The following are pre-publication drafts of articles from the Morbidity and Mortality Weekly Report dated September 22, 1995. Late-breaking articles, and final editorial revisions are not included; therefore, these articles should be considered preliminary, and not to be released to the public. --CDC -------------------------------------------------------------- State-Specific Pregnancy and Birth Rates Among Teenagers -- United States, 1991-1992 Pregnancy and childbearing rates for teenagers remain high in the United States despite well-documented associated adverse health, social, and economic consequences for many of these teenagers and their children. In 1990, approximately 835,000 (10%) teenagers aged 15-19 years became pregnant and either gave birth or had an abortion (CDC, unpublished data, 1995); an estimated 95% of such pregnancies are unintended (1). This report presents estimates of pregnancy rates among women aged less than or equal to 19 years for each state and the District of Columbia (DC) by age group, pregnancy rates for women aged 15-19 years by race, and birth rates for women aged 15-19 years by race and by Hispanic ethnicity for 1991-1992, and compares pregnancy rates for 1991 and 1992. The numbers of pregnancies for 1991 and 1992 were estimated as the sum of live births and legal induced abortions among women aged less than or equal to 19 years (data were analyzed for women aged less than 15, 15-17, 18-19, and 15-19 years); estimates of spontaneous abortions and stillbirths were not included. Births were reported by state of residence; because abortion data by residence were not available for all states, abortions were reported by state of occurrence.* Denominators for rate calculations were obtained from intercensal population estimates provided by the U.S. Bureau of the Census (2). Rates for 15-19-year-olds were calculated as the number of pregnancies, abortions, or births per 1000 women aged 15-17, 18-19, and 15-19 years. Because almost all pregnancies (97% of births and 94% of abortions) among girls aged less than 15 years occur among those aged 13-14 years (3; CDC, unpublished data, 1993), the number of girls aged 13-14 years was used as the denominator when calculating rates for the less than 15-year age group. For each state included in rate calculations, the number of women who had abortions for whom age or race information was missing and the number who gave birth for whom ethnicity information was missing were included in age, race, or ethnicity categories based on the known distributions for abortions or births in that state.** Differences in pregnancy, abortion, and birth rates for 1991 and 1992 were calculated and tested for statistical significance at p less than 0.05. Although abortion totals were available for all states, age-specific abortion data were available from 42 states for both 1991 and 1992; race-specific data were available from 31 states for 1991 and from 34 states for 1992. Because Hispanic ethnicity information for women who had abortions was available for only 18 states for 1991 and 20 states for 1992, pregnancy rates by ethnicity are not included in this report. Information was available for birth rates by age and by race for all 50 states and DC and by ethnicity for 49 states and DC. For both years and for all states for which data were available, pregnancy rates were higher for older teenagers than for younger teenagers. Rates for 15-19-year-olds in 1991 ranged from 54.3 per 1000 women (North Dakota) to 109.2 (Georgia) (Table 1),*** and in 1992, from 53.7 (Wyoming) to 106.9 (Georgia). For those aged less than 15 years, pregnancy rates in 1991 ranged from 1.8 per 1000 (Idaho) to 10.6 (Georgia) and in 1992 from 2.0 (Idaho) to 10.9 (Mississippi). From 1991 through 1992, pregnancy rates for 15-19-year-olds decreased significantly in 31 of the 42 states for which age-specific data were available (range: 2%-15% decrease) (Table 1). In two states, rates increased significantly. Among states with decreases in pregnancy rates, the percentage decrease was generally greater for 15-17-year-olds than for other age groups. For those aged less than 15 years, pregnancy rates decreased significantly in one state and increased significantly in two. Decreases in teenage pregnancy rates were reflected in both birth and abortion rates. More states had decreases in abortion rates than had decreases in birth rates, and the decreases generally were greater for abortion rates than for birth rates. For the 50 states and DC, birth rates for 15-19-year-olds decreased significantly in 20 states: by less than 5% in 11 states and by 5%-9% in nine states. Abortion rates decreased significantly in 31 of the 42 states for which data were available: rates decreased by less than 5% in two states, 5%-9% in six states, 10%-14% in eight states, and 15%-27% in 15 states. Except in one state in 1992, pregnancy rates and birth rates by state for 15-19-year-olds were higher for blacks than for whites (Tables 2 and 3). Among states with decreases