Although malpractice lawsuits are frequently related to a delayed breast cancer diagnosis in symptomatic patients, information on claims at European screening mammography programs is lacking. We determined the type and frequency of malpractice claims at a Dutch breast cancer screening region. We included all 85,274 women (351,009 screens) who underwent biennial screening mammography at a southern breast screening region in The Netherlands between 1997 and 2009. Two screening radiologists reviewed the screening mammograms of all screen detected cancers and interval cancers and determined whether the cancer had been missed at the previous screen or at the latest screen, respectively. We analyzed all correspondence between the screening organization, clinicians and screened women, and collected complaints and claims until September 2011. At review, 20.9% (308/1,475) of screen detected cancers and 24.3% (163/670) of interval cancers were considered to be missed at a previous screen. A total of 19 women (of which 2, 6 and 11 women had been screened between 1997 and 2001 (102,439 screens), 2001 and 2005 (114,740 screens) and 2005 and 2009 (133,830 screens), respectively) had contacted the screening organization for additional information about their screen detected cancer or interval cancer, but filed no claim. Three other women directly initiated an insurance claim for financial compensation of their interval cancer without previously having contacted the screening organization. We conclude that screening-related claims were rarely encountered, although many screen detected cancers and interval cancers had been missed at a previous screen. A small but increasing proportion of women sought additional information about their breast cancer from the screening organization.
Many countries have introduced screening mammography programs with the aim to reduce breast cancer mortality.1 Essential for reducing morbidity and mortality is the early detection of breast cancers, as a diagnostic delay lowers breast-conserving treatment options and worsens prognosis.2, 3 Unfortunately, a delayed diagnosis resulting from a missed cancer at screening is not rare. Certain cancers are just not visible at screening mammography, whereas others are misinterpreted or overlooked.4, 5
Interpretation of mammograms is one of the most difficult tasks in radiology and the sensitivity of screening mammography for breast cancer detection ranges from 70% to 80%.6, 7 Nevertheless, the public's expectations of the efficacy of screening mammography are high, and diagnostic errors can have major legal consequences for the screening radiologist. An Italian study observed, over a period of 12 years, a marked rise in malpractice claims related to diagnostic mammography in symptomatic women.8 In the United States, a delay in breast cancer diagnosis is nowadays the most prevalent and the second most expensive condition resulting in malpractice lawsuits.9, 10 The most common defendant in these lawsuits is the interpreting radiologist and as a consequence the number of radiologists willing to read mammograms in the United States is decreasing.11–13 Recall rates in most European screening programs vary from 3% to 6%,1 which is in line with the recommended recall rate in the European guidelines for quality assurance in breast cancer screening and diagnosis.14 In contrast, recall rates in the United States frequently exceed 10%,15 and these higher recall rates may be the result of the fear for lawsuits.16 The combination of manpower shortage and the high risk of lawsuits may thus have a detrimental effect on breast cancer screening and health care in general.
A majority of claims are related to mammographic misinterpretation and failure to communicate openly with patients about these errors.10, 11 Discussing errors with patients could enhance their satisfaction and may reduce the number of malpractice claims. However, fear of lawsuits makes disclosure of medical errors to patients difficult, and recent studies suggest that communicating openly about errors is the exception rather than the rule.17, 18
To our knowledge, no data have been published about malpractice claims involving screening mammography in Europe. In the current study, we determined the type and frequency of malpractice claims at a Dutch breast cancer screening region over a 12-year screening period.
Material and Methods
We included all 351,009 screening examinations of 85,274 women who underwent screening mammography at two specialized analogue screening units in a southern breast cancer screening region of The Netherlands between January 1, 1997 and January 1, 2009. All women in our study had given written informed consent to use their data for evaluation purposes before participating in the screening program. Approval by our local Institutional Review Board was not required for this study, according to the Dutch Central Committee on Research involving Human subjects (CCMO).
