Solution
= Solution
Let $A_i$ be treatment, $Y_i=(Y_{i0},\ldots,Y_{i,12})$, and $R_i\in\{0,\ldots,12\}$ the last observed visit. The <missing at random> assumption is
$$
\boxed{
\mathbb P(R_i=r\mid A_i,Y_i)
=\mathbb P(R_i=r\mid A_i,Y_{i0},\ldots,Y_{ir})
}
$$
for every feasible $r$. Equivalently, each conditional dropout hazard may depend on treatment and observed QoL history but not on current or future unobserved QoL after conditioning on that history.