Take a simple example: Look at how many times waiters get sick with the flu, say a million hours per year, and then multiply by the average amount of money a waiter earns - say 20 bucks an hour -- and conclude that the flu costs the restaurant economy 20 million dollars.
Well that assumes that if the waiters didn't call in sick, that an extra 20 million dollars of tips/earnings would have been available. Yet that amount is determined by the demand for meals, not availability of waiters, and when one waiter is sick, others cover for them so all the meals get served. But then the extra labor of waiters needed to cover for each other is a cost, no? Well, again no, as collectively the pool of waiters that cover for sick waiters is the same pool that call in sick -- it is some extra volatility in hours worked, but total hours worked is unchanged. So at the end of the day, it's hard to determine how much damage to the economy is created by waiters calling in sick and covering for each other. It's certainly much, much lower than what you get by multiplying hourly wages by time spent sick.
Now come to depression. Depression may be just the mental price of brains capable of risk assessment or deep thought. Like sleep is not a dead weight cost, but just the price of having a developed brain. It's not clear that Depression is "costing" us anything anymore than sleep is costing us lost labor, and it could be that a world without depression would also be a world without genius, realism, and creativity.