Fowler, Jr FJ, McNaughton Collins M, Albertsen PC, Zietman A, Elliott DB, Barry MJ. Comparison of Recommendations by Urologists and Radiation Oncologists for Treatment of Clinically Localized Prostate Cancer. JAMA. 2000;283(24):3217-3222. doi:10.1001/jama.283.24.3217
Author Affiliations: Center for Survey Research, University of Massachusetts, Boston (Dr Fowler and Ms Elliott); Medical Practices Evaluation Center (Drs McNaughton Collins and Barry) and the Department of Radiation Oncology (Dr Zietman), Massachusetts General Hospital, Boston; and the Division of Urology, University of Connecticut Health Center, Farmington (Dr Albertsen).
Context Multiple treatment options are available for men with prostate cancer,
but therapeutic recommendations may differ depending on the type of specialist
Objective To define and contrast the distribution of management recommendations
by urologists and radiation oncologists for a spectrum of men with prostate
Design, Setting, and Participants Mail survey sent in 1998 to a random sample of physicians in the United
States, who were listed as urologists (response rate 64%, n=504) and radiation
oncologists (response rate 76%, n=559) in the American Medical Association
Registry of Physicians and practicing at least 20 hours per week.
Main Outcome Measure Questionnaire addressing beliefs and practices regarding prostate cancer
Results Forty-three percent of radiation oncologists vs 16% of urologists would
recommend routine prostate-specific antigen testing for men aged 80 years
and older. For men with moderately differentiated, clinically localized cancers,
and a more than 10-year life expectancy, 93% of urologists chose radical prostatectomy
as the preferred treatment option, while 72% of radiation oncologists believed
surgery and external beam radiotherapy were equivalent treatments. For most
tumor grades and prostate-specific antigen levels, both specialty groups were
significantly more likely to recommend the treatment in their specialty than
the other treatment. Both groups reported giving patients similar estimates
of the risks of complications due to surgery and radiation. Neither group
favored watchful waiting in their treatment management except for a subset
of men with life expectancies of less than 10 years and cancers with very
favorable prognoses (Gleason score of 3 or 4 and prostate-specific antigen
level ≤5 ng/mL).
Conclusions Based on this study, while urologists and radiation oncologists do agree
on a variety of issues regarding detection and treatment of prostate cancer,
specialists overwhelmingly recommend the therapy that they themselves deliver.
About 180,400 men will be diagnosed as having prostate cancer in the
United States this year, most with clinically localized disease.1
The majority of these men will choose among 3 primary therapies: radical prostatectomy,
external beam radiotherapy, or brachytherapy.
The choice among these therapies is not easy. Because most prostate
cancers are found in men in their 60s and 70s, and because these cancers generally
grow slowly, many prostate cancer patients are destined to die of competing
medical problems.2 For others, especially men
with poorly differentiated tumors or high prostate-specific antigen (PSA)
levels, these therapies may not be curative. Moreover, while cohort studies
following surgery and external beam radiotherapy patients for 10 to 15 years
have been done,3 patients receiving contemporary
brachytherapy have not been followed up that long.4
Meanwhile, all of these therapies have the potential to create adverse effects
The clinical judgment of the physicians who counsel patients can play
a critical role in the treatments chosen. In a 1988 survey, urologists and
radiation oncologists were asked what they personally would do if they were
diagnosed as having clinically localized prostate cancer. In that survey,
79% of US urologists said they would choose a radical prostatectomy, while
92% of radiation oncologists said they would choose external beam radiotherapy.5 Obviously, depending on which physicians they consult,
patients might well expect to get different counsel about optimal management.
We were interested in extending our understanding of the differences
between specialties beyond simply their preferences for treatment. We also
wanted to see whether the different specialties continued to have such polar
views regarding treatment 10 years later, well into the era of PSA testing.
To these ends, in 1998, we surveyed a nationwide random sample of practicing
US urologists and radiation oncologists.
A random sample of physicians who listed their specialty as either urology
or radiation oncology were selected from the American Medical Association
Master List of Physicians. Subsequently, the offices of the sampled physicians
were contacted by telephone to verify the address and specialty, that they
were not in residency training, and that they were in clinical practice for
at least 20 hours weekly.
Physicians then were sent a pretested questionnaire, a cover letter,
and $10. A reminder postcard was sent to all sampled individuals; another
survey instrument and cover letter were sent to nonrespondents after about
3 weeks. Finally, for those physicians who had not responded to the questionnaire,
follow-up telephone calls were made to the physicians' offices to encourage
response and to identify subjects who needed another questionnaire.
The survey instrument for urologists contained questions about the diagnosis
and treatment of benign prostatic hyperplasia and prostatitis, as well as
prostate cancer. The instrument for radiation oncologists focused almost exclusively
on prostate cancer. Whenever it made sense, comparable questions were asked
of both specialties.
