Clinical & Policy Updates:
SMMGP Clinical Update August 2009
|Download the PDF version of this Update here! (PDF*, 94K)|
What could the program have done differently? A qualitative examination of reasons for leaving outpatient treatment
Laudet AB, Stanick V, Sands B. Journal of Substance Abuse Treatment 2009; 37. 182-190.
The goal of this paper was to examine clients' reasons for leaving treatment and to seek their opinions on what could have been done to retain them in services. It was conducted in New York between September 2003 and December 2004.
A total of 278 were recruited at the start and they managed to interview 250 of them again at the end of the treatment program. Overall only 40% managed to complete the treatment and allowing for a few that got transferred to another agency there were, in their terminology, 135 'drop-outs'. This group of 135 people who did not complete their treatment were the sample.
They had semi-structured interviews and views were recorded verbatim in a computer-assisted interviewing package. There was not any recording or transcription. Questions asked included: what is/are the most important reason(s) why you dropped out of the program? Is there anything the program could have done differently so you would have continued attending? If yes: what could have been done differently so that you would have continued attending?
The key themes identified as reasons for leaving the program were: disliking the program, interference with other activities, substance use, practical considerations, not wanting help, personal issues, finances and not finding the service helpful. However, only 33% said something could have been done differently: unmet social need was the commonest theme with 54% mentioning it. After that, the most common themes were wanting more supportive staff and greater scheduling flexibility.
SMMGP comment: There can't be many healthcare professionals working in substance misuse who haven't wondered what they can do improve the DNA rates amongst patients. It's an impressive effort to have interviewed 135 people that have dropped out of care and a qualitative study exploring this issue is a welcome addition. Interestingly, only a third of those that left the program felt something could have been done differently.
The authors highlight a limitation of this study that they feel clients may seek to externalise reasons for failing to complete the program rather than facing their own issues. However, we think that doesn't devalue this study and it emphasises the highly personalised nature of barriers to retention.
A summary of a qualitative study can only give the very essence - this one is worth diving into and reading in a little more depth. It won't give anyone definitive answers but it will give you plenty to reflect on and it may help develop services to keep people in treatment.
This paper is worth reading along side the NTA's recent publication "Towards successful treatment completion: A good practice guide", which discusses the improvement in this area and the downward trend in unplanned discharge but there is still 48% (down from 71% in 2004/5) unplanned discharge, with treatment drop out of 28% being the commonest reason.
It shows that most clients drop out of treatment between initial assessment and start of treatment or in first few weeks.
Diacetylmorphine versus methadone for the treatment of opioid addiction
Oviedo-Joekes E, Brissette S, Marsh D et al. New England Journal of Medicine 2009;361:777-86.
This paper in the illustrious NEJM looks at the rather pointed issue of prescribing injectable heroin to dependent users and inevitably it has the ubiquitous tortured acronym: North American Opiate Medication Initiative (NAOMI).
There were 251 patients entered into the randomisation phase of the study with 111 getting oral methadone, 115 getting injectable diacetylmorphine and 25 receiving injectable hydromorphone. The hydromorphone arm was there to act as a control. The primary outcomes were retention in treatment and illicit drug use/illegal activity (measured by the European Addiction Severity Index).
Participants must have had at least two previous treatment episodes with at least one attempt on methadone of doses greater than 60mg for one month. Those receiving diacetylmorphine attended a clinic up to three times daily where they self-administered a maximum daily dose of 1000mg of diacetylmorphine.
There seems to be some benefits with 88% retained in treatment compared with 54% of the methadone arm. The reduction in rates of illicit drugs were 67% in the heroin group and 48% in the methadone group.
There were a number of serious adverse events: a total of 18 incidents in the oral methadone arm but they were all assessed as being unrelated to the treatment. There were 51 events in the diacetylmorphine group and 24 of these were deemed to be directly related to the drug. These included overdoses, seizures and sepsis. There were no deaths in the diacetylmorphine groups but there was a death in the methadone group. This was reported as an opioid overdose but no further details are available.
SMMGP comment: The role of injectable heroin to treat dependent users remains, to say the least, controversial. NAOMI couldn't even be conducted in the USA and took place over the border under the more liberal gaze of the Canadians.
As the authors highlight, it's very likely that users entered this trial in the hope of getting injectable diacetylmorphine. They were recruited knowing there was around a 50/50 chance of getting injectable heroin. Those that were randomised to heroin would have been more likely to be retained and so it proved. It's difficult to get past this enormous bias in the study.
