Randomized Trial of Information for Older Women About Cessation of Breast Cancer Screening Invitations

Background Older women receive no information about why Australia’s breast screening program (BreastScreen) invitations cease after 74 years. We tested how providing older women with the rationale for breast screening cessation impacted informed choice (adequate knowledge; screening attitudes aligned with intention). Methods In a three-arm online randomized trial, eligible participants were females aged 70–74 years who had recently participated in breast screening (within 5 years), without personal breast cancer history, recruited through Qualtrics. Participants read a hypothetical scenario in which they received a BreastScreen letter reporting no abnormalities on their mammogram. They were randomized to receive the letter: (1) without any rationale for screening cessation (control); (2) with screening cessation rationale in printed-text form (e.g., downsides of screening outweigh the benefits after age 74); or (3) with screening cessation rationale presented in an animation video form. The primary outcome was informed choice about continuing/stopping breast screening beyond 74 years. Results A total of 376 participant responses were analyzed. Compared to controls (n = 122), intervention arm participants (text [n = 132] or animation [n = 122]) were more likely to make an informed choice (control 18.0%; text 32.6%, p = .010; animation 40.5%, p < .001). Intervention arm participants had more adequate knowledge (control 23.8%; text 59.8%, p < .001; animation 68.9%, p < .001), lower screening intentions (control 17.2%; text 36.4%, p < .001; animation 49.2%, p < .001), and fewer positive screening attitudes regarding screening for themselves in the animation arm, but not in the text arm (control 65.6%; text 51.5%, p = .023; animation 40.2%, p < .001). Conclusions Providing information to older women about the rationale for breast cancer screening cessation increased informed decision-making in a hypothetical scenario. This study is an important first step in improving messaging provided by national cancer screening providers direct to older adults. Further research is needed to assess the impact of different elements of the intervention and the impact of providing this information in clinical practice, with more diverse samples. Trial Registration ANZCTRN12623000033640. Supplementary Information The online version contains supplementary material available at 10.1007/s11606-024-08656-3.

ABSTRACT BACKGROUND: Older women receive no information about why Australia's breast screening program (BreastScreen) invitations cease after 74 years.We tested how providing older women with the rationale for breast screening cessation impacted informed choice (adequate knowledge; screening attitudes aligned with intention).METHODS: In a three-arm online randomized trial, eligible participants were females aged 70-74 years who had recently participated in breast screening (within 5 years), without personal breast cancer history, recruited through Qualtrics.Participants read a hypothetical scenario in which they received a Breast-Screen letter reporting no abnormalities on their mammogram.They were randomized to receive the letter: (1) without any rationale for screening cessation (control); (2) with screening cessation rationale in printedtext form (e.g., downsides of screening outweigh the benefits after age 74); or (3) with screening cessation rationale presented in an animation video form.The primary outcome was informed choice about continuing/stopping breast screening beyond 74 years.RESULTS: A total of 376 participant responses were analyzed.Compared to controls (n = 122), intervention arm participants (text [n = 132] or animation [n = 122]) were more likely to make an informed choice (control 18.0%; text 32.6%, p = .010;animation 40.5%, p < .001).Intervention arm participants had more adequate knowledge (control 23.8%; text 59.8%, p < .001;animation 68.9%, p < .001),lower screening intentions (control 17.2%; text 36.4%,p < .001;animation 49.2%, p < .001),and fewer positive screening attitudes regarding screening for themselves in the animation arm, but not in the text arm (control 65.6%; text 51.5%, p = .023;animation 40.2%, p < .001).