in pregnancy rates, the percentage decrease generally was greater for whites than for blacks. Birth rates generally were higher for Hispanics than for non-Hispanics in both 1991 and 1992. Reported by: Behavioral Epidemiology and Demographic Research Br, Statistics and Computer Resources Br, Div of Reproductive Health, National Center for Chronic Disease Prevention and Health Promotion; Natality, Marriage, and Divorce Statistics Br, Div of Vital Statistics, National Center for Health Statistics, CDC. Editorial Note: The estimates in this report indicate pregnancy rates for women aged 15-19 years decreased in 1992 from 1991 in many states; these changes were reflected in both abortion and birth rates. As a result, the national birth rate for 15-19-year-olds decreased by 2% in 1992 following a 24% increase from 1986 through 1991 (4). In conjunction with a continuing decrease in abortions (5), this change may signify a turning point in pregnancy trends among teenagers. Changes in pregnancy rates for women aged 15-19 years may reflect shifts in the proportion of those who have had sexual intercourse and the proportion who use contraception effectively. The percentage of women in this age group who were sexually experienced remained stable from 1990 through 1993, while the percentage of sexually experienced teenagers who used condoms increased, and the percentage using oral contraceptives remained stable (6). The finding that significant decreases in pregnancy rates occurred both in states with relatively low 1991 rates and states with relatively high rates suggests that potential exists for all states to achieve lower rates of pregnancy among this age group. School-based programs that focus on the risks of unprotected sexual intercourse and assist students in developing appropriate values, self-efficacy, and negotiation skills appear to be effective in postponing initiation of sexual activity and in decreasing rates of unprotected intercourse (7). Some community-based programs emphasize development of self-esteem and orientation toward the future and provide sex education and family-planning services; such programs may be effective in decreasing rates of unprotected intercourse and potentially can reach teenagers who are not enrolled in school. Pregnancy and birth rates did not decrease equally for all groups of teenagers. For example, pregnancy rates for those aged less than 15 years decreased significantly in only one state. In addition, percentage decreases in pregnancy rates for 15-19-year-olds were greater for whites than for blacks; consequently, both pregnancy and birth rates remained higher for blacks than whites. In addition, birth rates remained higher for Hispanics than for non-Hispanics. Differences in these rates by race and ethnicity may reflect differences in factors such as income, education levels, sexual activity, reproductive preferences, contraceptive use, and access to health care. Analysis of these factors could identify reasons for the differences in rates and suggest possible interventions to reduce these rates. The estimation in this report of pregnancy rates for teenagers was limited by the lack of complete abortion data for some states. In addition, pregnancy totals based on births and legal induced abortions reported to CDC may underestimate the actual number of pregnancies. In 1990, approximately 835,000 live births and abortions among 15-19-year-olds were reported in the United States to CDC (CDC, unpublished data, 1995). In contrast, a recent analysis using alternative methodology and including fetal losses estimated approximately 1 million pregnancies among 15-19-year-olds in the United States during 1990 (8,9). Accurate monitoring of pregnancy trends requires complete reporting of age, race, and Hispanic ethnicity for those who have abortions as well as those who give birth. Birth patterns cannot be used to estimate trends in pregnancy because the ratio of pregnancies to births varies by state, age, race, and ethnicity. Therefore, efforts by states to collect complete abortion data are essential for evaluating the progress of pregnancy-prevention programs for teenagers. References 1. Piccinino LJ. Unintended pregnancy and childbearing in the United States, 1973-1990. Hyattsville, Maryland: US Department of Health and Human Services, Public Health Service, CDC, NCHS. (Advance data) (in press). 2. Bureau of the Census. Estimates of the population of states, by age, race, sex, and Hispanic origin, 1990-92. Washington, DC: US Department of Commerce, Bureau of the Census, Statistical Information Office, Population Division (in press). 3. NCHS. Vital statistics of the United States, 1992. Vol 1--natality. Hyattsville, Maryland: US Department of Health and Human Services, Public Health Service, CDC (in press). 4. Ventura SJ, Martin JA, Taffel SM, et al. Advance report of final natality statistics, 1992. Hyattsville, Maryland: US Department of Health and Human Services, Public Health Service, CDC, 1994. (Monthly vital statistics report; vol 43, no. 5, suppl). 