If you can tie a definition of "heavy drinker" to "causative factor in 1 in 10 deaths of working-age adults", I don't care whether it agrees with the common def

If you can tie a definition of "heavy drinker" to "causative factor in 1 in 10 deaths of working-age adults", I don't care whether it agrees with the common definition, it's still useful. If anything, it's the reverse - if some quantity is enough to cause that much loss of life, then it's probably "heavy".