If I’m reading it correctly, the actual study found decreases in drinking days and drinks per day for all groups during the holidays compared to non-holiday periods. Heavy drinking days was the only measure that was worse on holidays for the placebo group.
The original study was also pretty underwhelming to me. Again, if I’m reading it correctly, it found an advantage of about 10 days over a 6 month period for the drug over placebo. It didn’t report on other measures and was done by researchers who receive funding from the manufacturer. It also framed its support as a public health measure (I read drunk driving reduction, though they could also be referring to medical problems.) That, combined with the intimation of court ordered Vivitrol in the commentary smacks a little too much of methadone arguments based on crime control and reduced drug use. Reduced use is a fine step in the right direction, but we need to be wary of success being redefined downward.