National survey of mammographic facilities in 1985, 1988, and 1992

PURPOSE: To determine trends in mammography in the United States. MATERIALS AND METHODS: A sample of mammographic facilities was selected for each year of the Nationwide Evaluation of X-ray Trends. The same protocol was followed for the 1985, 1988, and 1992 surveys. Data were collected with use of the same imaging phantom for all three surveys and also with a different phantom in the 1988 and 1992 surveys. RESULTS: Of the 356 facilities surveyed in 1992, 59% claimed to be in compliance with the Health Care Financing Administration (HCFA) mammography requirements, 42% were accredited by the American College of Radiology (ACR), and 23% did not hold credentials from either the HCFA or the ACR. Since 1985, there has been a 34% improvement in acceptable phantom image quality score and a 20% decrease in the mean glandular dose. CONCLUSION: Mammography as practiced today is essentially a screen-film technique. Mammographic phantom image quality has improved considerably. The overall mean glandular dose has decreased primarily because of the elimination of xeroradiography.     

Basic characteristics of radiology groups in the United States: results of a 1991-1992 census

OBJECTIVE: The American College of Radiology (ACR), the principal professional organization of United States radiologists, receives numerous requests for information on the characteristics of radiology groups. This report describes the basic characteristics of radiology groups in the United States. We defined radiology groups as any practice with two or more radiologists or radiation oncologists, including academic departments, units in multispecialty groups, and staff of government facilities. MATERIALS AND METHODS: To collect basic information on radiology groups, the ACR conducted a mail census of all identified radiology groups in the United States during late 1991 and early 1992. Follow-up was conducted by mail and telephone. To make the responses accurately representative of all radiology groups, we weighted the approximately 2000 responses to correspond to known control totals for the number of groups of each of seven size categories in each of the four census regions (Northeast, Midwest, South, and West). These control totals were obtained from the ACR's 1990 Manpower Survey, which showed a total of approximately 3200 radiology groups. RESULTS: Approximately one fourth of all groups have two radiologists, one fourth have three or four radiologists, one fourth have five to seven radiologists, and one fourth have eight or more radiologists. Academic groups were relatively large; almost 50% had 11 or more radiologists. Nonmetropolitan areas had very few large groups, and metropolitan center cities had relatively few small groups. Ninety-two percent of all groups practiced at hospitals, and 73% of all groups practiced at nonhospital offices or centers. The median number of practice sites for all groups was three, including both hospital and nonhospital sites. Eighty-eight percent of all groups provided diagnostic radiology services, 23% provided radiation oncology, 12% offered both, and 11% were oncology-only groups. Relatively many academic groups (25%) were oncology-only groups; very few radiology groups (2%) in multispecialty practices were oncology-only groups. The diagnostic radiology techniques available from the largest percentages of groups were general radiography (plain film), sonography, mammography, and CT. One eighth of academic groups that provided diagnostic services did not report providing mammography, compared with only a few percent of all groups in the United States that provided diagnostic services. CONCLUSION: Half of all groups have two to four radiologists, and this has not changed since at least 1986. A substantial percentage of groups that perform diagnostic radiology do not provide MR, interventional, or nuclear medicine services. This is particularly true of relatively small groups. These characteristics may become the source of some problems as managed care becomes more prominent and larger groups, offering a full range of services and practicing at several sites, are favored by managed care organizations that seek to contract with one group for all their radiology services.     

Trends in breast cancer screening in Missouri from 1987 to 1995, and predictions for the years 2000 and 2010

BACKGROUND: Breast cancer is a major cause of morbidity and mortality in the United States (U.S.) and Missouri. In 1992, 3,915 new breast cancer cases were diagnosed and in 1995, 1,006 deaths from breast cancer were reported in Missouri. Although breast cancer incidence has increased in Missouri in the past 20 years, there are indications that early detection has also increased during the same period. Knowledge about which segments of the population have experienced the greatest increase in mammography screening rates helps in planning and implementation of breast cancer control programs at the state level. OBJECTIVES: Examine the prevalence and trends of lifetime mammography and 2-year mammography compliance in Missouri by age, race, and education from 1987 to 1995 and make predictions for the years 2000 and 2010. METHODS: We used data from the Missouri Behavioral Risk Factor Surveillance System (BRFSS), 1987 to 1995, to estimate the prevalence of ever having had a mammogram and compliance with mammography screening guidelines within two years by race, age, and education status among Missouri women over age 18. Using linear models, we regressed breast cancer screening prevalence estimates on time to obtain trends and predictions. RESULTS: Overall, African-American women were more likely to have had a lifetime mammogram than white women. However, we found a steady increase in the prevalence of ever having had a mammogram for all groups of women defined by age and education status, except among African Americans. Increase in the prevalence of ever having had a mammogram was much higher in women age 50 and older and slightly higher among women with a high school education or less. The average prevalence of 2-year mammography screening compliance was about 60% for all groups, a rate which did not significantly change between 1987 and 1995. By the year 2000, white women will have mammography rates equal to or higher than African-American women, and the majority of all women age 50 and older (98.3% to 100%) will have had a lifetime mammogram. CONCLUSION: Missouri target populations are predicted to attain Year 2000 National Health Objectives concerning lifetime mammography. Current efforts should be continued in order to maintain levels of mammography, particularly among African-American women.     

