Mammogram Order Completion Rates Among Women with Diabetes

Discussion

Of the 16,668 patients that were included in this study, 54.5% successfully completed mammography orders prescribed by their primary care physician. Overall, our study found that women with diabetes were less likely (51.6%) to complete their mammogram compared with women without diabetes (55.3%), consistent with prior studies.^8,9 These findings suggest that interventions may be needed to improve mammography screening among women with diabetes, who are at elevated risk for breast cancer.^6,7

The results also indicate that patient race/ethnicity were associated with mammogram completion rates. Healthy People 2020 data suggest that Black women are less likely to undergo cancer mammography screening compared with White women; however, our study did not observe this disparity in screening.^12,13 It is important to note that Black women are more likely to die from breast cancer than are White women in the United States.¹⁴ Black women in this study may have generally been insured or had better access to testing facilities, which may have increased the overall compliance in screening. It is also possible that Black women are more likely to complete their mammogram orders but less likely to receive orders in the first place. The 40 to 49 age group had the highest completion rate at 57.6%. These results are different from what previous studies have reported about women 55 and older being more adherent to mammogram screening.¹⁵

We identified a discrepancy between mammogram orders issued and completion of these orders. Patients with diabetes were less likely to complete mammograms than patients without, with the higher prevalence of breast cancer and mortality among the aforementioned group. This study was conducted among a group of patients who generally have access to insurance, which could be related to the discrepancy. Elsewhere, socioeconomic factors could have had an impact on mammogram completion rates, such as with affordability or transportation to imaging center(s), and these may have been barriers as seen in previous studies.⁹ Patients inadequately educated on the importance of the screening could have also resulted in lower completion percentages. Current literature has noted that certain interventions may enhance screening order completion: reminder-based interventions, culturally tailored interventions, and multifaceted interventions.¹⁶ These interventions do increase compliance to screening, especially for White patients of higher socioeconomic status.¹⁶ For patients within the lower income and underinsured group, provider-targeted interventions were found to increase compliance over patient-targeted interventions; physicians that were given reminders/prompts in the electronic health record would always talk to their patients about getting screening.¹⁶

The study was limited to mammograms that were ordered and completed within the USF Health/Tampa General Hospital System. The study does not consider orders that were completed at other institutions or ordered at other clinics; the completion rates may have been higher than reported if patients completed them externally, as is generally the case. We also note availability of free mammogram services apart from the USF system possibly impacting internal completion rates. However, we were unable to account for sociodemographic differences associated with screening disparities. Another limitation to consider is that patients with undiagnosed diabetes may have been included in the patient pool, underestimating the rate of diabetes.¹⁷ Future studies should explore whether sociodemographic differences (AlC level) help to explain observed screening disparities among women with and without diabetes.