EFFECTIVENESS BANK BULLETIN 21 August 2009: reviews and meta-analyses
EFFECTIVENESS BANK BULLETIN 21 August 2009: reviews and meta-analysesThe entries below are our accounts of documents selected by Drug and Alcohol Findings as particularly relevant to improving outcomes from drug or alcohol interventions in the UK. Entries were drafted after consulting related research, study authors and other experts and are © Drug and Alcohol Findings. Permission is given to distribute these entries or incorporate passages in other documents as long as the source is acknowledged including the web address http://findings.org.uk. However, the original documents were not published by Findings; click on the Titles to obtain copies. Free reprints may also be available from the authors; if displayed, click Request reprint to send or adapt the pre-prepared e-mail message. Abstracts are intended to summarise the findings and views expressed in the study. Below are comments from Drug and Alcohol Findings. Links to source documents are in blue. Hover mouse over orange text for explanatory notes.
If you have not found what you want you could:
● Try a subject or free text search instead. Searches include bulletin entries and all other documents on this site.
● Try browsing other bulletins or back issues of the magazine.
● Try searching the libraries of
Alcohol Concern or
DrugScope (opens new window).
● Documents are regularly added. Use the e-mail update service to monitor additions.
● Return to the home page.
Click HERE and enter e-mail address to receive alerts of new bulletins
Acupuncture does not aid treatment of alcohol dependence ...
Unproven verdict on hospital-based brief alcohol interventions ...
Guidance on managing pregnant opiate-dependent women ...
Effective strategies for extending evidence-based practice ...
Acupuncture for alcohol dependence: a systematic review.Cho S-H, Whang W-W. Request reprint
Alcoholism: Clinical and Experimental: 2009, 33(8), p. 1305–1313.
An exhaustive multi-country and multi-language trawl for randomised trials of acupuncture in the treatment of alcohol dependence found just 11 studies which overall offered little support for any form of the therapy.
Abstract Acupuncture has been used in the treatment of substance-related disorders for the past 30 years. However, there has not yet been a systematic review to assess the effectiveness of various types These is a German-language review relating to auricular acupuncture. See http://direct.bl.uk/bld/PlaceOrder.do?UIN=152361543&ETOC=RN of acupuncture in the treatment of alcohol dependence. The present review aimed to fill this gap by analysing results from trials which randomly allocated alcohol dependent patients to any type of acupuncture versus a comparison (or no) treatment. To find the studies the authors searched 19 electronic databases (including English, Korean, Japanese, and Chinese databases) for studies in any language published up to June 2008, as well as manual searching relevant journals, symposia, and conference proceedings.
The search found 11 such studies involving 1110 participants, of which just two satisfied all the review's methodological quality standards. In relation to the primary outcomes assessed by the trials, six found no significant advantages for acupuncture. Among the three trials to report these, adverse side effects were mostly minimal.
Eight of the trials reporting relevant data compared 'real' acupuncture with either a sham or placebo procedure which simulated real acupuncture but was not intended to be an active treatment. Of the five trials which assessed this, two found a significant impact on craving for alcohol, but one was available only in abstract form and the other was able to follow-up only a minority of patients. Of the two studies which looked at this, neither reported an effect on withdrawal symptoms. Treatment completion rates were improved by real acupuncture in one of five studies. However, when the results from all five were combined in a meta-analysis, A study which uses recognised procedures to summarise quantitative results from several studies of the same or similar interventions to arrive at composite outcome scores. Usually undertaken to allow the intervention's effectiveness to be assessed with greater confidence than on the basis of the studies taken individually. there was no overall advantage for real versus simulated acupuncture.
Among the five studies which tested acupuncture as a supplement to conventional treatments, three of four found reduced craving for alcohol. Just one study of five reported an improved treatment completion rate and when results from all five were combined, there was no statistically significant improvement in completion rates among patients allocated to supplementary acupuncture.
A further study compared acupuncture with aromatherapy and found no significant differences in craving or withdrawal symptoms.
