Clinical & Policy Updates:
SMMGP Clinical Update October-November 2012
|Download the PDF version of this Update here! (PDF*, 485K)|
A record-linkage study of drug-related death and suicide after hospital discharge among drug-treatment clients in Scotland, 1996-2006
Merrall ELC, Bird SM, Hutchinson SJ. Addiction 2012, published online ahead of print
This study investigated the relationship between time after hospital discharge, drug-related death (DRD) and suicide in Scotland. They took 69,457 individuals registered for drug treatment in Scotland between 1996 and 2006. The results were split into time periods after discharge and expressed in DRD rates per 1000 person-years: during hospitalisation the rate was 87; within 28 days is was 21; 29-90 days it was 12; and 91 days to 1 year was 8.5. After one year the DRD rate per 1000 person-years was 4.2 and for those who were never admitted it was 1.9. These can be translated into adjusted hazard ratios (compared against those never admitted) of 9.6 within 28 days of discharge and 5.6 for days 29-90. Although drug-related suicides were less frequent than DRDs a similar risk pattern was observed.
Commentary: We have previously reported (in the December 2011 Clinical Update) on a similar study that looked at overall mortality of those attending drug services in Scotland over the same ten-year period from 1996 to 2006. That study showed increased rates of death in older people and those diagnosed with hepatitis C. It also showed an association between stimulant use and increased suicide risk. (Suicide was the second most frequent cause of death after drug-related deaths.)
This study now looks at the more specific period of time after hospital discharge and it found an increased risk of death. We've known about the increased risk of death for people transitioning in and out of treatment. It seems we should also be adding hospital admission to that list. It may seem relatively obvious - people who have sufficient medical problems to warrant admission will be carrying more risk factors. However, the authors make the point in a quarter of the admissions there wasn't even an overnight stay involved. When presented like this these data look stark, and the obvious clinical implication is that we should be ensuring our services have explicit follow up and review systems for people who've just been discharged from hospital.
Opioid dependence during pregnancy: relationships of anxiety and depression symptoms to treatment outcomes
Benningfield MM, Dietrich MS, Jones HE, et al. Addiction 2012;107 Suppl 1:74-82
This study reports on a secondary analysis of the Maternal Opioid Treatment: Human Experimental Research (MOTHER) randomised clinical trial. There were a total of 175 opioid-dependent pregnant women and 131 completed their treatment. Symptoms of anxiety and depression were measured with the 15-item Mini International Neuropsychiatric Interview (MINI) screen.
The results showed that women who reported only anxiety symptoms at the start of treatment were more likely to discontinue treatment (OR 4.56, 95% CI: 1.91-13.26, p=0.012). Those who reported only depression symptoms were less likely to discontinue compared to women who reported neither depression or anxiety. Just over 61% of women used some kind of psychotropic medication during the course of the study, including medications such as benzodiazepines, antidepressants and antipsychotics.
Jones HE, Heil SH, Baewert A, et al. Buprenorphine treatment of opioid-dependent pregnant women: a comprehensive review Addiction 2012;107 Suppl 1:5-27
This paper reviews the published literature in five areas related to the use of maternal buprenorphine - maternal efficacy, foetal effects, neonatal effects, effects on breast milk and longer-term developmental effects.
Maternal efficacy. Buprenorphine seems to be comparable to methadone. There are some issues with poorer retention in the buprenorphine groups and this may be related to induction procedures. Foetal effects. The evidence suggests that there is less suppression of foetal heart rate and movements than with methadone. There are recurring reports of intrauterine growth retardation with buprenorphine but the impact of other factors (e.g. smoking) hasn't been fully disentangled. Neonatal effects. The percentage of babies who will have some kind of neonatal abstinence syndrome is the same as for those on methadone - around half. However, there is a clinically significant reduction in the severity of neonatal abstinence syndrome with buprenorphine compared to methadone. Breast milk. There is only limited evidence that relies on case reports. What does exist suggests that breast feeding, as with methadone, is an option and is not contra-indicated. Longer-term developmental effects. So far, there have been no data to point to any harmful effect on development from buprenorphine.
Commentary: The first paper is a secondary analysis from the MOTHER study - a randomised controlled trial that weighed up buprenorphine versus methadone in pregnancy. Although neonatal outcomes were better in the buprenorphine arm, it did show that there was greater difficulty in retaining women in treatment who were on buprenorphine. Retention is a key factor and this secondary analysis now shows that women with just anxiety were more likely to discontinue. If this holds then there is an important clinical message here. We should be assessing anxiety and depression early on - and it may well be that if we concentrate on managing the anxiety that may help to reduce premature discontinuation.
