Friday, February 03, 2012

Daily Mail and the Cochrane review on heroin prescribing



I've just sent the following email:

Dear [Daily Mail] Corrections editor

There appears to be a quite serious factual error in the piece 'More heroin for heroin addicts – is the Department of Health quite off its head?' by Kathy Gyngell online here:

I am unclear if this is only in the Mail Online or also in the print version. Regardless, in this piece the author states that:
"The last Cochrane Review on the matter concluded that, ‘No definitive conclusions about the overall effectiveness of heroin is possible’. The DoH, though, was persuaded that UK trials would prove this wrong."

Although not cited or linked directly (which, as an aside, would be useful - in fact I would suggest should be standard practice when referencing scientific literature) this appears to be quoted from the conclusion of July 2009 Cochrane study on 'heroin maintenance for chronic heroin dependents' which is available here:

http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD003410.pub2/abstract

Its conclusion in full:

"No definitive conclusions about the overall effectiveness of heroin prescription is possible. Results favouring heroin treatment come from studies conducted in countries where easily accessible Methadone Maintenance Treatment at effective dosages is available. In those studies heroin prescription was addressed to patients who had failed previous methadone treatments. The present review contains information about ongoing trials which results will be integrated as soon as available."
This study has, however, subsequently been updated twice. All Cochrane reports are subject to regular review, with any new study findings incorporated, and conclusions updated accordingly.

An updated version of the study was published in August 2010

http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD003410.pub3/abstract

In this updated version, the conclusion was also updated to:

"The available evidence suggests an added value of heroin prescribed alongside flexible doses of methadone for long-term, treatment refractory, opioid users, to reach a decrease in the use of illicit substances, involvement in criminal activity and incarceration, a possible reduction in mortaliity; and an increase in retention in treatment. Due to the higher rate of serious adverse events, heroin prescription should remain a treatment for people who are currently or have in the past failed maintenance treatment, and it should be provided in clinical settings where proper follow-up is ensured."

The statement quoted by Gyngell, that ‘No definitive conclusions about the overall effectiveness of heroin is possible’ is no longer in the conclusion.

It was reviewed/updated a second time in December 2011, and the conclusion did not change:

http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD003410.pub4/abstract

The quote used would therefore appear not be from the 'the last Cochrane review' - indeed the most recent two Cochrane reviews do not include this quote. Please bear in mind that Cochrane is regarded as the gold standard in evidence based analysis of health interventions, and thus particular care needs to be taken when quoting them in support of a particular view.

Given that the updated conclusions no longer include the quoted statement, and the meaning of the conclusion is now somewhat different to what is implied in the Gyngell piece, I feel a correction should be made and the piece updated (or removed),  along with an apology to both Cochrane Library for misrepresenting their work, and to readers for potentially misleading them.

I look forward to hearing from you

regards

Steve Rolles, etc...

Ive also pointed out the mistake in the online comments thread. 

Just to note, there are certainly important debates to be had about the cost benefit analysis of heroin prescribing (Transform have made our own contribution here for example) and, as the Cochrane report makes clear - it is an intervention that current evidence only supports for a fairly small group of problematic injectors (although some newer evidence suggest wider applicability). The contributions of Kathy and others is certainly a valid part of the debate, whether you agree with them or not.

Clearly different interventions will work for different people, and these are decisions that need to be made by doctors and their clients based on the best possible evidence. That the Government has made this particular option more available based on new evidence (something you can obviously only get by doing the trials the Mail seems so critical of) is surely a welcome development. All other debates aside, making sure treatment providers should have the widest possible range of options available, informed by the best possible evidence, is unambiguously a good thing.  

Anyway - this is really just to flag up that the Cochrane library has apparently been misrepresented in the Mail, and whatever one may feel about the heroin / drug policy debate, that is not a good thing.

Finally, if I've made a mistake about this - and these things do happen, Im happy to put up a correction and apology. 



17 comments:

David Raynes said...

For once I agree with you Steve, there is an important debate here but in trying to pick off Kathy Gyngell you are in my view pin dancing.

