Trends in screening mammography rates: Variations by patient age and other demographics and across data sources

Title

Trends in screening mammography rates: variations by patient age and other demographics and across data sources

Research Objective

Despite widespread use, policy on optimal mammography screening is controversial and rates have been decreasing in recent years. Since the December 2009 release of the USPSTF's revised recommendations, there has been heated debate regarding the reported benefits, risks and cost effectiveness of screening mammography in women under age 50. There is concern that screening rates will be suboptimal in populations with lack of access, and inadequate or no insurance coverage.

In this context, our objective is to carefully examine trends in mammography screening, using several national data sources, and to identify at-risk subpopulations; specifically: 1) measure trends in mammography screening since 1996 and 2007 for sub-groups by patient age and other demographics, with controls for socio-economic characteristics, and 2) attempt to reconcile variations in findings across data sources.

Data and Methods We will use data from the Medical Expenditure Panel Surveys (MEPS) for 1996-2007 (including a short-recall period report), National Health Interview Survey (NHIS) for 1998 and 2006, and the Behavioral Risk Factor Surveillance Survey (BRFSS) for 1996-2006

Analyses will be performed separately for women over and under 50 (if sample sizes permit) to determine whether recent guidelines changes may have different impacts on women in different age categories, controlling for year, patient demographics, and socioeconomic characteristics.

Principal findings Preliminary analysis on data through 2004 show that mammography does vary by patient characteristics, for example, older women are more likely to have mammography (odds ratio (OR) of 1.23-1.32 across datasets, p<0.01.), less educated women are less likely to have mammography. (OR 0.66-0.68, p<0.01)

Using updated data, we will report on utilization rates for each year, by age group and data source, including event-reported rates in MEPS. Also, we will report regression findings from each data source on effect of patient characteristics separately for women older and younger than 50.

Conclusions

Changes in guidelines may have different effects on different population subgroups.

Implications for policy, delivery, and practice

In light of the change in USPSTF recommendations, and concerns that screening will decrease in both women age 40-49 and over 50, it will be extremely important to continue to monitor and address disparities in access to and utilization of screening services.