs, and other tripping hazards). The most effective fall prevention programs have reduced falls in select populations by 30%-50% by using a multifaceted approach that includes education, exercise (e.g., Tai Chi to improve strength, balance, and coordination), medication review, risk factor reduction, and home safety modifications (14,40,41). For these interventions to be effective, older adults must take an active role to reduce their risk for falling. Because most older adults live independently, fall prevention programs must include effective strategies to promote behavioral changes. Innovative and effective fall prevention strategies are needed to reduce morbidity and mortality associated with fall-related injuries, to increase independence, and to improve the quality of life for the growing number of older adults. In addition to behavioral and environmental interventions to prevent falls (14,41-43), new approaches, including protective hip pads (44) and impact-absorbing floor materials, offer promise to reduce the incidence of hip fractures among older adults.

Rates of motor vehicle-related morbidity and mortality increased slightly for older adults during 1990-1997, and rates for males were higher than those for females. During the 8-year study period, approximately 55,000 adults aged greater than or equal to 65 years died as a result of traffic crashes, and an estimated 1,869,308 others were nonfatally injured. These numbers represent a costly burden to the health-care system and to society.

The increased risk for motor vehicle-related injury to older adults has many possible explanations, including visual deterioration, declines in cognitive skills (e.g., processing speed and ability), and deterioration of motor skills (45). Also, physical frailty increases susceptibility to injury in a crash. The same severity of crash might cause the death of an elderly vehicle occupant yet result in nonfatal injuries to a younger person. Other factors tend to moderate the overall magnitude of the problem. For example, a smaller proportion of older adults have driver's licenses compared with younger groups, and they drive fewer miles per licensed driver (5). In addition, adults aged greater than or equal to 65 years are the most likely group to wear seat belts and the least likely group to report drinking and driving or riding with a drunk driver (46-49).

To prevent these deaths and injuries, specific risk factors for older adults must be considered and interventions must be tailored to reduce risk. Interventions could be aimed at changes in the vehicle, the traffic and road environment, or the behavior of older adults. For example, drivers could be aided by increasing the size and illumination of instrument panel dials and road signs for better nighttime readability (50,51). Also, drivers could be assisted in assessing their own driving ability. Many older drivers impose partial driving restrictions on themselves (e.g., driving during nonpeak traffic hours or driving only on familiar roads) (52). These restrictions tend to decrease their amount of driving and, consequently, their overall crash risk. For drivers who cannot adequately assess their driving ability because of functional or cognitive impairment, a more rigorous screening and testing program might be necessary (53-56).

Older pedestrians might be aided by environmental alterations (e.g., changing the timing of traffic signals or installing median islands for refuge on wide roadways) (57). In addition, problems judging traffic speed and the time required to cross a road might require remediation (e.g., visual correction, walking aids, or changing crossing styles) (57,58). The efficacy of interventions should be evaluated to reduce the risk for motor vehicle crashes and injuries involving older adults.

This report documents a change in the trend of suicide rates among older adults. For the period 1990-1996, suicide rates for adults aged greater than or equal to 65 years decreased. Men had a higher overall suicide rate than women (38 vs. 5.7 per 100,000 persons). Among men and women, suicide rates varied by age.

Some risk factors and protective factors for suicide are similar for older and younger adults; however, the importance of these factors might differ by age group (e.g, the intensity of depressive symptoms, use of highly lethal methods, and social isolation) (59). Participation in religious services has been identified as a protective factor against suicidal behavior (60). Religious participation and its associated belief system might exert a differing influence among older and younger adults. In addition, older adults make fewer attempts per completed suicide, have a higher male-to-female ratio, have more often visited a health-care provider shortly before their death, and have more physical illnesses and affective disorders than younger persons (61).

The declining suicide rate among adults aged greater than or equal to 65 years could be related to changes in the effect or type of risk factors traditionally occurring among older adults (e.g., depression, social isolation, chronic illness). Perhaps the importance of these factors has changed or the prevalence of protective factors has increased.

Because older adults have the highest suicide rates, prevention research should focus on factors associated with suicide among older adults. For example, prevention strategies might need to be tailored to specific age, sex, and racial/ethnic groups. The role of protective factors in preventing suicide among older adults should also be more closely explored and should be integrated into prevention strategies. Recent cohort studies indicate that suicide rates are higher among younger adults today than they were when their grandparents were young adults (62). As these younger adults age, their suicide rates might increase above current rates among older adults (62). However, in some birth cohorts, suicide rates might be higher because of the relative size of the group (i.e., larger cohorts might face increased stressors because of greater competition for resources and a disparity between their expectations and the means to satisfy those expectations) (62).

