Mortality Patterns—United States, 1997 FREE

MMWR. 1999;48:664-668

1 figure, 1 table omitted

In 1997, a total of 2,314,245 deaths were registered in the United States—445 fewer than the record high of 2,314,690 in 1996.¹ The overall age-adjusted death rate* was 479.1 per 100,000 standard (1940) population, the lowest ever recorded. In 1997, nearly two thirds of deaths resulted from heart disease, cancer, and stroke. This report summarizes mortality patterns in 1997¹ and compares them with patterns in 1996.

National death statistics are based on information from death certificates filed in state vital statistics offices and are compiled by CDC into a national database. Cause-of-death statistics are based on the underlying cause of death.† Causes of death are recorded on the death certificate by the attending physician, medical examiner, or coroner using a format specified by the World Health Organization and endorsed by CDC.

Compared with 1996, death rates decreased for all age groups except persons aged ≥85 years. The largest percentage decreases occurred in persons aged 25-34 years (9.2%), 35-44 years (8.2%), and 1-4 years (6.5%).

From 1996 to 1997, age-adjusted death rates declined among whites (from 466.8 to 456.5) and among blacks (from 738.3 to 705.3).‡ In 1997, the overall death rates for the black population were higher than for the white population; for seven of the 15 leading causes, age-adjusted death rates were at least 1.5 times greater for blacks than for whites. The largest differences in rates were for human immunodeficiency virus (HIV) infection (7.5 times) and homicide (6.0 times).§ Death rates were lower for blacks than whites for three leading causes: chronic obstructive pulmonary disease (0.8 times), Alzheimer's disease (0.7 times), and suicide (0.6 times). The 1997 age-adjusted death rates declined 4.3% from 1996 for the Hispanic population (from 365.9 to 350.3). The three leading causes of death for Hispanics were heart disease, cancer, and unintentional injuries.

In 1997, age-adjusted death rates for males were higher than for females. From 1996 to 1997, age-adjusted death rates declined for males (from 623.7 to 602.8) and for females (from 381.0 to 375.7). Of the 15 leading causes of death, the greatest difference between the rates for the sexes was for suicide; the suicide rate was more than four times greater for males than for females. Also higher for males was the death rate for homicide (3.8 times) and HIV infection (3.5 times).

In 1997, 327 women died from maternal causes, including complications of pregnancy, childbirth, and the puerperium§* within 42 days after pregnancy termination. The maternal mortality rate was 8.4 deaths per 100,000 live births, and was more than three times higher for black than for white women.

In 1997, the infant mortality rate was 7.2 infant deaths per 1000 live births; in 1996, infant mortality was higher but the difference was not statistically significant. Among the 10 leading causes of infant death,§† only pneumonia and influenza decreased by a statistically significant amount during 1996-1997. The infant mortality rate was two times higher for black infants than for white infants.¹

From 1996 to 1997, mortality increased from septicemia (2.4%) and kidney disease (4.4%); however, mortality decreased for the three leading causes of death: heart disease (3.0%), cancer (1.8%), and stroke (1.9%). HIV-infection mortality dropped in ranking from the eighth leading cause in 1996 to the 14th in 1997 (Table 1). The age-adjusted death rate for HIV infection decreased 47.7%, the largest decline among the 15 leading causes of death. In 1997, 16,516 deaths were attributed to HIV infection. Age-adjusted death rates for HIV were highest for black males (38.5), black females (13.3), white males (5.6), and white females (1.0). HIV infection continued to be the fifth leading cause of death for black females aged 15-24 years, the sixth for black males aged 5-14 years, the sixth for black males aged 15-24 years, and the leading cause for black males aged 25-44 years.

In 1997, overall life expectancy (LE) at birth was 76.5 years. The overall LE increased by 0.4 years from the 1996 LE primarily because of decreases in mortality from HIV infection, heart disease, cancer, stroke, and homicide. White females continue to have the highest LE at birth (79.9 years), followed by black females (74.7 years), white males (74.3 years), and black males (67.2 years). All four race-sex groups had increases in LE during 1996-1997 and achieved record high life expectancies. The gap between the white and black population is 6.0 years, down from 6.6 years in 1996. The gap between men and women is 5.8 years, down from 6.0 years in 1996.

Mortality Statistics Br, Div of Vital Statistics, National Center for Health Statistics, CDC.

This report is based on all the death records registered in the United States in 1997 and indicates that decreases have occurred in the risk for death from the top three causes and from HIV infection. Progress in preventing and treating these conditions, however, is offset by increases in mortality from septicemia, kidney disease, and drug-induced causes. The differences in LE by race and sex narrowed in 1997 but disparities remain large and may reflect such factors as socioeconomic status, access to medical care, and the prevalence of specific risks.

Advances in treatment for HIV and acquired immunodeficiency syndrome (AIDS), such as the use of triple combination antiretroviral therapy, resulted in decreases in AIDS incidence and HIV mortality and increases in the number of persons living with HIV and AIDS.^2,3 During 1987-1994, HIV infection mortality increased an average of 16% annually. In 1995, the age-adjusted death rate for HIV infection was approximately the same as in 1994. Then mortality began to decrease: in 1996 by 28.8% and in 1997 by 47.7%.