INTRODUCTION
There is uncertain benefit of screening mammography for older women, as women aged ≥ 74 years have not been included in randomized controlled trials. 1 At a population level, potential harms become more likely with increasing age, including false positives and anxiety due to additional tests following detection of an abnormality, as well as overdiagnosis (diagnosis of a cancer that would never have caused problems) and overtreatment. 1 In one Australian modeling study, if screening were extended to 1000 women aged 70-74 years (compared to inviting women up to 69 years), one additional breast cancer death would be averted, eight breast cancers would be overdiagnosed, and 102 additional tests would include 78 additional false positives. 2However, a US modeling study indicated that continued screening to 78 or 80 years for women with no comorbidity could have similar benefit-harm ratios as screening women aged 50-74 years at average risk. 3][5][6][7][8] Screening is not recommended for individual women with limited life expectancy (< 10 years) due to the 10-year lagtime to benefit 9 and greater chance of harm. 10 However, over 50% of women aged ≥ 74 years report recent screening in US national survey data, 11,12 and around 38% of women with limited life expectancy continue breast screening. 13In Australia, the nationally funded screening program (Breast-Screen) invites women to participate in mammography screening for breast cancer up to 74 years of age through local health districts. 14Women aged ≥ 74 years can continue free screening and are encouraged to speak with their primary care clinician (general practitioner [GP]) if they wish. 14Screening rates by life expectancy are unknown in Australia.However, since BreastScreen extended invitations to women aged 70-74 years in 2013, annual participation rates in this age group increased from 25.9 to 55.8%, and data suggests 7.6-10% of women continue to be screened beyond 74 years. 15,16One reason for this may be the high trust people living in Australia have in the healthcare system. 17lder women should be supported to make informed breast screening decisions (i.e., decisions with adequate knowledge of potential benefits/harms, consistent with their values/preferences).However, many have limited knowledge of the potential harms of screening 17 and believe that they are no longer invited to screen for reasons they consider are ageist and related to government cost-saving or reduced risk of developing cancer. 18,19Ps also find it challenging to explain these recommendations. 20][23] We conducted a randomized controlled trial to test how different messaging about the rationale for breast screening cessation impacts older women's decision-making.We included intervention arms via printed-text or animation video.Animations are increasingly used for communicating health information and may result in greater knowledge improvements compared to printed-text materials, 24,25 particularly for people with low health literacy. 26

Study Design
This study was an online, three-arm randomized controlled trial.We followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline for randomized clinical trials.Participants were randomly assigned to one of three study arms using a balanced allocation ratio of 1:1:1.The protocol for this trial was registered with the Australian New Zealand Clinical Trials Registry (ANZC-TRN12623000033640) and ethical approval was obtained from The University of Sydney's Human Research Ethics Committee (2022/809).

Participants
Eligible participants included women aged 70-74 years, living in Australia, who had engaged in mammography screening at least once in the past 5 years.Women who had self-reported a current or previous diagnosis of breast cancer were excluded.This age range was chosen as these women would be approaching the upper age limit for cessation of invitation to the program.They were recruited through Qualtrics, a social research company who partners with many panels to recruit individuals who have already signed up to receive surveys.Upon registration, sample providers verify respondent address, demographic information, and email address.They target eligible participants based on specific parameters including age, gender, and location, but do not guarantee national representation among the invites sent or responses received.They use an incentivization system for participation, for which rewards vary (e.g., cash, gift cards, redeemable points, charitable donations).

Procedure
Qualtrics' online survey platform was used to administer the questionnaire and piloting was conducted to test the scenarios and questionnaire.Potential participants were sent an email invitation via Qualtrics with general study information, directed to the study landing webpage where they were able to read the participant information statement, screened for eligibility, and then consented to participation.Participants were presented with a hypothetical scenario in which they imagined they had received a standard BreastScreen letter after a recent mammogram which reported that no breast cancer could be seen.The delivery method of this letter was not specified.This format best represents current Australian BreastScreen practice, reflecting the usual result letters sent to women aged 70-74 years.After reading the letter, participants were randomly allocated to one of three arms: 1) Control: standard letter with no additional information.2) Intervention-printed-text: standard letter with additional information on the following page of the survey explaining in bullet points why they will not receive reminders to screen after 74 years.3) Intervention-animation video (1 min 51 s long): standard letter with additional information in a video on the following page of the survey explaining why they will no longer receive reminders to screen after 74 years.The video included subtitles and narration.A professional animator developed the video based on a script and reference videos provided by the lead researcher.
The messaging in both intervention arms was developed by the research team, including older female health consumers and experts in health communication (JS, RD, JJ, KM), breast cancer screening (NH, MS, WV), geriatric medicine (VN), and general practice (KW).The information explained why women will not receive reminders to screen after 74 years, due to both potential benefits and downsides of screening (overdiagnosis and overtreatment), and because with increasing age and other health issues the chance of experiencing the downsides of screening outweighs the chance of benefit (see Supplementary Material).The interventions were designed to be brief and feasible for delivery at scale.We therefore chose to focus on the harm of overdiagnosis as it is important in older age 27 and poorly understood by women.In contrast, the harm of false positives is better understood and tolerated by women. 28The information also emphasized the decision was up to them but that they could speak with their GP if they were unsure.Iterations were revised based on feedback from the research team, including two consumer experts who advised on acceptability and understandability.
Randomization was conducted automatically using the randomizer function within Qualtrics, which uses the Mersenne Twister pseudorandom number generator.Both participants and researchers were blinded to the allocation sequence until after data collection was completed.