These findings underscore the need for suicide prevention activities directed at older adults. Strategies for reducing suicide rates among older adults include training primary-care providers to better recognize suicidal risk factors, including depressive disorders, and to make appropriate referrals. These strategies have been effective in reducing suicide risk among older adults (63). Community-based interventions to identify and treat persons at risk also have been shown to be effective (64). Other prevention strategies include senior peer-counseling programs; suicide prevention efforts that target persons at high risk; improvements in mental health services through suicide prevention centers; and programs that increase awareness of risk factors and protective factors among persons who have frequent contact with older adults (63).

The findings in this report confirm that homicide rates among older adults decreased from 1987 through 1996 in the United States. Moreover, homicide rates among older adults decreased at a time when rates were increasing in all other age groups. The increasing rates observed among adults aged less than or equal to 64 years until 1993 were driven by increases in firearm-related homicides among younger persons (aged 15-24 years) (65). The differences in rates between older and younger adults in this analysis might be related to the substantially lower proportion of firearm-related homicides among older adults (39%) than in the total U.S. population (72%) (65).

The racial patterns among older homicide victims were similar to those among younger adults. Homicide rates were higher among blacks than whites and remained highest among black males. However, the disparity between blacks and whites was lower among older adults (4:1) compared with the U.S. population (6:1) (1). Reasons are unclear as to why blacks consistently have higher rates of homicide than whites; however, research indicates that racial differences in homicide rates nearly disappear when socioeconomic status is considered (66). Other commonly raised factors for consideration include prior exposure to violence, involvement with drugs, and the accessibility of firearms. The increased black-white homicide differential seen among younger adults reflects the disproportionately high rates among young, black, male victims; these rates decrease with age, beginning at approximately 35 years of age.

Half of the older homicide victims included in this analysis were killed by someone they knew, usually someone of the same race. Older adults tend to spend less time engaged in activities outside of the home than younger adults, thus limiting their exposure to strangers. Increased interpersonal contact with potential offenders by older adults, along with their increased fragility, might be important reasons for this finding. In cases of elder abuse, the abusers are most often adult children, followed by spouses and other relatives. These findings, however, vary among different studies (67).

Perceived risk of being a victim of violence among older adults might be greater than actual risk (68). This "victimization-fear paradox" among older adults is related to their perception of having a higher physical vulnerability than younger persons (69). When this fear of being a victim diminishes their quality of life, intervention is necessary. Interventions must be informed by an accurate description of the problem and identification of risk factors.

Homicide and violence prevention strategies require integration of approaches from various disciplines (e.g., criminal justice, social services, education, community advocacy, and public health). Conventional approaches to reduce homicide are described elsewhere (65). For violence among older adults, interventions are needed that promote public and professional awareness of the problem; promote good mental health (e.g., reduce depression), independence, and social contact; buffer stressful life events; and teach conflict resolution without violence (70). Also recommended are interventions that provide chore social services (e.g., meals, homemaking), which can help alleviate the stress and resentment of the caregiver (71).

The links between homicide, assault, and elder abuse are not well defined, but considerable evidence, including findings in this report, bolsters the need to target family members and persons known to the victim. In addition, teaching family members and friends who are not direct caregivers to recognize signs of physical abuse and neglect is important because these persons can facilitate reporting and referral when appropriate. Education for older adults should inform them of their actual risk -- including whom they are most at risk from -- and promote healthy coping strategies that will reduce their fear and help keep them safe from violence.

Issues of family integrity and care for older adults pose significant challenges to efforts by public health and criminal justice organizations to design appropriate and effective intervention strategies. Continued surveillance for homicide, the removal of reporting barriers for all forms of violence, and expanded epidemiologic research and evaluation should help to reduce the burden of fatal and nonfatal forms of violence among older adults.