Screening procedure and referral
Details of our breast cancer screening program have been described previously19–21 In summary, five regional breast-screening organizations offer the Dutch screening program, providing biennial screening mammography to all Dutch women aged 50–75 years. All mammograms are performed by specialized screening mammography radiographers. In our screening region, the screening examinations are independently double read by a group of 12 certified screening radiologists. From 2003, in addition to radiologist double reading, the radiographers also actively participate in the assessment of the screening mammograms. Each of the screening radiologists evaluates at least 5,000 screening mammograms yearly. In case of subsequent screening, prior screening mammograms are always available for comparison. Women with normal or benign mammographic findings or with a nonspecific minimal sign are not referred. Minimal sign lesions are present in about 10% of screening mammograms and have a less than 1% chance of malignancy.22 If screening mammography shows a suspicious or malignant lesion, women are referred by their general practitioner to a surgical oncologist or breast clinic for further analysis of the mammographic abnormality.
For each referred woman, we collected data on radiology, pathology and surgical procedures at the hospitals where the mammographic screening abnormalities of referred women were evaluated, with 2-year follow-up. A majority of interval cancers (interval cancers are breast cancers that are diagnosed in women after a negative screening examination, defined as no recommendation for referral) were identified by linking the records of screened women to those of the regional cancer registry and radiotherapy laboratory. Some interval cancers were traced by the occasional reports on interval cancers provided by general practitioners or medical specialists to the screening centre, whereas other interval cancers were identified by inquiry about pathology specimens at the various regional pathology laboratories, some months after a hospital had requested the screening mammograms of a screening participant who had not been referred. Communications between the screening organization and screened women that were related to screening procedures, screening outcome or diagnostic procedures after referral were routinely recorded by the organization.
Review of screen-detected cancers, interval cancers and delayed cancer diagnosis after referral
Two experienced screening radiologists (LD and FJ) reviewed the screening mammograms of all screen detected cancers and interval cancers. For cancers detected at subsequent screening, they determined whether the cancer had been missed or whether it had shown a non-specific minimal sign at the previous screening mammogram. For interval cancers, the two radiologists determined whether the cancer had been missed or had been present as a minimal sign lesion at the latest screening mammogram. To determine the main reasons for diagnostic delay after referral, the two radiologists reviewed the diagnostic breast images of all women who had breast cancer pathologically confirmed more than three months following a positive screen.23, 24 Each reviewer classified the lesions according to the American College of Radiology BI-RADS.25 The radiologists were blinded to each other's review and discrepant assessments were followed by consensus reading. To determine whether the delay was due to a false negative pathology report, a pathologist reviewed the specimen of those women who had undergone more than one breast biopsy procedure needed for breast cancer confirmation.
Communication between the coordinating screening radiologist and screened women
Our screening organization routinely asked the coordinating screening radiologist to contact women with a request for additional information about her screen detected cancer or interval cancer. Contact between the radiologist and the woman was established first by telephone, and then, if desired, by personal contact. In all cases, the radiologist specifically informed the woman whether or not her screen-detected cancer had been visible at the previous screen, or in the case of interval cancer, whether the malignancy had been missed at the latest screen. The conclusions of these communications were routinely documented and were also recorded at a database that had been developed for this study. For the current study, we included all communications regarding the screening period 1997–2009, that had been recorded until September 1, 2011.
Overall screening results
A total of 351,009 analogue screens in 85,274 women were acquired between January 1, 1997 and December 31, 2008 (Fig. 1). Altogether, 4,450 screens (1.3%) required further evaluation because of a mammographic screening abnormality. Six of these women (0.2%) had either not been referred by their general practitioner or their type of diagnostic procedures was unknown. Breast cancer was diagnosed in 1,773 referred women, yielding an overall cancer detection rate of 5.1 per 1,000 screening examinations and a true positive referral rate of 39.8%. Within 2 years of follow-up, interval cancers had been diagnosed in 670 women who had been screened negative, resulting in a 72.6% (1,773/2,443) screening sensitivity for breast cancer detection.
In 1,586 (59.4%) of 2,671 referred women with benign follow-up, evaluation of the abnormality detected at screening mammography merely consisted of one or several radiologic examinations (i.e., additional mammographic views, breast ultrasonography and/or magnetic resonance mammography). Evaluation also included percutaneous biopsy in 830 (31.1%) women, invasive surgical biopsy in 133 (5.0%) women or a combination of percutaneous and surgical biopsy in 122 (4.5%) women (Fig. 1).