Trends in cigarette smoking: the Minnesota Heart Survey, 1980 through 1992

OBJECTIVES: The purpose of this study was to describe trends in the prevalence of cigarette smoking between 1980 through 1982 and 1990 through 1992 in Minneapolis and St. Paul, Minn. METHODS: Three population-based surveys were conducted among adults 25 to 74 years of age in 1980 through 1982, 1985 through 1987, and 1990 through 1992. RESULTS: Overall age-adjusted prevalences of cigarette smoking declined significantly between 1980-1982 and 1985-1987 and between 1985-1987 and 1990-1992. Serum thiocyanate, a biochemical marker for tobacco use, also declined significantly over the 3 periods. CONCLUSIONS: Favorable trends in smoking prevalence and cigarette consumption among smokers were observed, but disturbing trends in some smoking behaviors were also noted.     

Factors associated with women's adherence to mammography screening guidelines

OBJECTIVE: To examine individual and environmental factors associated with adherence to mammography screening guidelines. DATA SOURCES: A unique data set that combines a national probability sample (1992 National Health Interview Survey); a national probability sample of mammography facility characteristics (1992 National Survey of Mammography Facilities); county-level data on 1990 HMO market share; and county-level data on the supply of primary care providers (1991 Area Resource File). STUDY DESIGN: The design was cross-sectional. DATA EXTRACTION/ANALYSIS: Data sets were linked to create an individual-level sample of women ages 50-74 (weighted n = 2,026). We used multipart, sequential logistic regression models to examine the predictors of having ever had mammography, having had recent mammography, and adherence to guidelines. We categorized women as adherent if they reported a lifetime number of exams appropriate for their age (based on screening every two years) and they reported having had an exam in the past two years. PRINCIPAL FINDINGS: Only 27 percent of women had the age-appropriate number of screening exams (range 16 percent-37 percent), while 59 percent of women had been screened within two years. Women were significantly more likely to adhere to screening guidelines if they reported participating with their doctor in the decision to be screened; were younger; had smaller families, higher education and income, and a recent Pap smear; reported breast problems; and lived in an area with a higher percentage of mammography facilities with reminder systems, no shortage of primary care providers, higher HMO market share, and higher screening charges. CONCLUSIONS: A small percentage of women adhere to screening guidelines, suggesting that adherence needs to become a focus of clinical, programmatic, and policy efforts.     

Human immunodeficiency virus seroprevalence in community-based primary care practices, 1990-1992. A report from the Ambulatory Sentinel Practice Network

OBJECTIVE: To estimate the seroprevalence of human immunodeficiency virus type 1 (HIV-1) infection in primary care practices. METHODS: Fifty-four practices in the United States participated in an anonymous, unlinked HIV seroprevalence study between January 1990 and December 1992. Residual blood samples drawn for routine clinical tests from patients 15 to 49 years of age were centrally tested for the HIV-1 antibody for 1 month of each quarter. Information about patient demographics, clinician-recognized risk factors, the known HIV status of the patient, and whether the blood was drawn for HIV testing was recorded with each specimen. RESULTS: Of 21,998 specimens collected, 99 (0.45%) were seropositive. Of these 99 seropositive persons, 31.3% (a seroprevalence of 0.15%) were not suspected by their clinicians of being infected with HIV. Seroprevalences in men (0.96%) exceeded those in women (0.22%), and rates in rural practices (0.18%) were lower than in urban practices (0.71%). Among patients with unsuspected HIV infection, however, the gender differences, especially in rural areas, were less pronounced. Risk factors for HIV infection were infrequently noted. There was an increase in the overall seroprevalence during the 1990 to 1992 study period (0.36% to 0.53%); however, this trend was not statistically significant. CONCLUSIONS: Within a 3-year period, clinicians in at least two of five primary care practices can expect to encounter patients infected with HIV, regardless of practice location. Also, nearly one third of the patients with HIV infection will not be suspected of having this condition by their clinician.     