References

1. Peters KD, Kochanek KD, Murphy SL. Deaths: final data for 1996. National Vital Stat Rep 1998;47(9).
2. Baker SP, O'Neill B, Ginsburg MJ, Li G. The injury fact book. 2nd ed. New York, NY: Oxford University Press, 1992.
3. Campbell AJ, Reinken J, Allan BC, Martinez GS. Falls in old age: a study of frequency and related clinical factors. Age Aging 1981;10:264-70.
4. Sattin RW. Falls among older persons: a public health perspective. Annu Rev Public Health 1992;13:489-508.
5. National Highway Traffic Safety Administration. Crash data and rates for age-sex groups of drivers, 1996. Research note. Washington, DC: National Center for Statistics and Analysis, National Highway Traffic Safety Administration, 1998.
6. Fife D, Barancik JI. Northeastern Ohio Trauma Study III: incidence of fractures. Ann Emerg Med 1985;14:244-8.
7. Kraus JF, Black MA, Hessol N, et al. The incidence of acute brain injury and serious impairment in a defined population. Am J Epidemiol 1984;119:186-201.
8. Alexander BH, Rivera FP, Wolf ME. The cost and frequency of hospitalization for fall-related injuries in older adults. Am J Public Health 1992;82:1020-3.
9. Fife D, Barancik JI, Chatterjee BF. Northeastern Ohio Trauma Study II: injury rates by age, sex, and cause. Am J Public Health 1984;74:473-8.
10. Englander F, Hodson TJ, Terregrossa RA. Economic dimensions of slip and fall injuries. J Forensic Sci 1996;41:733-46.
11. Cooper C, Campion G, Melton LJ III. Hip fractures in the elderly: a world-wide projection. Osteoporos Int 1992;2:285-9.
12. Melton LJ III, Riggs BL. Epidemiology of age-related fractures. In: Avioli LV, ed. The osteoporotic syndrome. New York, NY: Grune & Stratton, 1983:45-72.
13. Cummings SR, Rubin SM, Black D. The future of hip fractures in the United States: numbers, costs, and potential effects of postmenopausal estrogen. Clin Orthop Rel Res 1990;252:163-6.
14. Tinetti ME, Baker DI, McAvay G, et al. A multifactorial intervention to reduce the risk of falling among elderly people living in the community. N Engl J Med 1994;331:821-7.
15. CDC. Incidence and costs to Medicare of fractures among Medicare beneficiaries aged greater than or equal to 65 years--United States, July 1991-June 1992. MMWR 1996;45:877-83.
16. US Bureau of the Census. Estimates of the population of the U.S., regions, and states by selected age groups and sex: annual time series, July 1, 1990 to July 1, 1997. Washington, DC: Population Estimates Program, Population Division, 1998.
17. National Highway Traffic Safety Administration. Traffic safety facts 1996: a compilation of motor vehicle crash data from the Fatality Analysis Reporting System and the General Estimates System. Washington, DC: National Center for Statistics and Analysis, National Highway Traffic Safety Administration, 1997.
18. National Highway Traffic Safety Administration. Traffic safety facts 1997: older population. Washington, DC: National Center for Statistics and Analysis, National Highway Traffic Safety Administration, 1998.
19. National Center for Injury Prevention and Control. Ten leading causes of death, United States, 1996. Atlanta, GA: US Department of Health and Human Services, CDC, National Center for Injury Prevention and Control, 1999. Available at http://www.cy118119.com/ncipc/osp/states/101c96.htm. Accessed November 29, 1999.
20. Ross CE. Fear of victimization and health. J Quantitative Criminol 1993;9:159-75.
21. Joseph J. Fear of crime among black elderly. J Black Studies 1997;27:698-717.
22. National Center for Health Statistics. Vital statistics of the United States, mortality, underlying and multiple causes of death [Public Use Files]. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics, 1998.
23. National Center for Health Statistics. Physicians' handbook on medical certification of death. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics, 1996.
24. National Center for Health Statistics. Medical examiners' and coroners' handbook on death registration and fetal death reporting. Hyattsville, MD: US Department of Health and Human Services, National Center for Health Statistics, 1987.
25. National Center for Health Statistics. Technical appendix. In: Vital statistics of the United States, 1995, mortality. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics, 1999.
26. National Center for Health Statistics. Mortality data from the National Vital Statistics System. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics, 1999. Available at http://www.cy118119.com/nchswww/about/major/dvs/mortdata.htm. Accessed July 27, 1999.