It's impossible to see a role for injecting diamorphine in primary care in the near future. Not least, the most terrifying aspect is the incidence of serious adverse incidents and the clear need for facilities to manage overdose and seizures mean, at best, it will remain a specialist intervention. There were 20 overdoses and seizures occurring in 18 patients. In a cohort of 115 patients these are not insignificant numbers though it could be argued that they may have occurred in users on the streets with a far more serious outcome.
The authors claim this is an option for a small subset of patients but their liberal recruiting criteria would encompass most patients. Clearly, there are also economic implications, and the conclusion of the authors that "diacetylmorphine is a safe and effective adjunctive treatment" remains contentious.
With the UK's own RIOTT (randomised injectable opiate treatment trial) soon to give first results, it will be interesting to see what the future for heroin prescribing in UK is.
Acupuncture for alcohol dependence: a systematic review
Cho SH and Whang WW. Alcoholism: Clinical and Experimental Research. 2009;33(8):1305-1313.
This was a systematic review of the literature that has investigated the effects of acupuncture on alcohol dependence. The authors scoured nineteen electronic database in 4 languages and assessed the papers according to Cochrane criteria. In total, they managed to find eleven studies to systematically review.
Overall, the review suggests that the existing evidence does not support the use of acupuncture for the treatment of alcohol dependence. They described the results as equivocal and they could not draw any further conclusions.
SMMGP comment: The problem, as confirmed by the authors, for much of the evidence around acupuncture is simply that it is so fundamentally weak. Small sample sizes and generally poor methodology with over-optimistic interpretation have conspired to produce an evidence base that is less than impressive.
Predictably, the authors call for more high quality research into acupuncture but it's difficult to imagine future studies will show a hitherto unequivocal dramatic clinical effect that can be distinguished from bias.
Concerns about consensus guidelines for QTc interval screening in methadone treatment
Annals of Internal Medicine 2009; 151(3):216-219.
Several letters and authors' reply to this original paper: QTc interval screening in methadone treatment (Krantz MJ, Martin J, Stimmel B et al. Annals of Internal Medicine 2009;150(6):387-395).
The discussion in one of the 'Big Five' medical journals on this controversial issue around methadone treatment merits reading. In March 2009 the AIM published a clinical guideline on the subject of QTc screening and came up with several recommendations:
- Clinicians should inform patients of arrhythmia risk when they prescribe methadone.
- Clinicians should ask patients about any history of structural heart disease, arrhythmia and syncope.
- Obtain a pretreatment ECG for all patients to measure the QTc interval and a further ECG within 30 days then annually. Additional ECGs are recommended if more than 100mg methadone is prescribed or there is unexplained syncope or seizures.
- If the QTc is 450-500ms then discuss the potential risks/benefits and monitor more frequently. If the QTc is >500ms then consider discontinuing or reducing the methadone dose; eliminating contributing factors, such as drugs that promote hypokalaemia; or using an alternative therapy.
- Clinicians should be aware of the interactions between methadone and other drugs that prolong the QT interval or slow the elimination of methadone.
This month, the letters page of the AIM as been rumbling with the palpable discontent of a number of doctors working with addictions. They have highlighted some methodological concerns but the overall feeling was that the guidelines were more likely to disrupt and damage care than to protect patients.
SMMGP comment: This important debate in one of the leading medical journals follows on the heels of the recent Addiction paper (reported in the SMMGP Policy Update June 2009 summary) that looked at the mortality associated with QTc interval prolongation. The Addiction paper suggested that the overall mortality risk with methadone remained very low indeed and we should all be reassured by this.
While performing ECGs in all patients on methadone is perhaps not immediately practical we do need to be vigilant to identify the high risk groups. Doctors treating substance users in primary care are ideally placed as ECGs are generally easy to access. Interestingly, the guidelines suggested that the evidence pointed toward automated QTc interval interpretation as being adequate in primary care and specialist manual interpretation by a cardiologist as being unnecessary. This simple piece of evidence may go a long way to breaking down some
Prevalence of problem alcohol use among patients attending primary care for methadone treatment
Ryder et al. BMC Family Practice 2009, 10:42.
