ArticlePDF Available


November 2009’s announcement of the USPSTF’s recommendations for screening for breast cancer raised a firestorm of objections. Chief among them were that the panel had insufficiently valued patients’ lives or allowed cost considerations to influence recommendations. The publicity about the recommendations, however, often either simplified the actual content of the recommendations or bypassed significant methodological issues, which a philosophical examination of both the science behind screening recommendations and their import reveals. In this article, I discuss two of the leading ethical considerations at issue in screening recommendations: respect for patient autonomy and beneficence and then turn to the most significant methodological issues raised by cancer screening: the potential biases that may infect a trial of screening effectiveness, the problem of base rates in communicating risk, and the trade-offs involved in a judgment of screening effectiveness. These issues reach more broadly, into the use of “evidence-based” medicine generally, and have important implications for informed consent.
Journal of Medicine and Philosophy, 0: 1–14, 2012
© The Author 2012. Published by Oxford University Press, on behalf of the Journal of Medicine and Philosophy Inc.
All rights reserved. For permissions, please e-mail:
Ethical Issues in Cancer Screening and
The University of Utah, Salt Lake City, Utah, USA
*Address correspondence to: Anya Plutynski, PhD, The University of Utah, Salt Lake City,
UT, USA. E-mail:
November 2009’s announcement of the USPSTF’s recommendations
for screening for breast cancer raised a firestorm of objections.
Chief among them were that the panel had insufficiently valued
patients’ lives or allowed cost considerations to influence recommen-
dations. The publicity about the recommendations, however, often
either simplified the actual content of the recommendations or
bypassed significant methodological issues, which a philosophical
examination of both the science behind screening recommendations
and their import reveals. In this article, I discuss two of the leading
ethical considerations at issue in screening recommendations:
respect for patient autonomy and beneficence and then turn to the
most significant methodological issues raised by cancer screening:
the potential biases that may infect a trial of screening effective-
ness, the problem of base rates in communicating risk, and the
trade-offs involved in a judgment of screening effectiveness. These
issues reach more broadly, into the use of “evidence-based” medicine
generally, and have important implications for informed consent.
Keywords: base rate fallacy, cancer screening, risk communication,
trade-off neglect, USPSTF
Cancer is the second leading cause of death in the United States. Under-
standably, with so many peoples’ lives affected by this devastating disease,
early detection has become the mantra of many cancer activists. Screening
for disease in asymptomatic individuals has an intuitive appeal. If all cancers
have a linear progression toward morbidity and mortality, then halting the
progress of the disease early seems the best strategy. Many are hopeful that
Journal of Medicine and Philosophy Advance Access published May 7, 2012
at University of Utah on May 8, 2012 from
Anya Plutynski
more and better screening modalities will be provided to a wider public. If not
all cancers have linear progression (as seems clear, cf. Croswell, Ransohoff,
and Kramer, 2010), more, however, may not always be better. The recent
upset over the USPSTF’s recommendations for screening for breast cancer
brought the intuitive thinking behind early detection into question and raised
a host of ethical issues, besides. The object of this essay will be to examine
the intuitive thinking behind support for frequent screening, consider the
nature of the evidence for the USPSTF’s recommendations, and whether the
recommendations can be made consistently with respect for autonomy and
beneficence. I will argue that the recommendations can and should be
followed consistently with respect for autonomy and beneficence but argue
that safeguards need to be put in place to ensure this.
The USPSTF recommended that women under 50 consult with their clinicians
about whether to have annual screening (US Preventive Services Task Force,
2009). A meta-analysis of multiple clinical trials appeared to show that the
benefit for most women under 50 may be relatively small, and the cost—in
terms of false positives, unnecessary biopsies, and overdiagnosis (unnecessarily
treating cancers that never would have caused illness during that person’s life-
time), is relatively high. For women over 50, they found that biannual screening
may be as effective as annual screening for reducing age-adjusted mortality
(Mandelblatt et al., 2009; Nelson et al., 2009). As might have been predicted by
more savvy publicists, the response was not positive. Many breast cancer activ-
ists, and the American Radiological Association, as well as the American Cancer
Association dismissed the findings. Some argued that the USPSTF was condemn-
ing women to early death. Moving past the outrage, we need to look closely at
both the salient scientific and ethical issues at stake in this heated exchange.
The USPSTF recommended individual women in their forties decide whether
to screen in consultation with clinicians. Why does this create such outrage?
Perhaps some worry that insurance companies will not cover screening
if screening is not standard care for some age group, or that even those
patients at elevated risk (perhaps due to family history) will not seek screening
based on recommendations for the population at large. However legitimate
these concerns, the matter of whether or not most women under 50 should get
screened should depend upon whether the risk to those screened is out-
weighed by the benefit. What is the evidence for the benefits for women in this
age group? We should not simply assume that more screening is better, and that
all screening modalities for all cancers are risk free for all patients. As the
authors of a recent article put it, we should be sure to “protect ourselves from
our intuitions” (Croswell, Ransohoff, and Kramer, 2010). Not all cancers prog-
ress linearly, and some may never cause clinical symptoms—in other words,
preventative screening methods may not always be for the benefit of patients,
for some cancers and in some populations. Although many of us assume that
in the era of modern medicine, no one prescribes either tests or treatments that
do more harm than good, we should make sure to check the evidence.
at University of Utah on May 8, 2012 from
Ethical Issues in Cancer Screening and Prevention 3
Although it is surely correct that clinicians and patients (and insurance
companies) find it easier to follow general recommendations, when screen-
ing will only benefit a minority at greater risk, it seems that individual-based
decision making, rather than universal screening, is a better option. Indeed,
the USPSTF was clear that women should consult individually with physicians;
their recommendations were not “broadly utilitarian,” in the sense of simply
considering costs for insurance companies, but instead, emphasized individual
assessment and weighing of risks, with clinicians. However, individual-based
decision raises another suite of issues about risk communication and deci-
sion making under uncertainty. Some have argued that patients are simply
bad at making choices based on, for example, misunderstanding of the dif-
ferences between absolute versus relative risk (Schwartz and Meslin, 2008).
As Schwartz and Meslin explain, pamphlets encouraging mammography
tend to cite relative over absolute risk as the apparent “risk reduction” will
always look higher described as relative risk. Moreover, they support this
strategy. They explain:
Absolute risk reduction (ARR) is calculated by subtracting the mortality rate in the
screened group from that of the unscreened group. Thus, according to one study,
the ARR of performing mammography regularly for 20 years is 4.9 per 1,000
(or 0.49%), i.e., 14 per 1,000 (1.4%) minus 9.1 per 1,000 (0.91%) (Barrat et al., 2005).