5. CDC. Abortion surveillance: preliminary data--United States, 1992. MMWR 1994;43:930-3,939. 6. CDC. Trends in sexual risk behavior among high school students--United States, 1990, 1991, and 1993. MMWR 1995;44:131-2. 7. Kirby D, Short L, Collins J, et al. School-based programs to reduce sexual risk behaviors: a review of effectiveness. Public Health Rep 1994;109:339-59. 8. Henshaw SK, VanVort J. Abortion services in the United States, 1991 and 1992. Fam Plann Perspect 1994;26:100-12. 9. Ventura SJ, Taffel SM, Mosher WD, et al. Trends in pregnancies and pregnancy rates: estimates for the United States, 1980-92. Hyattsville, Maryland: US Department of Health and Human Services, Public Health Service, CDC, 1995. (Monthly vital statistics report; vol 43, no. 11, suppl). * For 47 reporting areas, data were provided from the central health agency (state health departments and the health departments of DC, upstate New York, and New York City). Data from upstate New York and New York City were combined to produce totals for the state. For the other five states, data were provided from hospitals and other medical facilities. The word "state" in this report refers to both states and DC except where DC is mentioned explicitly. Wisconsin and DC reported age for those who had abortions among residents only. ** Reasons for the exclusion of rates in some states are listed in the tables; the following hierarchy was used to determine whether data were excluded: 1) abortion data by age or race or birth data by Hispanic ethnicity were not reported by state; 2) less than 20 births or pregnancies or less than 1000 women were in the group; and 3) for greater than 15% of the abortion data, age or race of the women was unknown. *** DC is not included in these comparisons because it is an urban area, and pregnancy rates are generally higher in urban areas than in states. Syringe Exchange Programs -- United States, 1994-1995 As of December 1994, approximately one third (35.3%) of the 435,319 cases of acquired immunodeficiency syndrome (AIDS) reported among adults to CDC were associated with injecting-drug use (1). In addition, injection of illegal drugs is the risk behavior most frequently associated with heterosexual and perinatal transmission of human immunodeficiency virus (HIV) in the United States. The goal of syringe exchange programs (SEPs) is to reduce HIV transmission associated with drug injection by providing sterile syringes in exchange for used, potentially HIV-contaminated syringes. This report presents data from a recent survey of U.S. SEPs about their activities during January 1994-April 1995 and compares the findings with those of a 1993 survey (2).* In April 1995, the North American Syringe Exchange Network (NASEN), in collaboration with the U.S. Conference of Mayors and Beth Israel Medical Center (New York City), mailed questionnaires to the directors of each of the 68 U.S. SEPs that were members of NASEN. Directors of SEPs from which a completed questionnaire was not returned within 3 weeks were contacted by telephone. Data collected included information about the SEP operations, legal status, services offered, number of syringes exchanged in 1994, and outreach efforts. In the April 1995 survey, 60 (88%) SEPs provided data (47 [78%] by mail and 13 [22%] by telephone). These 60 SEPs reported operating in 46 cities in 21 states**. Forty-two (70%) of the SEPs were located in five states (California, New York, Washington, Connecticut, and Hawaii); in nine cities, at least two SEPs reported operating. In the 1993 survey, a total of 33 SEPs reported operating in 29 cities in 12 states (2). The 55 SEPs operating in 1994 reported exchanging approximately 8 million new, sterile syringes for used syringes during January-December 1994 (median: 39,014 syringes per SEP; mean: 145,914). The seven most active SEPs (i.e., those that exchanged greater than or equal to 500,000 syringes; two SEPs in New York City and one each in Chicago; Philadelphia; San Francisco; and Seattle and Tacoma, Washington) exchanged nearly 5.5 million syringes, representing 68% of all syringes exchanged by SEPs in 1994 (Table 1). The San Francisco SEP reported exchanging the largest number of syringes (1.5 million) in 1994. Some SEPs reported exchanging relatively small numbers of syringes in 1994: 31 SEPs (56%) exchanged less than or equal to 55,000 syringes each while 12 SEPs (22%) exchanged less than 10,000 syringes each. In comparison, approximately 2.4 million syringes were exchanged by U.S. SEPs in 1992 (2). In addition to syringe exchange, services provided by SEPs included provision of latex condoms (45 SEPs), HIV counseling and testing (23), tuberculin skin testing (12), primary health care (10), and directly observed tuberculosis therapy (six). Most (45 [85%] of 53) SEPs reported counseling injecting-drug users (IDUs) to follow medical hygiene standards when injecting illegal drugs (i.e., prepare the injection site with an alcohol swab; use a new, sterile needle