Of the 1,773 referred women with a diagnosis of breast cancer at follow-up, the mammographic abnormality had been detected at initial screening in 298 women and at subsequent screening in 1,475 women (Fig. 1). Review of the latter group showed that 308 (20.9%) cancers had been missed 2 years earlier, at the previous screening mammogram, and 322 (21.8%) had been visible as a minimal sign. The remaining 845 (57.3%) cancers detected at subsequent screening were either mammographically occult or not yet present at the previous screening examination.
A total of 670 interval cancers had been diagnosed among the screened population (Fig. 1). The reviewers reported that 163 (24.3%) of these cancers had been missed at the latest screening mammogram, 164 (24.5%) had been visible as a minimal sign and 343 (51.2%) were not visible at the latest screen.
Requests for additional information and malpractice claims
Between January 1997 and September 2011, a total of 19 screened women had contacted the screening organization for questions related to their cancer detected at screening (eight women) or interval cancer (11 women) (Fig. 2). Of these 19 women, two had been screened between 1997 and 2001 (102,439 screens, including 498 screen detected cancers and 169 interval cancers), six between 2001 and 2005 (114,740 screens, including 599 screen detected cancers and 244 interval cancers) and 11 between 2005 and 2009 (133,830 screens, including 676 screen detected cancers and 257 interval cancers). The resulting contact between the women and the coordinating radiologist was limited to conversation by telephone in five cases and comprised three screen-detected cancers (previous screen: two minimal sign lesions, one not visible) and two interval cancers (latest screen: one minimal lesion, one not visible). Additional face to face contact occurred in 14 women, of whom five had advanced cancer detected at subsequent screening (previous screen: one missed, three minimal sign lesions, one not visible) and nine had a diagnosis of advanced interval cancer (latest screen: two missed, five minimal sign lesions, two not visible). In these patients advanced cancer was defined as invasive cancer >20 mm and/or lymph node positive cancer. None of the 19 women started a malpractice lawsuit or insurance claim for financial compensation.
Apart from the 19 women mentioned above, another three women directly initiated an insurance claim for financial compensation of their interval cancer without previously having contacted the screening organization (Fig. 2). The time span between the latest screen and the start of the claim in these three cases was respectively 14, 25 and 40 months, respectively. One claim was rejected (latest screen: minimal sign lesion), whereas the verdicts of the other two claims (two interval cancers that had been missed at the latest screen) still have to be finalized.
There was one special case of a referred woman who initially demanded from the screening organization to be compensated for the €155 expenses made for additional clinical mammography and breast ultrasonography. She refrained from further steps after being informed by the coordinating radiologist that only the actual costs for screening mammography are covered by the nation-wide breast screening program.
Delayed cancer diagnosis after referral
Of the 1,773 referred women with breast cancer, 89 (5.0%) experienced a more than 3-months delay in cancer diagnosis (Fig. 2). An incorrect BI-RADS classification at clinical mammography, especially categorization of suspicious lesions (BI-RADS 4) or malignant lesion (BI-RADS 5) as probably benign lesions (BI-RADS 3), comprised 57.3% (51/89) of diagnostic delays. The pathologist did not encounter any incorrectly classified biopsy reports. However, false negative percutaneous biopsy results, i.e., retrieval of nonsuspicious cells from a malignant lesion, accounted for 20.2% (18/89) of diagnostic delays. The remaining 21 (23.6%) diagnostic delays were due to a variety of reasons, including false negative open (surgical) biopsy or noncompliance of a surgical oncologist with the pathologist's advise to excise a lesion with suspicious cytology or histology at percutaneous biopsy. There were no women-based reasons of delay (e.g., refusal to undergo additional evaluation after referral). The delay was 3–12 months, 13–24 months or more than 24 months in 58 (65.2%), 19 (21.3%) and 12 (13.5%) women, respectively. An advanced tumor stage at the time of definitive surgery was diagnosed in 27.0% (24/89) of women with a delayed diagnosis, compared to 19.3% (325/1,684) in women who experienced no diagnostic delay after referral. None of the 89 women with diagnostic delay filed a malpractice claim.
