Difference-in-difference regression with extra variable

Hi all,

I'm working on my thesis and I want to use a difference-in-difference model. To explain in brief: I am comparing the effect of child care support (WKO) on the effect on labor supply (EMPL) on single- and two-parent households. Parents are not eligible for receiving this benefit anymore once their child is 12, making households whose youngest child is 12 or older the control group (CHILD12). Without incorporating WKO or household composition, I would get this model:

EMPLdt = a + b * CHILD12dt + y * REFORMt + s (CHILD12d * REFORMt) + Edt

I want to somehow incorporate the actual loss in the amount of WKO (budgets were cut, so people received less), and not just the before/after reform dummy. Aside from that, I want to control for household composition. Can anyone explain how I can put these two variables in my regression?