and syringe for each injection; avoid reuse of syringes [even by the same person]; use clean [ideally sterile] water to prepare drugs for injection; and return used syringes to the SEP for safe disposal). In both the 1993 and 1995 surveys, the legal status of SEPs was categorized as legal, illegal-but-tolerated, and illegal/underground. An SEP was defined as legal if it operated in a state that had no law requiring a prescription to purchase a hypodermic syringe (i.e., a "prescription law") or had an exemption to the state prescription law allowing the SEP to operate; illegal-but-tolerated if the program operated in a state with a prescription law and had received a formal vote of support or approval from a local elected body (e.g., a city council); and illegal/underground if the program operated in a state with a prescription law but had no formal support from local elected officials. Of the 60 SEPs in the 1995 survey, a total of 33 (55%) reported that they were legal; 19 (32%), illegal-but-tolerated; and eight (13%), illegal/underground. Reported by: D Paone, EdD, DC Des Jarlais, PhD, J Clark, Q Shi, MS, A Orris, Beth Israel Medical Center; M Krim, PhD, M Reinfeld, American Foundation for AIDS Research; SR Friedman, PhD, National Development and Research Institutes, New York. D Purchase, H Smith, North American Syringe Exchange Network, Tacoma, Washington. P Jones, US Conference of Mayors, Washington, DC. P Lurie, MD, Univ of California, San Francisco. Div of HIV/AIDS Prevention, National Center for Prevention Svcs, CDC. Editorial Note: Practices associated with injection of heroin, cocaine, meth- amphetamine, and other drugs can be linked to transmission of HIV, hepatitis B virus (HBV), and hepatitis C virus (HCV) (1,3). During intravenous injection, IDUs usually draw blood into the needle and syringe to verify the needle is in a vein. If that blood contains HIV, HBV, HCV, or other bloodborne pathogens, subsequent use of the syringe by another drug injector may result in transmission of these pathogens (4). To assist in reducing transmission of HIV, HBV, HCV, and other bloodborne pathogens (5), SEPs provide new, sterile syringes for IDUs and collect used, blood-contaminated syringes. In addition, nearly all SEPs provide alcohol swabs, latex condoms, and counseling services to clients, and many assist clients in obtaining health and social services (e.g., HIV counseling and testing, tuberculin skin testing, or admission to drug-treatment centers) either on-site or by referral (2). SEPs have been widely implemented as an HIV-prevention intervention in Australia, Canada, Netherlands, and the United Kingdom. In the United States, the first SEP was established in Tacoma, Washington, in 1988 (2). Based on the survey in April 1995, the number of SEPs operating in the United States increased 82% over that in 1993; in addition, the number of syringes exchanged by SEPs increased threefold from 2.4 million in 1992 (2) to 8 million in 1994. The findings in this report are subject to at least two limitations. First, the extent of SEP activities probably is underestimated because of incomplete participation in the survey by known SEPs and because some operational SEPs may be unknown to the NASEN. Second, some SEPs that participated in the 1995 survey included in their reported data information from separate, independent SEPs. Previous studies demonstrate the effectiveness of SEPs and other interventions that increase access to sterile syringes in preventing HIV infection. For example, participation by IDUs in SEPS in Tacoma was associated with substantially lower risk for hepatitis B and hepatitis C among IDUs (sixfold and sevenfold lower, respectively) (6). The National Academy of Sciences recently reviewed research on SEPs and, in a September 1995 report, concluded that SEPs should be regarded as an effective component of a comprehensive strategy to prevent infectious disease (7). In addition, in Connecticut, simultaneous partial repeal during 1992 of a law that required a prescription to purchase syringes and a law that specified possession of syringes as illegal was followed by increased purchasing of syringes from pharmacies by IDUs and decreased sharing of injection equipment (6,8). References 1. CDC. HIV/AIDS surveillance report, 1994. Atlanta, Georgia: US Department of Health and Human Services, Public Health Service, 1995:11-2. (Vol 6, no. 2). 2. Lurie P, Reingold AL, Bowser B, et al. The public health impact of needle exchange programs in the United States and abroad. Vol I. San Francisco, California: University of California, October 1993. 3. Alter MJ. Epidemiology of hepatitis C in the west. Seminars in Liver Disease 1995;15:5-14. 4. Jose B, Friedman SR, Curtis R, et al. Syringe-mediated drug-sharing (back loading): a new risk factor for HIV among injecting drug users. AIDS 1993;7:1653-60. 5. Haverkos HW, Jones TS. HIV, drug-use paraphernalia, and bleach. J Acquir Immune Defic Syndr 1994;7:741-2. 