The second paper really is, as the title suggests, a comprehensive review. Since the MOTHER study in 2010 buprenorphine has now moved into the mainstream as an option for pregnant women dependent on opioids. The evidence suggests that in most respects buprenorphine is at least as good as methadone and in some areas better. As discussed above, there are still some issues with retention in treatment. The authors in this review note that nearly 30% of the dropouts from the buprenorphine side of the MOTHER study occurred on the day of study entry. This strongly suggests that the induction is a problem for many women. There is a need for more definitive research into induction for pregnant opioid-dependent women but until then clinicians all need to be particularly wary of the challenges.
"Subutex is safe": Perceptions of risk in using illicit drugs during pregnancy
Leppo A. Int J Drug Policy 2012;23:365-73
This qualitative paper explored how pregnant, drug-using women perceive the risks involved in using illicit drugs. The study was based on fourteen ethnographic interviews with women in Finland who had used during pregnancy. They were informal, semi-structured interviews focussing on the women's experiences. Each interview lasted around one hour and they were then transcribed and inductively analysed using thematic coding.
The results found that women were not primarily concerned with health risks. The areas of biggest concern were: the baby having withdrawal symptoms; child protection interventions and child removal; and encountering negative attitudes in seeking professional help. However, the results also showed that the women didn't regard abstinence as risk-free - there were potential consequences of physical pain and mental difficulties, as well as the loss and disruption of important social bonds. They regarded buprenorphine as not being harmful to the baby if the doses were steady and injecting equipment clean and they drew this conclusion from having seen other women having healthy babies while using buprenorphine. One woman herself had condemned women using drugs while pregnant and then continued to use buprenorphine while pregnant: I can remind myself of how critical I used to be of pregnant women and mothers who do drugs, and I know how people generally feel about it... It does not necessarily mean that you don't care about your child or that you are irresponsible; there are other things involved.
Commentary: There are some useful issues in this qualitative paper that have the potential to shift the emphasis of our consultations and interactions with pregnant, drug-using women. While the main concern of clinicians is often about the risk of foetus and child, or as the author puts it "the biomedical discourse", this paper points out that the women have more complex perceptions about safety and risk. This paper highlights the importance of how our professional approach impacts on these challenging consultations.
Heroin users' experiences of depression: a qualitative study
Cornford CS, Umeh K, Manshani N. Fam Pract 2012;29:586-92
This study took place in a general practice in Middlesborough that specialises in the treatment of substance misuse. In total they have around 1100 patients registered and, at any one time, about 700 of them are receiving treatment for opioid addiction. The practice also treats around 600 people with opiate dependence from other local practices - but these weren't included in this study. The general practice setting means that problems such as depression are also treated within the practice. A total of 17 semi-structured interviews were conducted with patients on opiate substitution therapy (OST) and antidepressants. The inclusion criteria were: a current record of antidepressant medication and current treatment with either methadone or buprenorphine for opioid dependence.
The results showed that many different thoughts and emotions were described as depression by the participants. The authors highlighted that many participants commented on adverse childhood events. Many examples of stigma were also given and these were thought to be a cause for depression: Yeah, you're not as clean and tidy and hygienic as you used to be and you know that yourself, you don't need other people to tell you that. You are passing them in the street and they go "hey, you scruffy smack head" and you think "I am".
The participants often felt isolated and many had particular beliefs about the antidepressants which: "incorporated ideas about blocking out thoughts, stopping thoughts racing and keeping emotions level".
Commentary: This paper comes from the Fulcrum Medical Practice that produced the excellent BJGP paper on deep vein thrombosis (and was reviewed in the December 2011 Clinical Update). It's clearly a remarkable practice. The authors quote one cohort study that showed 24% of new entrants to treatment for opioid dependency had evidence of depression and that 31% of the group were on antidepressants at some point in the 3-year follow up. Overall, this paper showed that the participants experiences with drugs were closely linked to their experience and beliefs about depression. The drugs had depressive consequences such as stigma and isolation. There was, in the minds of the individuals, a circular relationship between drugs and depression with each causing each other, and sharing common root causes. Perhaps the greatest lesson from this paper is that it is utterly artificial, and ultimately unhelpful, to try to separate the depression from all the other issues and events in people's lives. They are inextricably linked.
Depression and prescription opioid misuse among chronic opioid therapy recipients with no history of substance abuse
Grattan A, Sullivan MD, Saunders KW, et al. Ann Fam Med 2012;10:304-11
This study, based in North California, had the aim of trying to tease out the factors that are associated with people without any substance misuse issues, who go on to misuse prescription opioids. They used a telephone survey to interview 1334 patients who were on long-term opioid therapy for non-cancer pain. They were asked about three forms of opioid misuse: self-medicating for symptoms other than pain, self-increasing doses, and giving opioids to/getting opioids from others. Depression was evaluated using the 8-item Patient Health Questionnaire (PHQ-8).