In contrast, Relative Risk Reduction (RRR) is calculated by dividing the decrease in
morbidity or mortality by the baseline rate. Using the same model as above, mam-
mograms performed over 20 years would provide a RRR of 36%, i.e., 4.9 per 1,000
decrease in death from breast cancer divided by the baseline rate (in the unscreened
group) of 14 per 1,000. In this case as in others, the RRR (36%) is a much larger
number than the ARR (0.5%), and so it is not surprising that brochures that aim to
encourage patients to get regular mammograms generally present RRR if numbers
are given at all. (Schwartz and Meslin, 2008, 868)
As mentioned above, the difference apparent reduction of risk for relative
versus absolute risk reduction may be quite stark, particularly when screen-
ing for disease that is not very prevalent. Schwartz and Meslin endorse the
practice of describing relative risk reduction, as opposed to absolute risk
reduction, to patients, to encourage mammography. However, we should
be careful to not slip into paternalism, particularly in “gray areas” where
the risk reduction is small, and the costs of screening may be relatively high
(Quanstrum, 2009). Although a narrow utilitarian or expected utility perspective
might simply attach values to lives saved, and so use any strategy (including
representing risk deceptively, or discounting or hiding cost to patients due
to unnecessary biopsies or overdiagnosis), there are broader issues at stake.
Questions of risk and benefit need to be supplemented by some discussion
of the reasonable variability in values patients attach to different risks, the
norms of clinician-patient relationships, and what respect for autonomy and
informed consent requires. Questions about risk and benefit can be better
answered once we know how sensitive or specific are our screening tools,
at University of Utah on May 8, 2012 from
Anya Plutynski
how prevalent the disease, and thus what the risks are of false positives.
However, we also need to assess the values behind general versus individu-
ally tailored recommendations, and arguments for and against individual
consultation with clinicians, versus general recommendations that may benefit
some, but not most patients.
Even the best clinical trials often leave some gray areas where it is not
clear that benefit outweighs cost (Quanstrum and Hayward, 2009). Evidence-
based medicine is surely important, but it cannot always draw a bright clear
line between when universal screening is advisable and when it is not. In
this paper, I will argue, along with Malm (1999), that public health recom-
mendations should be such that the benefit on balance for the patient who
receives a recommendation is greater than the harm.1 By this, I mean that
that benefit for any particular patient is higher than risks of harm, taking into
account the probability that this harm will occur. However, of course, how
risks or benefits are assessed is not merely a matter for science to decide.
Although EBM may give us information that we can use to make judgments
about policy, in some cases, the evidence underdetermines the recommenda-
tions. In just such cases, to make public policy recommendations, absent con-
sideration of the patient’s (a) particular risk, and (b) weighting of preferences,
is unsound.
Further, we need to also consider the role and limitations of risk commu-
nication. When communicating with a patient about their preferences, we
should make clear when the recommendations are not ‘black and white’;
some women need to receive false positive tests, have unnecessary biopsies,
or be unnecessarily treated on balance to save lives. How often will women
be willing to take the risk of being in the first category to offset the chance
of landing in the latter? Similarly, how many men should receive diagnoses
of “indolent” prostate cancer that may never have presented clinically to
save one life? Arguably, individual patients need to decide for themselves
whether they will be willing to go through the risk of unnecessary post-
screening tests.
Unnecessary tests—test that, by and large, do not benefit the vast majority
of patients, and may harm them—should not be touted as a human right and
a necessity for health and survival. Screening does have downsides, as well
as positive outcomes (see, e.g., Gøtzsche and Nielsen, 2009). Patients should
know that they may be exposed to unnecessary radiation, that false positives
are more likely when a disease is rare, and that in many cases, screening for
early disease is neither always necessary nor beneficial. For instance, many
more prostate cancers are detected and treated today than 30 years ago due
to the promotion and use of PSA tests, and yet it is not clear that quality of
life for the vast majority of men treated for prostate cancer is thereby
improved (Esserman, Shieh, and Thompson, 2009). Since most prostate cancers
are slow growing, such that many men may never have clinical symptoms of
these cancers, and treatment can cause impotence and incontinence, it is not
at University of Utah on May 8, 2012 from
Ethical Issues in Cancer Screening and Prevention 5
clear that the vast increase in number of prostate cancers diagnosed benefits
most patients or extended lives.
However, those who administer and profit from these tests and treatments
do benefit. We cannot always assume that makers of mammography and other
technologies are motivated entirely by altruism. There must be oversight of
medical technologies and procedures to ensure that they help and do not
harm patients (Groopman, 2007; Quanstrum and Hayward, 2009). The
following is an overview and discussion of the general principles that should
be put in place for ethical practice of cancer screening and prevention and
a call for areas where we need to be more attentive to the variable biology
of different cancers, as well as the variable values of patients.
Respect for individual decision making and informed consent are central to
a commitment to patient autonomy and are, of course, required by law for
federal research (National Institutes of Health, 2005). However, respect for au-
tonomy is not simply a matter of giving patients choices; it requires that
they be informed about both risks and benefits of screening, and it requires
recognizing that patients may value risks and benefits differently. PSA tests
should be accompanied with some explanation, for example, of disease
prevalence, morbidity and mortality, test sensitivity and false positive rates,
and side effects of further screening and treatment. As mentioned above, for
many prostate cancers, “catching it early” may not be for the best, at least
in terms of quality of life. Many men die of heart disease, but die with pros-
tate cancer, having never experienced clinical symptoms. A better understand-
ing of the distinct pathology and genetics of more and less aggressive
forms of such cancers is important to assessing the value of such tests. Clini-
cians are understandably frustrated when, even when screening does
appear to have a proven benefit, some patients refuse screening because
they are afraid of medical procedures, do not like being put in the role of a
“patient,” or simply do not trust the evidence. In such cases, clinicians might
provide the patient with more information or better explanations of benefits
and risks of screening. However, they also must recognize that it is not un-
reasonable to value some choices over others; choices to not be screened
may not be due to ignorance but differences in values.
Informing patients about risk and benefit, as many have argued, is not a
straightforward matter. Even those with PhDs in statistics are subject to
heuristics and biases that make it difficult to understand or distinguish between
absolute and relative risk, for instance (Kahneman and Tversky, 1979; Weinstein,
1980; McKenna, 1993; Baron, 2000; Bowling and Ebrahim, 2001). Availability
heuristics and base rate fallacies are problems that plague researchers as well
as the average Joe. Clinicians may be as subject to bias as patients; a clinician
at University of Utah on May 8, 2012 from
Anya Plutynski
does not want to think of the possibility that the tests (or treatment) are
either not effective or may even harm their patient unnecessarily.
With this in mind, respect for autonomy requires that those who manufacture
and profit from diagnostic tests actively seek to document and acknowledge
risk of screening tests. Federal regulations should require them to provide
effective ways of communicating such risk to both clinicians and patients.
Ideally, third parties without an interest in misinformation should assess the
science behind such risk/benefit assessments. Most of us assume (correctly)
that an X-ray or even a CT or two is not a serious risk. However, it has become
apparent that such machines may malfunction due to operator errors as well
as improper maintenance. Moreover, a lifetime of tests can add up. Patients
must be made aware that multiple CT scans and X-rays, over time, are not
“risk free.” Risks should be communicated in a way that is user friendly
and transparent, and patients and clinicians should communicate about the
cumulative effects of such tests over a lifetime.