LE has increased every year since 1993, the major reasons during 1996-1997 being reduced risk for homicides among teenagers and HIV infection among working age adults, and reduced risk for deaths attributable to heart disease, cancer, and stroke among older persons.

Data in this report are subject to at least two limitations. First, death rates for the American Indian/Alaskan Native and Asian/Pacific Islander populations are not included because of inaccuracies on death certificates and in population censuses that result in reported death rates being lower than actual death rates.⁴ Similar but less severe problems affect the Hispanic population.⁴ Targeted research and evaluation is needed to assess reporting problems and to identify methods that would compensate for inaccuracies.⁴ A second limitation is the quality of medical cause-of-death information on the death certificate. Physicians, medical examiners, and coroners sometimes are not trained in the correct completion of this form. Approaches to address this problem include expanded availability of continuing medical education, instructional materials,^5-7 and World-Wide Web resources.§‡

Mortality data from the National Vital Statistics System have been used to document public health trends since 1900 and are key indicators for monitoring groups at risk for death from specific diseases and injuries.⁸ Additional information is available from the National Center for Health Statistics, CDC, 6525 Belcrest Rd., Room 1064, Hyattsville, MD 20782; telephone (301) 436-8500; or from the World-Wide Web, http://www.cdc.gov/nchswww/about/major/dvs/mortdata.htm.

References: 8 available

*Age-adjusted to the 1940 U.S. population. Age-adjusted death rates indicate changes in the risk for death more effectively than crude death rates and are better indicators for comparing mortality of population groups with different age structures.

†Defined by the World Health Organization's International Classification of Diseases, Ninth Revision (ICD-9) as "(a) the disease or injury which initiated the train of morbid events leading directly to death, or (b) the circumstances of the accident or violence which produced the fatal injury."

‡Hispanic ethnicity is independent of race, therefore, Hispanics are included in the race categories. Data for other racial groups were not included because of reporting inaccuracies on death certificates and population censuses.

§"Motor-vehicle accidents" and "all other accidents and adverse effects" are not included as causes of death for which the rate has decreased because these causes are subcategories of the leading cause "accidents and adverse effects." When a death occurs under "accidental" circumstances, the preferred term within the public health community is "unintentional injury."

§*ICD-9, codes 630-676.

§†Congenital anomalies; disorders relating to short gestation and unspecified low birthweight; sudden infant death syndrome; respiratory distress syndrome; newborn affected by maternal complications of pregnancy; newborn affected by complications of placenta, cord, and membranes; infections specific to the perinatal period; accidents and adverse effects; intrauterine hypoxia and birth asphyxia; and pneumonia and influenza.

§‡From the National Center for Health Statistics, information on writing cause-of-death material is available at http://www.cdc.gov/nchswww/about/major/dvs/handbk.htm, and from the National Association of Medical Examiners, information is available at http://www.thename.org/main.htm. References to sites of nonfederal organizations on the World-Wide Web are provided as a service to MMWR readers and do not constitute or imply endorsement of these organizations or their programs by CDC or the U.S. Department of Health and Human Services. CDC is not responsible for the content of pages found at these sites.

MMWR. 1999;48:777-779

In October 1998, the Maryland Department of Health and Mental Hygiene (MDH) was notified that a public sexually transmitted disease (STD) clinic in a county (county A) had used a nonrecommended preparation to treat syphilis patients during January-October 1998. The clinic had been inadequately treating syphilis patients or syphilis contacts with Bicillin®* C-R (a mixture of 1.2 million units [MU] benzathine penicillin G [BPG] and 1.2 MU procaine penicillin G), rather than with Bicillin® L-A (2.4 MU BPG). Compared with short-acting procaine penicillin G, BPG has a longer half-life considered essential for effective syphilis treatment because it yields sustained spirochetecidal levels needed to treat the slowly reproducing agent of syphilis, Treponema pallidum. The inadvertent use of Bicillin C-R, which contains only half the recommended dose of BPG for syphilis, was recognized by a health-care provider at the STD clinic in a neighboring county (county B) approximately 1 month after county B had borrowed BPG from county A. This report summarizes the investigation of the use of Bicillin C-R to treat STD patients in county A and discusses the frequency of Bicillin C-R use in STD clinics nationwide. Findings of this investigation indicate that inadvertent Bicillin C-R use is more frequent than previously known and that preventive measures should be taken to minimize such use.

Three BPG-containing products are marketed by Wyeth-Ayerst Laboratories (Philadelphia, Pennsylvania): Bicillin L-A, Bicillin C-R, and Bicillin® C-R 900/300 (a mixture of 0.9 MU BPG and 0.3 MU procaine penicillin G). Besides having similar proprietary names, the package and label for Bicillin C-R and Bicillin L-A have similar lettering and colors. Bicillin L-A is recommended for treating syphilis patients and upper respiratory tract infections caused by susceptible streptococci.¹ The efficacy of Bicillin C-R to treat syphilis is unknown. The package insert for Bicillin C-R states that this product should not be used to treat syphilis, gonorrhea, yaws, bejel, or pinta.