Outcomes
Immediately after the intervention, participants completed the primary outcome measure of informed choice, which is a binary composite measure (informed/uninformed). 29n this hypothetical scenario, participants were considered to have made an informed choice if they had adequate knowledge and reported attitudes regarding screening for themselves that aligned with their intention to stop or continue screening (Box 1).0][31] Adequate knowledge was determined using the median score (i.e., participants had adequate knowledge if their total score was the within-sample median or greater).Personal screening attitudes were assessed using six items scored on a total scale from 6 to 30, with the within-sample median score used to dichotomize into "positive attitudes" and "less than positive attitudes," as has been done in prior decision aid work. 32Screening intention was assessed by a single item with a 5-point scale (1 = Definitely will to 5 = Definitely will not).We decided a priori to dichotomize responses to intending to screen (definitely or probably will) or not intending to screen (unsure or definitely or probably will not), similar to prior decision aid work. 29Secondary outcomes included intention to speak to a GP, perceived risk of developing cancer, cancer worry, and emotional response to the letter they received.See Supplementary Table I for further details.*Within-sample median threshold was chosen for adequate knowledge because a typical "pass" (or 50%) cut-off would not be appropriate.Participants could, on average, answer 50% of the questions correctly by guessing the answers.
5][36] An attention check question and consistency check question were also included to assess the quality of respondent data.

Statistical Analysis
We conducted a sample size calculation assuming a conservative estimate that 50% of participants in one study arm would make an informed choice.We required 110 participants per arm (330 total) to detect a 20% difference in informed choice between the arms, which would have 80% power (α of 5%, two-sided test).Descriptive statistics were calculated for sociodemographic and health characteristics of the sample, and the primary and secondary outcome measures (mean [SD] for continuous variables and frequency and relative frequency for categorical variables).Statistical significance of study arm differences was tested for using chi-squared tests for binary categorical outcomes, ordinal regression models for ordered categorical outcomes, and linear regression models for continuous outcomes.The assumption of parallel lines for ordinal regression models was satisfied.All statistical models included study arm (control, printed-text information, animation video information) as categorical covariates.Where overall differences were identified, planned simple contrasts (e.g., control vs. text intervention, control vs. animation intervention) were conducted.A Bonferroni-adjusted significance threshold of p = 0.017 was used for multiple pairwise comparisons.
We conducted sensitivity analyses where participant responses from Western Australia were removed to test whether the outcomes differ, as BreastScreen Western Australia provides older women with information about the upper age of screening programs through their website and final screening letter.This information states, "There is no scientific evidence that for women over 75 participating in a population breast cancer screening program results in more health benefits than harms." 37Other states in Australia do not provide older women with information about why screening invitations stop after 74 years.Sensitivity analysis was also conducted with adjusted thresholds set for primary outcomes (see Supplementary Table IV).SPSS (version 28) was used for all analyses.

RESULTS
A total of 938 respondents accessed the first page of the survey from March 8 to 29, 2023, of whom 451 were randomized and 376 were included in the final analysis (Fig. 1).Characteristics of the participants are presented by study arm in Table 1.Most participants were born in Australia (69.4%) and located in the states of New South Wales, Victoria, or Queensland (80.6%).Half of the participants had an education level of high school or less (51.1%), over half had private health insurance (60.4%), and most reported adequate health literacy (94.1%) using the single-item health literacy screener. 38esults for primary and secondary outcomes are presented in-text below, including pairwise comparisons where overall differences were identified between study arms.Table 2 presents results for the primary outcomes by control and intervention arms, including a breakdown of each item contributing to the composite "informed choice" measure.Secondary outcomes by study arm are presented in Table 3.

