Abstract: PD32-03
Sources of Funding: This work was supported by the American Cancer Society (RSG 12-323-01-CPHPS), the National Cancer Institute (R01 CA168691, R01 CA174768, T32 CA180984) and the National Institute on Aging (R01 AG048071).

Introduction

Prostate cancer is the most common and among the most costly cancer in US men. Uncertainties regarding optimal management lead to treatment variations and increase cost. Accountable care organizations (ACO) can potentially improve care by decreasing variation (i.e. avoidance of treatment in low value settings) and constraining costs. Our objective was to determine the effect of Medicare Shared Savings Program (MSSP) ACOs on prostate cancer care.

Methods

Using a 20% Medicare sample we perform a retrospective cohort study of men with newly diagnosed prostate cancer from 2010-2013. We assigned patients to ACOs based on their primary care provider's MSSP ACO participation. We then performed a difference-in-differences analysis comparing the impact of ACO participation on initial prostate cancer curative treatment, treatment of men with a very high 10-year non-cancer mortality risk (i.e. those least likely to benefit) and per beneficiary payments. Outcomes in the post-implementation period were compared to outcomes in the pre-implementation.

Results

We identified 33,461 men with incident prostate cancer of which 5,015 (15%) were assigned to an ACO. Overall, 58% of men were diagnosed in the pre-ACO implementation period. We noted secular trends in the non-ACO group from pre- to post-implementation in overall curative treatment (4.2% decline, p<0.001), treatment of men with the highest non-cancer mortality risk (6.2% increase, p=0.11) and annual per beneficiary payments 4.0% decrease (p<0.001). ACO participation had no significant effect beyond the secular trend (Figure) on overall treatment or annual payments (difference-in-differences estimator p=0.8, p=0.09, respectively). There was a significant relative decrease in treatment among men with the highest mortality risk of 17% (p=0.03), however did this not lead to differences in cost.

Conclusions

Curative treatment of prostate cancer and annual per beneficiary payments decreased significantly between 2010 and 2013. For men diagnosed with prostate cancer, ACO participation did not impact trends in treatment or cost. However, among men least likely to benefit, ACOs resulted in a decline in treatment of prostate cancer.

Funding

This work was supported by the American Cancer Society (RSG 12-323-01-CPHPS), the National Cancer Institute (R01 CA168691, R01 CA174768, T32 CA180984) and the National Institute on Aging (R01 AG048071).