Physician reimbursements for androgen suppression
therapy were reduced by 64% between 2004 and 2005 as a result of the 2003
Medicare Modernization Act. According to recent data, this deduction in
reimbursements resulted in a reduced rate of men overtreated for low-risk
prostate cancer without reducing the number of needed treatments.
“A major reduction in physician reimbursement for
[androgen suppression therapy] was associated with a 39% decrease in the odds
of receiving non-indicated [androgen suppression therapy] but not indicated [androgen
suppression therapy],” the researchers wrote. “These findings may
help inform how payment changes will affect health care utilization in other
disease models.”
Using the SEER database, researchers identified 72,818
men who were diagnosed with prostate cancer between 1992 and 2005. According to
Medicare claims data, indicated treatment was defined as 3 or
more months of androgen suppression therapy in the first year among men with
metastatic disease (n=8,030); non-indicated treatment was defined as androgen
suppression therapy alone, with no radical prostatectomy or radiation, in those
with low-risk disease (n=64,788).
According to the researchers, the unadjusted annual
proportion of men receiving therapy was plotted against the median Medicare
androgen suppression therapy reimbursement. In addition, logistic regression
models for metastatic and low-risk cohorts were developed; the models included
three different definitions of androgen suppression therapy use: at least 1
month, at least 3 months and at least 6 months. Calendar year of payment change
was the predictor of interest; additional covariates included age in 5-year
categories, clinical tumor stage, WHO grade, Charlson comorbidity, race,
education, income and tumor registry site.
In 2003, androgen suppression therapy among patients in
the metastatic group peaked at 64% and decreased to 58.5% in 2005. In the
low-risk group, between 1999 and 2003, therapy use was stable, then peaked at
10.2% in 2003; use declined to 7.1% in 2004 and 6.1% in 2005. Compared with
2003, the odds of receiving non-indicated therapy in 2004 and 2005 decreased
significantly after adjusting for tumor and demographic covariates (OR=0.70;
95% CI, 0.61-0.80 in 2004 and OR=0.61; 95% CI, 0.53-0.71 in 2005). According to
the researchers, therapy use was stable at 60% during the payment change in the
metastatic group. Between 2004 and 2005, the OR for receiving therapy was
unchanged among this group.
In an accompanying editorial, Nancy L. Keating, MD,
MPH, Brigham and Women’s Hospital and Harvard Medical School, Health
Care Policy, wrote: “This work adds to findings from a prior study
demonstrating that actual Medicare payments to physicians for
[gonadotropin-releasing hormone] agonists decreased by 65% between 2003 and
2005. The nice contribution by Elliot et al was in demonstrating that the
decreased reimbursement for [gonadotropin-releasing hormone] agonists was
associated with a substantial decrease in their use for an indication that very
likely reflected overuse (primary therapy for very low-risk tumors) but no
change in use for an indication that reflected appropriate use of this
therapy.”