Harms of Breast Cancer Screening: Systematic Review to Update the 2009 U.S. Preventive Services Task Force Recommendation

In 2009, the U.S. Preventive Services Task Force (USPSTF) recommended biennial mammography screening for women aged 50 to 74 years (1) on the basis of evidence of benefits and harms (2, 3). The USPSTF concluded that screening decisions for women aged 40 to 49 years should be based on individual considerations and that evidence was insufficient to assess benefits and harms for those aged 75 years or older (1). Although there is general consensus that mammography screening is beneficial for many women, benefits must be weighed against potential harms to determine the net effect of screening on individual women. Determining the balance between benefits and harms is complicated by several important considerations that are unresolved, including defining and quantifying potential harms; the optimal ages at which to begin and end routine screening; the optimal screening intervals; appropriate use of various imaging modalities, including supplemental technologies; values and preferences of women in regards to screening; and how all of these considerations vary depending on a woman's risk for breast cancer. This systematic review updates evidence for the USPSTF on the harms of breast cancer screening, including false-positive mammography results, overdiagnosis, anxiety, pain during procedures, and radiation exposure, and how these adverse effects vary by age, risk factor, screening interval, and screening modality. Systematic reviews of the effectiveness of screening (4), performance characteristics of screening methods (5), and the accuracy of breast density determination and use of supplemental screening technologies (6) are provided in additional reports. Methods Scope, Key Questions, and Analytic Framework The USPSTF determined the scope and key questions for this review by using established methods (7, 8). A standard protocol was developed and publicly posted on the USPSTF Web site. A technical report further describes the methods and includes search strategies and additional information (4). Investigators created an analytic framework outlining the key questions, patient populations, interventions, and outcomes reviewed (Appendix Figure 1). Key questions include the harms of routine breast cancer screening and how they differ by age, risk factor, screening interval, and screening modality (mammography [film, digital, or tomosynthesis], magnetic resonance imaging [MRI], and ultrasonography). Harms include false-positive and false-negative mammography results, overdiagnosis, anxiety and other psychological responses, pain during procedures, and radiation exposure. Overdiagnosis refers to women receiving a diagnosis of ductal carcinoma in situ (DCIS) or invasive breast cancer when they have abnormal lesions that are unlikely to become clinically evident during their lifetime in the absence of screening. Overdiagnosed women may be harmed by unnecessary procedures and treatments as well as by the burden of receiving a cancer diagnosis. Appendix Figure 1. Analytic framework and key questions. KQ = key question. * Excludes women with preexisting breast cancer; clinically significant BRCA1 or BRCA2 mutations, Li-Fraumeni syndrome, Cowden syndrome, hereditary diffuse gastric cancer, or other familial breast cancer syndrome; high-risk lesions (ductal carcinoma in situ, lobular carcinoma in situ, atypical ductal hyperplasia, or atypical lobular hyperplasia); or previous large doses of chest radiation (20 Gy) before age 30 y. False-positive and false-negative mammography results, biopsy recommendations due to false-positive mammography results, overdiagnosis and resulting overtreatment, anxiety, pain, and radiation exposure. Family history; breast density; race/ethnicity; menopausal status; current use of menopausal hormone therapy or oral contraceptives; prior benign breast biopsy; and, for women aged >50 y, body mass index. Mammography (film, digital, or tomosynthesis), magnetic resonance imaging, ultrasonography, and clinical breast examination (alone or in combination). The target population for the USPSTF recommendation includes women aged 40 years or older and excludes women with known physical signs or symptoms of breast abnormalities and those at high risk for breast cancer whose surveillance and management are beyond the scope of the USPSTF recommendations for preventive services (preexisting breast cancer or high-risk breast lesions, hereditary genetic syndromes associated with breast cancer, and previous large doses of chest radiation before age 30 years). Risk factors considered in this review are common among women who are not at high risk for breast cancer (9) (described in Appendix Figure 1). Data Sources and Searches A research librarian conducted electronic searches of the Cochrane Central Register of Controlled Trials, the Cochrane Database of Systematic Reviews, and Ovid MEDLINE through December 2014 for relevant studies and systematic reviews. Searches were supplemented by references identified from additional sources, including reference lists and experts. Studies of harms included in the previous systematic review for the USPSTF (2, 3) were also included. Study Selection Two investigators independently evaluated each study to determine eligibility based on prespecified inclusion criteria. Discrepancies were resolved through consensus. We included recently published systematic reviews; randomized, controlled trials (RCTs); and observational studies of prespecified harms. When available, studies providing outcomes specific to age, risk factors, screening intervals, and screening modalities were preferred over studies providing general outcomes. Studies that were most clinically relevant to practice in the United States were selected; relevance was determined by practice setting, population, date of publication, and use of technologies and therapies in current practice. Studies meeting criteria for high quality and with designs ranked higher in the study designbased hierarchy of evidence were emphasized because they are less susceptible to bias (for example, RCTs were chosen over observational studies). Data Extraction and Quality Assessment Details of the study design, patient population, setting, screening method, interventions, analysis, follow-up, and results were abstracted by one investigator and confirmed by another. Two investigators independently applied criteria developed by the USPSTF (7, 8) to rate the quality of each RCT, cohort study, casecontrol study, and systematic review as good, fair, or poor; criteria to rate studies with other designs included in this review are not available. Discrepancies were resolved through consensus. Data Synthesis Studies meeting inclusion criteria were qualitatively synthesized. Most studies in this review had designs for which quality rating criteria are not available, which limited data synthesis. When possible, we assessed the aggregate internal validity (quality) of the body of evidence for each key question (good, fair, or poor) by using methods developed by the USPSTF based on the number, quality, and size of studies; consistency of results between studies; and directness of evidence (7, 8). Role of the Funding Source This research was funded by the Agency for Healthcare Research and Quality (AHRQ) under a contract to support the work of the USPSTF. The investigators worked with USPSTF members and AHRQ staff to develop and refine the scope, analytic frameworks, and key questions; resolve issues during the project; and finalize the report. The AHRQ had no role in study selection, quality assessment, synthesis, or development of conclusions. The AHRQ provided project oversight; reviewed the draft report; and distributed the draft for peer review, including to representatives of professional societies and federal agencies. The AHRQ performed a final review of the manuscript to ensure that the analysis met methodological standards. The investigators are solely responsible for the content and the decision to submit the manuscript for publication. Results Of the 12004 abstracts identified by searches and other sources, 59 studies met inclusion criteria for key questions in this report, including 10 systematic reviews of 134 studies and 49 additional studies (Appendix Figure 2). Appendix Figure 2. Summary of evidence search and selection. RCT = randomized, controlled trial. * Cochrane Central Register of Controlled Trials and Cochrane Database of Systematic Reviews. False-Positive Mammography Results Two new observational studies estimated the cumulative probability of false-positive results after 10 years of screening with film and digital mammography, based on data from the Breast Cancer Surveillance Consortium, a large population-based database in the United States (Appendix Table 1) (10, 11). When screening began at age 40 years, the cumulative probability of receiving at least 1 false-positive mammography result after 10 years was 61% (95% CI, 59% to 63%) with annual screening and 42% (CI, 41% to 43%) with biennial screening (10). Estimates were similar when screening began at age 50 years. The cumulative probability of receiving a biopsy recommendation due to a false-positive mammography result after 10 years of screening was 7% (CI, 6% to 8%) with annual screening versus 5% (CI, 4% to 5%) with biennial screening for women who initiated screening at age 40 years and 9% (CI, 7% to 12%) with annual screening versus 6% (CI, 6% to 7%) with biennial screening for those who began at age 50 years. Appendix Table 1. U.S. Studies of Cumulative False-Positive Mammography and Biopsy Results In a separate analysis, rates of false-positive mammography results were highest among women receiving annual mammography who had extremely dense breasts and either were aged 40 to 49 years (65.5%) or used combination hormone therapy (65.8%) (11). The highest rates of biopsy due to false-positive mammography results were related to similar characteristics and ranged from 12% to 1

