Abstract
Introduction
Interval breast cancers are detected symptomatically after a non-suspicious mammogram, but before the patient’s next scheduled screen. Interval breast cancers are often diagnosed at a later stage and larger size, and have poorer prognostic factors and survival than screen-detected breast cancer. Our qualitative study heard from women with interval breast cancer to describe their symptoms and their reactions to the diagnosis; and identify themes for educational messaging.
Methods
We conducted 20 in-depth interviews with participants who were between the ages 40-69, had all screening, diagnostic, and treatment services completed within the same hospital system, and had a negative mammogram screen followed by breast cancer diagnosis before the next screen.
Results
Fifteen women noticed a lump cyst during a breast self-exam or when dressing. Most women reached out to their gynecologist or their primary care provider. Main themes from their reactions to the diagnosis included: unaware that interval breast cancers could occur; surprised that screening tools could not see all cancer; worried to being seen as ridiculous based on previous experience with non-cancerous breast issues, and disappointed that they waited to reach out for care. Ideas for messaging included: listen to your body, prioritize your health, and keep doing breast self-exams.
Conclusion
Breast cancer prevention programs should focus on the awareness of interval breast cancers and the importance of breast self-exams and self-awareness in conjunction with screening mammograms.
Plain Language Summary
Introduction. Interval breast cancers are defined as cancers that emerge after a non-suspicious mammogram, but before the patient’s next scheduled screen. Interval breast cancers are often diagnosed at a later stage and larger size, and have poorer survival than screen-detected breast cancer. Purpose. Our study talked with women with interval breast cancer to describe their symptoms and their reactions to the diagnosis; and identify ideas for educating other women. Methods. We recorded 20 in-depth interviews with women who were between the ages 40-69, had all their health services completed within the same hospital system, and had a negative (normal) mammogram screen followed by breast cancer diagnosis before the next mammogram. Results. The symptoms were all first discovered by the woman. Seventy-five percent of women noticed a lump cyst during a self-breast exam or when dressing. Most women reached out to their gynecologist or their primary care provider. Main themes from their reactions to the diagnosis included: unaware that interval breast cancers could occur; surprised that screening tools could not see all cancer; worried that they were seen as ridiculous based on previous experience with breast issues, and disappointed that they waited to reach out for care. Ideas for messaging included: listen to your body, prioritize your health, and keep doing self-breast exams. Discussion. Breast cancer prevention programs should focus on the awareness of interval breast cancers and the importance of self-breast exams and self-awareness in conjunction with screening mammograms.
Introduction
In the United States, screening recommendations for breast cancer include annual and biennial mammograms starting at age 40. 1 Even when women follow these recommendations, cancers can develop. These cancers, known as interval breast cancers, are detected symptomatically after a non-suspicious mammogram but before the patient’s next scheduled screen. Interval breast cancers include missed cancers, true interval cancers, or occult cancers. 2 Missed cancers are false-negatives where the cancer is visible on the screening mammogram but is not recalled or is misinterpreted. True interval cancers are where the cancer is not visible on the screening mammogram but becomes visible at the diagnostic mammogram. Occult cancers are not visible on either the screening or diagnostic mammogram, but may be symptomatic or diagnosed through other modalities such as ultrasound. The incidence and proportion of interval breast cancers increase with longer screening intervals.2–4 According to Orsini et al, the proportion of interval cancer in annual screening is 12.5%, biennial screening is 28.5%, and 42.5% in triennial screening. 3
While recommended screening can identify the majority of breast cancers, approximately 8% to 30% of breast cancers are discovered between screening in the United States.5–8 According to the sensitivity of mammogram machines, false negative are low – around 1% - and have improved with improved technology. However, the proportion of interval breast cancers are larger than just the false-positives because they also include both true or occult cancers. The proportions of interval cancers among women diagnosed with breast cancer range in the US from 8-30%5–8 with a similar range observed in other countries.9–16 Interval breast cancers are often diagnosed at a later stage, at a larger size, and have poorer prognostic factors than screen-detected breasts cancers such as significantly more lymph node involvement and lobular histology.2,6,10,17–22 Interval breast cancers are associated with a 2-to-3 fold increased risk of breast-cancer-specific mortality compared with screen-detected cancers, even after adjustment for clinically relevant variables.2,18–24 Risk factors include breast density, hormone replacement therapy, young age, family history, and a previous false positive.2,25,26
Most previous studies of interval breast cancers employed epidemiological designs such as cohort or case-control and used quantitative data abstracted from medical records or large cohort studies. 19 Program evaluation of screening programs have found that women who were diagnosed with an interval breast cancer raised concerns about their trust in screening programs and mammography technology and wanted more education about interval breast cancer.27,28 Our research focused on women’s experience with interval breast cancer, how or when symptoms are noticed, and their perception of risk following a normal mammography screen. We asked women for their advice on messages that could be used in an educational campaign about interval breast cancer.
