Post-Multicriteria Decision Analysis (MCDA) analysis of drug-related harms – Prof David Nutt

Posted on: June 18, 2019
Last Updated: June 19, 2019

David Nutt is currently the Edmond J. Safra Professor of Neuropsychopharmacology and director of the Neuropsychopharmacology Unit in the Division of Brain Sciences. In addition to several prestigious appointments in the past, David is currently Chair of DrugScience (formally the Independent Scientific Committee on Drugs (ISCD) and President of the European Brain Council.

He has edited the Journal of Psychopharmacology for over two decades and acts as the psychiatry drugs advisor to the British National Formulary. David  broadcasts widely to the general public both on radio and television including BBC science and public affairs programmes on therapeutic as well as illicit drugs, their harms and their classification.  He has published over 400 original research papers, a similar number of reviews and books chapters, eight government reports on drugs and 27 books.

Author Quotes:  

The point about MCDA and this is the critical aspect, it allows us to compare (in the case of drugs) harms which occur in such different aspects of life that the metrics are fundamentally different. So, you are going from the economic metric pounds lost, cost of drugrelated behaviour, to a completely separate measure, say the likelihood of dying.

MCDA is without question the most powerful way of assessing something like drug harms, but it’s also the most powerful way of understanding what you know about any topic.

Alcohol is the most harmful drug to other people [vs the user] because it is widely used and has such huge economic costs – costs in terms of personal violence, social violence, healthcare costs, etc.

Summary and slides:  

Professor Nutt begins this segment by describing his work on the analysis of the MCDA drug model.

He explains that decision conferencing has become the standard way of dealing with very complicated questions (e.g. how to deal with nuclear waste). 

He discusses how each of the 20 drugs was ranked on each of the 16 parameters of harm. Each parameter was measured on a zero-100 scale to determine the most harmful drug and to provide a ratio scale by which all other drugs could be measured for harm. 

An example is given of two separate measures of harm; drug-specific mortality and drug use when repeated over time (i.e. chronic use of a drug). Moving on, Professor Nutt presents the criteria used for the weighting.

Prof Nutt concludes this video excerpt explaining the findings that alcohol, which had only been fourth in the previous ninepoint scale, came out as the most harmful drug in the UK and how these results correlate with European data. 

Take Home Points:  

  • As all variables (i.e. 16 parameters of drug harms) are on a ratio scale and not absolute numbers, this allows comparison between all drugs being scored. 
  • Weighting is group specific, probably culturally specific, and likely related to your chosen professional groups. 
  • A repeated study by European scientists revealed results supporting the use of MCDA analysis as an extremely robust process. 
Rate this post