6. Valleroy LA, Weinstein B, Jones TS, et al. Impact of increased legal access to needles and syringes on community pharmacies' needle and syringe sales--Connecticut, 1992-1993. J Acquir Immune Defic Syndr 1995;10:73-81. 7. Normand J, D Vlahov, LE Moses, eds. Preventing HIV transmission: the role of sterile needles and bleach. Washington, DC: National Academy Press, 1995. 8. Groseclose SL, Weinstein B, Jones TS, et al. Impact of increased legal access to needles and syringes on practices of injecting-drug users and police officers--Connecticut, 1992-1993. J Acquir Immune Defic Syndr 1995;10:82-89. *Single copies of this report will be available until September 22, 1996, from the CDC National AIDS Clearinghouse, P.O. Box 6003, Rockville, MD 20849-6003; telephone (800) 458-5231 or (301) 217-0023. **California (18 SEPs), New York (nine), Washington (eight), Connecticut (four), Hawaii (three), Illinois and Minnesota (two each), and one each in Alasaka, Colorado, Indiana, Louisiana, Maryland, Massachusetts, Michigan, Missouri, New Jersey, New Mexico, Oregon, Pennsylvania, Texas, and Wisconsin. Fatalities From Motor-Vehicle Collisions With Trains -- Kansas, 1990-1994 During 1983-1992, a total of 5831 deaths in the United States were attributed to motor-vehicle collisions with trains.* During that same period, Kansas had the third highest death rate in the United States from motor-vehicle collisions with trains, and the annual rate for the state (0.8 per 100,000 persons) was approximately four times the national rate (0.2 deaths per 100,000 persons). To identify approaches for preventing such collisions, the Kansas Department of Health and Environment (KDHE) characterized all fatal motor-vehicle collisions with trains at highway-rail grade crossings** in the state from 1990 through 1994. This report summarizes the results of that study. Information about motor-vehicle collisions with trains was obtained from the Federal Railroad Administration and the Kansas Corporation Commission, which receive incident reports from railroads following motor-vehicle collisions with trains. Specific information about drivers involved in fatal collisions with trains (e.g., age, sex, and blood alcohol content [BAC]) was obtained from the Kansas Department of Transportation, the Office of Vital Statistics in the KDHE, and the Kansas Bureau of Investigation. Additional information was obtained from published newspaper reports. During 1990-1994, a total of 510 motor-vehicle collisions with trains occurred in Kansas, representing annual rates of 4.0 collisions per 100,000 persons and 4.2 collisions per 1 billion vehicle-miles driven. Injuries to 233 persons (167 nonfatal and 66 fatal) were reported in 186 (36%) collisions, of which 53 (10%) resulted in at least one fatality. Of these 53 collisions, drivers were killed in 52. Forty-four (83%) of the 53 motor-vehicle drivers involved in fatal collisions were male; in comparison, 50% of all licensed drivers in the state during 1992 were male (rate ratio=4.9; 95% confidence interval=2.4-10.0). The median age of drivers was 33 years (range: 14-86 years); three (6%) were aged less than 18 years and nine (17%) were aged greater than or equal to 65 years, a pattern of age distribution similar to that for all licensed drivers in Kansas. Of the 53 collisions that resulted in fatalities, 35 (66%) occurred during daylight, 50 (94%) occurred during fair weather (i.e., clear or cloudy), and 44 (83%) occurred in rural areas. The number of collisions involving fatalities was similar by day of the week and by month of the year. Fifty (94%) fatal collisions involved freight trains, and three involved the one passenger train service in the state. These trains had a median of three locomotives (range: one to eight locomotives) and 56 cars (range: one to 127 cars) and were traveling at an average estimated speed of 45 mph (range: 0-90 mph) at the time of the collision. Of the 53 motor vehicles involved in fatal collisions, 32 (60%) were automobiles; 16 (30%), trucks; two (4%), farm tractors; and three (6%), other types of vehicles. Five motor vehicles were stopped or stalled on the railroad tracks at the time of the collision. For the 48 motor vehicles moving at the time of the collision, the median estimated speed was 25 mph (range: 4-75 mph). In 48 (91%) collisions, the motor vehicle was struck by or struck the lead engine of the train; in the remaining five (9%), the motor vehicle struck the side of the train behind the lead engine. In these five side-impact collisions, three occurred at night, one was an apparent suicide, and one occurred after the vehicle skidded 176 feet in an attempt to stop before reaching the rail crossing. Thirty-two (60%) drivers were killed in a collision in their county of residence, and six (11%) drivers were killed while working. Of the 28 (53%) drivers who were tested postmortem for BAC, detectable levels (greater than or equal to 0.02 g/dL) were present in 10 (19%), including six (11%) who were legally intoxicated (greater than or equal to 0.10 g/dL). The manner of death was