This study uses a one-time point, cross-sectional design to listen to women diagnosed with interval breast cancer to: • Describe symptoms and reactions to the diagnosis of interval breast cancer; and • Identify themes for educational messaging.
Methods
We conducted 20 in-depth interviews over Zoom which lasted on average 32 minutes (range 21-56). Women aged 40 and older were recruited in three counties from the same multi-county community hospital system in the United States. Women are typically referred annually by their primary physician or gynecologist for their screening mammogram. Interval breast cancers were defined as a breast cancer diagnosed after a negative mammogram, but before the patient’s next scheduled screening mammogram. Study procedures were approved by the Institutional Review Board at the Samaritan Health Services (IRB20-032) and the Oregon State University IRB ceded oversight (IRB-2021-1059). Forty-four participants were eligible based on data in electronic medical records and twenty women completed interviews between June and December 2021. See Figure 1 for the consort diagram. The age of diagnosis ranged from 40-67 with a mean of 58.8 (SD = 8.1) years. Consort diagram of women eligible, contacted, and interviewed
Eligibility
Staff at the community hospital (SH, KM, AF) identified potential participants through electronic medical records using the following inclusion criteria: female; age at screen between 40-69; lived in one of the three specified counties in the United States; all screening, diagnostic and treatment services done within same community hospital system; medical records January 2013 - May 2021, and negative mammogram screen followed by breast cancer diagnosis before next screen. Women needed to have participated in more than 1 round of mammography to be included. Exclusion criteria included a previous breast cancer diagnosis; screened or diagnosed outside of available electronic medical record review; and under age 40 or over 70+ at mammogram.
Recruitment
Letters were mailed to women in either English or Spanish based on their preferred language identified in medical records (SH, KM) with a signature and contact information from the doctor (DB) who followed up with each potential participant. If the participant agreed, the staff from the community hospital (SH, KM, AF) emailed the university investigator (VI) and shared names and contact information through a secure process. The university investigator called participants to describe study procedures. If interested, participants were emailed the consent and release of protected health information form to sign.
Interview Process and Questions
Participants were offered interviews in English and Spanish conducted by the female university investigator. No participant requested Spanish language.
Interviews used a semi-structured interview guide to assess: how the interval cancer was noticed and by who (i.e. the woman, a provider); reactions to diagnoses; if and why women delayed or sought care; barriers to care; ideas for future health messaging. At the end of the in-depth interview, we asked close-ended items: checklist of symptoms, approximate wait time to seek care, personal and family cancer history, age at diagnosis, type of health insurance. See Appendix for the full interview guide.
Staff from the community hospital provided dates of the last screening or routine mammogram, positive mammogram, and diagnosis from the medical record.
Statistical Analyses
Study procedures follow the COREQ (COnsolidated criteria for REporting Qualitative research) procedure. Interviews were recorded accompanied by detailed written notes as a back-up. Recordings and transcriptions were done using Zoom, de-identified at the point of data collection, and stored on a password protected computer and university-managed cloud storage. Study co-author JCI compared each transcript to the audio recording to ensure accuracy. Interviewers read each transcript and drafted a memo of key points or themes that they took away from each interview before the coding process began. A deductive thematic approach was engaged, and a series of preliminary structural and descriptive codes were generated a priori on anticipated responses. Next, a small group of female, co-authors (VI, MLG, JCI) and students from the university reviewed 2 interviews based on the list of a priori codes, provided feedback, and finalized the code book. Two female co-authors (VI and MLG) coded each interview individually using the structural and descriptive codes in Atlas.ti. We reviewed for agreement and calculated percent agreement in Atlas.ti for the following discrete variables: age started screening, screening guidelines, length of time waited, who contacted, symptoms, experience with navigator, history of dense breast, non-cancer breast issue, and family history. All of these variables had inter-rater agreements greater than 85%. Study authors VI and MLG met to review discrepancies, developed and compared themes and reviewed themes with other co-authors.