Hiring preferences of diagnostic radiology groups: results of a 1994 survey

RATIONALE AND OBJECTIVES: We surveyed diagnostic radiology group preferences and considerations in hiring radiologists and compared these findings with those of a survey performed in 1990. We sought to identify changes in hiring practices that might have occurred because of socioeconomic changes. We also sought to identify features of job candidates that make them more attractive to hiring groups. METHODS: One hundred surveys were mailed to a stratified random sample of diagnostic radiology groups identified by the American College of Radiology. We solicited information on the importance of various attributes and the level of experience of a candidate, the fellowship training considered most desirable, and the effect of changes in the health care socioeconomic environment. The responses were weighted by group size and geographic location to estimate what results might have been obtained if we had surveyed all groups in the United States. RESULTS: Seventy-five groups returned the survey. The two most important factors in choosing a candidate were motivation and radiologic knowledge. The fellowships that groups that were hiring considered to be the most desirable were body imaging, neuroradiology, and angiography/interventional radiology. Groups overwhelmingly preferred recent training over long experience. CONCLUSION: Fellowship training increases a candidate's marketability, but the two factors that hiring groups consider the most important are motivation and radiologic knowledge.     

Technology transfer in digital mammography. Report of the Joint National Cancer Institute-National Aeronautics and Space Administration workshop of May 19-20, 1993

Digital mammography is one of the most promising novel technologies for further improvement of early detection of breast cancer, offering important potential advantages: 1) improved image quality; 2) digital image processing for improved lesion contrast; 3) computer-aided diagnosis for enhanced radiologic interpretation; and 4) teleradiology for facilitated radiologic consultation. The Diagnostic Imaging Research Branch of the National Cancer Institute (NCI) recently funded an international, multidisciplinary, multi-institutional Digital Mammography Development Group for collaborations between NCI, the academic community, and industry to facilitate the integrated development and implementation of digital mammographic systems. Currently, however, digital mammography faces a number of fundamental technological roadblocks: 1) cost-effective digital detectors and displays for imaging systems; 2) the need for novel algorithms for image processing and computer-aided diagnosis; and 3) high performance, low cost digital networks to provide an "information superhighway" for teleradiology. To solve some of these technological problems, the Diagnostic Imaging Research Branch of NCI joined efforts with the Technology Transfer Division of the National Aeronautics and Space Administration to pursue a federal technology transfer program in digital mammography. The authors discuss the findings and recommendations of the workshop entitled "Technology Transfer in Digital Mammography," which was organized and held jointly by the NCI and the National Aeronautics and Space Administration in May, 1993. Numerous innovative technologies of varying degree of promise for digital mammography were presented at the conference. In this article, specific technologies presented at the workshop by the federal and federally-supported laboratories are described, and critiques of these technologies by the leaders of the medical imaging community are presented.     

Trends in alcohol consumption patterns among whites, blacks and Hispanics: 1984 and 1995

OBJECTIVE: To report national trends in alcohol consumption patterns among whites, blacks and Hispanics between 1984 and 1995, in relation to the recent decline in per capita consumption in the United States. METHOD: Data were obtained from two nationwide probability samples of U.S. households, the first conducted in 1984 and the second in 1995. The 1984 sample consisted of 1,777 whites, 1,947 blacks and 1,453 Hispanics; the 1995 sample consisted of 1,636 whites, 1,582 blacks and 1,585 Hispanics. On both occasions, interviews averaging 1 hour in length were conducted in respondents' homes by trained interviewers. RESULTS: Between 1984 and 1995, the rate of abstention remained stable among whites but increased among blacks and Hispanics. Frequent heavy drinking decreased among white men (from 20% to 12%), but remained stable among black (15% in both surveys) and Hispanic men (17% and 18%). Frequent heavy drinking decreased among white women (from 5% to 2%), but remained stable among black (5% in both surveys) and Hispanic women (2% and 3%). White men and women were two times more likely to be frequent heavy drinkers in 1984 than in 1995. CONCLUSIONS: The reduction in per capita consumption in the U.S. is differentially influencing white, black and Hispanic ethnic groups. The stability of rates of frequent heavy drinking places blacks and Hispanics at a higher risk for problem development than whites. This finding is, therefore, a concern to public health professionals and others interested in the prevention of alcohol-related problems among ethnic groups in the United States.     

