DODGY DOSSIERS OF ADDICTION NON-TREATMENT

In the week of the Labour Party Conference, Jack Straw publicly called for heroin on the NHS – based on publicity for a project called Riott which had not been independently evaluated nor peer reviewed (click here for critique). 

The Goliath of Labour’s National Treatment Agency for Substance Misuse, to guard its large well-fed body, also funded a document – Effectiveness of community treatments for heroin and crack cocaine addiction in England: a prospective, in-treatment cohort study – proclaiming its own effectiveness, deflecting attention from the fact that it has failed to set up a treatment system in this country offering drug-free lives. Professor John Strang was named as lead researcher on the first document, his long-time protégé John Marsden on the second. Both have rich links with the NTA.

Labour’s ‘dodgy dossier’ led this country into war and an unacceptable loss of lives – but the loss of lives which could be incurred due to questionable addiction-treatment ‘research papers’ could be many multiples of the lives lost in the Iraq and Afghanistan wars combined.

INFLUENCING THE MEDIA

Neither flaws in the contents of Effectiveness of community treatments... nor the intertwined interests were spotted by the medical journal Lancet, which published the NTA document in what appears to be a leap of faith.  Due to the Lancet ‘s prior reputation, national and international media picked up the story in good faith. Yet seven of the report authors are employed by the NTA, the other two receive benefit financially from it.

“Scientists: drug treatment does work. Therapy or counselling help heroin and crack addicts to quit their habit,” wrote Jeremy Laurance in the Independent.  Even the BBC took the bait: “Drug therapy record encouraging,” it states its website. So, too, did Sky: “A recent study has shown that there is indeed a way out for users of crack cocaine and heroin”. Medical News Today and academic health-publication websites around the world carry a similar message.

Well... we have seen a “way out” of drugs for those getting rehab and day services based on similar psychosocial and mutual-aid principles. Sadly, under the NTA’s regime only about 2% of patients manage to get into rehab and the ilk, with a similar percentage becoming drug free. So why, instead of improving the numbers of patients onto this drug-free path, would the NTA publicise a more expensive, unproven route of non-treatment?

“Two-thirds of heroin and crack cocaine addicts on drug treatment programmes either abstain or substantially reduce their use of street drugs during the first six months, according to the biggest study of outcomes ever attempted,” lauded Sarah Boseley in the Guardian.

It builds on our hopes – if only that was true!  We would embrace the news with joy. But, as with Labour’s other media-grabbing dossier, all is not what it seems: we must look at both the content and the interests behind the report.

FIRST NOTE OF CAUTION

In a paper accompanying Effectiveness of community treatments... Dr Thomas McLellan, deputy of the White House Office of National Drug Control Policy, warned that short-term interventions might not produce long-lasting, positive effects. "It may be more reasonable to expect enduring improvements through sustained outpatient clinical management with drugs and behavioural therapies – like expectations we have for the management of diabetes and hypertension."

CONFLICTS OF INTEREST

Seven of the nine authors in the NTA document  – Effectiveness of community treatments for heroin and crack cocaine addiction in England: a prospective, in-treatment cohort study – are employed by it and the other two received funding from the NTA, whose employees also designed the study, gathered data and interpreted it. Study ‘treatment’ settings depend on the NTA for funding. The TOP form (Download TOP form) measuring the results, which feeds into the faulty National Drug Treatment Monitoring System from which results were extracted, was created by the report’s lead author as a commission for the NTA some years ago; we have not had time to confirm the suggestion that the NTA paid more for it than the yearly salaries of all Addiction Today’s staff combined. So the report seems envisaged to make the NTA look good, and divert attention away from hard questions about non-treatment.

“This is scandalous! The Lancet, which accepted the piece and thus provided a base for the NTA to hand it to newspapers, should be asked if this is its standard practice. This is like a tobacco company funding a piece of research where the authors are in their company,” complained one professor to Addiction Today.

