I like David Nutt (see his old blog and new one), well I've never met him personally but I like his work. Dr. Nutt was the head of the UK's Advisory Council on the Misuse of Drugs (ACMD) until he criticised the government's arbitrary classification scheme. The classification scheme in the UK is similar to the one's used in the US and world over, which means it inexcusably leaves out alcohol and tobacco while criminalizing basically everything else. Dr. Nutt's great sin, for which he lost his job, was to publicly question the relationship between the actual harms of drugs and the law. His dismissal occurred right after he released this pamphlet.
Nevertheless I do have a bone to pick with David Nutt. Specifically two articles he co-authored that were published in the medical journal The Lancet. The first, published in 2007, claims to rank drugs according to the harm they potentially cause the user. The second, published in 2010, ranks the harms done to users and greater society and then adds the values producing a value supposedly representing the total harm done to society.
Now this is not an easy task and necessarily simplifies confounding variables. For example it does not distinguish pharmacokinetics, especially with regard to route of administration. The article calculates the values based on the way the drugs are usually administered. For heroin they assume this means by injection. I challenge this assumption but more important than that is this line:
However, direct comparison of the scores for tobacco and alcohol with those of the other drugs is not possible since the fact that they are legal could affect their harms in various ways, especially through easier availability.
The major problem with their analysis is that they compare illicit drugs with licit ones but fail to acknowledge the harms caused by prohibition. They correctly note that a legal market may mean easier availability (its hard to see how providing other drugs in a free-market context would not result in increased use given basic economic principles of supply and demand), however the authors do not take into account how a drug's illicit status affects its harms. Heroin users do not know what dosage a given preparation is, nor do they know what it may be adulterated with. Furthermore the social effects of a given drug are greatly impacted by its legal status and artificially high price. This leads to acquisitive crime to fund a habit as well as social costs due to law enforcement.
The long term health effects of opiate drugs on the human body are very mild. Aside from being addictive, there are very few long term health consequences from using opiates even over long periods. The scale lists 'street methadone' as less harmful than heroin. This is ironic since methadone is far more dangerous than heroin. Allow me to explain.
Methadone is more potent than morphine, up to 6X as potent orally. By itself this means nothing as you would just take a lower dose to compensate. However methadone also takes a long time to reach peak blood concentration, 2-3 hours. Because real education about opiates is virtually nonexistant, what happens is people underestimate the strength of methadone and the time needed for it to 'kick in.' Let's compare methadone to percocet. Suppose someone assumes they're relatively equal in strength and is used to getting high off 20mg oxycodone (orally) so they take 20mg of methadone. Oxycodone begins working within 20 minutes so after half an hour this individual isn't feeling anything so they take another 20mg. After an hour they still aren't feeling anything so they take an additional 20mg. After 2 and a half to three hours the methadone really begins to work and it totally kicks this individual's ass. Unfortunately they not only misjudged the strength of the original dose but also took two booster doses. Can you say overdose?
In the figure above the various factors that were weighed are represented in different colors and summed to get a total value of harm. It is important to note two things about heroin, first its total score is below alcohol and second the largest contributors to heroin's harms are crime and drug-related mortality. Both crime and drug-related mortality are both products of prohibition. As I have states many times before opiates are actually very gentle to the user's health. It is very hard to make the case that they are pathogenic at all. Unlike alcohol or tobacco, it is hard to truthfully say (though the truth never stops people from doing so) that opiate users are "slowly killing themselves," which is in itself a vastly oversimplified analysis of drug use and how it fits into the lifestyle of the user. For example there are shamans in South America who systematically addict themselves to tobacco, the acquired nicotine tolerance allows them to take massive doses that induce visions. Even if the use of tobacco reduces their life expectancy how much value do you assign to the role of tobacco use in the religion of the user and in the larger context of the culture?
The crime aspect of heroin is due entirely to prohibition and its "tax" on black market opiates. If anything opiates generally have a calming effect on users and so would be expected to reduce crime. Furthermore alcoholics were often treated with morphine in the 19th century, doctors judging that an addiction to opiates was less harmful to the alcoholic and society at large than an addiction to alcohol. The largest drawback to an addiction to opiates is their dependence potential, though I can think of no reason why methadone should have less dependence than heroin. If anything experience indicates that methadone is harder to kick than heroin.
My point is that it is nearly impossible to make real comparisons of licit and illicit drugs. While I greatly admire Nutt's work and statements against drug prohibition, the papers in the Lancet are drawing a false equivalency between drugs. There are two approaches to looking at the true effects of opiates in an age of prohibition. The first is at opiate maintenance programs, particularly heroin maintenance as used in Europe. The second is to compare the social situation prior to the criminalization of opiate users to the modern prohibition regime. In both cases the welfare of the users as well as prohibition-caused crime were greatly reduced. A completely legal market in narcotics, far from a nightmare, is the answer to the most pressing problems facing society around these drugs.
We saw no clear distinction between socially acceptable and illicit substances. The fact that the two most widely used legal drugs lie in the upper half of the ranking of harm is surely important information that should be taken into account in public debate on illegal drug use. Discussions based on a formal assessment of harm rather than on prejudice and assumptions might help society to engage in a more rational debate about the relative risks and harms of drugs. 
 Development of a rational scale to assess the harm of drugs of potential misuse
Prof David Nutt FMedSci,Leslie A King PhD,William Saulsbury MA,Prof Colin Blakemore FRS
The Lancet - 24 March 2007 ( Vol. 369, Issue 9566, Pages 1047-1053 )
 Drug harms in the UK: a multicriteria decision analysis
Prof David J Nutt FMedSci,Leslie A King PhD,Lawrence D Phillips PhD,on behalf of the Independent Scientific Committee on Drugs
The Lancet - 6 November 2010 ( Vol. 376, Issue 9752, Pages 1558-1565 )
Thanks to anonymous commenter for the pdf links.