A clinical trial compares a new treatment vs. control using a two-group design. The effect size (Cohen's d) is 0.6 (medium-to-large), and the researchers want power = 0.85 at α = 0.05 (two-sided). Approximately how many participants per group are needed?

(Use the approximation: $n \approx \frac{2(z_\alpha + z_\beta)^2}{d^2}$ where $z_{0.05} = 1.96$ and $z_{0.15} = 1.04$)