The conclusions the authors could draw were limited by the poor methodological quality and the limited number of the trials. Overall their results did not support the use of acupuncture for the treatment of alcohol dependence. In particular evidence was lacking that 'real' acupuncture actually conferred any benefits compared to a simulated procedure. As a supplement to conventional therapies, there were promising findings indicating that acupuncture can reduce craving for alcohol.
This review is particularly valuable for its rigorous approach and its exhaustive multi-country and multi-language
trawl of the relevant research. It shows that the limited research to date has been unable to demonstrate consistent
benefits, except perhaps in relieving craving during treatment. The pattern of findings suggests that if there are
any benefits from acupuncture, they are caused not by the intended mechanisms, but by non-specific factors such as
extra therapist contact time or the placebo effect of receiving what seems to be an active therapy. The featured
review's 'unproven' verdict for the treatment of alcohol dependence follows similar verdicts in respect
of the treatment of dependence on opiates, and on cocaine using auricular acupuncture.
The inadequacy of the evidence for a specific effect of acupuncture also however applies to some recognised psychosocial therapies, which are held to a lesser standard for demonstrating their efficacy. In many of the studies acupuncture is tested to the same standard as medications, using a (to the patient) indistinguishable simulation as the comparator. This research design gives the comparator treatment the full benefit of the placebo effect and other non-specific therapeutic influences, making it difficult for 'real' acupuncture to further improve outcomes. Such rigorous testing is not possible for psychosocial therapies, which are generally tested against either no therapy or a clearly different alternative. The difference the comparator can make is clear for example in respect of cognitive-behavioural therapies, which record a much greater advantage when compared to no treatment than when compared to an active alternative therapy. None of the studies analysed in the featured review compared acupuncture to no treatment at all. When the comparator was merely treatment as usual, there appeared to be some benefits in terms of reduced craving. When the potential for a placebo effect was elevated by simulated acupuncture, the real version was less able to demonstrate benefits.
Despite these findings, the possibility remains that offering something concrete like acupuncture (even if it is a 'sham' procedure) helps attract people to services. Some studies have also suggested that doing something both clients and staff believe is worthwhile can help retain patients in treatment. If this is the case, acupuncture could indirectly improve outcomes by increasing the patient's exposure to treatment's active ingredients. Just such a role was specified Complementary therapies were defined as "Any non medical intervention which regardless of therapeutic value enhances client access and retention in services, such as auricular acupuncture." in recent guidance from the National Treatment Agency for Substance Misuse on treatment intervention costing and on treatment systems. Such considerations may explain why despite no convincing evidence of efficacy, acupuncture continues to feature in many of the treatment plans As revealed in March 2009 by a search for term 'acupuncture' on the web site of the National Treatment Agency for Substance Misuse, http://www.nta.nhs.uk. developed by local partnerships responsible for commissioning treatment services in England.
Last revised 15 August 2009 Comment on this entry Give us your feedback on the site (one-minute survey) Back to contents list at top of page
Traditional medicine in the treatment of drug addiction REVIEW 2009
Review of treatment for cocaine dependence STUDY 2010
Acupuncture yet to convince NUGGET 2000
Naltrexone prevents return to heavy drinking NUGGET 2001
Mindfulness meditation for substance use disorders: a systematic review STUDY 2009
Failure to replicate positive findings on acupuncture NUGGETTE 2003
'Real-world' studies show that medications do suppress heavy drinking NUGGET 2005
Convincing evidence that acamprosate and naltrexone help prevent alcohol relapse NUGGET 2002
The state of pharmacotherapy for the treatment of alcohol dependence REVIEW 2009
Brief interventions for heavy alcohol users admitted to general hospital wards.McQueen J., Howe T.E., Allan L. et al.
Cochrane Database of Systematic Reviews: 2009, 3, Art. No. CD005191.
Heavy drinking hospital inpatients, many with alcohol-related disorders, ought to be prime candidates for advice on their drinking. But this synthesis of studies found impacts were often lacking; the jolt of a serious injury may have been the best platform for intervention.