The issue (and you know and acknowledge this full well) is not if heroin can be used in
marginal situations for a "very few absolutely chaotic individuals" which is the
thrust of what John Strang told me was his personal view at Cumberland Lodge. This is something with which I have agreed with since the 60s and the days of Byng Spear. This is in line with Cochrane.

This is something the UK has always had to some extent, even post the end of the "British System".

That this could and might well be done in the context of
(IMPORTANTLY) other support, to help these chaotic addicts to stabilise and then to
work towards a drug free state, is accepted by most commentators.

The REAL issue is about a massive open endeed extension of maintenance heroin on the State to just about anyone who wants it, with all the extra costs that a network of centres would bring, out of a finite NHS budget and to the detriment of other NHS services. That would just be massively more expensive version of the failed "British System".

That is being covertly pushed by some people. It will mean an ever rising cost against the NHS,-
for the lifetime of these addicts. The cost is not just the cost of the heroin and the treatment centres it is also in the lost opportunity costs of those NHS resources. A cost that is rarely described, except by me.

Giving addicts heroin and keeping them addicted, preserving the dangerous injecting habit is assuredly not treatment. These addicts deserve better from our society.

We did it in the 60s, we as a country know more about the downsides than anyone else, it did not stop an illegal market.

We have seen the damage that open ended maintenance methadone has done. We must surely not repeat the errors of the past.

Evert said...

David, substance addiction is a chronic medical illness. Patient's with a chronic illness often have to take medicine to treat their condition for the rest of their lives (despite sometimes deleterious side-effects). Anti-retroviral therapy for HIV is one such example.

The Swiss have been treating heroin addiction like this for some time. According to their Office of Public Health:

"At the end of 2009, 1356 patients were undergoing HAT [heroin assisted treatment] at 21 outpatient centres and in 2 prisons..." and "Each year, between 180 and 200 patients discontinue HAT." http://www.bag.admin.ch/themen/drogen/00042/00629/00798/01191/index.html?lang=en

You argue that heroin addicts "deserve better". I agree. Some patients don’t respond to our current treatments. This makes the availability of other types of treatment (2nd line treatment) a good idea. Diamorphine is a effective 2nd line treatment for addiction to the corresponding street drug. Denying patients 2nd line treatment is inhumane.

Steve Rolles said...

I dont know anyone pushing an open ended roll out. The problem has been political restrictions on available treatments and researching new approaches. Prescribed injectables may only be appropriate for the 10% most problematic injectors but that would still be 20k - a big number compared to the 2-400 who recieve it now.

Opening up research opportunities could also mean that prescribing of non njectables could be explored - possibly oral or smoked preparations, or even more traditional opium preparations (we touched on some of this in Blueprint)

Ultimately this os about experimenting to see what works in terms of delivering the outcomes we all seek (what is being tried today is very different to what was done in the 60s so we shoudl reject that comparison). Helping people stay safe, improving health and minimising wider social harms (That obviously includes impacts on criminality associated with users and the domestic/international trade)

That is the moral course IMHO We can call it treatment or something else - its not the most important question. The inevitable debates wont be settled in the Daily Mail or on blog discussions but by ongoing scientific research and experimentation.

The reality is that demand for opiates exists - we can and should try and reduce that demand - but whilst its here we need to deal with it as effectively (and cost effectively) as possible.

Evert said...

There are already a variety of formulations:

Diamorphine nasal spray like they give to children: http://www.bmj.com/content/322/7281/261.abstract#alternate

From the Swiss Office of Public Health:

"The following dosage forms were again available in 2007:

• Diaphin®, 10 gm vials of diacetylmorphine hydrochloride monohydrate for intravenous injection.
This product was registered by Swissmedic on 7th December 2001.

• 200 mg IR tablets of diacetylmorphine hydrochloride anhydride (containing 213 mg diacetyl-
morphine hydrochloride monohydrate) with immediate release. The supplier submitted an ap- plication to Swissmedic for a sales licence in December 2005. A decision on this application is expected in 2008.