specified on the death certificate for 49 drivers: of these, 47 (96%) were considered unintentional injuries or "accidents,"*** and two (4%) were considered suicides. Thirty-three (62%) drivers did not stop at the highway-rail crossing before the collision; two (4%) drivers stopped and then proceeded before the collision. Five (9%) drivers drove behind or in front of a train and struck or were struck by a second train on a parallel set of tracks, and seven (13%) motorists drove around or through crossing gates. For the 51 grade crossings at which collisions involving fatalities occurred, 49 (96%) crossings had one fatal collision each, and two (3.9%) crossings had two each. All crossings had some type of warning device. At 37 (73%) crossings, passive warning devices were present, including 32 (63%) at which the crossings were marked only by crossbucks (i.e., black and white X-shaped signs that read "Railroad Crossing"). At 14 (27%) crossings, at least one type of active warning device (e.g., gates or flashing lights) was present. In five (9%) collisions, the view of the railroad track was obstructed at the crossing by standing railroad equipment, a passing train, topography, or vegetation. Reported by: B George, Federal Railroad Administration. N Mattson, D Vialle, Kansas Dept of Transportation; V Wenger, Kansas Corporation Commission; D Rundle, Kansas Bur of Investigation; K Sommer, Center for Health and Environmental Statistics, G Pezzino, MD, Acting State Epidemiologist, Kansas Dept of Health and Environment. Div of Unintentional Injury Prevention, National Center for Injury Prevention and Control; Div of Field Epidemiology, Epidemiology Program Office, CDC. Editorial Note: Although the number of motor-vehicle collisions with trains in the United States has decreased substantially since 1976, each year approximately 500 persons die as the result of such collisions (1). The findings from a previous study indicate that many motorists are unaware of, or chose to ignore, the need for caution at highway-rail crossings (2). Two strategies employed to prevent collisions have included educational campaigns (e.g., "Operation Lifesaver") to alert the public to the hazards of highway-rail crossings, and engineering controls (e.g., installation of active warning systems) to improve the safety of crossings (3). The findings from Kansas suggest that vigorous use of both approaches might assist in decreasing the occurrence of fatal collisions at highway-rail crossings. In particular, educational efforts should encourage drivers to approach every highway-rail crossing as if a train were approaching and to take appropriate precautions. The findings in Kansas also indicate that drivers may fail to heed passive warning systems and that the presence of obstructed views may contribute to some collisions. Of the 8040 public highway-rail crossings in Kansas in 1993, a total of 6502 (81%) had only passive warnings (1). However, even though collisions at crossings with passive warning devices accounted for 70% of fatal crashes, the risk for fatal collisions at crossings in relation to the presence of warning devices cannot be determined without estimates of both train and motor-vehicle traffic volume. Because the probability of a collision is two to 40 times higher at crossings without gates (4,5), unnecessary crossings should be eliminated and remaining crossings should be upgraded. However, upgrading is expensive; installation of active warning devices costs approximately $150,000-$200,000 per crossing in Kansas (Kansas Department of Transportation, unpublished data). This analysis did not assess data about collisions involving nonfatal injuries; therefore, the findings may not be representative of all collisions involving injuries. Further research is necessary to identify risk factors for fatal and nonfatal collisions. An action plan to improve highway-rail crossing safety by the Federal Highway Administration, the Federal Railroad Administration, the Federal Transit Administration, and the National Highway Traffic Safety Administration (6) includes six major initiatives (i.e., promoting enforcement of traffic laws, encouraging safety reviews, supporting public education, reviewing private-crossing issues, conducting research and making data available, and fostering trespass prevention) encompassing 55 specific proposals. In addition, this plan has targeted as a goal for the year 2004 reduction nationally of the number of fatalities from motor-vehicle collisions with trains per year by at least 50%. References 1. Federal Railroad Administration. Highway-rail crossing accident/incident and inventory bulletin, no. 16, calendar year 1993. Washington, DC: US Department of Transportation, 1994. 2. Wigglesworth EC. The effects of local knowledge and sight restrictions on driver behavior at open railway crossings. J Safety Res 1978;10:100-7. 3. National Transportation Safety Board. Safety study: passenger/commuter train and motor vehicle collisions at grade crossings (1985). Washington, DC: National Transportation Safety Board, 1986; report no. NTSB/SS-86/04. 