Results
Participants adhered to mammography screening guidelines. Twelve started screening in their 40 s and 6 reported that they started screening earlier than age 40 most of whom reported a family history of breast cancer. Seventeen participants followed annual screening recommendations and fifteen had been consistently screening annually. Some women had only screened for 1-2 years, others had adhered to screening for over 20 years before ever being diagnosed with cancer. However, all the women had completed more than one round of mammography screening.
Half the women lived in a self-described rural area. All had insurance coverage during their care. More than half of women had a family history of breast cancer (n = 12) which included 6 with a mother, 2 with a sister and the remainder with another female relative. Eight women reported having dense breasts, four reported fibrous breast tissue and five had a prior cyst or fluid-filled cyst. Three women reported receiving care during the covid-19 pandemic and did not report any barriers to care because of the pandemic.
Symptoms and Reaching Out for Care
All cancer symptoms were first discovered by the woman. The most common symptom was a lump – fifteen women noticed a lump during a breast self-exam or when dressing. Other symptoms included pain or burning sensation (n = 4), their breast did not look typical (n = 2), or red (n = 1). Most women reached out to their gynecologist (n = 9) or primary care provider (n = 7), and 1 reached out to her oncologist because she was previously diagnosed with a different type of cancer. According to medical records, the number of days between their last negative screening mammogram and their initial positive diagnostic imaging ranged from 61 – 363 days with an average of 239.5 days, standard deviation of 36.4 days.
Reactions to the Diagnosis of Interval Breast Cancer
Qualitative Themes and Exemplar Quotes From Women With Interval Breast Cancer
First, most participants had not heard about interval breast cancers. Many women did not realize that these cancers could emerge so quickly after a normal mammogram and could be so aggressive. For example, participant ID #7 noted “Absolutely didn’t believe it could be and that’s why I waited a full month before I even called the doctor.” Second, most women in our study had received their screen using a 3D mammogram (n = 11) with the remainder being screened by a 2D mammogram (n = 8) or ultrasound (n = 1). Women with dense breasts were surprised or shocked that dense breasts meant technicians or radiologists could not see tumors as well on these machines. Participant ID # 3 stated “I had a cancer, that is not seen with the mammogram which I can’t believe in this day and age, that happens.” Information may be available to women, but they might not recognize the importance at that time. Participant ID # 3 reflected on information received in mammogram results, “ You have to read the fine line on your mammogram results where not all cancers are found with a mammogram”.
Five women had previously experienced symptomatic benign breast issues such as fluid-filled cysts. These women had previously reached out to a provider and were told they did not have cancer. When they noticed symptoms this time, they were reluctant to contact a provider because the symptom was in the same area where the cyst had been. Several participants expressed that they doubted and waited to call because they did not want to appear foolish or ridiculous as described by participant ID #20, “ I didn't want them to think I was this crazy hypochondriac because it was right in the area where the cyst always was… so I didn't want to look like the crazy lady whose screaming that the sky is falling, when I just had a cyst that’s filling up.”
Lastly, women were disappointed that they waited to reach out to their provider. The lack of awareness of interval breast cancer combined with the perception that screening would catch all tumors led to women waiting. This waiting led to feeling disappointed with themselves and guilt. An example is provided by participant id # 1, “I was very disappointed in myself that I waited the six months to get that ultrasound maybe it wouldn’t have gone into the lymph nodes but everybody was saying no harm no foul. Don’t beat yourself up.” Some participants assured themselves that their symptoms were nothing to worry about because they recently had a normal mammogram. After noticing symptoms, some women called their health care provider within a few days (n = 3), but most waited weeks (n = 6) or months (n = 6) or couldn’t remember how long it took them to call. Six women waited a full month to call a provider after noticing their symptoms.
Messages for Health Education About Interval Breast Cancers
We asked participants about additional resources for interval breast cancer. Many advocated for more availability for evening and weekend appointments and more access for un or under-insured. Nine participants had worked with a nurse navigator and 8 thought that a nurse navigator would be useful for all cancer patients.
Most participants suggested resources on awareness about interval breast cancers such as pamphlets, interactive displays or video demonstrations in the waiting rooms of primary care, gynecology and radiology. These resources should be available in languages beyond English and available in languages used by the community. Participants also suggested TV or radio coverage about interval breast cancer and breast self-exams.