To our knowledge, no previous data have been published with respect to screening mammography claims in Europe. We found that, although more than one-fifth of screen-detected cancers and interval cancers had been missed at a previous screen, only three insurance claims for financial compensation had been filed following a diagnosis of interval cancer. Moreover, none of the women who experienced a delay in cancer diagnosis after referral filed a claim against any of the involved hospital physicians.
A delay in breast cancer diagnosis is the most common reason for medical malpractice claims in the United States and the radiologist is the most frequently litigated physician in these cases.9 Similar observations were made in Italy.8 Consequently, the number of qualified screening radiologists, especially in the United States, is decreasing and this may seriously impede access to breast cancer screening in the future. Besides a reduction in the number of breast radiologists and screening mammography radiologists, malpractice lawsuits may also result in a lower quality of mammographic interpretation. Lawsuits may cause uncertainty of the interpreting radiologist, which is associated with high recall rates at screening and lower positive predictive value rates.15, 16 Higher recall and false positive rates also increase anxiety among referred women and increase workup costs.26
Several authors have focused on the issue of increasing medical malpractice claims regarding mammography in the United States.10, 11 One explanation for the high number of claims seems to be the public's high expectations of mammography performance. The limitations of screening are often poorly understood, which is probably due to campaigns encouraging women to undergo screening mammography and the media, which tend to emphasize only the benefits of screening. Berlin suggested that improving public education about the shortcomings of mammography may reduce the number of lawsuits.10 Moreover, jury verdicts are influenced by the public's perception and a better public understanding of the limitations of screening mammography may not only reduce the number of lawsuits but may also lower the costs of medical malpractice claims by reducing the awards paid to patients following a jury verdict.
In our study, only 19 women had contacted the screening organization for additional information about their cancer detected at screening or interval cancer. This is a remarkably low number, especially when taking into account the fact that a large proportion of screen-detected cancers and interval cancers could have been diagnosed at a previous screening round and showed an unfavorable, advanced tumor stage at surgery. None of these women filed a claim, even if they had been informed by the coordinating screening radiologist that the cancer had been missed (three cases) or had shown a minimal sign abnormality (11 cases) at a previous screen. The open communication between the coordinating radiologist and these women may probably have prevented them to initiate a litigation procedure. A majority of the 19 women (11/19; 58%) had been screened between 2005 and 2009, and this finding suggests that the proportion of women who contact a screening organization after a diagnosis of breast cancer will probably increase in the future.
So far, only three women have filed a claim for financial compensation of their interval cancer and they did so without first contacting the coordinating screening radiologist. Apart from the open communication, several other factors may also partly explain the very low number of contacts between the screening organization and screening participants and sporadic claims in our study. With each invitation, women receive written information that breast cancer may be missed at screening mammography, and they should always seek medical attention in case of breast complaints following a negative screen. Although data about payouts for missed and delayed breast cancer diagnoses in The Netherlands are lacking, these payouts may probably be much lower when compared to the United States. Moreover, Dutch lawyers are not allowed to practice a no-win no-fee policy. Finally, women may have refrained from contacting the coordinating screening radiologist and starting malpractice claims after having received satisfactory information from their physicians in the hospitals to which they had been referred.
False positive referral negatively affects quality of life26 and almost 1 out of every 10 referred women in our study (9.5%, 255/2,671) with benign follow-up underwent invasive surgical biopsy. Surgical biopsy harbors the risk of infection and cosmetic drawbacks and may decrease the accuracy of future screening mammography.20 Nevertheless, only one referred woman complained about her false positive referral, and she wanted the screening organization to compensate her for the costs of clinical mammography and breast ultrasonography. The Dutch nation-wide screening mammography program provides free screening mammography, but it does not cover any costs related to additional diagnostic procedures. An addition of this information to the screening invitation letter probably prevents invited women to be confronted with unexpected financial expenses in case not all hospital related costs are covered by their medical insurance.