• 200 mg SR tablets of diacetylmorphine hydrochloride anhydride (containing 213 mg diacetyl- morphine hydrochloride monohydrate) with slow release. For this product too Swissmedic’s de- cision on whether to license it for sale is still outstanding."

Don't forget old fashioned drinkable preparations: http://goo.gl/TnBGD

David Raynes said...

Steve
There are many people who argue for NHS prescription of heroin, Peter Carter for one. It has been put to me in several media interviews.

Chris Mullen as Chairman of HASC argued with me about open ended maintenance years ago.

There are many and Carter is one, who have linked the idea with reducing crime.

It is in my view a very great mistake to link the idea of medical treatment with "reducing crime". That was the false prospectus sold to Tony Blair.

Medical treatment should be based on individual clinical need and because it produces the best outcome for the patient and reduces the spread of infection to society. Open ended maintenance does not do that.

It is disingenuous to pretend that no one is pushing for open ended maintenance. Open ended, unlimited maintenance creates a client state of zombies and an ever increasing bill.

It might suit the pharmaceutical companies and because it normalises drug use it might also, dare I suggest, suit Transform.

It will not suit the taxpayer, it will not suit society. We have been there, it was a failure.

Too many people forget the history.

Steve Rolles said...

Its not about what suits Transform - its about what delivers the best outcomes. Health interventions have impacts beyond the health arena - crime being an obvious one of them; particularly under prohibition; and particularly regards heroin (and crack). As the Cochrane paper highlights heroin prescribing has a positive impact on offending levels - and given the scale such offending this can be hugely signifincat social positive. Obviously there are opportunity costs to enforcement as well - something we have written about often. Other forms of crime and anitsocial activities associated with the illegal trade will also be lessened - street dealing and public injecting for example. Legal heroin demonstratbly also does not profit organised crime or cause its associated harms/costs.

Its certainly not about 'normalisation' - indeed evidence suggests the medicalisation of responses to dependent injectors has had the exact opposite effect in Switzerland and the Netherlands where they have more extensive programs.

We should learn form history certainly, but a lot has happened since the 60s and knowledge on effective responses is not frozen in time - that why we do, and should continue to do a range of research on which interventions - established or pioneering - deliver the results we all want on a range of indicators. This obviously applies to the legal frameworks in which interventions take place, as well as the interventions themselves.

I have no love of pharmaceutical companies but see them, on blalance, as preferable to transnational organised crime networks. Regulation of any commercial entities could be a strict as we deem appropriate, for certain elements of the market for certain substances deemed sufficiently risky - potentially even state monopolies.

Ann Stoker said...

It would seem to me to be cheaper and more humane in the long term to treat the "very few absolutely chaotic individuals" to long term residential rehab - with after care, housing and continued counselling. In Germany at Synanon there is open-ended rehab. for those who cannot cope with life without drugs. Some stay for four or more years - but the aim is to enable the individual to eventually live an abstinent life outside. To have heroin prescribing for all who want (not need) it is simply wrong - the costs to the NHS would be unsustainable without cutting drugs to others suffering illness like prostrate cancer.

David Raynes said...

Steve
You say:
"Legal heroin demonstratbly also does not profit organised crime or cause its associated harms/costs".

The situation is much more complex than that bald statement. We in the UK have more practical experience of heroin prescribing than any other country, in fact all the files on the British System still exist I understand.

Giving heroin away to virtually any addict who was prepared to register did not stop an illegal market, it did not stop criminal supply.

For well worn reasons you and I have debated many times, prescription/legalisation/regulation, call it what you will, cannot achieve that.

This is just part of the history from the 60s. It used to grieve Byng Spear, he initially believed what you appear to believe.

The truth is, that normalisation of heroin use & addiction and consequent easy availability on the State via NHS prescription, (as another medical condition with heroin as the appropriate treatment) CAN actually SUPPORT a wider criminal market in the long term.

First time users do not tend to get their supply from clinicians, nor do they tend to get their initial supply from the most chaotic addicts.

The heroin using social infection spreads because use is normalised.

The social pressure & taboo NOT to use heroin, is lessened in effect.