I n 2009, the U.S. Preventive Services Task Force (USPSTF) recommended biennial mammography screening for women aged 50 to 74 years (1) on the basis of evidence of benefits and harms (2,3). The USPSTF concluded that screening decisions for women aged 40 to 49 years should be based on individual considerations and that evidence was insufficient to assess benefits and harms for those aged 75 years or older (1).
Although there is general consensus that mammography screening is beneficial for many women, benefits must be weighed against potential harms to determine the net effect of screening on individual women. Determining the balance between benefits and harms is complicated by several important considerations that are unresolved, including defining and quantifying potential harms; the optimal ages at which to begin and end routine screening; the optimal screening intervals; appropriate use of various imaging modalities, including supplemental technologies; values and preferences of women in regards to screening; and how all of these considerations vary depending on a woman's risk for breast cancer.
This systematic review updates evidence for the USPSTF on the harms of breast cancer screening, including false-positive mammography results, overdiagnosis, anxiety, pain during procedures, and radiation exposure, and how these adverse effects vary by age, risk factor, screening interval, and screening modality. Systematic reviews of the effectiveness of screening (4), performance characteristics of screening methods (5), and the accuracy of breast density determination and use of supplemental screening technologies (6) are provided in additional reports.
harms of routine breast cancer screening and how they differ by age, risk factor, screening interval, and screening modality (mammography [film, digital, or tomosynthesis], magnetic resonance imaging [MRI], and ultrasonography). Harms include false-positive and false-negative mammography results, overdiagnosis, anxiety and other psychological responses, pain during procedures, and radiation exposure. Overdiagnosis refers to women receiving a diagnosis of ductal carcinoma in situ (DCIS) or invasive breast cancer when they have abnormal lesions that are unlikely to become clinically evident during their lifetime in the absence of screening. Overdiagnosed women may be harmed by unnecessary procedures and treatments as well as by the burden of receiving a cancer diagnosis.
The target population for the USPSTF recommendation includes women aged 40 years or older and excludes women with known physical signs or symptoms of breast abnormalities and those at high risk for breast cancer whose surveillance and management are beyond the scope of the USPSTF recommendations for preventive services (preexisting breast cancer or highrisk breast lesions, hereditary genetic syndromes associated with breast cancer, and previous large doses of chest radiation before age 30 years). Risk factors considered in this review are common among women who are not at high risk for breast cancer (9) (described in Appendix Figure 1).