Although the screening organization is not responsible for the quality of clinical assessment of referred women, we also determined whether any of the referred women with a delay in breast cancer diagnosis filed a claim. About 5% of women in our screened population experienced a diagnostic delay of more than 3 months after their referral, and we previously found that this delay is most frequently due to erroneous interpretation or classification of breast lesions at clinical breast imaging.23, 24 To our surprise, none of the women filed a malpractice claim, not even when they were diagnosed with advanced cancer 2 years after referral.
In our screening program, several measures have been taken in the past to minimize the risk of delayed cancer diagnosis. First, our team of screening radiologists routinely reviews newly diagnosed interval cancers with the aim of reducing the proportion of interval cancers at future screening. Second, as part of the nation-wide quality assurance of screening mammography, the National Expert and Training Centre for Breast cancer screening (NETCB) evaluates our screening results every 3 years. This evaluation not only includes feedback on screening outcome parameters such as referral rate and cancer detection rate but also a comprehensive review of interval cancers and advanced breast cancers detected at subsequent screening. Finally, our radiographers have been encouraged to interpret screening mammograms. During quality assurance sessions, they bring mammographic abnormalities that may require additional work-up to the attention of a supervising breast radiologist.6
Our study has certain limitations. We realize that the eagerness of patients to start a lawsuit will be influenced by national legislation and the beforehand probability and height of insurance payouts. The Dutch nation-wide breast screening program differs in several aspects from other European programs and U.S. programs. For example, the referral rate in The Netherlands is very low when compared to a 3–6% referral rate in other European countries and a 10% referral rate in the United States.1, 15 In The Netherlands, all mammograms are routinely double read by two screening radiologists, which may not be standard practice in other screening programs.7, 27 Also, many European programs, including the Dutch one, offer biennial screening for women aged 50–75 years. In contrast, women are screened every 3 years in the United Kingdom, and U.S. programs usually offer annual screening from the age of 45. Screening interval length may influence screening outcome parameters such as screening sensitivity and tumor size of screen-detected cancers and interval cancers.28, 29 Furthermore, it was not possible to compare the low number of claims in our screening region with a nation-wide number of claims following breast cancer screening. Details of claims are not routinely registered by screening organizations in The Netherlands. We found, however, one published claim, regarding a woman with interval cancer.30 Her claim was rejected after concluding that both the screening radiologists had made the right decision by not referring the woman for her minimal sign lesion. Despite the absence of a national registry of claims, we have no reason to believe that the frequency of claims in our region differs from those in other parts of The Netherlands.
Two interval cancer claims in our study are still under discussion. In 1991, the screening organization of the Southern Netherlands (Bevolkingonderzoek Zuid, BOZ) started with the implementation of biennial screening mammography. More than 2,500,000 screens have been performed since then and it is estimated that about 220,000 women will attend screening mammography in the Southern Netherlands in 2011. So far, the screening organization had to compensate for a claim only once; € 5,000 was paid in advance for temporary compensation of an interval cancer diagnosed in a woman who had been screened at another unit than ours. The general practitioner and clinical radiologist were also held responsible for the delay in cancer diagnosis in this particular case and the final decision of the claim has not yet been set. The insurance policy of the screening organization covers a maximum payout of €5,000,000 per claim and a maximum annual payout of €10,000,000. Acceptance of our two claims will probably have no impact on the communication and referral strategy as the insurance policy sufficiently covers payouts for missed and delayed cancer diagnosis and screening radiologists are not personally liable for any financial compensation of claims. Moreover, insufficient communication is a leading factor in a majority of radiologic lawsuits and an open and clear communication with screened women may thus potentially prevent a litigation procedure.31
Finally, only mammograms obtained with analogue units were included in the current study, whereas most mammography units are now digital. Digitization of the breast cancer screening program in the Netherlands has recently been completed and the proportion of cancers missed at a previous screen may hopefully decrease in the future as digital screening tends to yield higher cancer detection rates when compared to analogue screening.32, 33
In summary, we conclude that women rarely filed a malpractice claim, although a substantial proportion of screen-detected cancers and interval cancers had been missed at a previous screen and 5% of women experienced a delay in breast cancer diagnosis after referral. An open communication between the screening organization and women who seek additional information after having been diagnosed with breast cancer may help to refrain women from starting a litigation procedure.