The results showed that patients who were moderately depressed (PHQ-8 of 10 to 14) and severely depressed (PHQ-8 score of 15 or more) were 1.8 and 2.4 times more likely, respectively, to misuse their opioid medications for non-pain symptoms when compared against patients who scored 4 or less on the PHQ-8. They were also more likely to misuse their medication by self-increasing the dose with the most severely depressed over 3.1 times more likely. There was no association between depression and giving opioids to others.
Development of dependence following treatment with opioid analgesics for pain relief: a systematic review
Minozzi S, Amato L, Davoli M. Addiction 2012, published online ahead of print
Data were extracted from 17 studies involving a total of 88,235 people. Most studies involved people with chronic non-malignant pain. Two also included patients with cancer pain and only a single study included people with a previous history of dependence. Overall, the incidence of dependence ranged from 0 to 24% (median 0.5%). The prevalence ranged from 0 to 31% (median 4.5%).
Commentary: These two papers looked at similar issues. The first, in the Annals of Family Medicine, was a specific study conducted to find out how those who were on long-term chronic opioid treatment used their medication. They showed that there was an association between increasing levels of depression (or at least higher PHQ-8 scores) and misuse (as defined by them). Arguably, one might have to be a little careful about characterising "self-increasing doses" as misuse - it could simply suggest under-treatment, perhaps due to doctor anxiety about opioids; or, the complex relationship between mood and pain.
The second paper aimed to quantify the development of dependence when opioid analgesics were used for pain relief. What was certainly clear from the second paper was that the evidence that does exist is mixed and there is a severe lack of good quality studies. The studies that were included were highly variable and this is perhaps responsible for the wide differences in reported incidence and prevalence. Allowing for this, for the moment the evidence suggests, for all the concerns about the use of opioids, the overall risk of dependence for people using long-term opioids for chronic pain with no history of substance misuse probably remains quite modest.
The challenges of reducing tobacco use among prisoners
Richmond RL, Butler TG, Indig D, et al. Drug Alcohol Rev 2012;31:625-30
This paper discusses the general challenges of tobacco use in prison and it summarises three Australian studies with which the authors have been involved The first study looked at the feasibility of delivering a multicomponent cessation intervention to prisoners. They achieved abstinence rates of 26% at six months but they also recognised some of the reasons for relapse. These included: transfers to other prisons, boredom, prolonged periods in cells, and family stresses or bullying. The second study was qualitative and explored the role of tobacco in prison. Prisoners reported that tobacco was used as currency and in exchange for goods, debts etc. Prisoners wanted smoking cessation to offer treatments like nicotine patches, access to a telephone helpline, non-smoking cells and more access to physical activity. The third study used some of the lessons learned from the first two studies and was an RCT of a smoking cessation intervention among 425 male prisoners in 17 prisons across New South Wales and Queensland. All prisoners received nicotine replacement, brief CBT and a support package to help cope with particular stressors known to trigger smoking. The intervention group also received nortriptyline. There were no statistically significant differences between the two groups at 0, 3, and 6 months. At 12 months the abstinence rates were 12.1% and 14.6% in intervention and control group respectively.
"Do more, smoke less!" Harm reduction in action for smokers with mental health/substance use problems who cannot or will not quit
Baker AL, Callister R, Kelly PJ, et al. Drug Alcohol Rev 2012;31:714-7
This paper points out that smoking rates are in the order of 74-98% in substance misusing populations and it outlines some of the harm reduction issues to consider. They make the point that intermediate smoking goals - such as reduced consumption - are valid and have been shown to be associated with an increased probability of future abstinence. They also point out that increased attempts to improve a wide range of lifestyle behaviours can also help someone to stop smoking.
Commentary: Both of these papers highlight the ongoing challenges of managing tobacco use in difficult groups with alarmingly high smoking rates. It's important the problem doesn't get labelled as being unmanageable and therefore reduced in priority out of simple defeatism. The issue of harm reduction is one that is likely to be developed in the next couple of years and the NICE draft guidance on this topic, Tobacco-harm reduction, is now out for consultation and is available at http://guidance.nice.org.uk/PHG/Wave23/23.
Widening access to treatment for alcohol misuse: description and formative evaluation of an innovative web-based service in one primary care trust
Murray E, Linke S, Harwood E, et al. Alcohol Alcohol 2012;47:697-701
This study was a description of the development and evaluation of a web-based service for people with hazardous or harmful levels of drinking. The patients were identified in general practice in what was described as a "relatively affluent area" in a single PCT on the outskirts of London. However, there were still 1863 hospital admissions due to alcohol-related harm in that time. The GPs had to identify adults aged 18 or over who were drinking at hazardous or harmful levels. There were three elements to the intervention. Firstly, the participants were assessed using tools such as past week alcohol consumption (TOT-AL) and the Alcohol Use Disorders Identification Test (AUDIT). They also measured for possible dependence, overall mental health and health-related quality of life. Secondly, there was a web-based self-help programme known as Down Your Drink (DYD). Thirdly, the participants received support in using the self-help programme.