For example, a patient admitted for a CT scan should be provided a pamphlet
showing exactly how much an excess of a certain number of such scans
will, on average, increase risk for cancer. Expression of risk in the simplest
possible fashion—for example, 1 in 100 individuals who have more than
N CT scans are at elevated risk for cancer to extent x—would be most effec-
tive. Clinicians, unfortunately, are incentivized to offer more tests, and many
do not have accurate knowledge about the risks of such tests, either because
manufacturers play down the risks or because there is simply insufficient
data. Either this incentive structure for clinicians needs to change or inde-
pendent parties need to provide some form of evidence-based regulation of
unnecessary and risky tests.
Clinicians and their professional organizations have a vested interest in argu-
ing that their screening and treatment protocols are safe and effective. An
independent third party should watch over such claims and make sure
that the claimed benefits genuinely outweigh the risks. Sometimes this will
involve sponsoring more and better clinical trials. However, epidemiologists
may estimate when and whether a test is more or less effective, even with-
out a clinical trial, if they know how prevalent a disease is and how sensi-
tive and specific the available screening modality.
A general rule of thumb is that if a disease is not very prevalent, false posi-
tives will be relatively common. A simple example of how prevalence, for
instance, is relevant to the effectiveness of screening regimes follows (from
Buys, 2011). Suppose that you have a screening test that has very high speci-
ficity (the probability that a negative test means the disease is absent) and
high sensitivity (the probability that the test will be positive if the disease is
at University of Utah on May 8, 2012 from
Ethical Issues in Cancer Screening and Prevention 7
present). Let us say that the test is 95% specific (95% of positive tests are due
to disease) and 95% sensitive (95% of disease is identified). Suppose 10,000
individuals are tested. The performance of the screening test will vary
depending upon whether the disease is common (say, 5% prevalence) or very
uncommon (.05%).
Consider table 1: For the disease with high prevalence, the positive pre-
dictive value of the test (or the proportion of those who test positive who
have the condition) = ½. You can calculate this number yourself by dividing
the number of true positives by the number of true and false positives
(475/950). This means that half of those who screened positive were in fact
diseased. That may seem like a very high false-positive rate, but consider
table 2 for a much less prevalent disease.
Consider table 2: For the disease with lower prevalence, the positive
predictive value of the same test is 1/101. This means that only one out of
101 individuals who screened positive were in fact diseased (see table 2).
Thus, the effectiveness of the very same screening modality is significantly
less for less common diseases. Cancer, on average, is more prevalent in older
The lesson of this example is fairly simple: when a disease is not very
prevalent (as is the case for most cancers in most persons under 50), even
the most sensitive and specific screening test is not as effective, and may be
more harmful, than not being screened.
What are the costs of a false positive test? Perhaps a few days of worry,
followed up by a relatively noninvasive biopsy, are, some argue, not a serious
problem. However, sometimes, according to a recent report, unnecessary
open biopsies are done when needle aspirated biopsies would be equally
effective and safer (Gutwin et al., 2011). Moreover, for many cancers,
the only recourse after a positive diagnosis is a very serious surgery. For
instance, in the case of ovarian cancer, ovarectomies are not a minor proce-
dure and one with some serious side effects long term. Further, pathologists
are not immune to error. So, positive biopsies, that is, claims of malignancy,
can be in error. How often does this occur? According to some (Moss, 2005;
Table 1. Hypothetical disease no. 1—high prevalence
No. with positive test No. with negative test Totals
Disease 475 25 500
No disease 475 9,025 9,500
Table 2. Hypothetical disease no. 2—low prevalence
No. with positive test No. with negative test Totals
Disease 5 0 5
No disease 500 9,495 9,995
at University of Utah on May 8, 2012 from
Anya Plutynski
Paci et al., 2006; Welch and Black, 2010) pathologists may overdiagnose some
cancers as much as one third of the time. Pathologists’ judgments about the
“invasiveness” or “metastatic potential” of a tumor vary; one pathologist’s
atypical ductal hyperplasia, a pathological finding that increases the risk of
subsequent cancer but is not itself cancer, may be another pathologist’s ductal
carcinoma in situ, which may be treated with surgery and radiation therapy.
Although on the side of the clinician, it is better to be “safe than sorry,” it is
not always safer for the patient to undergo ever earlier screening and pos-
sibly unnecessary surgical procedures. The upshot is: patients should be
able to decide what risk they are willing to accept, and not all screening
options are risk free.
In light of these concerns, there needs to be more and better clinical trials of
screening modalities and other preventative measures. However, even the
best clinical trial is not free of bias. A variety of concerns have been raised
about evidence-based medicine, from concerns about meta-analyses and
their limitations to the variety of limitations of RCTs, randomized clinical trials.
Screening RCTs are subject to a number of systematic biases, and their impli-
cations for decision making under risk and uncertainty are not simple.
To start with some simple examples, consider lead-time bias—when detection
does not on average extend life but only makes patient aware of a diagnosis
earlier. This is likely a confounding factor in many clinical trials of new
screening modalities. More sensitive detection methods mean that cancers
are detected earlier, which leads to higher estimates of time to mortality since
diagnosis. However, if undetectable metastases have occurred prior to the
cancer being detected by the screening test, the cancer may cause death at
the same time as if it had never been detected. Thus, the measure or “end
point” of screening tests should be the “age-adjusted mortality,” rather than
years of survival after diagnosis.
Length bias—the fact that while many indolent (slow growing) cancers
show up on screening exams, but more aggressive cancers tend to show up
between screenings—is a serious problem for some cancers. Length bias leads
to the conclusion that screening and therefore early treatment improves
survival, whereas screening has simply identified less aggressive tumors.
Finally, overdiagnosis bias (the extreme of length bias) is the most damn-
ing problem. It is possible that many people are screened and treated for
cancers that may never have caused clinical symptoms. This is particularly
true in populations of patients over 80, who are more likely to die of other
causes and so may never have suffered any clinical symptoms of a slower
growing cancer. Older adults with significant comorbidities should not be
subject to the toxic effects of cancer treatment unnecessarily.
at University of Utah on May 8, 2012 from
Ethical Issues in Cancer Screening and Prevention 9
What the case of overdiagnosis in older populations illustrates is that context
matters. Not all screening tests are best for all patients at all ages, sexes, or
ethnicities. Underrepresented groups should be studied for variability in age
of onset of different cancers, as well as in aggressiveness of different can-
cers. It is well known that African Americans (particularly men) have higher
rates of mortality from cancer. How much of this is because they suffer from
more aggressive cancers and how much of it is due to the fact that African
Americans on average do not have adequate access to screening and/or
treatment? We need to remedy both inadequate access and inadequate or
ill-suited treatment regimens, taking human variability into consideration.
More inclusive and comprehensive RCTs of screening for different cancers
in different ethnic groups may help resolve some of these dilemmas (for a
discussion of the dilemmas of difference, see King, 1992).
In addition to the possible confounding factors of clinical trials, there are
a variety of methodological and broader policy concerns that some raise
about evidence based medicine.
