Trends in Cancer Screening-United States, 1987 and 1992

Screening methods and programs are critical strategies for theearly detection and timely treatment of some cancers. Establishedmethods for early detection of cancer include mammography, clinicalbreast examination (CBE), the Papanicolaou (Pap) test, proctosigmoidoscopy,fecal occult blood test (FOBT), and digital rectal examination(DRE). To examine changes in the reported use of selected cancerscreening tests, the National Cancer Institute analyzed data fromCDC's National Health Interview Survey (NHIS) for 1987 and 1992to calculate rates of use and compared these rates with the nationalhealth objectives for the year 2000. This analysis suggests that,although the use of these tests increased, substantial progressis needed to meet the objectives.

The NHIS is a continuing nationwide household survey that collectsinformation from a representative sample of the US civilian, noninstitutionalized population age equal to or more than 18 years.The overall response rate for the 1987 and 1992 surveys was 95.3%(N = 122,859) and 95.7% (N = 128,412), respectively. In 1987 and1992, questions were included to determine respondents' knowledgeand practices regarding cancer screening. Respondents were askedwhether they had ever had a Pap test, CBE, mammography, DRE, FOBT,or proctosigmoidoscopy. Respondents who answered "yes"to any of the questions were asked when their most recent testhad been performed. Screening tests were defined as tests performedfor any reason other than as the result of a health problem.

For CBE, mammography, DRE, and FOBT, screening was consideredrecent if it had been performed during the year preceding theinterview; for the Pap test and proctosigmoidoscopy, within thepreceding 3 years. Data about CBE and mammography are presentedfor women age equal to or more than 40 years; for DRE, FOBT, andproctosigmoidoscopy, persons age equal to or more than 40 years,and for the Pap test, women age equal to or more than 18 yearswith an intact uterus.

From 1987 to 1992, the overall percentage of women age equal toor more than 18 years who re-ported having had a recent Pap testremained stable (Table 1). The percentage increased slightly forHispanic women, and remained low for women aged equal to or morethan 70 years.

The increase in the percentage of women ever tested was greaterfor women aged equal to or more than 50 years (85% to 89%) thanwomen aged equal to or more than 50 years (90% to 92%), and forblack (88% to 92%) and Hispanic women (75% to 83%) than whitewomen (91% to 92%).

During this period, the percentage of respondents who reportedrecent mammography increased at least twofold for women in everyage and racial/ethnic group. The greatest increases were for blackand Hispanic women; as a consequence, in 1992, screening rateswere similar for white, black, and Hispanic women. However, womenage equal to or more than 70 years in 1992 remained less likelyto have had a recent screening and to have ever been tested thanwomen age less than 70 years. From 1987 to 1992, the percentageof respondents who reported having had a recent CBE also increased;in 1992, at least 75% of women in each age group reported everhaving the test.

From 1987 to 1992, the percentage of respondents who reportedever having had a DRE increased from 49% to 54% for men and from51% to 54% for women. Although increases were greater for menthan women, rates for recent DRE were lower for men than women(22% vs 21% in 1992). Rates of recent FOBT remained stable; however,the rate for black men increased more than twofold, from 7% to15%. The overall percentages of respondents who reported everhaving had proctosigmoidoscopy increased for men (24% to 30%)and for women (21% to 26%), and the percentage screened recentlywas higher for men than women in both 1987 and 1992 (7% and 11%,respectively, vs 5% and 7%, respectively).

Editorial Note From the CDC

The analysis described in this report estimates use of cancerscreening tests based on a representative sample of the US population,and four of these tests have been targeted as national healthobjectives for the year 2000 (objectives 16.11-16.14). Althoughthe findings indicate an increase in the recent use of all cancerscreening tests (except the Pap test) from 1987 to 1992, percentagesare substantially lower than the national objectives. For example,one objective is to increase the rate of mammography among womenage equal to or more than 50 years to 60% every 2 years and amongwomen age equal to or more than 40 years to 80% ever (objective16.11). Based on this survey, the rate of recent mammography amongwomen age equal to or more than 50 years was 44% in 1992 and everhaving had mammography was 70% for women age equal to or morethan 40 years.

The differences in the screening rates and the national healthobjectives may, in part, reflect for respondents a lack of (1)health insurance coverage, (2) a primary-care physician, or (3)clear communication between physicians and patients about theimportance of routine screening. For example, the lower rate ofmammography use by women age equal to or more than 50 years (whoare at greatest risk for breast cancer) may reflect the findingthat these women are less likely to visit gynecologists, and ofall physician specialists, gynecologists are most likely to recommendmammograms. In addition, for women with low incomes, the mammographyobjectives are unlikely to be met because facilities that performmammography may not accept women without a referral from a primary-carephysician, and a disproportionate number of women with low incomesdo not have a regular health-care provider. To promote mammographyscreening among older women, since 1990, Medicare has reimbursedthe cost of biennial mammograms. Although the reimbursement feeis substantially less than the median price of mammograms in theUS, the fee is feasible if mammograms are delivered using moreefficient methods and established mass-production techniques.

In the United States, managed care and the increased use of healthmaintenance organizations (HMOS) are likely to increase the useof all preventive-care services (including screening examinations),particularly if primary-care physicians are encouraged to screenpatients routinely and recommend screening tests they currentlydo not perform. In addition, however, the importance of some screeningexaminations, such as the Pap test, may need to be emphasizedregularly in public health messages.

Reported by: N. Breen, PhD, L. Kessler , PhD, Applied ResearchBr., NCI, NIH, Div.of Cancer Prevention and Control, NationalCenter for Cronic Disease Prevention and Health Promotion; NationalCenter for Health Statistics, CDC.

Adopted from Morbidity and Mortality Weekly Report, vol 45,no. 3, 1995