NEJM: Inappropriate use of prostate cancer therapy drops as Medicare funding plummets
As Medicare reimbursement for androgen-deprivation therapy (ADT) to treat prostate cancer was slashed in half in 2005, inappropriate use of the therapy fell by 13 percent while appropriate ADT for prostate cancer held steady, according to a study published Nov. 4 the New England Journal of Medicine.

ADT use as a hormone treatment for patients with prostate cancer doubled in the 1990s, with nearly half of all patients with prostate cancer receiving the therapy either as primary or adjuvant treatment. In men 80 years or older with low-risk or localized prostate tumors, patients who "would almost certainly be asymptomatic and die from causes unrelated to prostate cancer," ADT treatment increased eightfold, according to Vahakn B. Shahinian, MD, at the University of Michigan in Ann Arbor, and co-authors.

Shahinian and co-authors conjectured that this dramatic increase in ADT use was largely influenced by Medicare's high-reimbursement rate for the therapy, especially in private practice, fee-for-service urology, as studies have found that patients of private practice physicians are 60 percent more likely to receive ADT than those receiving treatment in academic departments when benefit is uncertain.

The authors aimed to measure the trends in ADT use by physicians between 2003 and 2005, following Medicare's 50 percent ADT reimbursement cut as part of the Medicare Modernization Act of 2003.

Shahinian and colleagues evaluated the treatment of 54,925 men enrolled in Medicare who were diagnosed with incident prostate cancer between 2003 and 2005. Information was pulled from the Surveillance, Epidemiology, and End Results (SEER) Medicare database, and patients were divided into groups according to appropriate use, discretionaryl use or inappropriate use of ADT.

Patients in whom ADT treatment was considered inappropriate consisted of men who received no radiotherapy or prostatectomy and were diagnosed with T1 or T2 tumors and low to moderate grade Gleason scores, which together indicate slow disease progression as well as small and low- to moderate-risk tumors. The authors argued that no clinical evidence of ADT efficacy exists in these patients and no reasonable basis for expecting benefits from ADT existed.

Appropriate use of ADT was as an adjuvant therapy for T3 or T4 tumors, where clinical efficacy had been demonstrated and alternative treatments were limited. The discretionary use group consisted of patients in whom ADT appropriateness was less certain, and analysis of trends within this group was more limited.

Of the 54,925 patients diagnosed with prostate cancer between 2003 and 2005, 43.2 percent received ADT within six months of diagnosis. The rate of appropriate use of ADT grew from 40.7 percent in 1994 to a peak of 85.1 percent in 2004. Administration of ADT to the inappropriate use group also increased, from 30.1 percent in 1994 to a peak of 44.9 percent in 2002. However, beginning in 2004 and through 2005 inappropriate use fell nearly 10 percent to 35.5 percent.

The appropriate use group experienced no decline in treatment with ADT between 2003 and 2005, while the discretionary use group experienced a gradual decline in 2004 and a larger drop in 2005, falling from approximately 70 percent usage to 60 percent (rates differed slightly for discretionary use of ADT as a primary versus adjuvant therapy).

The odds of receiving ADT for all patients dropped nearly 20 percent between 2003 and 2005, with an odds ratio of 0.82. "For the inappropriate-use group, there were substantial and significant decreases in the odds of use both in 2004 and in 2005, with a nearly 30 percent decline in 2005, as compared with 2003," Shahinian and colleagues observed.

No significant decrease in the odds of patients receiving appropriate or discretionary ADT occurred.

"We found a substantial decline in the use of ADT in close association with reductions in reimbursement for GnRH agonists [the most common form of ADT] in 2004 and 2005. Notably, reductions were most dramatic for indications that were not compatible with available evidence of efficacy," the authors noted. "These findings are consistent with previous research on the influence of financial incentives on the delivery of health care. Financial incentives are most likely to have an effect on physicians' behavior in cases in which medical uncertainty exists, as opposed to cases in which care is clearly lifesaving."

Shahinian and colleagues also found that inappropriate use of ADT was more common in black and Hispanic men, areas with high incidences of poverty and in unmarried men. Married men and areas with higher education rates and lower poverty rates were more likely to receive appropriate ADT.

While acknowledging that ADT administration to prostate cancer patients may simply have changed during the study period, the authors said controlling for this by including any ADT within a 6 month period should obviate such an effect. Shahinian and co-authors did say that the association between falling ADT and Medicare cuts may be confounded with a 2005 study that correlated bone fractures with ADT administration.

"Our findings suggest that reductions in reimbursement may influence the delivery of care in a potentially beneficial way, with even the modest changes in 2004 associated with a substantial decrease in the use of inappropriate therapy. The corollary is that reimbursement policies should be carefully considered to avoid providing incentives for care for which no clear benefit has been established."

The authors concluded by saying that "[t]he extreme profitability of the use of GnRH agonists during the 1990s probably contributed to the rapid growth in the use of ADT for indications that were not evidence based."