The most accessible financial relationships can be viewed in the NTA Accounts (Download NTA annual accounts 2008-9): it gave £577,000 to Strang’s Riott, and discloses a “material transaction” of £419,696 as ‘consultancy’ between the NTA and South London and Maudsley (SLAM) NHS Foundation Trust. “Disclosures in accounts are due to the fact such transactions are between bodies that are legally related,” explained an accountant. Strang is clinical director of SLAM. Marsden works at the National Addiction Centre in the Maudsley Hospital, where other authors also gain work.

Is there also a financial relationship between the NTA and the Lancet’s unnamed report reviewers?  If the process was transparent, and their names given, we could answer that question.

“Research is at its best when it is dispassionate from any specific set of industry or sectoral interests,” commented Professor Neil McKeganey, professor of Drug Misuse Research at the University of Glasgow. “It is striking in this context that employees of the study-sponsor [NTA] ‘contributed to the study design, data analysis, data collection, data interpretation and writing of the report’. The NTA is the agency responsible for funding most drug-abuse treatment in England and for overseeing the effectiveness of drug treatment provision. It is remarkable that it should have been so deeply involved in all aspects of this effectiveness assessment.

"If one were to substitute the NTA initials for the name of a drug company in the statement above, it is questionable if such a paper would have been accepted by such a prestigious journal as the Lancet. The fact that seven of the nine authors on this paper are either employed by or have received funding from the NTA should have sounded a note of caution.

“And both the initial assessments of drug-use behaviour and the follow-ups to assess the effectiveness of drug treatment agencies were collected by drug treatment agency staff themselves – who will have had a particular interest in the judgment as to the effectiveness of the services they were working within.”

Perhaps the peer review should have been more rigorous? “Peer review has long been criticised for failing to identify flaws in research,” stated Professor Peter Bacchetti in the British Medical Journal (BMJ 2002;324:1271-1273 ( 25 May ). “The judgment of colleagues plays a critical part in how grants are distributed, journal articles are selected, and careers are formed. Yet this system may raise ethical dilemmas,” states a Columbia University article in 21stC on Ethics in Research.  

“Controversies regarding potential conflicts of interest have led some journals to re-examine their rules for revealing the financial relationships of published researchers,” wrote the New Atlantic Journal of Technology & Society. “Some notable journals concerned about fraudulent research have reportedly improved the screening of manuscripts under consideration, in an attempt to catch those who would misrepresent or ‘beautify’ their data.”

“CREDIBILITY OF NTA CALLED INTO QUESTION AFTER DRUG STUDY”

This was the title of a news-agency release Addiction Today received on 5 October. “Last year the success rate of the addiction services was just 3.6% [getting into rehab, off drugs]. However, this year the NTA published a study in the Lancet medical journal [2.10.09] claiming two thirds (66%) of people stopped using these drugs within six months of treatment initiation. Either the addiction services have dramatically increased their powers of persuasion or the study is an utter sham,” wrote Tony Wilks of London- and Manchester-based New Ways Clinic.

“The information in the Lancet is flawed on many levels. The truth is, a few addicts might have temporarily stopped or reduced their use of ‘street drugs' only, because they transferred to the more-addictive but legal methadone – but they were still opiate dependant. Despite the additional drugs provided by the state at taxpayers’ expense, the vast majority were still street-drug dependent as well, at the end of the study,” the news-agency release continued.

“The only thing this study proves is that NTA treatment programmes do not work. They treated 14,656 addicts over six months and not a single person is reported as leaving the programme drug free. It is a waste of tax-payers’ money.”

Critics of the NTA-funded report believe it is desperate to keep its £20million funding (the vast bulk for bureaucracy) and its power to approve budgets of an estimated £800million a year. And senior NTA staff were given a 10% salary bonus if they were shown to meet targets on paper. A radical overhaul” or “total dismantling” of the NTA has been discussed among professionals and politicians alike.

QUESTIONS ON FINDINGS...

I can see no mention in the NTA Accounts (Download NTA annual accounts 2008-9), which must be formally submitted to parliament, of the NTA Goliath of a system getting people off drugs, unlike the impression given by Effectiveness of community treatments ... document disseminated by the Lancet. Indeed, although the NTA claimed to have treated 202,000 people in the previous year, the latest accounts reduce the number as 163,000 problem drug users “in effective treatment”. It also states that “30,000 individuals completed treatment satisfactorily” – whatever that means. It is a small percentage of the total, and no indication is given that they are drug free.