Data Sources and Searches
A research librarian conducted electronic searches of the Cochrane Central Register of Controlled Trials, the Cochrane Database of Systematic Reviews, and Ovid MEDLINE through December 2014 for relevant studies and systematic reviews. Searches were supplemented by references identified from additional sources, including reference lists and experts. Studies of harms included in the previous systematic review for the USPSTF (2, 3) were also included.

Study Selection
Two investigators independently evaluated each study to determine eligibility based on prespecified inclusion criteria. Discrepancies were resolved through consensus.
We included recently published systematic reviews; randomized, controlled trials (RCTs); and observational studies of prespecified harms. When available, studies providing outcomes specific to age, risk factors, screening intervals, and screening modalities were preferred over studies providing general outcomes. Studies that were most clinically relevant to practice in the United States were selected; relevance was determined by practice setting, population, date of publication, and use of technologies and therapies in current practice. Studies meeting criteria for high quality and with designs ranked higher in the study design-based hierarchy of evidence were emphasized because they are less susceptible to bias (for example, RCTs were chosen over observational studies).

Data Extraction and Quality Assessment
Details of the study design, patient population, setting, screening method, interventions, analysis, followup, and results were abstracted by one investigator and confirmed by another. Two investigators independently applied criteria developed by the USPSTF (7,8) to rate the quality of each RCT, cohort study, case-control study, and systematic review as good, fair, or poor; criteria to rate studies with other designs included in this review are not available. Discrepancies were resolved through consensus.

Data Synthesis
Studies meeting inclusion criteria were qualitatively synthesized. Most studies in this review had designs for which quality rating criteria are not available, which limited data synthesis. When possible, we assessed the aggregate internal validity (quality) of the body of evidence for each key question (good, fair, or poor) by using methods developed by the USPSTF based on the number, quality, and size of studies; consistency of results between studies; and directness of evidence (7,8).

Role of the Funding Source
This research was funded by the Agency for Healthcare Research and Quality (AHRQ) under a contract to support the work of the USPSTF. The investigators worked with USPSTF members and AHRQ staff to develop and refine the scope, analytic frameworks, and key questions; resolve issues during the project; and finalize the report. The AHRQ had no role in study selection, quality assessment, synthesis, or development of conclusions. The AHRQ provided project oversight; reviewed the draft report; and distributed the draft for peer review, including to representatives of professional societies and federal agencies. The AHRQ performed a final review of the manuscript to ensure that the analysis met methodological standards. The investigators are solely responsible for the content and the decision to submit the manuscript for publication.

RESULTS
Of the 12 004 abstracts identified by searches and other sources, 59 studies met inclusion criteria for key questions in this report, including 10 systematic reviews of 134 studies and 49 additional studies (Appendix Figure 2, available at www.annals.org).