First, a general worry is that EMB will result in a “one-size-fits-all” approach
to setting standards of care. For example, universal screening for a disease
starting at some specific age, when individual assessment of risk may be
more appropriate. Many worry that EBM and policy will reduce the power
of clinicians to make decisions on a case-by-case basis. Moreover, perhaps
clinicians need to have their enthusiasms (occasionally) reined in by the evi-
dence. As the case of open biopsies, reported above, and cases of back
surgery discussed by Groopman (2007) make strikingly clear, the culture of
medicine may be such that clinicians are trained in a practice—a surgical
procedure, screening modality, or treatment—that may or may not have
evidential support of benefit versus other (perhaps less costly) options.
To avoid cases of bias, clinicians may occasionally need to have their practices
and choices “second guessed” by epidemiologists.
Second, some may fear that where EBM points to “low” or “gray areas”
of benefit, third-party insurers may not be willing to pay for screening or
other tests or treatments. This is a legitimate concern, but perhaps this is an
argument for, rather than against, more public and open communication
about exactly when screening is and is not effective. For instance, if a clinician
needs to justify a screening test by giving evidence of increased patient risk,
this may actually reduce unnecessary harm to the vast majority of patients
who would receive little individual benefit from such a test.
Third, a variety of philosophers have raised questions about the “water-
tight” conclusions of RCTs. It is not clear, at least according to Worrall (2002,
2007), that randomized clinical trials are the “final word” on any medical
question (they are famously subject to reversals!), and he raises legitimate
objections to the claim that randomization (necessarily) eliminates all known
and unknown confounders. Further, some (Cartwright, 2010) argue that we
need to be careful of “exporting” the results of RCTs into contexts that are
at University of Utah on May 8, 2012 from
Anya Plutynski
significantly different. It is not clear that the results of even the most carefully
conducted RCT at time X in geographical location Y can be exported to
an entirely different location, time, or culture. Medical culture, as well as
patient population, varies by location, university setting, and rural versus
urban areas.
Fourth, critics of meta-analyses argue that it is not altogether clear that they
are free of bias (Stegenga, 2011). Bias can enter in at the stage where choice
of studies to include in a meta-analysis is made, either in terms of the effect
measured or inclusion or exclusion on the bases of assessments of quality
and other methodological criteria. Further, different RCTs in different loca-
tions (even though they claim to be measuring the same “effect”) may be
drawing upon different patient populations, may be different in terms of
rates of diagnosis at different institutions and may not have uniform measures
of either cause or effect variable (RWJF, 2010). For example, when looking
into the USPSTF’s meta-analysis of the effectiveness of screening, different
institutions included ductal carcinoma in situ in their “count” of cancer diag-
noses and some used different measure of end point; for example, relative
risk versus hazard ratio.
Finally, meta-analyses do not have policy recommendations “written on
their sleeves.” Even when all the issues surrounding RCTs and meta-analyses
are appropriately addressed, an empirical study will always underdetermine
policy recommendations. For instance, the USPSTF’s recommendations are
based on clinical trials where patients were “invited” to be screened. Of course,
compliance was not perfect. So, a recommendation of universal screening for
individuals based on less than perfect compliance may be doubtful.
Confounders, such as “healthy volunteer” bias, may also contribute such
that only those at less risk are actually seeking opportunities to be screened,
so that the effectiveness of screening may be downplayed. Issues surrounding
compliance require more empirical follow-up. If compliance is imperfect,
some worry that recommendations of less frequent screening may cause
some women to choose not to go at all. However, the opposite may be the
case as well; if women are told that they only need to go every other year,
they may be better able to comply. Empirical data on when and why patients
seek screening might help resolve this issue. Determining how or whether
annual versus biannual recommendations are more likely to be complied
with is surely relevant to determining what should be recommended. Whether
and why patients fail to go for screening seems to be very important to
understand, not simply whether screening itself saves lives.
Apart from the issues surrounding the empirical evidence for and against
screening, there is a further matter of risk communication and decision
at University of Utah on May 8, 2012 from
Ethical Issues in Cancer Screening and Prevention 11
making under uncertainty. How should physicians communicate about
screening, particularly when the benefits are small, or uncertain, and may be
outweighed by not insignificant risks? How should patients weigh the risks
and benefits of early screening? Ought one to make general clinical recom-
mendations drawing upon a precautionary principle and what does it mean
to do so? How, if at all, should individual medical decisions be conducted,
as distinct from public health recommendations? Is it necessary for clinicians
to communicate both absolute and relative risk to patients? What are we to
make of the fact that both patients (and clinicians!) are poor at deliberating
on decisions under risk? (see, e.g., Fagerlin et al., 2007; Schwartz, 2009).
As Resnik (2004) and Katz (1984) remind us in this context, we must be
attentive to both the science and values at stake in medical decision making.
The case of screening is not unique in this respect; many medical decisions
are made without complete evidence. Diagnosis and treatment decisions
often are made under time constraints; waiting for all the evidence to come
in might mean that a decision comes too late. Resnik (2004) argues that
where a risk is plausible and serious (e.g., if negative outcomes are irrevers-
ible), precaution is in order. This would seem to argue in favor of screening,
whatever the potential risks of false positives or overdiagnosis. However, as
Resnik remarks, given the sensitivity of the PSA test (and thus the high risk
of false positives), the patient has to make an individual decision about
whether he or she is willing to risk the need for follow up and possible treat-
ment for a disease that was indolent. Precaution is a double-edged sword.
Schwartz and Meslin (2008) argue that providing patients with absolute
rather than relative risk information may cause them to forgo a test or treatment
that may be beneficial. As mentioned above, absolute risk is calculated by sub-
tracting the mortality rate in the screened from the unscreened group. Relative
risk is calculated by dividing the decrease in morbidity or mortality by the base-
line rate. Schwartz and Meslin argue that providing relative risk information is
preferable and warn that absolute risk information can cause individuals to
forgo screening and the benefits thereof. If patients request further information,
they should certainly be provided it (Schwartz and Meslin, 2008).
Schwartz and Meslin acknowledge that their proposal may raise legitimate
concerns about informed consent and paternalism. There is, arguably, a
principle of “truth in advertising” that clinicians should follow. By way of
analogy, were the NRA to argue that carrying a firearm reduces one’s risk of
dying in a terrorist attack by 50%, one should be skeptical, and not only
because getting statistically significant data to support such a claim would
be nearly impossible. Relative risk reduction can be deceptive, particularly
when the baseline prevalence of some disease is exceedingly low. Clinicians
must be attentive to this and to the fact that there may be more than one
reasonable response to risk. Trade-off neglect, or the failure to acknowledge
and recognize reasonable trade-offs in decision, is a very common failure in
reasoning in such contexts. Different parties value precaution differently and
at University of Utah on May 8, 2012 from
Anya Plutynski
may value, for example, more years of life without worry of false positives
over early aggressive treatment. As we have seen from the examples discussed
above, we all (clinicians included) should not assume that our “intuitions” are
always correct; early frequent screening may not always benefit all patients.
Framing risk reduction in relative terms is, arguably, a form of coercion, par-
ticularly when considering cases of gray areas, where one could legitimately
choose one of several options and not be entirely “irrational.” Values differ,
and putting likely trade-offs “on screen” is at least plausibly a moral obligation,
even if putting every possible remote risk on screen for patients is not.