Trends in incidence rates of adenocarcinoma of the oesophagus and gastric cardia in New Zealand, 1978-1992

BACKGROUND: Increasing incidence rates for adenocarcinomas of the oesophagus and gastric cardia have been reported from the United States, Denmark, United Kingdom, Switzerland, and Sweden. This paper reports on the incidence of adenocarcinomas of the oesophagus and gastric cardia in New Zealand in the Maori (Polynesian), and non-Maori (predominantly European) populations. METHODS: Incidence data from the National Cancer Registry for 1978 through 1992 were used to compute age-adjusted rates by sex, ethnic group, anatomic subsite, morphology, 14 area health districts, and for three periods: 1978-1982, 1983-1987 and 1988-1992. Statistical tests for significance of trends and differences in frequencies were employed. RESULTS: Incidence rates for adenocarcinoma of the oesophagus are increasing in non-Maori men and women, but at a lesser rate than that reported for the US. The rate of 2.3 per 100000 population (1988-1992) for non-Maori men is similar to the rate for US white men of 2.5 (1988-1990). Rates for adenocarcinoma of the gastric cardia in non-Maori men declined from 2.5 in 1983-1987 to 1.9 in 1988-1992, and were stable at 0.4 in non-Maori women. However, rates for cases with unspecified anatomic subsite fluctuated over the 15-year period and probably caused a deflation in rates in the most recent 5-year period. CONCLUSIONS: The incidence patterns of adenocarcinomas of the oesophagus and gastric cardia in New Zealand should be monitored over the next decade for confirmation of the trends observed here. There is need to review the quality of the data in the New Zealand registry.     

The recent decline in mortality from coronary heart disease, 1980-1990. The effect of secular trends in risk factors and treatment

OBJECTIVE: To examine whether secular trends in risk factor levels and improvements in treatment can account for the observed decline in coronary heart disease mortality in the United States from 1980 to 1990 and to analyze the proportional contribution of these changes. DATA SOURCES: Literature review, US statistics, health surveys, and ongoing clinical trials. STUDY SELECTION: Data representative of the US situation nationwide reported in adequate detail. DATA EXTRACTION: A computer-simulation state-transition model of the US population between the ages of 35 and 84 years was developed to forecast coronary mortality. The input variables were estimated such that the combination of values led to an adequate agreement with reported coronary mortality figures. Subsequently, secular trends were modeled. DATA SYNTHESIS: Actual coronary mortality in 1990 was 34% (127,000 deaths) lower than would be predicted if risk factor levels, case-fatality rates, and event rates in those with and without coronary disease remained the same as in 1980. When secular changes in these factors were included in the model, predicted coronary mortality in 1990 was within 3% (10,000 deaths) of the observed mortality and explained 92% of the decline; only 25% of the decline was explained by primary prevention, while 29% was explained by secondary reduction in risk factors in patients with coronary disease and 43% by other improvements in treatment in patients with coronary disease. CONCLUSIONS: These results suggest that primary and secondary risk factor reductions explain about 50% of the striking decline in coronary mortality in the United States between 1980 and 1990 but that more than 70% of the overall decline in mortality has occurred among patients with coronary disease.     

Socioeconomic disparities in preventive care persist despite universal coverage. Breast and cervical cancer screening in Ontario and the United States