Questionnaires were returned by 76% of eligible radiation oncologists
(n=559) and 64% of eligible urologists (n=504). Table 1 compares the characteristics of the respondents in the 2
specialties. Responding urologists were older, more often male, more likely
to be in solo practice, and less likely to be salaried. When respondents and
nonrespondents were compared, using data from the file from which the sample
was drawn, no significant differences were noted in terms of age or region
of the country. Urologists who graduated from medical school less than 20
years previously were more likely to respond than earlier graduates. Physicians
in multispecialty groups were also more likely to respond than physicians
in solo practice in both specialties.
Figure 1 compares both groups
of specialists' answers to questions about their recommendations for routine
PSA testing. For men up through age 70 years, members of the 2 specialties
made similar recommendations. Urologists are slightly more aggressive about
screening than radiation oncologists but significantly more aggressive when
patients are men in their 50s. The clearer trend, however, is the much more
aggressive stance of radiation oncologists regarding PSA testing for men 70
years and older, and particularly for men older than 74 years. For example,
while only 16% of urologists recommend routine PSA testing for men older than
80 years, 43% of the radiation oncologists do so.
Figure 2 presents these specialists'
beliefs about when aggressive treatments actually have a survival benefit.
This figure presents the percentage of respondents saying that there is "definitely"
or "probably" a survival benefit for each of the 3 main therapies for men
with clinically localized, moderately differentiated prostate cancer and less
or more than a 10-year life expectancy, respectively. Although there are numerous
statistically significant differences between the 2 groups given the large
sample sizes, there is also considerable consistency in their answers. The
majority of both groups of specialists think that all 3 therapies have survival
benefit for men with a 10-year or longer life expectancy; only a minority
think there is benefit from any of the treatments for men with less than a
10-year life expectancy. More of both types of specialists, though, believe
that radiation, either by external beam or brachytherapy, is more likely to
offer a survival benefit for men with less than a 10-year life expectancy
than a radical prostatectomy. Indeed, that perception is slightly but significantly
more common among urologists than among radiation oncologists. Figure 2 also demonstrates that members of each specialty are generally
somewhat more likely than those in the other specialty to believe in the efficacy
of the treatments that they perform, and are also a little less positive about
the survival benefits of the treatment offered by the other specialty group.
Table 2 presents physicians'
responses to questions about head-to-head comparisons of the effectiveness
of the 3 major therapies. When comparing radical prostatectomy and external
beam radiotherapy for patients with life expectancies of 10 years or longer,
urologists are almost all convinced (93%) that radical prostatectomy is better.
On the other hand, radiation oncologists' perceptions (72%) are that surgery
and radiotherapy are equivalent. Almost a third of urologists think neither
treatment has survival value for men with less than a 10-year life expectancy,
while only 19% of radiation oncologists agree.
When deciding among treatment options, prostate cancer patients need
to consider not only the effectiveness of the options at extending life, but
also their adverse effects. Sexual dysfunction and incontinence are 2 potential
adverse effects of aggressive therapy for prostate cancer. How are the likelihoods
of these adverse effects presented to patients by urologists and radiation
oncologists? Table 3 presents
the risks of these complications that the respondents said they quote to their
patients. Table 3 is most striking
in how similarly the 2 groups estimate the probability of complications associated
with surgery and radiation.
Respondents were also asked to choose their preferred treatment for
patients with tumors of varying Gleason scores and PSA levels. In Figure 3 and Figure 4, the patient profiles are ordered based on the decreasing
likelihood that such a cancer would be organ-confined at surgical staging.6 Subjects were given the options of expectant management
(watchful waiting) and androgen deprivation (as primary therapy) in addition
to the potentially curative therapies. For this analysis, both forms of radiotherapy
were considered together.
Both groups of specialists have some members (10%-20%) who are willing
to consider watchful waiting for patients with cancers with Gleason scores
of 3 or 4 and PSA levels no higher than 5 ng/mL. Beyond that low-risk subset,
essentially no one in either group is willing to recommend watchful waiting
As one would expect from the preceding data, the majority of both groups
of specialists would recommend for most patients the therapy that they themselves
deliver. However, there is a subset of radiation oncologists who indicated
a preference for surgery for low-grade, low-PSA tumors. There are very few
urologists who prefer radiation for such tumors (Figure 4).
As Gleason scores reach 7 or 8, both groups start to consider androgen
deprivation as a primary therapy (Figure 3). Urologists also begin to recommend radiation more often in relationship
to surgery as Gleason scores and PSA levels increase (Figure 4).
As the probability of organ-confined disease decreases, urologists become
divided about the value of surgery. For tumors with Gleason scores of 8 or
higher, or a Gleason score of 7 with a high PSA level, they become as likely
to recommend androgen deprivation or radiation as they do surgery. However,
there is a substantial minority who continue to recommend surgery even when
tumors are likely to be extracapsular. Radiation oncologists, on the other
hand, continue to recommend radiation for higher-risk tumors.
Finally, physicians were asked whether they believed that the 3 main
potentially curative prostate cancer therapies are overused or underused in
the United States (Table 4). A
majority of radiation oncologists believe that radical prostatectomy is overused
(82%), and about half think that radiation and brachytherapy are underused.
