EURURO Vol. 71 No. 5

735 844

receive RP (AOR: 1.03, 95% CI: 1.01–1.05, p = 0.003). Higher
income was also associated with receipt of RP (AOR: 0.84,
95% CI: 0.83–0.86, for areas averaging < $30 000/yr vs $46
000, p < 0.001).
A second multivariate logistic regression model was
constructed for low-risk patients to identify factors associ-
ated with receipt of observation versus definitive local
therapy (RP, brachytherapy, EBRT, and cryotherapy; .
Advancing age was again a strong predictor of therapy choice
with an 8% increased odds of observation for each year of life
(p < 0.001). [26_TD$DIFF]Non-White race was also associated with a
higher odds of being observed (AOR: 1.26, 95% CI: 1.21–1.31,
for blacks and AOR: 1.08, 95% CI: 1.00–1.15, for Hispanics vs
Whites, p < 0.001 and 0.04, respectively). Lack of insurance
or insurance throughMedicaidwas similarly associatedwith
an increased odds of receiving observation (AOR: 2.87, 95%
CI: 2.61–3.15, and AOR: 1.88, 95% CI: 1.72–2.06, respectively
vs private insurance, p both < 0.001). [27_TD$DIFF]Income [28_TD$DIFF]was [29_TD$DIFF]again
[30_TD$DIFF]associated [31_TD$DIFF]with [32_TD$DIFF]treatment [33_TD$DIFF]choice with [34_TD$DIFF]increased odds of
observation[35_TD$DIFF] in those patients living in the lowest income
areas (AOR: 1.06, 95% CI: 1.01–1.11 for areas averaging
<$30,000 per year vs. $46,000, p = 0.02).
4. Discussion
Several novel findings from this study are worthy of
discussion. One of the most striking of these is the recent
increase in the receipt of RP across risk groups. Several
possible reasons for this change are likely. One may include
a perception by urologists that RP is the most ideal
treatment for most patients with prostate cancer. While
emerging data from some large retrospective series show a
potential benefit to RP over radiation therapy, obtaining
firm conclusions from such studies is difficult given the
possibility of unidentified confounding factors .
Indeed, a recent Agency for Healthcare Research & Quality
report suggests no strong evidence exists to favor one
treatment for localized prostate cancer over another
. The strong association between higher socioeconomic
status and receipt of prostatectomy in our study suggests
that economic concerns may be at least partially driving this
trend, perhaps even more than other clinical factors.
Another possible explanation is the recent rapid dissemi-
nation of robotic surgical platforms. Use of such platforms is
strongly associated with privately-insured patients and has
been shown to increase surgical volume and overall health
cost in locations that adopt them .
Recent rapid changes in patterns of care are perhaps
most clearly seen in patients with low-risk prostate cancer,
a subgroup for whom observation is often appropriate.
Rates of observation for low-risk patients in our study
increased significantly during the study period; however,
during this same period receipt of RP also increased. This
suggests that it is perhaps those low-risk patients that
would have previously received radiotherapy that are now
preferentially being observed. This conclusion is supported
by our data showing that in the oldest subgroup of patients
utilization of observation continues to climb while utiliza-
tion of radiotherapy falls off. Given the long natural history
of low-risk prostate cancer, this finding may be wholly
appropriate. Despite this, the rates of observation seen in
our study are significantly lower than those seen in other
studies . This discrepancy might be explained by more
strongly represented differential use of this approach in
particular geographic regions in the databases used in these
previous studies. All of these studies do point to a slower
adoption of observation in the USA compared with some
European countries
Another striking finding from our study is the significant
drop in the use of brachytherapy across risk groups. While
decreased use of brachytherapy monotherapy for those
patients with unfavorable intermediate-risk or high-risk
disease is in line with practice guidelines, the decrease seen
in low-risk patients remains unexplained. Brachytherapy
has been shown to be one of the least costly forms of
treatment for localized prostate cancer Given this,
economic incentives may again be at play. Alternatively,
decreased training of radiation oncologists in performing
brachytherapy implants may also be contributing
Brachytherapy has also recently received negative media
coverage due to several recent incidents of poorly performed
implants leading to significant complications, which may
have affected patient willingness to accept this form of
treatment
Radiation therapy, particularly EBRT, has historically
been the dominant form of treatment for patients with
high-risk prostate cancer; however, in our study it has been
overtaken by RP as of 2010. While the combination of
radiation therapy and ADT has been shown to be superior to
ADT[36_TD$DIFF] or radiation alone for high-risk patients, RP has not
been compared in a similar randomized fashion . De-
spite this, local therapy is of great importance for this
subgroup, even for patients with very advanced disease
. Previous studies from the CaPSURE database
identified high, although falling rates of primary ADT use
in this population . In our study, rates of primary ADT
were low and on the decline for patients with high-risk
disease. This discrepancy may be explained by differences
in center types captured by the two databases (with the
NCDB containing a higher proportion of large academic
centers vs community practices) or differences in how
primary ADT was defined (in our study, ADT followed by
some other form of treatment within 6 mo was not
considered primary ADT).
The strong association between socioeconomic status
and choice of RP over radiation therapy is of both clinical
and economic interest. Similar patterns are seen even in
countries with single-payer healthcare systems . It is
unclear from our study whether the lower rates of surgery
in those with markers of poor socioeconomic status
represent underuse in this population, an overuse in those
more socioeconomically advantaged, or a mixture of factors.
Regardless, these data signal better access to a broad range
of healthcare providers is needed for disadvantaged
populations. Similarly, historically disadvantaged popula-
tions with low-risk prostate cancer also appear to have a
higher likelihood of being observed. While active surveil-
lance is a good choice for many patients with low-risk
E U R O P E A N U R O L O G Y 7 1 ( 2 0 1 7 ) 7 2 9 – 7 3 7 735