They noted that there was a low level of referrals in the first year of the service with just 31 patients referred. Of these, 19 patients attended the assessment with a mean past week alcohol consumption of 75 units per week and AUDIT score of 23. Only six patients (32%) logged on at least once after the initial assessment appointment. In the course of the next year the Alcohol Project Coordinator had managed to contact 13 (68%) patients for at least one of the planned phone calls. Follow-up data were only available for seven of the patients. In those there was a significant reduction in past week alcohol consumption to 35 units (p=0.018) and mean AUDIT scores were also reduced significantly from 23 to 19 (p=0.018).
Commentary: The authors declared that the service "was shown to be feasible and acceptable to stakeholders" but the numbers of participants are worryingly low. There is an obvious need for interventions to address this area and the potential benefits of a low-resource intervention such as a web-based one are clear to see. The authors alluded to the challenges of implementing new and innovative services. Many GPs will be familiar with this story and the giant upheavals to NHS service provision will cast a long shadow. The authors also highlighted the challenges in getting primary care workers to engage in the process of tackling alcohol use that they perceive is at more social levels.
Opiate substitution treatment and HIV transmission in people who inject drugs: systematic review and meta-analysis
MacArthur GJ, Minozzi S, Martin N, et al. BMJ 2012;345:e5945-5
The aim of this review was to quantify the effect of opiate substitution therapy (OST) in relation to HIV transmission. They selected studies that directly assessed the impact of OST on HIV and, also, studies that assessed the incidence of HIV in people who inject drugs that might have collected data on OST but not reported it.
In total they found twelve published studies that were relevant and they also found unpublished data from three further studies. All of the OST studies were looking at methadone maintenance. In total, the studies amounted to 819 incident HIV infections over 23,608 person-years of follow up. OST was associated with a 54% reduction in risk of HIV infection among people who inject drugs (95% CI 0.32 to 0.67, p<0.001). There was some weak evidence that there was greater benefit associated with longer duration of exposure to OST.
Commentary: This study is further evidence of the benefits of OST. Arguably, this study has even more relevance throughout the rest of the world than in the UK, given that globally, in some places, the incidence of HIV is still rising in people who inject drugs. The authors highlight the evidence that the coverage worldwide of OST suggest that only 6-12% of people who inject drugs receive it. The study did note that they couldn't find any evidence of an association between detoxification and the risk of HIV infection. The authors did speculate on whether the findings could reflect comparatively high levels of motivation in people who go on to have opiate substitution therapy. Like any observation study, it can only highlight associations and the inference of causation should be approached with caution.
Benzodiazepine use and risk of dementia: evidence from the Caerphilly Prospective Study (CaPS)
Gallacher J, Elwood P, Pickering J, et al. J Epidemiol Community Health 2012;66:869-73
This paper is based around a prospective cohort in Caerphilly that has now been studied for 22 years. The men in the cohort have been seen on five occasions for full medication histories, repeat measures of cognitive function, as well as clinical diagnoses of dementia.
In total, there were 1134 men with complete data and 103 (9.1%) had been taking benzodiazepines regularly at one or more phases of the study. These men showed a marked increase in dementia (OR=3.50, 95% CI 1.57 to 7.79, p=0.002). This persisted after adjustment for psychological distress and covariates such as social class, education, smoking and cardiovascular disease. Three further models were tested with further adjustment for psychological distress, trait anxiety and daytime sleepiness and more proximal determinants of benzodiazepine use.
Commentary: Arguably, the most immediate and pressing problem for many GPs when considering benzodiazepines is not the illicit use associated with benzo binges and other substances but the potential insidious adverse effects when used long-term by an older population. This paper shows that there is certainly an association between benzos and dementia. Further work is still needed to unearth the full story.
The issue of reverse causation is discussed in the paper - this is the problems where an increased use of benzodiazepines is seen because men are taking them to get relief from various aspects of dementia. In this study reverse causation seems less likely as there was a greater association with dementia for those exposed early. However, if the benzos were causal then it is likely there would be worsening problems with increased use - i.e. a dose-response relationship. The study found no evidence of this and so it remains unclear on whether benzos are a cause of dementia or simply some kind of marker for it. This paper, while not yet definitive, will add to evidence about long-term cognitive impairment related to benzodiazepines.