Clinical & Policy Updates:
SMMGP Clinical Update September 2008
|Download the PDF version of this Update here! (PDF*, 47K)|
Drug Related Deaths
Male deaths from drug poisoning are at their highest level for five years, according to new data from the Office for National Statistics. The report is contained in the autumn edition of Health Statistics Quarterly, available online at www.statistics.gov.uk - page 82 - 88.
Worryingly there is a rise of some 16% in heroin and morphine deaths, (n=829 in 2007) and even more alarmingly, a rise of 35% in methadone deaths since 2006 (n=325). Cocaine deaths show a continuous, steady increase n=196. Deaths due to drug poisoning (all causes) peaked in 2001, and this is the first significant increase since then. The rise is almost wholly accounted for by males, whereas female deaths from drug poisoning have declined consistently over the time period. Interestingly the peak age has moved upwards from 20-29 in the 1990s to ages 30-39 now.
In the same bulletin, pages 14 - 20 another article presents variations in mortality from drug misuse in England and Wales for 1993 to 2006. ONS has collected data on Drug-related Deaths (DRDs) from death certificates since 1993 (but NB they counsel cautious interpretation due to inconsistency in reporting by coroners). There is useful discussion of definitions, and in-depth geographical and social analysis which will help to inform local policies.
SMMGP comment: Required reading for all those involved in local clinical governance and DRD confidential enquiries groups.
Wernicke-Korsakoff syndrome in Australia: no room for complacency
Feeney et al., Drug Alcohol Rev 2008;27:388-392
With the current recognition of the extent of alcohol problems in the UK this article from Australia is a timely reminder of one of its most insidious and harmful sequelae, yet one which is easily preventable. The paper starts by citing several studies from around the world which have shown a much higher population-wide incidence of Wernicke's in large-scale post mortem studies than were detected in life, confirming that only 20% of Wernicke's cases present with the full clinical picture, (classic triad of confusion, nystagmus, and ataxia) and as many as 30% present with mental symptoms only. Two case-studies illustrate the latter scenario, with, in one case, the diagnosis possibly having being missed on 4 occasions, resulting in the patient requiring lifelong residential nursing care. The paper includes very useful recommendations on more sensitive diagnostic criteria (including being underweight) and on the prevention and management of Wernicke-korsakoff syndrome, highlighting the ineffectiveness of oral thiamine and the need to give this in injectable form.
SMMGP comment: Anecdotally, many clinicians report under-treatment of Wernicke's. This could be a very useful topic for audit.
Self-report of illicit benzodiazepine use on the Addiction Severity Index predicts treatment outcome
Ghitza et al., Drug and Alcohol Dependence 97(2008) 150-157
This rather topical paper may be of interest to those thinking of implementing contingency management (CM) as it demonstrated that even low-level benzodiazepine use could blunt the effect of CM interventions in a large, randomised controlled trial. Opiate and cocaine dependent drug users were recruited, with those dependent on alcohol or benzodiazepines excluded, leaving a sample of 361. The sample was divided into those who had not used benzodiazepines at all in the past 30 days (BZD-N) and those with some non-dependent use (BZD-U). Methadone maintenance treatment and weekly counselling was given to all patients for 5 weeks. Those still testing positive for heroin or cocaine were then randomised to CM interventions (voucher or prize-based "incentivisation" for heroin or cocaine abstinence) or a control group who received incentives irrespective of test results. After 12 weeks the CM BZD-N group showed improvements in drug use, HIV risk behaviours, and quality of life scores, whilst the BZD-U group showed significantly worse outcomes. In the control (no CM) group, the BZD-U and BZD-N had similar outcomes, but BZD-U had a significantly higher rate of in-treatment cocaine use.
SMMGP comment: There is no cause and effect assumed in this study - BZ use predicts rather than causes poorer outcomes.