It is very simple stuff Steve.

You seem to be in a sort state of denial about some people arguing for heroin on the State for all addicts.

I assure you absolutely, I have commented on it several times because I have been asked about it several times, including by politicians of all parties.

See here for example:
http://news.bbc.co.uk/1/hi/uk/2002026.stm

A legal supply would plainly not stop deaths. For an evidence base for what I say, look at the deaths associated with (legal) methadone.

I stuck my neck out in the early 70in saying I supported heroin prescrition for a few chaotic addicts, when it was unfashionable to say that for someone in my profession. I have not changed my position one jot.

Strang told me that he broadly agreed with me. I was actually surprised at the degree of unanimity.

Despite your weasel words it does look as though you slip into "heroin for all" arguments, whenever you get the opportunity.

Given the position you get paid to promote, that is maybe understandable, but do not try to assert some moral high ground.

I said you were dancing on a pin when you had a go at Gyngell, so you were. You can expect me to jump on you whenever you try this nonsense.

Enough.

Jake said...

@Ann Stoker. You say "the costs to the NHS would be unsustainable without cutting drugs to others suffering illness like prostrate cancer". However, and maybe Steve can answer this, what is the cost per week for a prescription supply? Then look at the cost per week of prison, the police to chase down addicts and file reports, the insurance costs from associated crime etc. in obtaining the prohibited plant extract.

We currently spend around £18bn a year policing our failed drug policy, with savage Human costs (here and abroad). Surely a well managed, evidence-based (and that is the key) program for supporting addicts through prescribed a supply and compassionate health care system would be a net reduction in cost to the state i.e. money saved from policing could be directed to health care? You can't look at drugs policy in isolation regarding costs. Prohibition spawns a raft of expensive 'unintended consequences' whilst simultaneously never achieving its stated goals.

@David Raynes. If Heroin is prescribed less now than it was in the 1950/60's under the 'British System'(so more normalised back then), do we have more addicts per capita now? And if so, why?

Steve Rolles said...

"The situation is much more complex than that bald statement"

If an illegal heroin user moves to prescribed supply the profits asvailable to the illegal trade contract. The more times it happens the bigger the contraction - theres nothing complicated about that. Moving to a legal supply of any drug will obviously have an impact on the size of the opportunity available to criminals (note: the end of alcohol prohitbiion for example). How big this shift would be depends on the detail of how a given drug was made available - that involves complex questions certainly, and yes there may be other pressures operating in the opposite direction - but the basic premise is simple. I was writing a blog comment not a phd.


"We in the UK have more practical experience of heroin prescribing than any other country, in fact all the files on the British System still exist I understand."

The Swiss model of prescribing, and similar models seen in the NIOMI trials, RIOTT trials and similar programs in the Netherlands and Germany are quite different to the 'british system of the 60s'. I am certainly not suggesting a return to that system, and your expertise based on it is, Im afraid rather redundant now. A lot has happened in the last 40 years to inform the debate.Lessons have been learnt from the failing then - models have been dramatically revised and improved.

"Giving heroin away to virtually any addict who was prepared to register did not stop an illegal market, it did not stop criminal supply"

Straw man: Im not suggetsing that - by all means go and argue with someone else who is. However - since that system stopped use and the illegal trade that fules has expanded by several 1000% so Im unclear what policy are arguing for or on what basis. Prohibition certainly hasnt got rid of the illegal market. As Melanie Philips so elequently pointed out on QTime last week - the tobacco market is 20% illegal. That is of course an 80% improvement on 100% criminal. And use continues to fall due to better regulation (of a kind we cant do with illegal products)


"For well worn reasons you and I have debated many times, prescription/legalisation/regulation, call it what you will, cannot achieve that."

We havent claimed that the illegal trade would be eliminated - but that it would be substantially reduced over a period of time. Repeating something wrong, unfortuntaely wont make it correct.

"This is just part of the history from the 60s. It used to grieve Byng Spear, he initially believed what you appear to believe."

Dont tell me what I believe. I refer you to what I've written.

