Testing a Breast Cancer Screening Decision aid Designed for Health Literacy Accessibility

INTRODUCTION

Shared decision making (SDM) is important for breast cancer screening decision-making related to age of initiation, screening intervals, and modality,¹ especially in the setting of 2024 changes to the United States Preventive Services Task Force (USPSTF) guidelines.² To guide screening decisions, individuals require individual risk assessment, and a discussion of screening’s benefits and harms. Yet studies consistently demonstrate that few clinicians assess breast cancer risk or discuss screening harms with their patients, resulting in confusion, and low engagement in decision-making for many patients. Decision aids (DAs) support patients in making deliberate, informed, preference-aligned health decisions, but often have relatively high health literacy demands. We developed MyMammogram,³ an online mammographic breast cancer screening decision aid, to minimize cognitive demands and address preferences for risk communication.

METHODS

This study evaluated the acceptability and impact of MyMammogram on knowledge, decision-making, and screening intentions using a pre-post design. United States-dwelling females ages 40–54 were recruited using Prolific, an online survey panel. We identified and invited those at average or lower breast cancer risk with no mammogram in the prior 9 months, and no personal history of breast cancer, ductal, or lobular carcinoma in situ. Study participants completed a pre-intervention survey, accessed MyMammogram, and after use, completed a post-intervention survey.

The post-intervention survey measured the primary outcome of acceptability (ease of use, helpfulness, time, satisfaction), each on a 5-point Likert scale. Pre- and post-surveys assessed knowledge about mammograms (5 items), decisional conflict using the Decisional Conflict Scale, Low Literacy (9 items),⁴ and breast cancer screening intentions (1 item: start mammography now, do not start now, or unsure). The pre-intervention survey measured health literacy using the Health Literacy Skills Instrument-10 (< 7 indicates limited health literacy).⁵

Acceptability was summarized using descriptive statistics (a priori threshold of 80% for acceptability). Paired t-tests, McNemar’s tests, and Wilcoxon signed rank tests compared changes in knowledge, decisional conflict, and screening intentions, respectively. We explored differences in knowledge, decisional conflict, and acceptability by health literacy level using mixed effects logistic regression.

RESULTS

Two hundred fifty people completed the screening questionnaire; 166 were eligible and invited to complete the survey; ninety-eight participants were included after passing two attention check questions. Table 1 describes the sample characteristics. The median study completion time was 24 min.

Table 1.

Demographics of Study Sample

Characteristic	Frequency (n = 98)
Age
40–44	35 (36%)
45–49	27 (28%)
50–54	36 (37%)
Ethnicitya
White	76 (80%)
Black	9 (10%)
Asian	1 (1%)
Mixed race	6 (6%)
Other	3 (3%)
Native Languageb
English	94 (97%)
Other	3 (3%)
Health Literacyb
Inadequate	11 (11%)
Adequate	86 (89%)
NCI-BCRAT 5-year riskc
Mean(SD)	1.08% (0.83)
Perceived Breast Cancer Risk
Low	68 (69%)
Average	25 (26%)
High	5 (5%)
Prior Mammogram
Yes	47 (48%)
No	51 (52%)

^aMissing for n = 3; ^bMissing for n = 1; ^c5-year risk was self-reported after completing MyMammogram

MyMammogram was highly acceptable: Ease of use, helpfulness, and satisfaction were all > 90%. Most (82%) indicated the time required was very acceptable. Use of MyMammogram was associated with higher mean knowledge scores post- vs. pre-intervention (3.86 [95% CI, 3.69–4.03] vs. 3.06 [95% CI, 2.87–3.25], p < 0.001; mean change = 0.8 [95%CI: 0.6–0.99]). Those with lower health literacy had lower baseline knowledge relative to those with adequate health literacy (2.18 [95% CI, 1.29–3.07] vs. 3.16 [95% CI, 2.99–3.34], p < 0.001). Differences by health literacy persisted post-intervention (3.27 [95% CI, 2.53–4.01] vs. 3.94 [95% CI, 3.77–4.11], p = 0.01). MyMammogram improved dimensions of decisional conflict related to feeling informed about options, risks, and benefits (Fig. 1). Intentions to obtain mammography increased: 54% intended to obtain screening pre-intervention and 62% post-intervention (p = 0.02).

Figure 1.

Decisional conflict before and after decision aid use a informed subscale questions, b choice clarity questions, c support questions.

DISCUSSION

In this pre-post study, MyMammogram was acceptable, improved knowledge, and reduced decisional conflict among participants. MyMammogram helped to address mammography knowledge gaps: Participants reported greater awareness of available screening options and associated risks and benefits, which are infrequently communicated by healthcare providers. Knowledge gains were similar to that of other breast cancer screening decision aids⁶ and achieved across all levels of health literacy. Our results support that designing for limited health literacy populations can benefit individuals of all health literacy levels.

USPSTF guidelines no longer recommend shared decision-making, yet this does not abrogate the imperative of supporting informed decisions¹ as the balance of screening benefits and harms remains unchanged for females in their 40 s.⁷ Our data demonstrate a need to provide women with information about the implications of screening choices, as they experience uncertainty regarding screening recommendations from guidelines and healthcare providers. As this test of MyMammogram occurred outside of the clinical setting and we did not elicit preferences for screening interval, the magnitude of effect may be different than if used in clinical practice or if we had examined screening interval. Further testing is needed to evaluate differences in screening outcomes and test its optimal use within the context of clinical care.

Data Availability

Study data may be made available upon request from the corresponding author.

Declarations:

Conflict of Interest:

The authors have no conflicts of interest to disclose.