Mammography Use--Wisconsin, 1980-1993

In the United States, efforts to reduce mortality from breastcancer focus primarily on secondary prevention (ie, early detectionand treatment). Since 1980, private, public, and voluntary organizationsin Wisconsin have promoted screening mammography as a means forreducing the death rate from breast cancer. To assess the effectivenessof these efforts, the Division of Health, Wisconsin Departmentof Health and Social Services (DOH), analyzed data from annualstatewide surveys of mammography providers during 1989 through1993 and data about self-reported mammography use from the BehavioralRisk Factor Surveillance System (BRFSS) during 1987 through 1993.This report summarizes these analyses and trends in the numberof mammograms performed annually in Wisconsin during 1980 through1993.

An annual survey of all registered mammography providers in Wisconsinhas been conducted since 1989. During 1980 through 1993, the numberof mammography providers ranged from 76 to 236. Survey questionnairesare mailed annually to all mammography providers in conjunctionwith a mailing of DOH radiation-protection registration materials.The questionnaire asks each facility to estimate the total numberof mammograms performed during that year and to provide informationabout mammography referral and follow-up procedures, fees, andavailability of low-cost screening services.

The 1989 survey requested estimates of the total number of mammogramsperformed during 1980 through 1989. The response rates for thefive surveys con ducted during 1989 through 1993 were 75%, 89%,90%, 91%, and 91%, respectively. Data were adjusted for nonresponseto provide statewide estimates of the annual number of mammogramsperformed.

Trends in self-reported mammography use were determined by analyzingdata from the BRFSS, which has included questions about mammographyuse since 1987. The BRFSS is a random-digit-dialed telephone surveyof household residents age 18 years that provides population-basedsurveillance data about selected health behaviors. The total numberof mammograms performed in the state each year during 1987 through1993 was estimated from the BRFSS by multiplying the adult femalepopulation in Wisconsin by the proportion of women who reportedhaving had a mammogram during the preceding 12 months. The numberof adult women interviewed in Wisconsin for the annual BRFSS rangedfrom 673 (in 1990) to 857 (in 1993).

Analyses of the mammography provider surveys (1989 through 1993)and the BRFSS (1987 through 1993) indicated steady increases inthe number of mammograms performed each year during 1980 through1993. Based on the provider survey, the total number of mammogramsperformed each year increased nearly 17-fold, from 31,000 in 1980to 517,000 in 1993. Compared with the provider survey, estimatesbased on the BRFSS were consistently higher, varying from 81%higher in 1988 to 17% higher in 1993. In addition, the numberof mammography providers increased 310%, from 76 in 1980 to 236in 1993. Standard errors for annual BRFSS estimates ranged from±1.5% to ±1.7%.

Editorial Note from the CDC: The findings in this reportdocument a nearly 17-fold increase in the annual number of mammogramsperformed in Wisconsin during 1980 through 1993. Although theestimated number of mammograms performed each year differed substantiallyby data source, the trends were similar for both sources. Twoimportant factors probably contributed to the increase in mammographyuse during this period: (1) the substantial increase in the acceptanceand use of screening mammography among physicians during the 1980s,and (2) initiation of extensive efforts to educate the publicand health-care professionals about national screening mammographyguidelines, which were implemented during the late 1980s. Otherpossible contributing factors include the initiation of low-costmammography screening programs and wider availability of high-quality,low-cost mammography equipment beginning during the early 1980s.

The findings in this report are subject to limitations associatedwith the two data sources. Although the provider survey is anexample of a low-cost, efficient method to characterize trendsin mammography use, it does not collect information about sociodemographicvariables and may overestimate mammography use because data werebased on the number of mammograms performed and some women receivemore than one mammogram each year. In addition, the estimatesbased on the self-reported BRFSS data probably were higher thanthose from the provider survey for at least three reasons. First,some women who reported having had a mammogram during the previous12 months probably received the mammogram more than 12 monthspreviously. Second, women who participated in the BRFSS may notbe representative of the total population of women in Wisconsinbecause they had telephones, were better educated, or were morelikely to have had insurance coverage-factors related to an increasedlikelihood of having received a mammogram. Third, the BRFSS isa household survey that does not include institutionalized women(eg, those in long-term care facilities), who are less likelythan noninstitutionalized women to receive mammograms.

The findings in this report are being used in Wisconsin to furtheridentify groups of women who underuse mammography screening, developintervention strategies to increase mammography use, and assessprogress toward the year 2,000 national health objectives forbreast cancer and mammography (objectives 16.11 and 16.16). Thisapproach can be adapted for use by health departments in otherstates to assess the effectiveness of efforts to promote mammographyscreening.

Adapted from Morbisity and Mortality Weekly Report 44(41):754-756,1995.