In contrast, 51% of urologists think that radical prostatectomy is used at
about the right rate and 37% think that external beam radiation is overused.
Substantial percentages of both radiation oncologists and urologists believe
that brachytherapy is both overused and underused, and a higher proportion
express no opinion than for the other 2 primary therapies.
Although urologists and radiation oncologists differed in many of their
beliefs regarding prostate cancer treatment, they also demonstrated agreement
on a variety of issues. First, despite controversy over the value of PSA screening,7,8 responding physicians in both these
specialties are virtually unanimous in their recommendation that PSA testing
be done routinely at least until around age 75 years. For men older than 75
years, the 2 specialties differ, with radiation oncologists being considerably
more positive about testing older men. This position is consistent with radiation
oncologists' perceptions that they have a therapy to offer that (according
to nearly a majority) is beneficial to men even with less than a 10-year life
Second, despite the lack of published supporting evidence from randomized
clinical trials, the vast majority of physicians in both specialties believe
that all 3 therapies offer a survival advantage for men with a 10-year or
more life expectancy.
Third, with respect to patients with less than a 10-year life expectancy,
only a minority of members of either specialty thinks that any therapy offers
a survival advantage. Within both groups, however, there are more physicians
who think that radiation therapy offers a survival advantage than think that
Fourth, specialists generally agree on the probabilities of complications
of all 3 treatments. Both groups perceive surgery as being more likely to
produce incontinence and sexual dysfunction than radiotherapy. Both urologists
and radiation oncologists are generally convinced that nerve-sparing surgery
substantially reduces the rate at which patients experience sexual dysfunction.
Disagreements about probabilities of adverse effects, then, do not account
for the differences in treatment recommendations noted between the specialties.
Some recent studies actually suggest, however, that these estimated risks
of complications for all 3 treatments may be low, and particularly low for
nerve-sparing radical prostatectomy.9- 13
Fifth, physicians from the 2 specialties are remarkably similar in the
extent to which they would recommend either watchful waiting or androgen deprivation
as primary therapies for particular subsets of men based on Gleason scores
and PSA levels. Less than 25% of members of either specialty would recommend
watchful waiting to men with a tumor with a Gleason score of 3, despite the
fact that these patients appear to have an essentially normal life expectancy
without aggressive treatment.14 Members of
both specialties feel much more comfortable if patients with any degree of
prostate cancer receive one of the major primary therapies. When the Gleason
scores (7-8) and PSA levels (≥10 ng/mL) are higher, increasing numbers
in both specialties, eventually nearing half, would recommend androgen deprivation
as the primary therapy. Radiation oncologists continue to recommend radiation
for tumors with a higher likelihood of capsular penetration, while urologists
appear more dubious about the value of surgery. This finding may reflect a
difference in perspectives about when the 2 local therapies are likely to
be still curative; that is, radiation oncologists may believe tumors with
some degree of capsular penetration may still be effectively treated with
radiation, while urologists may doubt that surgery will cure such tumors.
Sixth, despite a relative shortage of long-term follow-up data, members
of both specialties generally seem to accept brachytherapy as being at least
as effective as external beam radiotherapy. Urologists seem to be slightly
more positive about brachytherapy than about external beam radiation therapy.
Of course, the most dramatic difference between these 2 groups of specialists
is that members of each specialty tend to believe in the therapy that they
themselves deliver. Radiation oncologists (72%) tend to believe that their
therapy is just as good as radical prostatectomy for men with moderately differentiated,
clinically localized cancer while urologists (93%) are overwhelmingly convinced
that radical prostatectomy is better. This difference is critical in understanding
what treatment recommendations patients are likely to hear. Presumably, the
radiation therapists can justify their preference for recommending external
beam radiotherapy on the basis that they believe it works just as well as
surgery. In contrast, urologists believe that while men who have surgery do
indeed have higher risks of sexual dysfunction and incontinence, cancer control
is better with surgery, and thus radical prostatectomy is preferred.
While clinical trials have not proven that patient outcomes are improved
by aggressive treatment with surgery or radiation, neither have they proven
these treatments ineffective. Given our findings, it is also important to
point out that the descriptions of the effectiveness of surgery and radiation
therapy that patients receive from urologists and radiation oncologists would
be expected to be quite different. Although some patients might find it confusing
to hear quite different treatment recommendations from experienced physicians
with access to the same medical literature, scheduling consultations with
a member of each specialty may be the best approach to ensure that patients
get a balanced picture regarding aggressive treatment options before making
Neither group of specialists studied was supportive of expectant management
for any but a limited subset of men; primary care physicians appear no more
enthusiastic.15 Interestingly, however, when
patients hear comprehensive presentations regarding the risks and benefits
of all potential treatments, a substantial minority appears to choose expectant
management.16,17 Given that many
more men are now being diagnosed as having prostate cancer than will ever
die of it, expectant management would appear to be appropriate for some men.
More research is needed to define those subgroups of men for whom an expectant
approach is a reasonable, or even optimal, management strategy. An added challenge
will be to ensure any such insights are effectively communicated to patients
facing a treatment decision for prostate cancer.