Nowhere in the Lancet document, either, can I find a reference to anyone becoming drug free  – merely that participants in the project no longer use their main drug of dependency. It does not state whether they have become addicted/dependent on other drugs or alcohol – for example, 40% of the people on the NTORS methadone maintenance became alcohol dependent.  Methadone manufacturers themselves advise that alcohol should never be taken in conjunction with the drug, particularly in pregnancy, and UK Advocates is on the path to a judicial review of medical negligence where heavy drinkers are prescribed methadone.

“We should regard the results set out in Effectiveness of community treatments... paper with considerable caution,” confirmed Professor McKeganey. “The paper reports that six-months after starting addiction treatment, 37% of heroin users and 52% of crack cocaine users were abstinent from those specific drugs in the 28-days before their first review. Recovery from dependent drug use is about sustained not temporary abstinence. And it is striking that the authors refer to individuals being abstinent from these drugs “in” the 28 days before review but we are not told how many days these individuals were free of all drugs over the 28-day period.” Or, indeed, afterwards.

The Lancet document states that “Psychosocial treatment is given to individuals with addiction to heroin or crack cocaine, or both, who might also use other substances”. There is a basic lack of understanding of dependency here: the US Diagnostic &Statistical Manual and the World Health Organisation International Classification of Diseases do not differentiate between types of drugs when diagnosing dependency/addiction.

The paper describes psychosocial treatment as if this was a standard part of treatment. But it is well known in the field that most agencies provide little if any psychosocial treatment; only about 2% of people seeing help manage to get referred to rehab; day programmes on similar lines are even rarer. Research from Dr David Best in Addiction Theory & Science shows that psychosocial support in community programmes averages only four hours a year.

There does not seem to be a control group to compare the study results against. The measurement tool depends on self-report: there is no independent check, such as daily urine or mouth-swab tests. A one-off swab, when the client knows s/he is to be interviewed and given that heroin leaves the system quickly, means little – recovered addicts report that they have lied about street use in order to maintain their script or even use it as a bargaining point for more methadone which they then sell on the black market.

The dossier contrasts markedly with that from McLellan and colleagues (2008) reporting long-term abstinence rates among a treated sample of substance-misusing clinicians, in which urine testing data were presented. 

THE STORY OF DRUG TREATMENT

To find that elusive success story, the NTA regional office sent a message to user representatives. “I am contacting you to ask for help in identifying individuals who would be willing to talk to the media about their personal experience of drug treatment services. This would be as part of the launch of the NTA annual report, on 8 October,” it wrote. “We are looking for the following... 25-29 year old man... preferably someone who doesn’t want or knows he is not ready for rehab... [and] a woman problem drug user who... didn’t want resi rehab.” Well, no attempt to skew the story here, then.

The NTA then requests the user representatives to “let me have some information on their [patients'] situation along with their name and contact telephone number”. This seems to seriously breach the Data Protection Act.

NTA’s own Service User Survey in 2006 showed that most service users aspire to abstinence, perceiving it as the goal of treatment, and found that there is no link between satisfaction and doses of substitute medication. The users appear to have been ignored.

CONCLUSION

This government organisation is trying to justify its expenditure and how it has influenced Local Authority and Primary Care Trust budgets, among others. Shouldn't research be conducted and funded independently? The problem is that, despite all the £££millions flushing around, no funding can be found for others to monitor if the system is working.

We must all be very careful about data taken from agencies which are potentially under threat if they are not seen to be getting good outcomes.  At the end of the day, these findings will be used to continue with the same ineffective system of non-treatment and denying patients, their families and society the promise of improved lives.

ADDENDUM: NTA’S HISTORY ON FACTS

In October 2008 Addiction Today posed questions to the NTA querying its statistics.   After more than a year, they remain unanswered – but inspiration for much of the NTA correspondence to Addiction Today. 

The lancet dossier can be found here.

Comments

[Duncan Raistrick]

Self evaluation is a risky business - a bad outcome is bad but a good outcome is viewed with suspicion. So it is with this study, albeit that the authors all declare their conflict of interest: this is the National Treatment Agency (NTA) evaluating itself.