False-Positive Mammography Results
Two new observational studies estimated the cumulative probability of false-positive results after 10 years of screening with film and digital mammography, based on data from the Breast Cancer Surveillance Consortium, a large population-based database in the United States (Appendix Table 1, available at www .annals.org) (10,11). When screening began at age 40 years, the cumulative probability of receiving at least 1 false-positive mammography result after 10 years was 61% (95% CI, 59% to 63%) with annual screening and 42% (CI, 41% to 43%) with biennial screening (10). Estimates were similar when screening began at age 50 years. The cumulative probability of receiving a biopsy recommendation due to a false-positive mammography result after 10 years of screening was 7% (CI, 6% to 8%) with annual screening versus 5% (CI, 4% to 5%) with biennial screening for women who initiated screening at age 40 years and 9% (CI, 7% to 12%) with annual screening versus 6% (CI, 6% to 7%) with biennial screening for those who began at age 50 years.
In a separate analysis, rates of false-positive mammography results were highest among women receiving annual mammography who had extremely dense breasts and either were aged 40 to 49 years (65.5%) or used combination hormone therapy (65.8%) (11). The highest rates of biopsy due to false-positive mammography results were related to similar characteristics and ranged from 12% to 14%. Rates of false-positive mammography results were lower among women aged 50 to 74 years who were receiving biennial or triennial mammography and had breasts with scattered fibroglandular densities (39.7% and 21.9%, respectively) or almost entirely fat breast density (17.4% and 12.1%, respectively), regardless of estrogen use.

Overdiagnosis
A meta-analysis of 3 RCTs (13, 14), a systematic review of 13 observational studies (15), and 18 new individual studies (16 -33) of overdiagnosis were identified for this update (4) (Appendix Table 2, available at www .annals.org). Estimates were primarily based on screening trials, screening programs and registries, or modeled data. Studies differed by patient populations; screening and follow-up times; screening policies, uptake, and intensity; and underlying cancer incidence trends. In addition, at least 7 different measures of overdiagnosis were reported (19). Estimates differed in their numerators and denominators, whether they included both invasive cancer and DCIS, their assumptions about lead time and progression of invasive cancer and DCIS, and whether they reported relative or absolute changes.
Various methods were used to estimate overdiagnosis. The most common methods determined the difference in cancer incidence in the presence and absence of screening (observed excess incidence approach) or made inferences about the lead time or natural history of breast cancer and estimated the corresponding frequency of overdiagnosis (lead-time approach) (35). How differences in study characteristics, measures, and methods affect estimates of overdiagnosis has been well-described (13, 14, 19, 35-37), yet there is no consensus about the appropriate approach (14) and there are no quality rating criteria to evaluate studies.

Estimates From RCTs
Data from 3 RCTs that did not screen control participants at the end of the trials were considered to be the least biased estimates of overdiagnosis in a comprehensive review (13, 14). The Malmö I trial and the Canadian National Breast Screening Study (CNBSS-1 and CNBSS-2) provided estimates from randomized comparison groups with follow-up that extended sufficiently beyond the screening period to differentiate earlier diagnosis from overdiagnosis (13). However, their approaches differed: The Malmö I trial included all breast cancer cases, and the Canadian trials included only those detected by screening.
Results of the Malmö I trial (34) and the 2 Canadian trials (38, 39) were used to compare the excess incidence of breast cancer (both invasive cancer and DCIS) in the screening population with the incidence in the absence of screening. Overdiagnosis was estimated at 10.7% (CI, 9.3% to 12.2%) (13, 14) when only cases identified during the screening period were included and 19.0% (CI, 15.2% to 22.7%) when cases identified throughout screening and follow-up were included. Estimates for women aged 40 to 49 years in CNBSS-1 (12.4% for shorter accrual and 22.7% for longer accrual) were higher than for those aged 50 to 59 years in CNBSS-2 (9.7% and 16.0%, respectively) and those aged 55 to 69 years in the Malmö I trial (10.5% and 18.7%, respectively). Recently published long-term follow-up of the 2 Canadian trials (15 years after enrollment) indicated a 22% overdiagnosis rate for invasive cancer for the combined age groups (31).

Estimates From Observational Studies
Unadjusted estimates from 13 observational studies included in a systematic review indicated overdiagnosis rates ranging from 0% to 54%, and 6 studies that adjusted for breast cancer risk and lead time indicated rates ranging from 1% to 10% (15). Estimates from other studies fall within this overall broad range.