We synthesized ideas for messaging suggested by participants. Three themes emerged - (1) Listen to your body and don’t doubt, (2) Prioritize you and take action, and (3) Keep doing your breast self-exams. The most common sentiment that women wanted to tell others was to listen to your body and don’t doubt. They wanted women to trust what they were feeling or sensing about their body. They knew their bodies the best and should not doubt what they were feeling even if they thought it could be a recurring non-cancer issue. Women noted another message was to urge other women to prioritize themselves and take action. Participants often waited out of an obligation to work or family and they urged women to prioritize their own health. They did not want others to wait, because the tumor can grow very rapidly and they wanted women to advocate for themselves and not to feel pushy or a burden on the system. Lastly, about half of women interviewed stressed focusing on breast self-exams and the need to continue to show women how to accurately do breast self-exams. Results letters from the participating hospital system tells women to perform monthly breast self-exams. However, participants remember receiving more educational resources around breast self-exams when they were younger. They wanted younger women today to learn about breast self-exams.
Discussion
All women in our study noticed their symptoms with most noticing a lump. Common reactions from participants were that they were unaware that interval breast cancers could occur or that screening tools could not see all cancers. Interval breast cancers occurred among women who followed screening guidelines, among women with and without dense breasts, and among women with and without a family history. Interval breast cancers can happen to any woman, and it is up to the woman to notice the symptoms and seek care soon. The importance of breast self-exams and self-awareness was mentioned by many participants with a focus to teach proper breast self-exam techniques and to remind women to continue with the breast self-exams in conjunction with their mammograms.
Our study findings matched those of other studies including symptomology,20,29–31 the relationship of breast density to interval breast cancer,14,25,32 and the theme of surprise or shock. 33 The most common symptom was a lump noticed during a breast self-exam or when dressing and similar symptoms were noted in other publications. 20,29–31 Approximately 35% of our sample included women with self-reported dense breast. Higher rates of interval breast cancers have been shown to be related to breast density.14,25,32 One of the main themes that emerged was surprise or shock at the diagnosis which has been documented in other cancer patient narratives. 33
Interviews with our participants highlight the trust that women had in mammograms. Two of the most common themes were that women were unaware of interval breast cancers and they were surprised that mammograms could not see all cancers. These concerns have been raised in program evaluations of screening programs.27,28 This was true especially for women with dense breasts. Recommendations from our participants for messaging focused on increasing awareness of interval breast cancers among all women and in multiple languages and to offer additional communication with physicians.
We learned that women waited to reach out - many participants waited a full month after symptoms to contact their provider. Women waited out of work or family obligations or because they doubted their symptoms since they had recently received normal results following a mammogram. When women found out that they had cancer, they were disappointed in themselves for taking too long. Recommendations were to craft messaging around the idea of women listening to their body and prioritizing their health.
An unexpected theme that we discovered was the relationship between a symptomatic benign breast issue, such as a cyst, among women with interval breast cancer. These women had previously reached out to a provider for a symptomatic breast issue and were told they did not have cancer. When they noticed symptoms this time, they were reluctant to contact a provider because they were worried about looking foolish again for a benign issue. Yet, given the limitations of screening, it is important to empower women to seek help. Women with a history of benign breast issues such as masses and microcalcifications were more likely to have mammography recalls including both false-positive and true-positive recalls. 34 Although women in our study reported symptomatic benign breast issues and not false-positives, studies have found similar hesitation to reach out for care following a false-positive recalls. 34 Previous studies have shown that a risk factor for interval breast cancer is a previous false positive and that women with a false positive were less likely to return to regular screening.25,29 Recommendations for care teams would be increased awareness and procedures for obstetrics, gynecology, primary care offices to reach out to women with a previous non-cancerous breast issue. Providers should discuss risk factors and benefits and harms of supplemental testing for women with dense breast tissue. 35
Limitations
We received only limited data from the medical record and we were unable to confirm that women had genetic risk, dense breasts or a previous false-positive recall. Potential sources of error could be subject recall where participants may have had difficulty remembering details of her breast cancer experience if it occurred several years ago. To mitigate these potential biases, we started the interview with the broader experience of her cancer and then moved to recalling specific details within that larger experience to help improve memory recall. The university investigator conducted interviews to reduce response bias because she was not involved in the woman’s care. Participants represented a higher socio-economic status 36 and had access to health care which may differ from other populations in the United States; yet they still lacked awareness of interval breast cancer. Lastly, we only interviewed women in English. Although Spanish-language services were offered, no women interviewed needed them. In order to get a perspective from women who prefer speaking Spanish, we reached out to two Latino community groups and presented aggregated findings from the English-language interviews and received their feedback on messages. The additional messaging advocated for this group was coverage for mammograms and cancer treatment regardless of citizenship status and for more appointments in evenings and weekends and access to a patient navigator.