27. World Health Organization. Manual of the international statistical classification of disease, injuries, and causes of death, 9th revision. Geneva, Switzerland: World Health Organization, 1977.
28. World Health Organization. International statistical classification of diseases and related health problems, 10th revision. Geneva, Switzerland: World Health Organization, 1992.
29. Graves EJ, Owings MF. 1996 summary: National Hospital Discharge Survey. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics, 1998. (Advance data from vital and health statistics; no. 301).
30. Public Health Service. The international classification of diseases, 9th revision, clinical modification. 4th edition. Washington, DC: US Department of Health and Human Services, 1991; DHHS publication no. (PHS)91-1260.
31. Rokaw WM, Mercy JA, Smith JC. Comparing death certificate data with FBI crime reporting statistics on U.S. homicides. Public Health Rep 1990;105:447-55.
32. Federal Highway Administration. Highway statistics summary to 1995. Washington, DC: Office of Highway Information Management, Federal Highway Administration, 1997; publication no. FHWA-PL-97-009.
33. CDC. Homicide surveillance, 1979-1988. MMWR 1992;41(No. SS-3).
34. Northridge ME, Nevitt MC, Kelsey JL, Link B. Home hazards and falls in the elderly -- the role of health and functional status. Am J Public Health 1995;85:509-15.
35. Baron JA, Barrett J, Malenka D, et al. Racial differences in fracture risk. Epidemiol 1994;5:42-7.
36. Cummings SR, Kelsey JL, Nevitt MC, et al. Epidemiology of osteoporosis and osteoporotic fractures. Epidemiol Rev 1985;7:178-208.
37. Griffin MR, Wayne AR, Fought RL, Melton LJ III. Black-white differences in fracture rates. Am J Epidemiol 1992;136:1378-85.
38. Public Health Service. Healthy people 2000: national health promotion and disease prevention objectives--full report, with commentary. Washington, DC: US Department of Health and Human Services, Public Health Service, 1991; DHHS publication no. (PHS)91-50212.
39. Fife D, Rappaport E. What role do injuries play in the deaths of old people? Accid Anal Prev 1987;19:225-30.
40. Wolf SL, Barnhart HX, Kutner NG, McNeely E, Coogler C, Xu T. Reducing frailty and falls in older persons: an investigation of Tai Chi and computerized balance training. Atlanta FICSIT Group. Frailty and injuries: cooperative studies of intervention techniques. J Am Geriatr Soc 1996;44:489-97.
41. Wolter LL, Studenski SA. A clinical synthesis of falls intervention trials. Top Geriatr Rehab 1996;11:9-19.
42. Stevens JA, Powell KE, Smith SM, Wingo PA, Sattin RW. Physical activity, functional limitations, and the risk of fall-related fractures in community-dwelling elderly. Ann Epidemiol 1997;7:54-61.
43. Sattin RW, Rodriquez JG, DeVito CA, Wingo PA. Home environmental hazards and the risk of fall injury events among community-dwelling older persons. J Am Geriatr Soc 1998;46:669-76.
44. Lauritzen JB. Hip fractures: epidemiology, risk factors, falls, energy absorption, hip protectors, and prevention. Dan Med Bull 1997;44:155-68.
45. Stutts JC, Stewart JR, Martell C. Cognitive test performance and crash risk in an older driver population. Accid Anal Prev 1998;30:337-46.
46. Nelson DE, Bolen J, Kresnow MJ. Trends in safety belt use by demographics and by type of state safety belt law, 1987 through 1993. Am J Public Health 1998;88:245-9.
47. Lui S, Siegal PZ, Brewer RD, Mokdad AH, Sleet DA, Serdula M. Prevalence of alcohol-impaired driving: results from a national self-reported survey of health behaviors. JAMA 1997;277:122-5.
48. Smith F, Remington PL. The epidemiology of drinking and driving: results from the Behavioral Risk Factor Surveillance System, 1986. Health Educ Quart 1989;16:345-58.
49. Dellinger AM, Bolen J, Sacks JJ. A comparison of driver- and passenger-based estimates of alcohol-impaired driving. Am J Prev Med 1999;16:278-82.
50. Graham JR, Fazal A, King LE. Minimum luminance of highway signs required by older drivers. In: Transportation Research Board, National Research Council. Safety and human performance, and highway operations, capacity, and traffic control; human performance in intelligent transportation systems, information systems, and highway design and older drivers. Washington, DC: National Academy Press, 1997. Transportation research record no. 1573;91-8.