To identify patients who might have been treated with Bicillin C-R at county A's STD clinic, investigators reviewed the clinic's invoice records and the penicillin injection log. MDH searched its STD surveillance database for residents from county A who were treated for syphilis or had a positive syphilis serology during January-October 1998.

During December 1997-May 1998, 150 doses of Bicillin C-R were shipped to county A's STD clinic after orders for Bicillin L-A were placed by telephone. During January-October 1998, 123 of 160 doses of penicillin administered for syphilis were Bicillin C-R. Sixty-three patients, including five pregnant women, might have received Bicillin C-R. Because the efficacy of Bicillin C-R for treatment of patients with syphilis is unknown, the clinic attempted to contact and treat all patients with Bicillin L-A. During this period, routine outreach activities were suspended, clinic hours were extended, and personnel were asked to work overtime.

Clinic workers contacted patients by telephone, and subsequent clinical evaluations were made by two nurses. STD field staff visited patients in their homes; multiple attempts were often needed to locate and counsel patients. Although the five pregnant women were located and treated with Bicillin L-A, four infants were treated for congenital syphilis because their mothers had not been treated adequately at least 30 days before delivery. None of the infants had congenital syphilis.

After 8 weeks of follow-up efforts, 52 (82%) of the 63 patients had been restaged and retreated; the remaining 11 patients either could not be located (one) or refused retreatment (10). The total estimated direct costs of follow-up efforts conducted by county A's clinic was approximately $24,000.

In county B, 10 syphilis patients received Bicillin C-R during an 11-day period according to the clinic's syphilis treatment records. Of these, eight were treated with Bicillin L-A, one was not located, and the other refused further treatment.

To determine the frequency of Bicillin C-R use in STD clinics nationwide and to educate STD program managers about the possible confusion between Bicillin C-R and Bicillin L-A, CDC surveyed 65 STD program areas during January-February 1999 about unintentional Bicillin C-R use from 1993 through 1998. Fifty-seven of the 65 program areas were state/city program areas, and the remainder were islands and territories; 55 (96%) of the state/city program areas responded to the survey. Of these, 45 (82%) used only Bicillin L-A to treat syphilis patients, three used Permapen® exclusively (a BPG product from Pfizer, Inc. [New York, New York]), and seven used both Permapen and Bicillin L-A. Besides the Maryland clinics, four program areas reported unintentional Bicillin C-R use at least once from 1993 through 1998. In two areas, Bicillin C-R was received at the state health department and was distributed to STD clinics statewide; the administration of a nonrecommended regimen subsequently occurred at many local STD clinics. Two other areas reported unintentional use of Bicillin C-R at individual clinics (one area reported multiple occurrences). In March 1999, unintentional use of Bicillin C-R was reported from a program area that had responded negatively to the earlier survey. The number of persons who received a nonrecommended regimen in this incident could not be determined.

Among the 55 state/city program areas that responded to the survey, 31 (56%) were unaware of the possible confusion between Bicillin C-R and Bicillin L-A; 24 (46%) program areas routinely ordered Bicillin L-A by telephone.

D Dwyer, MD, State Epidemiologist, Maryland Dept of Health and Mental Hygiene. Div of STD Prevention, National Center for HIV, STD, and TB Prevention; and an EIS Officer, CDC.

The inadvertent use of Bicillin C-R in county A's STD clinic disrupted routine public health functions and incurred substantial monetary costs to the clinic and unnecessary discomfort to patients. Such incidents may undermine the credibility of and trust in health departments on the part of affected patients and the broader community. Although no treatment failures or congenital syphilis cases were associated with this incident, treatment according to standard guidelines was missed for patients who either could not be relocated or refused retreatment.

In addition to Maryland, five program areas reported unintentional use of Bicillin C-R from 1993 through 1998. This number should be viewed as a conservative estimate because some program areas might have failed to report such use because of concerns over liability or performance evaluation. Because most program areas surveyed were unaware of the possible confusion between Bicillin C-R and Bicillin L-A, some unintentional Bicillin C-R use could have occurred that remained unknown.

Penicillin therapy is the mainstay of treatment and a core element of syphilis prevention in the United States.^2,3 However, declining syphilis rates may have caused providers to become less familiar with the penicillin regimens appropriate for syphilis. Less attention may have been paid to clinician outreach and training for medications used to treat a disease that has declined as sharply as syphilis (83% decline in primary and secondary syphilis from 1990 to 1997 in the United States).⁴

Sustained participation by manufacturers in providing diagnostic and therapeutic products is an essential element of emerging initiatives to eliminate syphilis transmission in the United States.⁵ Increased efforts are needed to re-educate clinic managers and providers about the existence of different penicillin preparations and their appropriate usage. Written rather than telephone orders may help to minimize ordering or shipment errors. Although the most important safeguard against medication errors is that providers carefully read package labels, some label and package modifications may help decrease confusion about Bicillin products and other pharmaceuticals with similar names and labels.

References: 5 available

*Use of trade names and commercial sources is for identification only and does not imply endorsement by CDC or the U.S. Department of Health and Human Services.