4. Meeks KD, Robertson LS. Study of road-rail crashes in Claremore, OK, and allocation of resources for preventive measures. Public Health Rep 1993;108:248-51. 5. Tenkink E, Van der Horst R. Car driver behavior at flashing light railroad grade crossings. Accid Anal Prev 1990;22:229-39. 6. Federal Railroad Administration. Rail-highway crossing safety action plan support proposals. Washington, DC: US Department of Transportation, 1994. * Classified according to the International Classification of Diseases, Ninth Revision (ICD-9), as E810. These data were obtained from CDC's Compressed Mortality File (CMF), which contains information from death certificates filed in the 50 states and the District of Columbia that have been prepared in accordance with external cause codes. CDC's Wide-ranging ONline Data for Epidemiologic Research (WONDER) computerized information system was used to obtain CMF data. ** Defined by the Federal Railroad Administration as a location where one or more railroad tracks intersect a public or private thoroughfare, a sidewalk, or a pathway. *** When a death occurs under "accidental" circumstances, the preferred term within the public health community is "unintentional injury." Economic Costs of Birth Defects and Cerebral Palsy -- United States, 1992 Birth defects are the leading cause of infant mortality in the United States and the fifth leading cause of years of potential life lost (1,2). Despite the substantial allocation of medical and nonmedical resources to the care and support of persons with birth defects, the economic costs of such defects have not been estimated accurately. Because estimates of the cost per new case of a birth defect represent the savings from preventing a case, an incidence-based approach enables assessment of the value of prevention strategies. This approach was used to estimate the cost of illness for cerebral palsy and for 17 of the most clinically important structural birth defects in the United States. This report uses data from California (adjusted to provide national estimates) and national data (Table 1) to estimate the costs of these 18 conditions occurring in the United States during 1992. Using a human capital approach,* estimates were made of the direct costs of medical, developmental,** and special education services and the indirect costs of lost work and household productivity attributable to premature morbidity and mortality of the cohort of persons born in California during 1988*** with any of the 18 conditions (6,7). Estimates were adjusted to reflect national costs in 1992 dollars and to avoid duplication when a child had more than one condition. Estimated costs of medical and other services used by children without these conditions were subtracted to yield the cost of each condition. The cost of associated conditions (e.g., cardiac anomalies with Down syndrome) were included because prevention of defects was presumed to prevent such conditions. The number of new cases of the conditions were estimated using data from the California Birth Defects Monitoring Program (CBDMP). Prevalence estimates were derived from CBDMP and from a combined sample of CDC's National Health Interview Surveys for 1985-1989 (8). For each condition, estimates of excess mortality through the first year of life were based on a CBDMP study linking birth and death records. Estimates of age-specific direct costs of the conditions were based on reported charges and expenditures for children with the conditions. For several conditions, limitations in the data restricted the incorporation of certain costs and the period of time during which costs could be assessed. For example, the long-term excess costs of education for persons with certain conditions was not available. For 1992, the combined estimated cost of the 18 conditions in the United States was $8 billion (Table 2). Costs ranged from $75,000 to $503,000 per new case. Conditions with the highest costs per case were characterized by relatively high levels of long-term activity limitations (e.g., cerebral palsy [$503,000], Down syndrome [$451,000], and spina bifida [$294,000]). In addition, these conditions had among the highest total lifetime costs ($2.4 billion, $1.8 billion, and $489 million, respectively), reflecting their relatively high incidences. The high cost per new case of major heart defects reflects the high medical costs associated with early surgical interventions for these defects and high costs of lost productivity attributable to deaths during the first year of life. Reported by: NJ Waitzman, PhD, Univ of Utah, Salt Lake City. PS Romano, MD, Univ of California, Davis; RM Scheffler, PhD, Univ of California, Berkeley; JA Harris, MD, California Birth Defects Monitoring Program, California Dept of Health Svcs. Div of Birth Defects and Developmental Disabilities, National Center for Environmental Health, CDC. Editorial Note: The findings in this report indicate that cerebral