Effectiveness of a Mobile Mammography Program.

Objective: Mobile mammography units have increasingly been used to address patient health care disparities; however, there are limited data comparing mobile units to stationary sites. This study aims to evaluate the characteristics of women who underwent mammography screening in a mobile unit versus those who underwent mammography screening at a cancer center.

Materials And Methods: In this retrospective study, we analyzed all screening mammography examinations performed in a mobile unit in 2014 (n = 1433 examinations). For comparison, we randomized and reviewed an equivalent number of screening mammography examinations performed at our cancer center in 2014 (n = 1434 examinations). BI-RADS assessment, adherence to follow-up, biopsies performed, cancer detection rate, and sociodemographic variables were recorded. An independent-samples t test was conducted to identify potential differences in age between cancer center patients and mobile unit patients. Chi-square analyses were used to test for associations between location and factors such as health insurance, race, marital status, geographic area, adherence to screening guidelines, recall rate, adherence to follow-up, and cancer detection rates.

Results: Patients visiting our cancer center (mean = 57.74 years; SD = 10.55) were significantly older than those visiting the mobile unit (mean = 52.58 years; SD = 8.19; p < 0.001). There was a significant association between location and health insurance status (χ = 610.92; p < 0.001) with more uninsured patients undergoing screening in the mobile van (cancer center = 3.70%, mobile unit = 38.73%). There was a significant association between screening location and patient race (χ = 118.75, p < 0.001), with more white patients being screened at the cancer center (cancer center = 47.28%, mobile unit = 33.30%), more black patients being screened in the mobile van (cancer center = 49.30%, mobile unit = 54.15%), and more Hispanic patients being screened in the mobile van (cancer center = 1.05%, mobile unit = 6.77%). There was a significant association between location and patient marital status (χ = 135.61, p < 0.001), with more married patients screened at the cancer center (cancer center = 49.16%, mobile unit = 38.31%), more single patients screened in the mobile van (cancer center = 25.17%, mobile unit = 34.47%), and more widowed patients being screened at the cancer center (cancer center = 8.09%, mobile unit = 4.47%). There was a significant association between location and geographic area (χ = 33.33, p < 0.001), with both locations reaching more urban than rural patients (cancer center = 79.99%, mobile unit = 70.62%). There was a significant association between location and adherence to screening guidelines (χ = 179.60, p < 0.001), with patients screened at the cancer center being more compliant (cancer center = 56.90%, mobile unit = 34.47%). Finally, there was a significant association between location and recall rate (χ = 4.06, p < 0.001). The cancer center had a lower recall rate (13.32%) than the mobile van (15.98%). Of those patients with BI-RADS 0, there was a significant association between location and adherence to follow-up (χ = 22.75, p < 0.001) with patients using the mobile unit less likely to return for additional imaging (cancer center = 2.65%, mobile unit = 17.03%).

Conclusion: Significant differences were found among patients visiting the cancer center versus the mobile mammography van. The cancer center's population is older and more adherent to guidelines, whereas the mobile mammography population exhibited greater racial and marital diversity, higher recall rate, and lack of adherence to follow-up recommendations. By identifying these characteristics, we can develop programs and materials that meet these populations' needs and behaviors, ultimately increasing mammography screening and follow-up rates among underserved populations.