51. Wood JM, Troutbeck RJ. Effect of age and visual impairment on driving and vision performance. In: Transportation Research Board, National Research Council. Safety and human performance, research issues on bicycling, pedestrians, and older drivers. Washington, DC: National Academy Press, 1994. Transportation research record no. 1438;84-90.
52. Benekohal RF, Michaels RM, Shim E, Resende PTV. Effects of aging on older drivers' travel characteristics. In: Transportation Research Board, National Research Council. Safety and human performance, research issues on bicycling, pedestrians, and older drivers. Washington, DC: National Academy Press, 1994. Transportation research record no. 1438;91-8.
53. Marottoli RA, Richardson ED, Stowe MH, et al. Development of a test battery to identify older drivers at risk for self-reported adverse driving events. J Am Geriatr Soc 1998;46:562-8.
54. Sims RV, Owsley C, Allman RM, Ball K, Smoot TM. A preliminary assessment of the medical and functional factors associated with vehicle crashes by older adults. J Am Geriatr Soc 1998;46:556-61.
55. Lucas-Blaustein MJ, Filipp L, Dungan C, Tune L. Driving in patients with dementia. J Am Geriatr Soc 1988;36:1087-91.
56. Gilley DW, Wilson RS, Bennett DA, et al. Cessation of driving and unsafe motor vehicle operation by dementia patients. Arch Intern Med 1991;151:941-6.
57. Langlois JA, Keyl PM, Guralnik JM, Foley DJ, Marottoli RA, Wallace RB. Characteristics of older pedestrians who have difficulty crossing the street. Am J Public Health 1997;87:393-7.
58. Oxley J, Fildes B, Ihsen E, Charlton J, Day R. Differences in traffic judgements between young and old adult pedestrians. Accid Anal Prev 1997;29:839-47.
59. Rosenberg ML, Gelles RJ, Holinger PC, et al. Violence: homicide, assault, and suicide. In: Amler RW, Dull HB, eds. Closing the gap: the burden of unnecessary illness. New York, NY: Oxford University Press, 1987:164-78.
60. Wasserman I, Stack S. The effect of religion on suicide ideology: an analysis of the networks perspective. J Sci Study Religion 1992;31:457-66.
61. Conwell Y, Rotenberg M, Caine ED. Completed suicide at age 50 and over. J Am Geriatr Soc 1990;38:640-4.
62. Blazer DG, Bachar JR, Manton KG. Suicide in late life: review and commentary. J Am Geriatr Soc 1986;34:519-25.
63. Institute of Medicine. The second fifty years: promoting health and preventing disability. Washington, DC: National Academy Press, 1990:202-3
64. Conwell, Y. Suicide in late life. Presented at the 1998 Suicide Prevention Conference: Developing a National Strategy for Suicide Prevention. Atlanta, GA: US Department of Health and Human Services, CDC, 1998.
65. CDC. Trends in rates of homicide--United States, 1985-1994. MMWR 1996;45:460-4.
66. Williams KR. Economic source of homicide: reestimating the effects of poverty and inequality. Am Soc Rev 1984;49:283-9.
67. Lachs MS, Williams C, O'Brien S, Hurst L, Horwitz R. Risk factors for reported elder abuse and neglect: a nine-year observational cohort study. Gerontologist 1997;37:469-74.
68. Greve W. Fear of crime among the elderly: foresight, not fright. Int Rev Victimology 1998;5:277-309.
69. Stafford MC, Galle OR. Victimization rates, exposure to risk, and fear of crime. Criminology 1984;22:173-85.
70. Committee on the Assessment of Family Violence Interventions. In: Chalk R, King PA, eds. Violence in families: assessing prevention and treatment programs. Washington, DC: National Academy Press, 1998.
71. United States General Accounting Office. Elder abuse: effectiveness of reporting laws and other factors. Washington, DC: Report to the Chairman, Subcommittee on Human Services, Select Committee on Aging, House of Representatives, 1991.

* Northeast=Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, and Vermont; Midwest=Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, and Wisconsin; South= Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, and West Virginia; and West=Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, and Wyoming.

Disclaimer   All MMWR HTML versions of articles are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the electronic PDF version and/or the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.

Page converted: 12/14/1999

HOME  |  ABOUT MMWR  |  MMWR SEARCH  |  DOWNLOADS  |  RSS |  CONTACT POLICY  |  DISCLAIMER  |  ACCESSIBILITY

Morbidity and Mortality Weekly Report Centers for Disease Control and Prevention 1600 Clifton Rd, MailStop E-90, Atlanta, GA 30333, U.S.A Department of Health and Human Services

This page last reviewed 5/2/01