Even clinicians may disagree as to the value of a treatment, procedure, or
screening modality, depending on their knowledge and values. As Quanstrum
argues, radiologists may be overly confident about the benefits of screening
(Quanstrum and Hayward, 2009). Especially where there are gray areas
concerning benefit, clinicians should watch that their own biases do not cause
them to frame the issues so that they gently urge one course of action.
However well meaning, such patterns run the risk of leading to overtreatment
and overdiagnosis, no small risk, particularly for older patients. What is at risk
is not simply worry or fear of cancer, but overtreatment, which is much more
serious, particularly for elderly women and men who may well die from other
causes before even experiencing clinical symptoms of disease.
Although clinical trials of the effectiveness of screening are important, under-
standing cancer’s basic biology and natural history is as, if not more, important
to both the development of smarter screening modalities and better treatment.
There needs to be more swift progress in understanding the variable prog-
nosis for varieties of “the same” cancers. The assumption that all cancers
progress in a linear fashion at the very same speed toward morbidity and
mortality is no longer reasonable (Zahl et al., 2008; Croswell, Ransohoff,
and Kramer, 2010). Although it is not news that not all breast cancers are
alike, as we learn more about the variety of the major forms of cancer, it ap-
pears that even some very aggressive and uniformly deadly cancers, such as
pancreatic cancer, vary in their genetic profile and prognosis. This variation
could make significant differences in recommendations for both screening
and treatment. Although sequencing the “genome” of these different cancers
promises much, it is also the case that targeted therapies need not await a
complete “genetic diagnosis” of each and every therapy. Some of the best
available targeted therapies were made possible without a complete under-
standing of the genetic bases of the disease—understanding not only the
genetic but also molecular character, epigenetics, and/or developmental
context of different cancers can suggest innovative treatment as well.
Cancer is not one disease, but several. Not all cancers progress in the same
fashion. Some are “indolent”—they never or only at long last produce clinical
at University of Utah on May 8, 2012 from
Ethical Issues in Cancer Screening and Prevention 13
symptoms. Such slow-growing cancers may call for a “wait and see” strategy.
With better and more evidence-based pathological information about the
prospective course of different varieties of this disease, as well as the devel-
opment of better, more targeted and less toxic therapies, this may be a case
where less is more.
1. As she puts it: “A necessary condition for an ethically justified screening recommendation is that the
test reasonably can be expected to be beneficial on balance for the person taking it. The qualifier ‘on balance’
takes into account the benefits and harms of the tests and outcomes as well as the probabilities that those
benefits and harms will occur. Thus it does not require that a given test be actually beneficial to everyone who
takes it. Such a requirement would be unreasonable since, given the nature of screening, the vast majority of
people screened will not have the disease in question and thus will not be able to benefit from the screening
(except by being told that they do not have a disease that they had no reason to suspect they had in the first
place). Instead, the standard requires only that the size of the benefit, taking into account the probability that
it will occur, outweighs the size of the risks, taking into account the probability that they will occur.
(Thus, an unlikely but great benefit can outweigh a very likely but small harm.) When this is the case, the test
can be said to be good for the recipient, all things considered or ‘beneficial on balance’” (Malm, 1999, p. 27).
Baron, J. 2000. Thinking and deciding. 3rd ed. Cambridge: Cambridge University Press.
Bowling, A., and S. Ebrahim. 2001. Measuring patients’ preferences for treatment and percep-
tions of risk. Quality Health Care 10(Suppl 1):i2–8.
Buys, S. 2011. “Why do we die from cancer?” University of Utah Medical School Lecture notes,
Cells, Molecules and Cancer Course, Spring 2011.
Cartwright, N. 2010. Philosophy of Science Association presidential lecture. Philosophy of
Science Association Meeting, Montreal, Canada.
Croswell, J. M., D. Ransohoff, and B. S. Kramer. 2010. Principles of cancer screening: Lessons
from history and study design issues. Seminars in Oncology 37:202–15.
Esserman, L., Y. Shieh, and I. Thompson. 2009. Rethinking screening for breast cancer and
prostate cancer. Journal of the American Medical Association 302:1685–92.
Fagerlin, A., B. J. Zikmund-Fisher, and P. A. Ubel. 2007. ‘If I’m better than average, then I’m OK?’:
Comparative information influences beliefs about risk and benefits. Patient Education
and Counseling 69:140–4.
Gøtzsche, P. C., and M. Nielsen. 2009. Screening for breast cancer with mammography. Cochrane
Database Systematic Review 4:CD001877.
Groopman, J. 2007. How doctors think. New York: Houghton Mifflin.
Gutwin, L. G., D. Gutwein, N. Ang, H. Liu, J. K. Marshall, S. N. Hochwald, E. M. Copeland, and
S. R. Grobmyer. 2011. Utilization of minimally invasive breast biopsy for the evaluation
of suspicious breast lesions. American Journal of Surgery. Available: http://www. (Accessed March, 2011).
Kahneman, D., and A. Tversky. 1979. Prospect theory: an analysis of decision under risk.
Econometrica 47:263–92.
Katz, J. 1984. The silent world of doctor and patient. New York: The Free Press.
King, P. 1992. The dangers of difference. Hastings Center Report 22:35–8.
at University of Utah on May 8, 2012 from
Anya Plutynski
McKenna, F. P. 1993. It won’t happen to me: Unrealistic optimism or illusion of control? British
Journal of Psychology 84:39–50.
Malm, H. 1999. Medical screening and the value of early detection: When unwarranted faith
leads to unethical recommendations. Hasting Center Report 1:26–37.
Mandelblatt, J., K. A. Cronin, S. Bailey, D. A. Berry, H. J. de Koning, G. Draisma,
H. Huang, et al. 2009. Effects of mammography screening under different screening
schedules: Model estimates of potential benefits and harms. Annals of Internal Medicine
Moss, S. 2005. Overdiagnosis and overtreatment of breast cancer: Overdiagnosis in randomised
controlled trials of breast cancer screening. Breast Cancer Research 7:230–4.
Nelson, H. D., K. Tyne, A. Naik, C. Bougatsos, B. K. Chan, and L. Humphrey. 2009. Screening
for breast cancer: an update for the U.S. Preventive Services Task Force. Annals of Internal
Medicine 151:727–37.
National Institutes of Health, Regulations and Ethical Guidelines. 2005. Title 45 CFR Part 46
Protection of Human Subjects. Code of Federal Regulations, Department of Health and
Human Services. (Accessed August, 2011).
Paci, E., G. Miccinesi, D. Puliti, P. Baldazzi, V. De Lisi, F. Falcini, C. Cirilli, et al. 2006. Estimate
of overdiagnosis of breast cancer due to mammography after adjustment for lead time. A
service screening study in Italy. Breast Cancer Research 8:R68.
Quanstrum, K., and R. A. Hayward. 2009. Lessons from the Mammography Wars. New England
Journal of Medicine 363:1076–9.
Resnik, D. 2004. The precautionary principle and medical decision making. Journal of Medi-
cine and Philosophy 29:281–99.