Recommendations
Recommend improving education for all women about symptoms and how to reach out to their provider.37–39 Women need to be better informed about symptoms and not wait to call their provider.31,40 Recommend increased outreach and education to women with non-cancerous breast issues with emphasis that they are not ridiculous and to continue to reach out if experiencing symptoms. Recommend provider training for increased patient-provider communication about cancer screening recommendations, breast density and relationship to cancer screening, and concerns related to benign breast issues and symptoms. Provision of culturally-responsive patient navigation services to support women through their cancer care journey.
Breast self-exam have failed to show evidence of benefit to mortality in randomized controlled trials1,41 Even though these trials have not statistically shown the benefit of breast self-exam, organizations such as World Health Organization, Breast Cancer Now, Union for International Cancer Control, and the American Cancer Society recommend mammography screening with breast self-exam. Women in our study mentioned how they no longer hear about the need for breast self-exams. However, everyone in our study found the symptoms on their own and not through a clinical breast exam or screen. Even in this time of new screening and artificial intelligence technologies,42-44 breast self-exams are still vital to finding cancer especially in geographical areas with limited access to mammography 45 or for women who are under or un-insured.
Conclusion
Breast cancer prevention programs should focus on the awareness of interval breast cancers and the importance of self-breast exams in conjunction with screening mammograms.
Supplemental Material
Supplemental Material - “Screening isn’t Your Ticket to Vacation”: In-depth Interviews With Women in the United States Who Experienced an Interval Breast Cancer
Supplemental Material for “Screening isn’t Your Ticket to Vacation”: In-depth Interviews With Women in the United States Who Experienced an Interval Breast Cancer by Veronica L. Irvin, Danielle Bertoni, Maritza Leon Gutierrez, Jimena Caballero Ignacio, Stephanie Hebert, Kristen Moylan, Anthony Franklin in Cancer Control.
Footnotes
Ethical Approval
The Institutional Review Board at the Samaritan Health Services (IRB20-032) approved the study protocol, participant consent, and recruitment materials. The Institutional Review Board at Oregon State University ceded oversight (IRB-2021-1059).
Consent to Participate
Letters were mailed to each woman in their preferred language identified in medical records with a signature and contact information from the doctor who followed up with each potential participant. If participant agreed, the staff from the community hospital emailed the university investigator and shared names and contact information through a secure process. The university investigator called participants to describe study procedures. If interested, participants were emailed the consent and HIPAA release form to sign.
Author Contributions
Study authors VI, SH, AF conceptualized the original study idea and submitted the grant application. All study authors reviewed and edited study protocols and assessment tools prior to data collection. Staff at the community hospital (AF, KB, SH) identified potential participants through Epic. Letters were mailed to women in their preferred language identified in medical records (SH, KB) with a signature and contact information from the doctor (DB) who followed up with each potential participant. If participant agreed, the staff from the community hospital (SH, KB, AF) emailed the university investigator (VI) and shared names and contact information through a secure process. The university investigator (VI) called participants to describe study procedure and conducted all interviews. Study co-author JCI compared each transcript to the audio recording to ensure accuracy. Co-authors (JCI, MLG, VI) finalized the code book. Two co-authors (VI and MLG) coded each interview individually and discussed and agreed up discrepancies. All study authors reviewed and had opportunity to provide interpretation for study results. Lead author (VI) wrote the first draft of the paper and all other co-authors reviewed and edited the paper as needed.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: funding for this project was provided through grant #20192032 by the OHSU Knight Cancer Institute Community Partnership Program, a grant program that supports communities across Oregon in addressing local cancer-related needs. Additional funding from Oregon State University, Celia Strickland Austin and G. Kenneth Austin III, the Celia Strickland Austin and G. Kenneth Austin III Endowed Professorship. The funder was not involved in deciding study purpose, design, methods, or results interpretation.
Declaration of Conflicting Interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr Irvin has financial interests in IZ Translational Sciences, a company which commercializes technology to increase decision-making skills and actions among community members. The conduct, outcomes, or reporting of this research could benefit IZ Translational Sciences and could potentially benefit her.
Data Availability Statement
This study was not required to register with 
and the analytic plan was not preregistered. The data generated in this study are not publicly available due to information that could compromise patient privacy or consent, but aggregate, de-identified data are available upon reasonable request from the corresponding author. Study author can share analytic code and summary information related to study.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