OBJECTIVE: To compare the association of income and education with breast and cervical cancer screening in Ontario, Canada, and the United States. DESIGN: Survey using data from the Ontario Health Survey and the US National Health Interview Survey. PARTICIPANTS: A multistage random sample of women aged 18 years and older living in households in Ontario (N = 23,521) and the United States (N = 23,932) in 1990. MAIN OUTCOME MEASURE: Persons were considered screened if they reported a Papanicolaou test within the previous 2 years, a clinical breast examination within the previous year, or a mammogram within the previous year. RESULTS: Papanicolaou test and clinical breast examination rates were similar between countries, but mammography rates were two to three times higher in the United States across all age groups. Compared with women with less than a high school degree, college graduates were more likely to receive screening (odds ratio [OR], 1.5; 95% confidence interval [CI], 1.2 to 1.7) and there was no difference between countries. Across all procedures, women with higher incomes were more likely to receive screening. For Papanicolaou test and clinical breast examination, there was no difference between countries. Compared with the lowest income, the OR was 1.7 (95% CI, 1.3 to 2.1) in Ontario and 1.9 (95% CI, 1.6 to 2.2) in the United States for Papanicolaou test and 2.1 (95% CI, 1.6 to 2.8) in Ontario and 2.1 (95% CI, 1.8 to 2.6) in the United States for the clinical breast examination for women with income greater than $45,600 (US dollars). For mammography screening, the association of income with use was greater in the United States: the OR was 1.8 (95% CI, 1.3 to 2.6) in Ontario and 2.7 (95% CI, 2.3 to 3.2) in the United States for women with income greater than $45,600 (US dollars). CONCLUSIONS: Despite the long-time presence of universal insurance coverage in Ontario the disparities in the use of cancer screening procedures by the poor were similar to the United States. Universal coverage is not sufficient to overcome the large disparities in screenings across socioeconomic status demonstrated in both countries.     

Hypopharyngeal cancer patient care evaluation

A survey was conducted to identify demographics and standards of care for treatment of hypopharyngeal squamous cell carcinoma in the United States. Data were accrued from voluntary submission of cancer registry and medical chart information from 769 hospitals representing 2939 cases diagnosed from 1980 to 1985 and 1990 to 1992. Clinical findings, diagnostic procedures employed, treatment practices, and outcome are presented. Overall, 5-year disease-specific survival was 33.4%, which segregated to 63.1% (stage I), 57.5% (stage II), 41.8% (stage III), and 22% (stage IV). Survival was best for patients treated with surgery only (50.4%), similar with combined surgery and irradiation (48%), and worse with irradiation only (25.8%). This analysis provides a standard to which current treatment practice and future clinical trials may be compared.     

Utilization of mammography--who will come and why?

Mammography has been used to detect early stages of breast cancer in women over the age of 40 years. Nationally organized mammography screening has not been initiated in Norway, but women have the possibility of undergoing mammography at many hospitals and private institutions. This article describes the use of mammography at R?ntgensenteret, a private institute, from January to July 1992. Women in the age group 40 to 50 years were the most frequent users. The most important reason for the women undergoing mammography was a recommendation from their physician. Living near to the examination unit was also an important factor.     

Are increases in mammographic screening still a valid explanation for trends in breast cancer incidence in the United States?

A number of studies have attributed much of the sharp increase in breast cancer incidence in the United States during the 1980s to the increased detection through mammography. The most recent breast cancer data from the US National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) program show that the incidence trend has slowed, while results from the National Health Interview Survey (NHIS) of 1987 and 1990 indicate that the percentage of women receiving mammograms continues to increase. This phenomenon suggested the need to reassess the relationship between increasingly early detection of breast cancer and overall incidence trends. A polynomial age-cohort model was used to establish the secular trend in incidence rates excluding the impact of recent increases in detection due to the rising use of mammography. Based on the model, the incidence trend in the youngest age group (40 to 49 years) would peak and then begin to decline in the early 1980s. This pattern would manifest itself later in successively older age groups as these younger cohorts age. Breast cancer trends are seen to be generally consistent with the impact of the increased use of mammography when its effect is superimposed upon the background of declining or slowing secular trends. These results support previous reports linking incidence rates with the increase in screening-mammography.     

Sensitivity of heat-denatured p24 antigen in the diagnosis of pediatric HIV infection

PURPOSE: Hamartomas are rare benign breast tumors which surfaced with the entrance of mammography in the diagnostic algorithm of breast lesions. In the present study the mammography findings of breast hamartomas were analyzed in relation to the histologic confirmation of the diagnosis. PATIENTS AND METHODS: During the seven-year period between January 1990 and January 1997 we studied 45 patients, aged from 32 to 66 years, with a clinical and/or mammographic finding of a breast lump, highly suspicious of hamartoma. In 29 (64.4%) the tumor was found in the left breast and in the remaining 16 (35.6%) in the right breast. RESULTS: The preoperative mammography of 41 patients revealed a nodular opaque mass of a nonhomogeneous composition and only four had the typical hamartoma picture. The tumor diameter ranged between 0.7-6.5 cm. All tumors were excised and the histological reports confirmed the diagnosis of hamartoma. CONCLUSION: Mammography helps in the diagnosis of hamartomas. Nevertheless, the final diagnosis must be based on the histological analysis of the tumor.     