RWJF. 2010. Who you are, where you live, affects medical treatment. Dartmouth study finds
[On-line]. Available: (Accessed
May 17, 2010).
Schwartz, P. 2009. Disclosure and rationality: Comparative risk information and decision-
making about prevention. Theoretical Medicine and Bioethics 30:199–213.
Schwartz, P., and E. Meslin. 2008. The ethics of information: Absolute risk reduction and
patient understanding of screening. Journal of General Inter nal Medicine 6:867–70.
Stegenga, J. 2011. Is meta-analysis the platinum standard of evidence? Studies in History and
Philosophy of Biological and Biomedical Sciences 42:497–507.
US Preventive Services Task Force. 2011. Screening for breast cancer: Recommendation
statement. AHRQ Publication No. 10-05142-EF-2, November 2009. Rockville, MD: Agency
for Healthcare Research and Quality. Available:
breastcancer/brcanrs.htm (Accessed February 15, 2011).
Weinstein, N. D. 1980. Unrealistic optimism about future life events. Journal of Personal Social
Psychology 39:806–20.
Welch, H. G., and W. Black. 2010. Overdiagnosis and cancer. Journal of the National Cancer
Institute 102:1–9.
Worrall, J. 2002. What evidence in evidence-based medicine? Philosophy of Science 69:
Worrall, J. 2007. Why there’s no call to randomize. British Journal for the Philosophy of Science
Zahl, P., J. Maehlen, and H. G. Welch. 2008. The natural history of invasive breast cancers
detected by screening mammography. Archives of Internal Medicine 168:2311–6.
at University of Utah on May 8, 2012 from
... This is essential, as it is ethically required that harms of cancer screening are communicated in a manner that could be understood by all. 28 Furthermore, the specific term overdiagnosis may have been replaced by other words due to the complexity surrounding its definition. 29 We also found that the estimates of overdiagnosis were reported in various ways, including the use of text (n = 3, 7%), presented visually as infographics/icon arrays (n = 2, 5%), tables (n = 1, 2.4%), or flow charts (n = 1, 2.4%) within the decision aids. ...
Full-text available
Background. Overdiagnosis is an accepted harm of cancer screening, but studies of prostate cancer screening decision aids have not examined provision of information important in communicating the risk of overdiagnosis, including overdiagnosis frequency, competing mortality risk, and the high prevalence of indolent cancers in the population. Methods. We undertook a comprehensive review of all publicly available decision aids for prostate cancer screening, published in (or translated to) the English language, without date restrictions. We included all decision aids from a recent systematic review and screened excluded studies to identify further relevant decision aids. We used a Google search to identify further decision aids not published in peer reviewed medical literature. Two reviewers independently screened the decision aids and extracted information on communication of overdiagnosis. Disagreements were resolved through discussion or by consulting a third author. Results. Forty-one decision aids were included out of the 80 records identified through the search. Most decision aids (n = 32, 79%) did not use the term overdiagnosis but included a description of it (n = 38, 92%). Few (n = 7, 17%) reported the frequency of overdiagnosis. Little more than half presented the benefits of prostate cancer screening before the harms (n = 22, 54%) and only 16, (39%) presented information on competing risks of mortality. Only 2 (n = 2, 5%) reported the prevalence of undiagnosed prostate cancer in the general population. Conclusion. Most patient decision aids for prostate cancer screening lacked important information on overdiagnosis. Specific guidance is needed on how to communicate the risks of overdiagnosis in decision aids, including appropriate content, terminology and graphical display. Highlights: Most patient decision aids for prostate cancer screening lacks important information on overdiagnosis.Specific guidance is needed on how to communicate the risks of overdiagnosis.
... Second, epistemological debates concerning the extent of this phenomenon and how to measure it (Marmot et al. 2013). Third, broadly ethical debates concerning how to balance the harms and benefits of population screening, particularly given widespread public misunderstanding of the risks involved (Plutynski 2012). Fourth, conceptual debates: for example, how the phenomenon relates to the distinction between "individual" and "population" perspectives in public health (Hofmann 2018). ...
Full-text available
There are many analogies between medical and judicial practice. This article explores one such analogy, between “medicalization” and “criminalization.” Specifically, drawing on an analogy between a judge’s speech act of delivering a verdict and a physician’s speech act of giving a diagnosis, it suggests a novel account of the phenomenon of “overdiagnosis.” Using this approach, we can make some headway in understanding debates over the early detection of cancer. The final section outlines the relationship between this approach and familiar debates in philosophy of medicine on the nature of disease and in philosophy of science on the “value-free ideal.”
... Second, epistemological debates concerning the extent of this phenomenon and how to measure it (Marmot et al. 2013). Third, broadly ethical debates concerning how to balance the harms and benefits of population screening, particularly given widespread public misunderstanding of the risks involved (Plutynski 2012). Fourth, conceptual debates: for example, how the phenomenon relates to the distinction between "individual" and "population" perspectives in public health (Hofmann 2018). ...
Full-text available
There are many analogies between medical and judicial practice. This article explores one such analogy, between “medicalization” and “criminalization.” Specifically, drawing on an analogy between a judge’s speech act of delivering a verdict and a physician’s speech act of giving a diagnosis, it suggests a novel account of the phenomenon of “overdiagnosis.” Using this approach, we can make some headway in understanding debates over the early detection of cancer. The final section outlines the relationship between this approach and familiar debates in philosophy of medicine on the nature of disease and in philosophy of science on the “value-free ideal.”
... Second, epistemological debates concerning the extent of this phenomenon and how to measure it (Marmot et al. 2013). Third, broadly ethical debates concerning how to balance the harms and benefits of population screening, particularly given widespread public misunderstanding of the risks involved (Plutynski 2012). Fourth, conceptual debates: for example, how the phenomenon relates to the distinction between "individual" and "population" perspectives in public health (Hofmann 2018). ...
Full-text available
There are many analogies between medical and judicial practice. This article explores one such analogy, between “medicalization” and “criminalization.” Specifically, drawing on an analogy between a judge’s speech act of delivering a verdict and a physician’s speech act of giving a diagnosis, it suggests a novel account of the phenomenon of “overdiagnosis.” Using this approach, we can make some headway in understanding debates over the early detection of cancer. The final section outlines the relationship between this approach and familiar debates in philosophy of medicine on the nature of disease and in philosophy of science on the “value-free ideal.”
... While there is a growing literature on the ethics of cancer screening in general, 14 the ethics of skin cancer screening in particular is less developed. One framework advanced by Stoff and Grant-Kels (2021) highlights the principles of utilitarianism, justice, and caring to guide our ethical thinking around skin cancer screening. ...
Cutaneous melanoma is the fifth most diagnosed cancer in the United States and the incidence is increasing yearly. At present, population screening for melanoma is not recommended by national guidelines on account of insufficient evidence to assess the balance of benefits and harms. Indeed, there remains significant controversy over whether screening for melanoma via increasing the frequency of routine skin checks leads to tangible long-term health benefits for patients. In this paper, we highlight how skin cancer screening can impose harms such as overdiagnosis and suggest that the principle of non-maleficence should play a greater role in the formulation of screening policies. We also explore the pressing issue of the underdiagnosis of melanoma in particular populations. In so doing, this paper underscores how the ethical duties of non-maleficence and justice must be balanced in current dermatological practice.