"The truth is, that normalisation of heroin use & addiction and consequent easy availability on the State via NHS prescription, (as another medical condition with heroin as the appropriate treatment) CAN actually SUPPORT a wider criminal market in the long term."

As I have already pointed out - the larger scale prescribing of heroin in Switzerland and the Netherlands has not had that effect. It has had the opposite effect - de-glamorising and medicalising the behaviour, falling numbers of injectors, an aging injecting population.

The opposite has happened here.

Steve Rolles said...

cont...

"First time users do not tend to get their supply from clinicians, nor do they tend to get their initial supply from the most chaotic addicts. The heroin using social infection spreads because use is normalised. The social pressure & taboo NOT to use heroin, is lessened in effect".

See previous comment - that simply hasn't happened where heroin has been more widely prescribed. You are going to have to respond to this observation with something more than:

"It is very simple stuff Steve."

< patronising comments like that.

"You seem to be in a sort state of denial about some people arguing for heroin on the State for all addicts".

Some may be (by all means go and argue with them) but not me or the Government (or Strang from what I can see). Read what Ive said.

"A legal supply would plainly not stop deaths."

Straw man - where has that been claimed? The argument is that deaths would reduce as indeed they havce wher prescribing has been increased. Worth pointing out that no one has ever died of an overdose in a swiss clinic, or contracted a BBV on one with a clean needle. No one has ever caught anthrax from prescribed heroin. Nor has anyone died in foriegn wars trying to destroy it, or been killed by gangsters trying to sell it.

"For an evidence base for what I say, look at the deaths associated with (legal) methadone".

How many of those deaths are from supervised methadone consumption? Is it more than zero? I believe that is what we are talking about here. This is another phantom arguments.

"I stuck my neck out in the early 70in saying I supported heroin prescrition for a few chaotic addicts, when it was unfashionable to say that for someone in my profession. I have not changed my position one jot."

good. well done

"Strang told me that he broadly agreed with me. I was actually surprised at the degree of unanimity."

Marvellous. I like unanimity.

"Despite your weasel words it does look as though you slip into "heroin for all" arguments, whenever you get the opportunity."

No I dont. ad hom. Read what Ive said/written.

"Given the position you get paid to promote, that is maybe understandable, but do not try to assert some moral high ground."

playground stuff

"I said you were dancing on a pin when you had a go at Gyngell, so you were."

I havent 'had a go at Gyngell' - Ive pointed out - correctly - that she misrepresented Cochrane in her column. Critical debate between positions that has happened elsewhere is natural, apporapriate and healthy part of developing policy.

"You can expect me to jump on you whenever you try this nonsense."

debate is always welcome - but please try and remain polite.

Steve Rolles said...

Anne - You make a quite reasonable point re costs, but as Jake points out CJS enforcement is expensive, and failing to respond to chaotic dependent opiate users (whichever apporach is adopted) creates massive secondary costs to society that eclipse the costs of the health responses. Ive explored some of the costs - at least made a first stab at it, in Transform's 2009 publication here: http://bit.ly/cAF7v7

In that context, I dont accept the suggestion that a single drug treatment health intervention takes place at the expense of something like cancer treatment. That said I dont think it is a helpful argument anyway - as the same could presumably be said about any Government spending (military, the arts, sports, aid etc etc). Personally I would be fine for the drug treatment budget to be ring-fenced and decisions made within that on how it was allocated.

I have absolutely no problem with residential rehab and believe it is an important option in the menu of responses. There is sometimes a mistaken sense that advocates of law reform are ideologically opposed to rehab or absitnence based treatments, or indeed that harm reduction is incompatble with them. This is simply incorrect and I hope that if you talk to some HR service providers this would become very clear.

However I do think that the choice of option or options appropriate for particular patient needs to be made by qualified proffessional and not by politicans or think tanks. Rehab doesnt work for everyone, just as maitenence isnt appropriate for everyone.

Such decisions to be informed by the best evidence and the process of collecting and reviewing that evidence is an ongoing one. The same critical scrutiny needs to be applied to all interventions.

Steve Rolles said...

