Investigators evaluated demographic and geographic patterns in the use of ranibizumab versus bevacizumab for newly diagnosed neovascular AMD among Medicare beneficiaries in the US.
Study design
The cohort comprised 195,812 beneficiaries who had Medicare coverage for at least 12 months prior to filing an index claim for wet AMD, and for at least 6 months thereafter. All claims took place between July 2006 and June 2009. The authors assessed the link between demographic factors and the first treatment decision (ranibizumab vs. bevacizumab) after initial diagnosis. The zip code for each index claim was linked to 1 of 9 census regions in the US.
Outcomes
The odds of receiving ranibizumab as a first treatment declined over time (OR 0.39 in 2009 vs. 2006) and was lowest among Blacks (OR 0.55 vs. non-Blacks) and individuals enrolled in both Medicare and Medicaid (OR 0.67).
Four factors increased the odds of receiving ranibizumab: urban areas (OR 1.12 vs. isolated rural towns), higher median incomes (OR 1.18 vs. lower incomes), New England and East-South-Central regions (OR 5.57 and 3.58, respectively, vs. Pacific census region), and preexisting cerebrovascular disease (OR 1.08).
The overall proportion of beneficiaries who received ranibizumab as a first treatment was 35%, but this percentage varied considerably across the US (range 0.9%–84.6%). Approximately 87% of individuals treated with bevacizumab and 76% of those treated with ranibizumab continued with the same medication over the 3-year study period.
Limitations
This retrospective study relies on administrative claims data, which may have been skewed if coding errors led to misclassification. Although investigators tried to identify only new cases of AMD, some preexisting cases may have been inadvertently included.
Furthermore, the impact of aflibercept on treatment patterns could not be ascertained because this study covered claims filed between 2006 to 2009. Bevacizumab costs significantly less and is used off-label, but the models used in this study could not account for the influence of these factors.
Clinical significance
This study found that the odds of Medicare beneficiaries receiving ranibizumab vs. bevacizumab as an initial therapy for neovascular AMD varied across demographic and geographic lines.
These findings raise significant questions: What factors drive variations in the use of these interventions? Do these variations reflect differential access to interventions? Do these differences impact patient outcomes? Future studies, especially those that include aflibercept, are needed to explore these important issues.