... Unfortunately, it may be impossible in advance to know who these people will be. 1 For example, we might know that a breast cancer screening programme will, overall, have a positive impact on population health, but, via overtreatment or overdiagnosis, will also have negative side effects for a very small, unidentifiable, number of people. 2 One key question in public health ethics concerns how to balance these trade-offs between harm to individuals and benefit to populations. 3 In this paper, I explore the relationship between this challenge and a very general question: whether it is morally problematic to vaccinate children against Covid-19, even when this might not be in their own medical interests, as a way of preventing the transmission of disease. ...
Public health policy often involves a trade-off between promoting population health and protecting the interests of identifiable individuals. This paper analyses this trade-off as it arises in the context of decisions about the vaccination of children against Covid-19, where vaccination may be in the interests of society as a whole, as a means to stopping transmission, but not in the interests of individual children. The paper argues that the UK's Joint Committee on Vaccination and Immunisation resolved this tension by appeal to a version of a non-maleficence principle. It argues that, while this principle can be a useful guide to some public health policy decision-making, it is inappropriate in the case of vaccination.
... 7,8 That change prompted strong opposition from some breast cancer activists and medical organizations, including the American College of Radiology and the American Cancer Society. 9 The Affordable Care Act of 2010 required private insurers to cover annual screening. 10 Women's actual screening frequencies changed little over this period. ...
Full-text available
Background We evaluate the construct validity of a proposed procedure for eliciting lay preferences among health care policy options, suited for structured surveys. It is illustrated with breast cancer screening, a domain in which people may have heterogeneous preferences. Methods Our procedure applies behavioral decision research principles to eliciting preferences among policy options expressed in quantitative terms. Three-hundred women older than 18 y without a history of breast cancer were recruited through Amazon MTurk. Participants evaluated 4 screening options for each of 4 groups of women, with varying risk of breast cancer. Each option was characterized by estimates of 3 primary outcomes: breast cancer deaths, false alarms, and overdiagnosis resulting in unnecessary treatment of cancers that would not progress. These estimates were based on those currently being developed by the Breast Cancer Surveillance Consortium. For each risk group, participants stated how frequently they would wish to receive screening, if the predicted outcomes applied to them. Results A preregistered test found that preferences were robust enough to be unaffected by the order of introducing and displaying the outcomes. Other tests of construct validity also suggested that respondents generally understood the task and expressed consistent preferences. Those preferences were related to participants’ age and mammography history but not to measures of their numeracy, subjective numeracy, or demographics. There was considerable heterogeneity in their preferences. Conclusions Members of the public can be engaged more fully in informing future screening guidelines if they evaluate the screening options characterized by the expected health outcomes expressed in quantitative terms. We offer and evaluate such a procedure, in terms of its construct validity with a diverse sample of women. Highlights A novel survey method for eliciting lay preferences for breast cancer screening is proposed and evaluated in terms of its construct validity. Participants were generally insensitive to irrelevant task features (e.g., order of presentation) and sensitive to relevant ones (e.g., quantitative estimates of breast cancer risk, harms from screening). The proposed method elicits lay preferences in terms that can inform future screening guidelines, potentially improving communication between the public and policy makers.
... One publication contained an explicitly ethically justified recommendation in favour of TSE [85]. However, 19 publications investigating ethical aspects of cancer screening in general were included [25,[85][86][87][88][89][90][91][92][93][94][95][96][97][98][99][100][101][102]. They were used to further specify criteria of the normative framework on ethical aspects of public health measures by Marckmann [25] for cancer screening (see Table 25 of the full report). ...
Technical Report
Full-text available
HTA key statements Research questions of the HTA report The aims of this investigation are  to assess the benefit of testicular cancer screening through clinical palpation and scrotal ultrasound or through testicular self-examination (TSE) in men from 16 years of age in comparison with no screening with regard to patient-relevant outcomes,  determine the cost (intervention cost) arising from testicular cancer screening in comparison with no screening in asymptomatic men from 16 years of age,  assess the cost-effectiveness of testicular cancer screening in comparison with no screening in asymptomatic men from 16 years of age, and  review ethical, social, legal, and organizational aspects associated with the screening. Conclusion of the HTA report Due to a lack of interventional studies on benefit and harm, the question of whether routine screening of asymptomatic men from 16 years of age (at average or higher risk) results in better treatment outcomes in testicular cancer cannot be answered in an evidence-based manner. There is no hint of (greater) benefit or (greater) harm. No studies are available on cost effectiveness. The theoretical maximum benefit indirectly derived from epidemiological studies in a supplementary presentation for the benefit assessment is relatively small in comparison with other cancers. Testicular cancer is rare, and even in the absence of routine screening, it is discovered in a relatively early stage in most cases and can be treated with correspondingly high cure rates. Therefore, routine screening for testicular cancer in men from 16 years of age cannot be recommended at this time. This applies to both TSE and clinical palpation / scrotal ultrasound. The low potential benefit is accompanied by potential harm due to unnecessary testicular exploration or removal. Nonmalignant testicular anomalies, which are frequently discovered as a result of targeted examinations, may worry affected men and sometimes involve unnecessary resource consumption. Particularly in case of clinical examinations, it is possible for the expected harm inflicted by additional unnecessary invasive evaluations to outweigh the expected benefit when looking at the entire target population. Therefore, clinical palpation and scrotal ultrasound for screening purposes should be offered neither as a standard statutory health insurance (SHI) benefit nor as an individual out-of-pocket health service. TSE is likely associated with less potential harm. It seems justifiable for young men worried about the risk of testicular cancer – of which there should be few according to psychosocial studies – to regularly practise TSE after they have been educated about the lack of direct evidence on their potential benefit and harm and instructed on how to perform the exam. The use of conventional health education channels to advise men to promptly see a physician for diagnostic evaluation in case of abnormalities of the testis can absolutely be recommended. Further, men should be educated about risk factors for testicular cancer and the generally more favourable benefit–harm ratio of screening for individuals at higher risk. Given that a relatively low potential benefit is expected due to the comparatively low incidence and often relatively good treatment outlook for testicular cancer, it seems hardly advisable to overcome the lack of evidence by conducting resource-intensive interventional studies of high methodological quality in men at average risk of testicular cancer. Since the benefit of screening measures increases with the risk of developing the disease, such interventional studies should be performed in high-risk groups, if at all.
... 15 Screening is having, at best, only a small effect on the rate of death from breast cancer. 16,17 In respect of ovarian cancer, a 30-year U.K. multicenter prospective study of the impact of screening on mortality rates in 200,000 women showed successful early detection of the cancer, but this did not translate into saving lives. 18 Noninvasive blood tests, based on circulating tumor methylation, have the potential to identify patients up to 5 years before their cancer becomes detectable by standard screening methods. ...
What, precisely, are we seeking to achieve in offering 'life-saving' treatment to patients with cancer? Research funding agencies and pharmaceutical industry media releases, and government cancer screening programs all promise that their cancer programs save lives. But everybody dies. The nature of life and death from cancer is explored philosophically in this essay, with particular reference to the quality of life, and its meaning, during the period of prolongation of survival by 'life-saving' cancer care.