This open access paper claims to be the first study that looks at the prevalence of problem alcohol use in current and former heroin users attending primary care for methadone treatment. It was a cross-sectional study that used a random sample from the Irish national database for patients receiving methadone in general practice. A non-medical interviewer administered a questionnaire which consisted of the usual socio-demographic, medical and substance use information. The crucial hook in this piece of research was that they were also all given the Alcohol Use Disorders Identification Test (AUDIT).
The national database suggested that there were 2585 patients attending primary care for methadone treatment. The power calculation suggested that they would need to sample 629 patients (which allows for a 30% non-response rate) and so they duly sampled 634 patients.
However, this is all academic as, mainly due to poor GP response rates, the numbers come crashing down with only 276 invited to participate. Once they picked through the further declinations they only managed to interview a paltry 196 patients.
Some notable findings were: 55% claimed to be hepatitis C positive and 5% HIV positive compared with reported rates in Ireland of 73% with hepatitis C and 9% HIV. The AUDIT questionnaire found 68 (35%) who scored 8 or more - the 'AUDIT positive' cases. Out of these 'AUDIT positive' cases there were 27 (14%) who had scores of 20 or higher suggesting possible alcohol dependence.
Further analysis of the data suggested some factors associated with the 'AUDIT positive' cases: they were significantly more likely to have attended a local Emergency Department in the last year (p=0.04) and more likely to be using benzodiazepines (p=0.02) and amphetamines (p=0.02). 'AUDIT positive' cases were also less likely to be on methadone maintenance (p=0.03) and less likely to have attended a hospital clinic in the past year (p=0.01). This statistic was also notable for its size - a whopping 25% of the 'AUDIT negative' cases attending secondary care rather suggesting complex medical needs.
SMMGP comment: The underlying aim of this paper is to prod at the major issue of alcohol in substance users in primary care. As with any cross-sectional prevalence survey the major limitation is that it is a snapshot and simply a freeze-frame in time offering no information on the natural progression for these individuals.
Any interpretation of this study has to be handled with care because it was statistically under-powered and this introduces some real concerns with selection bias in the sample. Their sample was a wee bit older, more likely to be employed and had less hepatitis C and HIV than has been reported in Ireland. However, it could be argued that this may be more likely to under-estimate the findings and it is unlikely to change the fundamental message - substance users in primary care are not simple, uncomplicated or necessarily stable.
The authors conclude: "Interventions that address problem alcohol use in this population should be considered as a priority, although the complex medical and psychological needs of this population may make this challenging". However, we should not lose heart. A recent systematic review on brief alcohol interventions in primary care settings reported that they can work in general practice (SMMGP Clinical Update April 2009).
This study reinforces the message that there is a massive need for alcohol issues to be addressed and the RGCP Part 1 Certificate in the Management of Alcohol Problems in Primary Care has never looked more relevant.
Saints not sinners? Young people bucking the trend of binge drinking. An analysis of the drinking trends of school age children in Sunderland
McInnes A and Blackwell D. Drugs: education, prevention and policy, 2009. 1-19, iFirst.
This was a study conducted in Sunderland that looked at 12-15 year olds and their self-reported drinking behaviour. No less than a total of 8008 schoolchildren completed a questionnaire every 2 years from 1996 to 2006. The questionnaire was administered by school teachers in the actual schools who had been specially recruited and trained.
The study reported a steady increase in the percentage of pupils, in all age groups and both genders, who had not consumed alcohol in the previous 7 days. There was a steady decline in the number of pupils who reported drinking on between 2 and 4 days of the week. The trend over the 10 year period was for a reduction in the number of units consumed. The much demonised 'alcopops' peaked in 1996 with 34% of females and 30% of males expressing a preference and by 2006 this had dropped to 15% of females and 7% of males.
SMMGP comment: This wasn't a true longitudinal study but rather a series of cross-sectional snapshots that rely on self-reported behaviour. Any interpretation in trends needs to bear this in mind but the results are interesting and they certainly challenge common perceptions. The findings were remarkably consistent - rather than a drink culture spiralling out of control there were signs, across the board, of reductions in drinking in both boys and girls of all age groups.
Why should we care? Well, in the authors' words: "the trend for binge drinking in schoolchildren can be seen as a harbinger of drinking behaviour and associated problems in older age group". It has been taken as gospel that alcohol use continues to increase amongst British schoolchildren and media outlets flood us with tales of lurid anti-social drinking.
This study contradicts previous findings with young people in Sunderland choosing not to drink at all or drinking within responsible limits. It may be the impact of specific health promotion measures in the NE of England or it may the first signs of a general change in attitudes.