Increasing Mammography Among Long-term Noncompliant Medicare Beneficiaries

Breast cancer is the second most frequently occurring cancer in women and the second leading cause of their death. The most effective approach to reducing breast cancer mortality is early detection and treatment through routine mammography and clinical examination. Despite the fact that breast cancer risk increases with age, women are actually less likely to undergo a mammography screening as they get older.

Health care organizations have successfully used centralized databases to identify women who have not received mammograms at recommended intervals and have increased rates among those women using personalized informational mailings. The purpose of this project was to develop a method to identify the mammography status of Medicare beneficiaries, to test the effects of a personalized targeted mailing to increase mammography among older women who have not had recent mammograms, and to assess the cost effectiveness of the intervention.

The intervention was based on the Health Belief Model and was targeted to older women using content that addressed the perceived susceptibility to breast cancer, the benefits of mammography, and barriers to obtaining a mammogram. Intervention components (mailed as a package) included: 1) a personally addressed letter on official stationery (e.g., Michigan Medicare) signed by the agency director. The letter emphasized that the recipient had not used her Medicare mammography screening benefit, explained the importance and benefits of being screened, and urged her to obtain a mammogram; 2) a brochure which emphasized the importance of mammography screenings, addressed common concerns of older women, and provided information on Medicare coverage. The brochure also included pictures and testimonials emphasizing the multi-ethnic and multi-generational benefits of screening (i.e., being in the lives of one's grandchildren); and 3) an information sheet explaining how to schedule a mammogram through the woman's physician or through a toll-free number to a referral service.

Each set of materials was mailed out in a standard Michigan Medicare envelope that was personally addressed to the recipient (i.e., not using a preprinted label) and affixed with a first-class postage stamp.

To implement this intervention, an organization must acquire or develop the materials, identify the targeted recipients, and conduct the mailings. Time required will vary depending upon availability of materials, efficiency of targeting procedure, and the number of recipients targeted.

The intervention was targeted to women aged 70 years and older who had no mammograms (no Medicare claim) for at least five years.

The intervention is suitable for organizations with sufficient access to the Medicare database and claims files, allowing for a targeted mailing to those individuals who have not had a recent mammogram.

Required resources include: 1) Personally addressed Medicare Part B letter, 2) Brochure, "Mammograms For Yourself, For Your Family, For Life", and 3) Information sheet, "How to Get Your Mammogram."

The cost associated with the study was $1.02 per recipient, primarily directed toward mailings and the development and printing of materials. For an organization replicating this intervention, costs would vary based on size of the organization and the number of patients it serves.  Additionally, existing use of Medicare data files and availability of appropriate computer equipment to facilitate the delivery of the intervention would affect the cost.

Participants of the study were women aged 70 years and older who had no Medicare claims for mammograms in the past five years.  They were classified into four groups based on race (black or not) from claims data and on residence location (urban or rural) using their zip codes and HCFA's designation of counties as urban or rural.  The study utilized a randomized design of paired intervention and control groups.  The intervention components consisted of a letter, brochure, and information sheet which were mailed to the participants. 

Beneficiary measures (e.g. age, race, zip code) used in the study were obtained from HCFA beneficiary files and HCFA's Michigan claim files (screening or diagnostic mammography, clinical diagnoses) during 1993-1998. 

Key findings include:

• Based on the number of claims submitted, previously non-adherent (no mammogram for at least 5 years) women who received the intervention (i.e., a personalized letter, fact sheet, and brochure) were more likely to obtain a screening or diagnostic mammography (8.1 percent) when compared to women in the control group (5.2 percent) with an Odds Ratio of 1.6 (p < 0.005). The majority of the increase in the intervention group was seen among women ages 70-79 (10.6 percent) when compared to the control group.

• The rate of diagnostic mammograms (i.e., more thorough screenings) increased more than screening mammograms. Diagnostic mammograms increased 2.2 percent where the intervention group (3.9 percent) was higher than the control group (1.7 percent).  Screening mammograms increased 0.8 percent in which the intervention group (4.5 percent) was higher than the control group (3.7 percent).