Surveillance began with counting the numbers of people in the population. At various times in history, numbers have been used to assess the overall strength of the population, to identify the march of dangerous contagion, or to determine needs for food or labor. But even simple counting of population numbers, vital statistics, or reports of disease has been controversial. Information is power and the most rudimentary surveillance can be used both for good and for harm. This chapter sets ethical questions about these basic surveillance methods in historical and epistemological context. It gives examples of uses of data about population numbers, vital statistics, or outbreaks that have been clearly beneficial, as well as examples that have bordered on the genocidal. Counting numbers, as a rudimentary epidemiological method, also presents the opportunity to explore ethical problems raised by epidemiology as a science, such as incomplete data, biased data, or false negatives or positives. Today, with increasing understanding of disease and availability of prevention or treatment, the advantages of outbreak detection may be shared far more widely and more equally. Nonetheless, outbreak detection can generate fear and hostility if patterns of disease track otherwise disfavored groups. COVID-19 has revealed the importance of demographic data about the distribution of disease burdens—data that may either generate mistrust as people see their disadvantage starkly, or that may foster trust if the result is increased attention to disparities in treatment and in health.
Background: This systematic review is an update of evidence since the 2002 U.S. Preventive Services Task Force recommendation on breast cancer screening. Purpose: To determine the effectiveness of mammography screening in decreasing breast cancer mortality among average-risk women aged 40 to 49 years and 70 years or older, the effectiveness of clinical breast examination and breast self-examination, and the harms of screening. Data sources: Cochrane Central Register of Controlled Trials and Cochrane Database of Systematic Reviews (through the fourth quarter of 2008), MEDLINE (January 2001 to December 2008), reference lists, and Web of Science searches for published studies and Breast Cancer Surveillance Consortium for screening mammography data. Study selection: Randomized, controlled trials with breast cancer mortality outcomes for screening effectiveness, and studies of various designs and multiple data sources for harms. Data extraction: Relevant data were abstracted, and study quality was rated by using established criteria. Data synthesis: Mammography screening reduces breast cancer mortality by 15% for women aged 39 to 49 years (relative risk, 0.85 [95% credible interval, 0.75 to 0.96]; 8 trials). Data are lacking for women aged 70 years or older. Radiation exposure from mammography is low. Patient adverse experiences are common and transient and do not affect screening practices. Estimates of overdiagnosis vary from 1% to 10%. Younger women have more false-positive mammography results and additional imaging but fewer biopsies than older women. Trials of clinical breast examination are ongoing; trials for breast self-examination showed no reductions in mortality but increases in benign biopsy results. Limitation: Studies of older women, digital mammography, and magnetic resonance imaging are lacking. Conclusion: Mammography screening reduces breast cancer mortality for women aged 39 to 69 years; data are insufficient for older women. False-positive mammography results and additional imaging are common. No benefit has been shown for clinical breast examination or breast self-examination.
Background: Despite trials of mammography and widespread use, optimal screening policy is controversial. Objective: To evaluate U.S. breast cancer screening strategies. Design: 6 models using common data elements. Data sources: National data on age-specific incidence, competing mortality, mammography characteristics, and treatment effects. Target population: A contemporary population cohort. Time horizon: Lifetime. Perspective: Societal. Interventions: 20 screening strategies with varying initiation and cessation ages applied annually or biennially. Outcome measures: Number of mammograms, reduction in deaths from breast cancer or life-years gained (vs. no screening), false-positive results, unnecessary biopsies, and overdiagnosis. Results of base-case analysis: The 6 models produced consistent rankings of screening strategies. Screening biennially maintained an average of 81% (range across strategies and models, 67% to 99%) of the benefit of annual screening with almost half the number of false-positive results. Screening biennially from ages 50 to 69 years achieved a median 16.5% (range, 15% to 23%) reduction in breast cancer deaths versus no screening. Initiating biennial screening at age 40 years (vs. 50 years) reduced mortality by an additional 3% (range, 1% to 6%), consumed more resources, and yielded more false-positive results. Biennial screening after age 69 years yielded some additional mortality reduction in all models, but overdiagnosis increased most substantially at older ages. Results of sensitivity analysis: Varying test sensitivity or treatment patterns did not change conclusions. Limitation: Results do not include morbidity from false-positive results, patient knowledge of earlier diagnosis, or unnecessary treatment. Conclusion: Biennial screening achieves most of the benefit of annual screening with less harm. Decisions about the best strategy depend on program and individual objectives and the weight placed on benefits, harms, and resource considerations. Primary funding source: National Cancer Institute.
In this thoughtful and important book, Jay Katz takes us on a journey through medical and legal history to establish convincingly one point— that medical and legal relationships and entitlements up to the 20th century have developed in a manner that has kept patients out of the process of decision in which the harms and benefits of therapeutic alternatives concerning them were weighed and balanced.In medicine, for centuries, such exclusion was justified on the basis of things such as the medical ignorance of patients, their intolerance of uncertainty, the need to maintain the patient's hope that things would improve, the simplicity that unilateral decision making gave to the doctor's life, the painfulness of discussing with patients dreaded events and hopeless prognoses, and the authority that silence in the medical relationship gave to physicians.In the law, too, the rights given to patients up to the mid-20th century were basically
Evidence‐Based Medicine is a relatively new movement that seeks to put clinical medicine on a firmer scientific footing. I take it as uncontroversial that medical practice should be based on best evidence—the interesting questions concern the details. This paper tries to move towards a coherent and unified account of best evidence in medicine, by exploring in particular the EBM position on RCTs (randomized controlled trials).
It has been sixty years since the beginning of the Tuskegee syphilis experiment and twenty years since its existence was disclosed to the American public. The social and ethical issues that the experiment poses for medicine, particularly for medicine's relationship with African Americans, are still not broadly understood, appreciated, or even remembered.[1] Yet a significant aspect of the Tuskegee experiment's legacy is that in a racist society that incorporates beliefs about the inherent inferiority of African Americans in contrast with the superior status of whites, any attention to the question of differences that may exist is likely to be pursued in a manner that burdens rather than benefits African Americans.
The evidence from randomized controlled trials (RCTs) is widely regarded as supplying the ‘gold standard’ in medicine—we may sometimes have to settle for other forms of evidence, but this is always epistemically second-best. But how well justified is the epistemic claim about the superiority of RCTs? This paper adds to my earlier (predominantly negative) analyses of the claims produced in favour of the idea that randomization plays a uniquely privileged epistemic role, by closely inspecting three related arguments from leading contributors to the burgeoning field of probabilistic causality—Papineau, Cartwright and Pearl. It concludes that none of these further arguments supplies any practical reason for thinking of randomization as having unique epistemic power. • 1Introduction • 2Why the issue is of great practical importance—the ECMO case • 3Papineau on the ‘virtues of randomization’ • 4Cartwright on causality and the ‘ideal’ randomized experiment • 5Pearl on randomization, nets and causes • 6Conclusion