Prevalence of multiple cardiovascular disease risk factors among women in the United States, 1992 and 1995: the Behavioral Risk Factor Surveillance System

We sought to examine the prevalence of self-reported multiple cardiovascular disease (CVD) risk factors (hypertension, high blood cholesterol, diabetes, overweight, and current smoking) among women in 1992 and 1995 in the United States using data from the Behavioral Risk Factor Surveillance System. In 1992, 37.5%, 34.4%, and 28.1% of women had zero, one, and two or more of the five risk factors, respectively. In 1995, the respective estimates were 35.5%, 34.3%, and 30%. In both years, the prevalence of two or more risk factors increased with age, decreased with educational level, was higher among black women (lowest among Hispanic women and women of other ethnic groups), and higher among women reporting cost as a barrier to healthcare. The percentage of women with two or more risk factors was higher in 1995 than in 1992 for 35 of 48 states, being statistically significant for 7 states. The percentage of women with at least two risk factors was not significantly lower in 1995 than in 1992 for any state. A higher percentage of women reported having multiple CVD risk factors in 1995 compared with 1992. A multifactorial approach to primary prevention and risk factor reduction should be encouraged to help reduce the prevalence and burden of CVD among women.     

A review of postmenopausal hormone therapy recommendations: potential for selection bias

OBJECTIVE: To survey the advice given to physicians in regard to postmenopausal hormone therapy, and thus to find out indirectly whether selective prescribing may have contributed to the findings of health effects. DATA SOURCES: Surveys on the health benefits of postmenopausal therapy were identified from previous reviews. Experts in various fields were consulted regarding relevant textbooks, and recommended booklists were used. METHODS OF STUDY SELECTION: Most of the surveys were conducted in the United States and concerned only estrogens; therefore, we focused on these data. Advice given to United States physicians on estrogen prescribing was abstracted from the Physicians' Desk Reference, five textbooks, and other sources written between 1945-1990. DATA EXTRACTION AND SYNTHESIS: In advice given to physicians, there was concern that estrogens could cause cancer. In addition, it was not recommended to give estrogens to women in poor general health. Advice concerning cardiovascular diseases and estrogens was contradictory and variable. CONCLUSIONS: If the advice observed reflects the prescribing practices of physicians, then surveys on the health impact of estrogen therapy may have underestimated the risk of breast cancer and overestimated the prevention of fractures. The impact on studies of cardiovascular diseases is less clear, and other mechanisms of selection bias may have been more important.     

Abortion surveillance--United States, 1995

CONDITION: Since 1990 (i.e., the year in which the number of abortions was highest), the annual number of abortions in the United States has decreased by 15%. REPORTING PERIOD COVERED: This report summarizes and reviews information reported to CDC regarding legal induced abortions obtained in the United States during 1995. DESCRIPTION OF SYSTEM: For each year since 1969, CDC has compiled abortion data received from 52 reporting areas: 50 states, the District of Columbia, and New York City. RESULTS: In 1995, a total of 1,210,883 legal abortions were reported to CDC, representing a 4.5% decrease from the number reported for 1994. The abortion ratio was 311 legal induced abortions per 1,000 live births, and the abortion rate was 20 per 1,000 women aged 15-44 years, the lowest ratio and rate recorded since 1975. Women who were undergoing an abortion were more likely to be young, white, and unmarried; most were obtaining an abortion for the first time. Approximately half of all abortions (54%) were performed at < or =8 weeks of gestation, and approximately 88% were performed before 13 weeks. Approximately 16% of abortions were performed at the earliest weeks of gestation (< or =6 weeks), approximately 17% at 7 weeks of gestation, and approximately 21% at 8 weeks of gestation. Few abortions were provided after 15 weeks of gestation -- approximately 4% of abortions were obtained at 16-20 weeks, and 1.4% were obtained at > or =21 weeks. Younger women (i.e., women aged < or =24 years) were more likely to obtain abortions later in pregnancy than were older women. INTERPRETATION: Since 1990, the number of abortions has declined each year. Since 1987, the abortion-to-live-birth ratio has declined; in 1995, it was the lowest recorded since 1975. This decrease in the abortion ratio reflects a trend that a lower proportion of pregnant women obtain induced abortion. ACTIONS TAKEN: The number and characteristics of women who obtain abortions in the United States should continue to be monitored so that trends in induced abortion can be assessed, efforts to prevent unintended pregnancy can be evaluated, and the preventable causes of morbidity and mortality associated with abortions can be identified and reduced.