Currently, the breast cancer screening strategy used in most countries (including Brazil), according to the best available evidence, is to perform biennial mammography, in the age group from 50 to 69 years (target population)7. Thus, there was a predominance in mammograms' number performed in the target population (63.4% of the total), in accordance with Ministry of Health guidance and guidelines for breast cancer early detection in Brazil7. According to the Breast Cancer Surveillance Consortium in the United States (BCSC), there was also a greater exams' number in population aged 50 to 69 years (about 48%)14. Despite this, the total time series of mammograms' number was decreasing in all age groups, including the target population. Even in the pre-pandemic period (2017 to 2019), the trend was stationary. Therefore, despite the greater exams' number performed between the ages of 50 and 69, the trend in mammograms' number has not followed the growth of the female population.
Screening women under 50 years or 70 years or older is questionable, considering that there is no clear evidence of a significant reduction in disease mortality5–6. Therefore, mammography is not performed routinely in these age groups, in most country, but according to the individual risk factors, after medical evaluation1,4,7. Even so, many exams are performed in these age groups, mainly from 30 to 49 years (30.1% of the total), a result similar to BCSC (29.4%)14. Among those aged 70 or older, the exams' number performed in Brazil (6.3%) is lower than that observed in BCSC (16.6%)14. In this case, as expected, the trend in mammograms' number was not increasing in these age groups, neither in total time series nor in the pre-pandemic period.
In 2020, the estimated women in Brazil were 107.5 million, with 21.1 million between 50 and 69 years old. About 12.5 million women in this age group (60%) depend on the Brazilian public health system. Considering the biennial mammography frequency, to meet the demand of the target population, in an ideal situation, there would be 6.2 million exams in 20207. However, there were 1.6 million mammograms performed, from 50 to 69 years old. Therefore, only 25.8% of the target population, which should undergo the exam, was screened in 2020. Even so, considering the average of 2.7 million mammograms performed per year in the pre-pandemic period (2017 to 2019), there were only 43.5% of the target population screened.
In Brazil, about 0.54% of women screened with mammograms are referred for cancer treatment7. In 2020, in the target population alone, there was a reduction of 1.1 million mammograms compared to the average exams' number performed in the pre-pandemic period. Therefore, there were about 5,940 women with undiagnosed breast cancer. However, in an ideal situation, considering the need for 6.2 million mammograms in the target population, in the public health system, there would be about 24,700 women not diagnosed early with breast cancer in 2020.
In 2020, there was a reduction in mammograms' number, identified in the segmented time series, lasting 3 months (from March to May), with a decreasing trend in women under 50 years old and over 69 years old. Specifically, in target population, this trend was stationary. This suspension or reduction period in mammograms' number also occurred in other countries, lasting from 1 to 6 months. In the United Kingdom, the study concluded that there would be an increase in 5-year mortality, from 6.3–22.3%, in case of interruption of screening at 3 and 6 months, respectively15. Furthermore, in 5 years there would be an increase in cases with advanced disease (stage III and IV) of 30% and 109%, in case of interruption of 3 and 6 months, respectively15.
Subsequently, there was an increasing period in mammograms' number in all age groups, lasting 5 months (from June to October), even during the COVID-19 pandemic. The North Carolina study identified a period of sustained recovery starting in April 2020 and, in September, it was already above expectations for a period without a COVID-19 pandemic16. Therefore, the recovery period in mammograms' number after the start of COVID-19 pandemic was not enough to change the decreasing trend of the series as a whole, in all age groups.
Breast biopsy is an essential procedure for the cancer diagnosis, and it is indicated according to the mammogram’s radiological classification. In Brazil, it is estimated that 1.6% of women screened with mammography are indicated for breast biopsy7. Therefore, the greater mammograms' number performed, proportionally, a greater breast biopsies' number is expected. However, in total time series, there was a dissociation between the increasing trend in breast biopsies' number and the decreasing trend in mammograms' number.
In addition, since the study beginning, there was a greater relationship in breast biopsies' number per mammogram performed, in women under 50 years old (2.84% on average) and over 69 years old (2.62% on average). On the other hand, women aged between 50 and 69 had a lower-than-expected ratio (1.16% on average). According to BCSC study, there was a smaller variation in this ratio, that is, 1.6% from 50 to 79 years, 1.8% from 40 to 49 years and 1.9% from 80 years or older14. Therefore, the breast biopsies' number grew disproportionately than mammograms' number, regardless of age group. Also, there was a greater breast biopsies' number performed outside the target population.
As mammography is not performed routinely in women under 50 years and over 69 years, but based on an individual medical evaluation, there would be a selection group with a higher cancer risk and a greater need for biopsies. On the other hand, there may also be inadequate procedure indication, that is, outside the Ministry of Health protocols7. Also, according to BCSC, there can be up to 12% of mammograms with false positive results, mainly in the population younger than 50 years14. Discordance in mammograms' radiological classification is also frequent and can result in unnecessary breast biopsies in up to 59.5% of cases17. Finally, another motivating factor for the increase in biopsies' number was a federal government incentive, increasing the values of procedures related to breast biopsy by 100% in November 201718.
There was a reduction in breast biopsies' number in the COVID-19 pandemic, regardless of age group. There was a decreasing trend, concomitant with the reduction in mammograms' number, lasting from 3 to 4 months, in all age groups, especially in women under 70 years old. However, the reduction in biopsies' number was not as important as the decrease in mammograms' number (monthly difference of 53%, on average). The recovery period was significant only in women under 50 years old, lasting 5 months. Both the reduction period and stationary trend in the recovery of procedures' number in women aged 50 years or older did not change the increasing trend of the time series as a whole. However, there was a reduction in the monthly percentage change compared to the pre-pandemic period. A North Carolina study also demonstrated a less intense reduction in breast biopsies' number than mammograms, similar to what was found in this study. However, the peak reduction in biopsies was not concomitant with that of mammography (it occurred 2 months later). Another difference was that 6 months after the beginning of the pandemic, the levels of all exams and procedures were already higher than those observed in the pre-pandemic.
Therefore, it is recommended to intensify breast cancer screening of women between 50 and 69 years of age in the country's public health system, to achieve an increasing trend compatible with the increase in this target population. Also monitor the quality of the exams and the BI-RADS radiological classification, to avoid performing unnecessary breast biopsies. It is essential to verify that the indications for breast biopsies are in accordance with the current protocols of the Ministry of Health of Brazil. New studies are important to assess the continuity of the time series of breast cancer screening and diagnosis in Brazil, from the end date of this study. This is essential to verify if there are favorable changes in trend of monitoring the target population.
Some limitations in DATASUS database may occur, such as lack of procedures registration performed, delay in forwarding, error or absence of information registered by service provider in system. It should also be considered that the data are not in real time. Therefore, there is a time interval between the performance of the procedure/service and the migration of information in the system (about 4 to 8 weeks, at least). Furthermore, only the exams performed in the country's public health system are available, that is, women who underwent mammograms and breast biopsies through the country's private health system are not in this analysis. Finally, epidemiological data on Brazilian population are available according to projection studies from the last census carried out, and therefore, they may not exactly represent the number of the real population.
At the beginning of the COVID-19 pandemic, there was a reduction in both the mammograms and breast biopsies' number performed, regardless of age group, lasting 3 to 4 months. The subsequent recovery period in mammograms' number was not sufficient to change the decreasing trend of exams in the time series as a whole, in all age groups. On the other hand, the trend of increasing in breast biopsies was maintained in the time series as a whole, however, with less intensity than the pre-pandemic period, in all age groups.