Table 1 Characteristics of Participants, Presented by Study Arm and Overall
*All values are numbers and proportions (i.e., N (%)), except for age which is reported as means (M) and standard deviation (SD) + Assessed using a single-item self-reported screener that asks whether the respondent needs help when reading written material from a healthcare professional 33 # Assessed using the combined Lee-Schonberg index which is available on the ePrognosis website (https:// eprog nosis.ucsf.edu/). 36Points are accumulated through answers to 15 questions regarding health status and functioning in order to calculate an estimate of risk < 10-year life expectancy 34
Regarding emotional response to the BreastScreen letter, those who received the text (mean difference [MD] = − 0.58, 95%CI = − 0.98, − 0.17, p = 0.006) or animation (MD = − 0.48, 95%CI = − 0.89, − 0.07, p = 0.023) were less likely to report feeling relieved compared to control (mean = 5.66).There were no significant differences in the degree to which participants were assured, informed, anxious, worried, or confused in response to the letter across intervention or control arms (all p > 0.05).

Sensitivity Analysis
All main effects for primary outcomes did not differ in a sensitivity analysis (n = 343) where all participants who reported being from Western Australia were removed (n = 33) (Supplementary Table III), as well as in analyses with adjusted thresholds for the primary outcomes (Supplementary Table IV).

DISCUSSION
In this online randomized controlled trial, women aged 70-74 years who received information that explained the reasons for screening invitation cessation were more likely to make an informed choice about screening beyond 74 years than women who did not receive this messaging.Women who received the interventions had greater knowledge, less positive screening attitudes, and reduced screening intentions.
The animation intervention increased the proportion of participants making an informed choice by 24% compared to controls, and the text by 15%.Similarly, a breast cancer screening decision aid for women aged 70 years resulted in a 25% increase in informed choice compared to controls. 31ur findings have significance for implementation given that our developed interventions were brief, which would be relatively easy to scale-up across a population screening program compared to a decision aid which requires time and significant discussion for adequate development and use, presenting considerable challenges for implementation. 39Decision aids may still be useful and appropriate for older adults who desire more detailed information, as interventions at patient, clinician, and system levels are key to improving decision-making around screening for older people. 21,22Screening providers communicating direct to consumer may be a feasible and effective way to counter the negative beliefs women hold regarding the current recommendations for no longer being invited to screen. 18,19Given how challenging it is for consumers to grasp the concept of screening harms such as overdiagnosis, 40 these findings also have wider implications for other screening programs, highlighting how overdiagnosis can be simply and effectively communicated.However, the context in which recommendations are communicated may have different impacts on how this messaging is received.For example, in the US, different levels of trust in the government-funded health insurance and controversy surrounding breast screening for older women may reduce the impact of this kind of messaging.
There are consequences of providing information that should also be considered.Firstly, our interventions led some older women to be more unsure about whether to continue or stop breast screening and more likely to want to speak with their GP.Clinicians may need support and training to navigate these discussions and provide recommendations if requested. 20,41Verbal scripts have been developed in the US that suggest how the idea of stopping cancer screening could be introduced, how a recommendation to stop could be explained with further reasoning, or how the benefits and harms of screening could be discussed in a shared decision-making conversation. 42These strategies for discussing stopping screening could be further adapted for use in contexts where national screening programs are implemented.Secondly, women who did not receive additional messaging reported greater relief in response to the BreastScreen letter.This may be explained by the relieving nature of receiving a letter from a trusted organization that provides a reassuring screen result without needing to consider additional information, or because information about screening benefits and harms reminds women that they are getting older and therefore feel a lack of control.Given women perceive mammograms as a way of looking after their health, this information could lead to feelings of distress. 43ealth animations have shown promise in communicating complex information to people with low health literacy. 26here were no statistically significant differences observed between text and animation study arms.However, the format of information delivery is worth exploring in future studies with larger, more diverse samples.Animations may be particularly useful for low health literacy audiences when communicating conceptual knowledge rather than numerical risk information. 44Regardless, both intervention formats improved informed decision-making which could provide helpful flexibility in allowing for potential individual differences in needs and preferences for text or animation formats.

