Yes, he is known for this graph (The Lancet, 2010): https://en.wikipedia.org/wiki/David_Nutt#/media/File:HarmCau...
Some of the criteria are that might fall under harm to others are:
Crime
Environmental damage
Family adversities
Economic cost
Community
As far as I know.... unlike the US, the UK doesn't have a large community of dependent meth users. The level of distribution is low and if some did become available, it would probably be used by poly-drug users as an alternative/combination to other party drugs and not give much of an on-ramp to habitual use. The measurable harm would be acute health events like heart issues etc. and not the consequences of long-term use.
I assume he was sacked for producing such a poor data visualization, or just the political "data" behind it?