A randomised controlled clinical trial of methadone maintenance for prisoners: findings at six months post-release
Gordon et al., (2008) Addiction 103: 1333-42
This US study showed that methadone maintenance, initiated prior to or immediately after release from prison, increased treatment entry and reduced heroin use post-release, compared to counselling only. 211 heroin dependent pre-release inmates were randomized into 3 groups: counselling only, counselling + transfer into treatment and counselling + methadone (administered during and after incarceration). At 6 months post-release, counselling + methadone participants were significantly more likely than both counselling only and counselling + transfer participants to be retained in drug abuse treatment (P = 0.0001) and significantly less likely to have an opioid-positive urine specimen compared to counselling only (P = 0.002). Furthermore, counselling + methadone participants reported significantly fewer days of involvement in self-reported heroin use and criminal activity than counselling-only participants.
Alcohol screening and brief counselling in a primary care hypertensive population: a quality improvement intervention
Rose et al., (2008) Addiction 103: 1271-1280
This study is based on established evidence on "facilitating physician behaviour change" (the 4 most promising strategies are listed as prompting, performance feedback, academic detailing and information technology). The researchers set out to show whether continuing support to primary care practices offering screening and brief interventions (SBI) for alcohol to hypertensive patients made a clinically worthwhile impact. Over a 2 year period, practices that received site visits from study personnel and were invited to annual network meetings increased the rate of alcohol screening to their hypertensive patients, compared to control practices, and also increased rates of alcohol counselling, leading to small but significant reductions in blood pressure (mean systolic 4.4 mm Hg, mean diastolic 3.3mm Hg). Practices' focus on alcohol screening did not detract from clinical care given in other domains.
SMMGP comment: this supports the robust evidence base in favour of SBI in alcohol and the DES for alcohol (now out September 2008) will help to get them used more in primary care.
Development of the Treatment Outcomes Profile (TOP)
Marsden et al., (2008) Addiction 103:1450-1460
Until the development of TOP a minority of services looked at the outcomes of care, using a variety of outcomes monitoring tools such as the Christo inventory or the Maudsley addiction profile, making it difficult to make comparisons, or to derive a comprehensive picture of how patients were benefiting from treatment. TOP is a first attempt at a common outcomes tool to use across all the providers in England, allowing us to say if the drug treatment strategy is really working.
This study written by the developers of the tool, takes us through the process of development and validation of the TOP. Results showed good correlation of TOP items with validated instruments and toxicology testing and good inter-rater reliability. At one month the TOP had detected reliable change in a host of measures including amounts of drugs used, sharing, housing measures and certain crimes. Twenty reliable valid measures have been developed for the first national outcomes monitoring tool creating one of the largest effectiveness outcomes monitoring databases in the field.
SMMGP comment: it is helpful to understand why certain measures which would normally be expected in an assessment are not present in TOP, especially history of overdose, benzodiazepine use and risky sexual behaviour. These could be for different reasons, for example overdose is relatively rare, benzodiazepine use would be likely to adversely affect recall, and risky sexual behaviour can be an uncomfortable topic for interviewers and interviewees alike. It is important to remember that TOP was never designed to be an assessment tool and services should supplement the TOP questions with other clinically important items at assessment. As the researchers acknowledge, there are several weaknesses: the subjects were not randomly selected, trained and motivated keyworkers were used to conduct the research and the follow up period was very short. SMMGP are attending the review of TOP on 23-24 October and are collecting comments from all. We have so far found there are more problems using this tool than were found in this brief study, hence send us your thoughts to Kate Halliday at firstname.lastname@example.org
Treating pregnant women dependent on opioids is not the same as treating pregnancy and opioid dependence
Winklbaur et al., Addiction, 103, 1429-1440
There is a dearth of good quality research into management of substance misuse in pregnancy, particularly randomised blinded controlled trials, for obvious reasons. This has left clinicians with little evidence on which to base practice. However these Vienna-based clinicians have pulled together a comprehensive literature review, and coupled with their experience, this distillation forms a useful evidence-base for good practice, in antenatal, labour and postnatal care, and management of the newborn.
SMMGP comment: clear and easy to read, there is largely good agreement with the Clinical Guidelines, with a few exceptions, for instance this review suggests neonatal abstinence syndrome (NAS) is reduced or absent after maternal buprenorphine treatment vs. methadone. This paper is helpful as confirmation and as a supplement to the orange book. All clinicians involved in drug misuse treatment should memorise the succinct paragraph of recommendations at the end.