Strengths and Limitations
This study was strengthened through randomization to study arms ensuring similar baseline characteristics of participants across arms.In addition, differences in outcomes can be attributable to the intervention.We also utilized survey software techniques to ensure quality responses (attention and consistency checks) and designed the messaging to potentially supplement existing BreastScreen communications and be easily implemented and scaled up.
There are limitations to this study.Recruitment through an online panel means that our sample is likely not representative of the women aged 70-74 years who would typically attend to BreastScreen services in Australia (i.e., underrepresentation of culturally and linguistically diverse and low socio-economic groups). 15Qualtrics randomly selects respondents but does not guarantee national representation.Our sample reported around 30% were not born in Australia, but approximately 40% of women attending BreastScreen are not born in Australia. 16Furthermore, this study was hypothetical in nature.Older women who receive this information may respond differently in real-life settings.Further research is needed, such as a clinical pilot trial conducted through BreastScreen services or other screening programs to understand the feasibility and impact of these interventions in practice, with more diverse samples.
We also cannot guarantee all participants paid attention to the entire intervention content despite restricting ability to proceed in the survey after a reasonable time.Additionally, there was a slight difference in the content provided in the two versions of the intervention.Pre-existing beliefs about why screening invitations stop were explicitly acknowledged in the animation version (i.e., government cost-saving, decreased risk), but not in the text version.Future research should test different elements of the information provided to understand which parts are more effective among different audience types, and whether directly addressing pre-existing beliefs is important in supporting informed choice.
Our interventions aimed to align with the BreastScreen Australia upper age limit for screening invitations (74 years), but we recognize there is no consensus on this matter.They were therefore limited in covering the nuanced information likely required for informed choice for some older women, but rather general information as a first step to be further elaborated with a primary care physician if desired.Further support is required for older women and clinicians navigating this decision.Another limitation is that intention rather than actual behavior was measured.However, lower screening intentions have been shown to be associated with lower screening rates. 45inally, there are ongoing discussions around what thresholds should be set to define an informed choice. 29Our sensitivity analyses showed that the overall main effects remained significant with different thresholds.Using a threshold of below 30 on the attitudes scale as "less positive screening attitudes" due to the high within-sample median is justified in the context of this study.The public has well-documented strong positive attitudes toward cancer screening, 46,47 which are particularly influential in older adults' cancer screening decisions 48,49 (likely due to a longer period of exposure to persuasive screening messaging).Scoring less than the maximum score on this scale may in fact have clinical significance.

CONCLUSIONS
Messaging outlining the rationale for cessation of breast screening invitation to women aged 70-74 years in this online randomized controlled trial improved informed choice and knowledge, reduced positive screening attitudes, and lowered screening intentions.This study is an important first step in improving the messaging provided by national cancer screening programs direct to older adults and supporting more informed choices.
Acknowledgements: Genevieve Webb and Deborah Fulwood were expert health consumers who contributed to the development of interventions and outcome measures.We would also like to thank Dr. Brooke Nickel for providing helpful assistance in study design.

Box 1
Informed Choice Framework.

Figure 1
Figure 1 Participant flow diagram.*Potential duplicates were flagged by Qualtrics based on duplicate IP addresses.

Table 2 Analysis of Primary Outcome. The p-Value Provided Is for the Main Effect of Study Arm
Data are number of participants (%), unless otherwise stated # Six attitude items were rated from strongly disagree (1) to strongly agree(5).Range of possible scores was 6-30 where higher scores indicated more positive attitudes *Informed choice defined as adequate knowledge and screening intentions aligned with screening attitudes

Table 3 Secondary Outcomes. The p-Value Provided Is for the Main Effect of Study Arm
Data are number of participants (%), unless otherwise stated.CI confidence interval, GP general practitioner Ns for emotion outcome differ due to missing data ).Dr. McCaffery is supported by an NHMRC Principal Research Fellowship (1121110).Dr. Houssami is supported by a National Breast Cancer Foundation (Australia) Chair in Breast Cancer Prevention (EC-21-001), and an NHMRC Investigator (Leader) Fellowship (1194410).Dr. Mara Schonberg is supported by a NIH/NIA K24 K24AG071906 grant.Data Availability Data are available upon reasonable request.Deidentified participant data will be made available upon request to anyone wishing to access it who provides a methodologically sound proposal to the principal investigator.The contact detail of the principal investigator is kirsten.mccaffery@sydney.edu.au.The authors declare that they do not have a conflict of interest.Disclaimer: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material.If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the