Starting the title with the word 'effectiveness' is somewhat disingenuous. The study shows what happens to a cohort of drug users entering two treatment modalities according to self report data. It might be proper to avoid a ‘no treatment’ control condition but then there needs to be a 'gold standard' reference treatment and herein lies the problem: process rating the delivery of ‘treatment as usual’ has shown that, despite models of care, there is not the standardisation of treatment modalities that the authors describe in the introduction. In this study we do not know what treatment was received although practitioners in the field will have a picture of the variation in what happens in substitute prescribing and psychosocial interventions.

There is also a problem with the data collection. The NTA has imposed a burdensome system which was created for political not clinical purposes. Fair minded people would support some form of national outcome measurement but this needs to be of universal application, simple, and preferably using international conventions such as EQ5D and ICD10. To impose something that is not seen to be useful, indeed is seen to be for another purpose, will not be embraced by clinicians. Furthermore, these same data are used to performance manage service providers – contracts depend on achieving good outcomes. In short, data quality will be uncertain.

The finding of 60-70% of heroin and crack users being abstinent or reliably improved at (a mean of) 19weeks is unexpectedly good. However, the difficulty with these data is that everybody, the service user, the agencies, the NTA, the government has a vested interest in these results being true, and yet there is no assurance that any of these sources of bias have been eliminated – for example what treatment was actually received and what checks were there on the self report drug use data?

In summary this looks like good analysis of bad data. A crucial finding, not discussed, appears to be that both heroin and crack users do as well or better in the psychosocial intervention as compared to the more costly pharmacotherapy (presumably substitute prescribing). If these findings are valid is it not logical that the NTA should be rapidly dismantling methadone programmes?

[Prof David Clark]

In 30+ years as a scientist, I have rarely been so concerned about a scientific publication as this paper published by The Lancet.

It is a piece of research that could have enormous implications for the drug treatment field in the UK and therefore for the lives of many people who are affected by substance use problems, either directly or indirectly. As the many points raised by Deirdre Boyd, Neil McKeganey and Duncan Raistrick reveal, the research itself and the conclusions derived are deeply flawed. There is clearly a vested interest, in that seven of the nine authors of this manuscript are employees of the NTA. The NTA is responsible for funding most drug treatment in England and trying to ensure good practice; it in effect tells commissioners and services what they can and cannot do; it develops systems to try to determine treatment effectiveness; it has helped create a situation whereby many services are worried that they will lose money if they do not obtain good outcomes. And now they are the sponsor of a piece of research to look at treatment effectiveness within their system where they ‘contributed to the study design, data analysis, data collection, data interpretation and writing of the report’. [I would like to know how much they also paid the first author of the manuscript as a consultant, either directly or indirectly].

The piece of research comes out just before the NTA annual report. It is accompanied by a mass of ‘positive’ media articles. In my humble opinion, it has quite simply been set up as a PR exercise by the NTA... Isn’t this the sort of approach that drug companies and the tobacco industry have been seriously criticised for?

How could the Editor of The Lancet judge this paper as acceptable for publication given that it is so flawed? Wasn’t he concerned about the clear vested interests? Didn’t the reviewers of this paper spot these problems?

[As an aside, I had to laugh at the section on psychosocial interventions, as it implies they are ‘part-and–parcel’ of treatment. People in the treatment industry know that this is rarely the case, as revealed by David Best’s research]

If we are going to examine the effectiveness of our treatment system, the research should be conducted by independent researchers. The research should be of a high quality and be focused on ‘real’ outcomes and also examine whether the well-being of clients is improved.

I am deeply concerned and angered by the fact that a piece of research has been conducted that is deeply flawed, has a clear vested interest, and is publicised widely in the media as showing the NTA is doing a good job. This whole agenda was about protecting the so-called integrity of the NTA (and the government-supported treatment system), rather than caring whether we are actually helping people whose lives have been so badly affected.

This is a disgraceful matter and should be looked into further. The implications for the future of drug treatment are potentially very serious.

Professor David Clark
Director, Wired In