Anxiety, Distress, and Other Psychological Responses
Four systematic reviews of 70 unique studies (40 -43) ( Table 1) and 10 additional observational studies (44 -53) ( Table 2) published after the systematic reviews described adverse psychological effects of screening.
Although several studies met criteria for fair or good quality, most were limited by enrollment of small numbers of narrowly selected participants, use of various self-reported measures, differential attrition or response rates, and low clinical applicability. No studies provided results by age, risk factor, screening interval, or screening modality.
Results of systematic reviews indicated that women who received clear communication of their negative mammography results had minimal anxiety, whereas
One study reported an increase in reattendance when women were given letters tailored to their last screening result (risk ratio, 1.10 [CI, 1.00 to 1.21]) (72).
Five new observational studies compared psychological outcomes in women receiving false-positive results versus those receiving normal results (44, 46 -48, 50) and reported findings similar to those of the reviews. Women with false-positive results had more breast cancer-specific worry (49% vs. 10%; P < 0.0001), more worries that affected mood or daily activities (31% vs. 2%; P < 0.0001) (48), and lower mental functioning (mean mental functioning score on the Short Form-36 at 6 months, 80.6 vs. 85.0; P = 0.03) and vitality (mean vitality score on the Short Form-36 at 6 months, 70.3 vs. 77.0; P = 0.02) (50). They also had increased measures  of depression (mean score on the depression subscale of the Hospital Anxiety and Depression Scale at 6 months, 3.2 vs. 2.4; P = 0.045); however, scores were below clinical thresholds for depression (50). An analysis of racial subgroups in a large study indicated increased depression scores among nonwhite women with false-positive results (odds ratio, 3.23 [CI, 1.32 to 7.91]) (44). Three studies found lower reattendance rates for women with false-positive results (51, 52) or biopsies (51,53), but reattendance sometimes varied by specific circumstances, such as age or type of biopsy (51).

Pain During Procedures
Two systematic reviews included 39 unique studies of pain associated with screening procedures (73,74), and a separate systematic review included 7 trials of interventions to reduce pain (75) (Appendix Table 3, available at www.annals.org). Results indicated that many women had pain (range, 1% to 77%) but few considered it a deterrent to future screening (73). In these studies, pain was associated with stage of the menstrual cycle, anxiety, and the anticipation of pain.
In a review of studies of pain or discomfort after screening mammography and their effect on screening reattendance (74), actual nonreattendance due to concerns about pain ranged from 11% to 46% (5 studies) and intended future nonreattendance ranged from 3% to 18% (2 studies). Fifteen studies that did not directly ask about reasons for nonreattendance found no differences in actual reattendance between women who had pain and those who did not (risk ratio, 1.38 [CI, 0.94 to 2.02]) (5 studies) (74). However, nonreattenders had significantly higher pain scores than reattenders in 2 of 3 studies (76 -78). Two studies reported lower intent to reattend among women with pain, whereas 3 others reported no differences in intended reattendance and pain (79 -83). A systematic review of trials of interventions to reduce pain associated with mammography screening (75) found that providing verbal or written information to women reduced discomfort in 2 studies (84, 85) but not in a third (86). Studies of different breast compression strategies (87,88) or premedication with acetaminophen (89) indicated no differences in discomfort, whereas use of a breast cushion reduced pain (90).

Radiation Exposure
No studies directly measured the association between radiation exposure from mammography screening and the incidence of breast cancer and death. Two-view digital mammography and screen-film mammography involve average mean glandular radiation doses of 3.7 and 4.7 mGy, respectively, and are considered to provide low-dose, low-energy radiation exposure.
Two modeling studies provided estimates of radiation exposure, breast cancer incidence, and death (91, 92) (Appendix Table 4, available at www.annals.org). A model predicting the number of breast cancer cases attributable to the radiation dose of a single typical digital mammogram estimated that the number of deaths due to radiation-induced cancer ranged from 2 per 100 000 in women aged 50 to 59 years screened biennially to 11 per 100 000 in those aged 40 to 59 years screened annually (92).

Differences Between Screening Modalities
Six observational studies compared false-positive recall rates with screening using mammography and tomosynthesis (93-97) or clinical breast examination (98) versus mammography alone (Appendix Table 5, available at www.annals.org). No studies evaluated MRI screening in women who were not at high risk for breast cancer.
Women receiving mammography and clinical breast examination had more recalls than those receiving mammography alone in a study from Canada (8.7% vs. 6.5%; 55 additional recalls per 10 000 women) (98).

