The 2009 update to the United States Preventive Services Task Force (USPSTF) screening recommendations did not decrease the mammography screening rates in women. This may be bad news for fans of the Task Force or may just reflect how well older messages are engrained and how difficult it is to change screening behaviors. (LOE = 4)
Study Design: Cross-sectional
In 2009 the USPSTF updated its recommendations for breast cancer screening, which were met with great controversy, a media firestorm, and other organizations issuing (or re-iterating) different recommendations. Recall that the Task Force’s mandate from Congress is to tell us what we know about screening and prevention, tell us what we don’t know, and do not speculate. To do this, they work with evidence-based practice centers who do high-quality systematic reviews to guide their recommendations. In essence, the Task Force is akin to meteorologists who report — not make — the weather. However, other organizations are allowed to speculate. So, the authors of this study decided to look at National Health Interview Survey (NHIS) data from 2005, 2008, and 2011. They identified nearly 28,000 women 40 years or older. The authors used the women’s self-report of having had a mammogram to estimate screening rates. They also collected data on other factors that might influence screening rates (eg, immigration status, socioeconomic status, race, and so forth). The crude screening rate in 2011 was actually higher (54%; P = .03) than in 2008 (52%) or 2005 (50%)! After taking into account race, income, education level, insurance, and immigration status, the difference was no longer significant. In particular, the screening rate among women between 40 and 49 years of age, the group stirring the greatest controversy, was also no different. Please note that the way the NHIS assessed mammography was slightly different in 2011 than in other years. The 2011 survey directly asked if a woman had a mammogram in the previous 12 months; the previous surveys used a more circuitous approach to the answer, requiring 2 or 3 questions.
Henry C. Barry, MD, MS
Michigan State University
East Lansing, MI