palsy and 17 of the most clinically important birth defects in the United States cause substantial economic burden. If all of the approximately 120,000 infants (3% of all live births) born each year in the United States with serious birth defects had been included in this analysis, the economic costs would have been higher. These cost estimates provide a basis for assessing prevention strategies using cost-benefit and cost-effectiveness analyses. Because the medical and nonmedical services provided to persons with the 18 conditions often continue into adulthood, the cost estimates for these conditions were particularly sensitive to the choice of discount rate (6). In this analysis, a discount rate of 5% was used to compute the present value of money to be spent or received in the future. The findings in this report are subject to at least four limitations. First, California data used to estimate incidence rates and treatment costs may not be representative of the United States; therefore, total costs per case may vary by state. Second, the contribution of time by family members to the provision of care was not estimated and may be substantial for some conditions. Third, the psychosocial costs of illness--which may exceed traditional human capital costs--also were not included (10). For these and other reasons, the use of the human capital approach underestimates what the public is willing to pay to prevent these conditions (9). Finally, excess medical and education costs probably were underestimated for some conditions because they could not be ascertained completely. Prevention of birth defects can substantially reduce their economic burden. In 1992, the Public Health Service recommended that all women capable of becoming pregnant consume 0.4 mg of folic acid (a B vitamin) to reduce their risk for a pregnancy affected by spina bifida or anencephaly (11). Based on the estimates in this report, if this recommendation were fully implemented, a substantial proportion of the $489 million in total costs associated with spina bifida could be averted. The high personal and societal costs of birth defects underscore the need to develop and implement effective primary-prevention programs. References 1. CDC. Years of potential life lost before ages 65 and 85--United States, 1987 and 1988. MMWR 1990;39:20-2. 2. CDC. Infant mortality--United States, 1989. MMWR 1992;41:81-5. 3. SRI International. The National Longitudinal Transition Study of Special Education Students: data documentation. Prepared by Kathryn A. Valdes for the Office of Special Education Programs, US Department of Education, under contract 300-87-0054, 1990. 4. Bureau of the Census, US Department of Commerce. Survey of income and program participation (SIPP), 1987 panel (Wave VI Rectangular Core and Topical Module File) [Computer file]. Washington, DC: US Department of Commerce, Bureau of the Census (producer), 1990. Ann Arbor, Michigan: Inter-University Consortium for Political and Social Research (distributor), 1991. 5. Rice DP, Max W. The cost of smoking in California, 1989. Sacramento, California: California State Department of Health Services, 1992. 6. Waitzman NJ, Romano PS, Scheffler RM. Estimates of the economic costs of birth defects. Inquiry 1994;31:188-205. 7. Waitzman NJ, Scheffler RM, Romano PS. The economic costs of birth defects, 1996. Lanham, Maryland: University Press of America (in press). 8. NCHS. Public-use data tape documentation, part I, National Health Interview Survey, 1988 [Machine-readable data file and documentation]. Hyattsville, Maryland: US Department of Health and Human Services, Public Health Service, CDC, 1989 (producer). National Technical Information Service, U.S. Department of Commerce, Springfield, Virginia (distributor), 1989. 9. Haddix A, Teutsch S, Shaffer P, Dunet D, Churchill E. A practical guide to prevention effectiveness: decision and economic analysis. Atlanta, Georgia: US Department of Health and Human Services, Public Health Service, CDC, 1994. 10. Hu T, Sandifer FH. Synthesis of cost of illness methodology. Washington, DC: Georgetown University, Public Services Laboratory, 1981; National Center for Health Services Research contract no. 233-79-3010. 11. CDC. Recommendations for the use of folic acid to reduce the number of cases of spina bifida and other neural tube defects. MMWR 1992;41(no. RR-14). * A method for estimating the economic cost of disease that includes the resources used for medical care and the productivity losses resulting from morbidity and premature mortality. Intangible costs (e.g., "pain and suffering") are excluded from estimates using this approach. ** Nonmedical services provided to children outside the educational system. Services were grouped into four categories: out-of-home services (e.g., community-care centers), day programs, camps (including day residential and respite care), and other services (e.g., training for independent living, driver training, and interpreters). *** The most recent year for which data were available.