Jake - Generally agree with your point but the £18billion figure you quote isnt correct. That figure is one that has been estimated for the total social and economic impacts of class A drug Use in England and Wales. The figre for enforcement within that is about 3-4 billion - which includes both pro-active supply side enforcement and reactive costs (prisons etc) across the CJS. A few people have been saying teh same thing and perhaps the Home office - who have thrown the figure around a lot (I think they use 16billion, and the Strategy unit used 20billion) have some responsibility for not making it clear what the figure represents.

See http://bit.ly/cAF7v7 for more detail.

Steve Rolles said...

apologies for typos - im not the worlds greatest typist esp when using this little netbook.

Jake said...

@Steve. Thanks, perhaps I have been better served saying that the financial costs of the prohibition policy cost society/citizens/taxpayers to the tune of around £17bn a year. Does that figure include policing Class B drugs i.e. Cannabis, as that must add another huge chunk of policing cost? Regardless, the point still stands (to which I'm sure you agree and to which the paper says), that financially, prohibition costs more than a legally regulated/healthcare regime, whilst simultaneously reducing the Human cost/suffering that prohibition causes.

The graph on page 11 is also interesting as it confirms what I alluded to regarding the number of dependant opiate/cocaine users - if prohibition serves to de-normalise drugs and treatment to normalise them, why the rise from 1971 onwards..

David Raynes said...

Steve
You say:
"If an illegal heroin user moves to prescribed supply the profits available to the illegal trade contract. The more times it happens the bigger the contraction - theres nothing complicated about that".

Er yes there is Steve and interesting that despite your earlier denial that anyone is pushing for massive expansion, to every addict. You slip in to language that tends to justify just that.

All your arguments with me over the years, have tended to be rather simplisticly based on static market size. You are still doing it.

If we can agree on one thing surely it is that markets are NOT static, demand is not static, drugs have fashions. Drug using culture changes over time.

Every step we take in drug policy needs in my view, to be judged in relation to what it will or might do, to the using culture. That is at the heart of why legalisation is such a risky strategy (see Reuter 1999).

Legal drug supply can increase the safety of an individual user episode yet fundamentally create much larger Total Harm from that drug. To coin a phrase: "There is nothing complicated about that!

And in respect of an addict given heroin by the State who was previously dealing to support their habit, there is increased opportunity for them to continue dealing, free from anxiety about their own core supply. We have had that problem before. It is an ever present risk.

The one thing that reduces opportunity for crime in relation to any drug, is a shrinking user and potential user base and even then there are other external factors which can override the effect of the shrinking user base and increase criminal opportunity.

Poverty and unemployment for example, affect the market opportunity for counterfeit & smuggled tobacco & alcohol goods.

Steve Rolles said...

Just to be absolutely clear - i said above that 'I dont know anyone pushing an open ended roll out'. This is in relation to the discussion of the Cochrane review (which doesnt talk about such a thing) and the recent Doh announcemetn which is also clearly framed as a useful secnd tier response. I dont advocate prescrbed heroin be available to anyone, and havent said that anywhere so please dont put words in my mouth. What Ive written, and linked to is very clear. If you want to argue with poeple advocating that Id probably join you. So lets move on.

re market size - the experience in countries that have significnatly expanded heroin prescribing (no comment from you on this evidence base yet) is not one of expanding markets. Infact the opposite - and health and social nuisance outcomes have also improved. There are cost benefit issues to be considered certainly in terms of impacts relative to other interventions but specifically the doomsday scenarios that you seem to be indicating simply havent appeared.

Re total harms - agreed - why we need to constantly evaluate which policies/ interventions work and which dont - as free from political interference as possible. Hence ongoing research informing decisions for the future. Im fully aware that any policy can have unintended consequences (see Blueprint chpt 4).

Essentially agree with you re poverty and unemployment - I suspect that addressing these wider social issues is where prevention efforts are most likely to deliver in the longer term(see the conclusions of our submission to the drug strategy consultation http://bit.ly/tHbOe5 ). That is ofcourse a much wider policy challenge.