Overdiagnosis and underdiagnosis in BreastScreen Norway
Status: Completed
One in eleven women in Norway will be diagnosed with breast cancer by age 75. Organized mammographic screening can detect early stage breast cancer and reduce deaths from this disease. However, screening involves harms, such as the potential detection of a slow growing breast cancer that would never present symptomatically during a woman’s lifetime (overdiagnosis). “Underdiagnosis” can also occur, which was defined as “failing” to diagnose a breast cancer that would present symptomatically during a woman’s lifetime. This thesis aimed to generate knowledge about these harms through three studies.
Main findings: The first study demonstrated that radiologists and pathologists preferentially round tumour diameter measurements, which can lead to understaging but not overstaging.
The second study considered whether women with “missed” cancers that were diagnosed at a subsequent screening examination could be underdiagnosed. Based on their tumour characteristics and survival profile, it was posited that these women received a timely diagnosis.
The last study indicated that women have relatively less knowledge about overdiagnosis than other topics related to mammographic screening. This study highlights some of the challenges related to communicating information about overdiagnosis and screening.
Background
Breast cancer is the most common type of cancer among women in Norway. About 3,500 new cases are diagnosed each year. Screening with mammography can detect breast cancer in an early stage. In Norway, women aged 50-69 are invited to BreastScreen Norway.
The principle behind mammography screening is to examine many otherwise healthy women to identify the few who have breast cancer. The goal of screening is to reduce breast cancer deaths and to offer women less aggressive treatment; however, screening also carries some risks. It is important to minimize the risks associated with screening and avoid unnecessary treatment.
BreastScreen Norway works continuously to minimize the risks of mammographic screening, including the risk of overdiagnosis or underdiagnosis of breast cancer, among others.
Overdiagnosis occurs when screening detects small, slow-growing tumours that would never have caused symptoms or been detected if a woman hadn´t been screened. It is currently not possible to determine which specific tumours are overdiagnosed and all women are offered treatment for their disease. Overdiagnosed women may undergo treatment that offers them little to no medical benefit. It is difficult to know how often overdiagnosis occurs because there is no agreement on how to best estimate it. Some studies have estimated that mammographic screening leads to a very low rate of overdiagnosis. Others have estimated relatively high rates of overdiagnosis.
Underdiagnosis can occur when a breast tumour is overlooked. These tumours can later be detected between two screening examinations because of symptoms. Tumours detected between two screening examinations often have a worse prognosis than breast cancers that are detected as a result of screening. Only a small percentage of breast cancer cases in a screening program are underdiagnosed. This means that many mammography images are needed to perform a robust study on this topic. Obtaining these data is challenging.
Awareness and knowledge about breast cancer, particularly overdiagnosis and underdiagnosis, can be important for women when they decide whether to participate in BreastScreen Norway.
Aim
The objective of this PhD project was to investigate overdiagnosis and underdiagnosis in BreastScreen Norway. The project consisted of three papers based on separate studies that explored concepts related to overdiagnosis, underdiagnosis, and women’s knowledge about breast cancer in general and overdiagnosis in particular.
Publications
Study 1: Evaluate whether measurement error was present among tumour diameter data registered at the Cancer Registry of Norway and whether this had the potential to lead to under- or overstaging.
Published: Tsuruda KM, Hofvind S, Akslen LA, Hoff SR, Veierød MB. Terminal digit preference: a source of measurement error in breast cancer diameter reporting. Acta Oncol. 2020;59(3):260-7.
Study 2: Determine whether tumour characteristics and survival are differentially associated with certain groups of cancers, and explore whether this could indicate if certain cancers are more likely to be over- or underdiagnosed on a group level.
Published: Tsuruda KM, Hovda T, Bhargava S, Veierød MB, Hofvind S. Survival among women diagnosed with screen-detected or interval breast cancer classified as true, minimal signs, or missed through an informed radiological review. Eur Radiol. 2021;31:2677–2686.
Study 3: Describe Norwegian women’s conceptual knowledge about mammographic screening in general, and overdiagnosis in particular. This was a cross-sectional study based on an online questionnaire sent to women aged 50-69 in Norway.
Published: Tsuruda KM, Veierød MB, Houssami N, Waade GG, Mangerud G, Hofvind S. Women's conceptual knowledge about breast cancer screening and overdiagnosis in Norway: a cross-sectional study. BMJ Open. 2021 Dec 14
Research team
The PhD candidate is Kaitlyn Tsuruda, MSc, Cancer Registry of Norway.
The primary investigator and supervisor of this project is Solveig Hofvind, head of the section for breast cancer screening at the Cancer Registry of Norway.
Marit Veierød, professor in biostatistics at the University of Oslo, is the co-supervisor.
External collaborators include:
- Lars A Akslen, Centre for Cancer Biomarkers CCBIO, University of Bergen and Haukeland University Hospital
- Sameer Bhargava, Vestre Viken Hospital Trust
- Gunvor G. Wåde, Oslo Metropolitan University
- Nehmat Houssami, University of Sydney (Australia)
- Tone Hovda, Vestre Viken Hospital Trust
- Solveig Roth Hoff, Ålesund Hospital