DISCUSSION
A summary of the evidence is provided in Table 3. Two large observational studies of women screened in the Breast Cancer Surveillance Consortium provided good-quality evidence about cumulative rates of falsepositive mammography results and biopsies over 10 years. In these studies, rates were higher with annual than biennial screening (mammography, 61% vs. 42%; biopsy, 7% vs. 5%) and for women with heterogeneously or extremely dense breasts, those aged 40 to 49 years, and those using combination hormone therapy. These results are consistent with those of an earlier study indicating cumulative 10-year rates of falsepositive mammography results of 49% overall and 56% for women aged 40 to 49 years, with an overall biopsy rate of 19% (12). The results of these highly clinically applicable studies can be used to inform women of the likelihood of false-positive results and additional procedures with mammography screening in the United States, particularly for women with characteristics associated with the highest rates of false-positive results.
Despite much research, the evidence for determining overdiagnosis is poor. There is no consensus definition of overdiagnosis, and there are no criteria on which to base critical appraisal of studies. Studies are highly heterogeneous, and estimates vary depending on the analytic approach. Possibly the least biased estimates were derived from 3 RCTs that indicated rates of 11% to 22%. Unadjusted estimates from 13 observational studies ranged from 0% to 54%, and 6 studies that adjusted for breast cancer risk and lead time found rates ranging from 1% to 10%. Until methodological standards for estimating overdiagnosis are more clearly defined, the correct estimate is uncertain.
Although overdiagnosis is an important outcome of screening, it is difficult to evaluate in individual women because it is based on knowing whether a specific lesion will progress and what its effect will be on a woman's health. Women who are overdiagnosed can be harmed by unnecessary procedures and treatments and by the burden of receiving a cancer diagnosis. The introduction of technology capable of detecting even smaller suspicious lesions may also lead to increased overdiagnosis. Understanding the concept of overdiagnosis is important to appropriately inform women about the benefits and harms of screening despite current limitations in determining its effect on individual women.
The effect of screening on anxiety and pain is supported by fair-level evidence that includes a large number of predominantly descriptive observational studies. In general, women with false-positive results have more anxiety and distress than those with normal results. Anxiety lessens over time for most women but persists for others, and some women with false-positive results do not attend subsequent screenings. Although many women have pain during mammography, the proportion of those who do not attend subsequent screenings varies. Studies indicate that the experiences of falsepositive results and pain during mammography differ widely among women but are important for many of them. Additional efforts to reduce false-positive results and improve how they are communicated and to recognize and reduce pain during procedures could improve the balance of benefits and harms of screening for many women. The harms of radiation exposure from mammography screening are based on only 2 modeling studies. The number of deaths due to radiation-induced cancer from screening with digital mammography was estimated to be 2 to 11 per 100 000 women, depending on age and screening intervals. As imaging technologies change, this estimate could improve or worsen depending on the uptake of supplemental imaging with tomosynthesis as well as additional imaging for falsepositive results. Reducing radiation exposure through more effective imaging is an important area of future research.
Five observational studies described false-positive results with the use of tomosynthesis. This evidence is limited by the lack of randomized trials, uncertainty about the comparability of comparison groups, and differences in outcome measures. A U.S. study comparing tomosynthesis and mammography versus mammography alone reported a significant reduction of 16 recalls but an increase of 1.3 biopsies per 1000 women. Available studies of screening with MRI or ultrasonography focus on high-risk women and are outside the scope of this systematic review. No randomized trials of the efficacy of the different imaging technologies for breast cancer screening have been published, and evidence on their benefits and harms for screening recommendations is lacking.
Limitations of this review include the use of Englishlanguage articles only, which could have resulted in language bias, although we did not identify non-English-language studies that otherwise met inclusion criteria in our searches. We included only studies that are applicable to current practice in the United States to improve clinical relevance for the USPSTF. The number, quality, and applicability of studies varied widely, and most studies were observational, with designs for which quality rating criteria are not available.
In conclusion, false-positive results are common and lead to additional imaging and biopsies, particularly with annual screening and among younger women and those with dense breasts. Although overdiagnosis, anxiety, pain, and radiation exposure may cause harm, their effects on individual women are difficult to estimate and vary widely.