Abstract Brief interventions are a time-limited intervention focused on changing behaviour, ranging from a few minutes of information and advice to up to three sessions of motivational interviewing or skills-based counselling involving feedback of assessment results. Risky drinkers among inpatients on hospital wards or in trauma centres are promising targets for brief interventions focused on drinking. These settings provide an opportunity to identify risky drinkers when they are accessible to health professionals, have time for an intervention, and can be made aware of any links between their hospitalisation and their drinking. Following a review supporting similar interventions in GPs' surgeries, this review aimed to assess whether they also reduce drinking and drink-related problems when conducted in hospital inpatient units.
A thorough search was conducted for studies which randomly allocated adult Aged 16 or over. patients to a brief intervention Of up to three sessions. versus no alcohol intervention Apart, of course, from being screened to identify them as heavy drinkers, plus any other research assessments, and in two studies a minimal intervention (leaflet or feedback on screening results). or usual care – or in some other way recruited an adequate comparison group – and then followed up the patients to see whether they had cut back on their drinking. Eligible studies were published in 2007 or earlier and concerned inpatients Except those admitted to units identified as psychiatric or addiction services. admitted to hospital not specifically for alcohol treatment, but who while there were identified as at risk due to their drinking. The search discovered 11 such studies involving 2441 participants. When possible Often it was not due to differences in measures. their results were amalgamated using meta-analysis. A study which uses recognised procedures to summarise quantitative results from several studies of the same or similar interventions to arrive at composite outcome scores. Usually undertaken to allow the intervention's effectiveness to be assessed with greater confidence than on the basis of the studies taken individually.
Combined results from three studies which re-assessed patients six months later showed that on average those allocated to a brief intervention were drinking about 102g Nearly 13 UK units. However, later the review says that "Meta-analysis of 3 studies showed that compared to a control intervention brief intervention reduced the amount of alcohol consumed per week by 69 grams ... at 6 months follow up". less alcohol per week than comparison patients, a statistically significant difference unlikely to have occurred by chance. However, results varied across the studies. One which recorded a much greater difference than the other two featured several follow-up advice sessions, and the people assessing the outcomes knew who had been allocated to the brief intervention. Leaving out this study, brief intervention patients were still drinking 55g Nearly 7 UK units. a week less, but now the difference just fell short of statistical significance. Across two studies which re-assessed patients not six months but a year later, the combined 39g Nearly 5 UK units. reduction in alcohol consumption due to the intervention was not statistically significant.
These results were based on amounts being drunk at the follow-up assessment. Other studies enabled a similar analysis, but based on reductions in drinking from before to after the intervention. Across two studies which assessed patients at 12 months, the extra reduction in drinking due to intervention was statistically significant, but this was not the case in another two studies which re-assessed patients at six months. No other differences in indicators of drinking severity were statistically significant.
The analysts concluded that the benefits of delivering brief interventions to risky drinkers in hospital wards are inconclusive. It was, they thought, conceivable that inconsistent benefits might be due to the fact that simply screening and assessing patients had an impact of its own, which it was difficult for a further brief intervention to improve on. There was however a possibility (based on differences from baseline at six months versus one year) that drinking reductions were more persistent if screening had been supplemented by intervention.
For the
reasons given
The setting provides an opportunity to identify heavy drinkers when they are accessible to health professionals, have time for an intervention, and may be made aware of any links between their hospitalisation and their drinking.
in the featured review, hospital inpatients seem promising candidates for alcohol interventions, yet the review found inconclusive evidence of an impact on their drinking over and above hospital admission and screening/assessment. The isolated positive findings related to drinking amounts at six months post-intervention and reductions from baseline drinking amounts at 12 months. Analysis of the five studies which underpinned these findings suggests that even these findings are questionable. The relevant issues are summarised below; for study-by-study details see background notes.
Of the three studies responsible for the six-month finding, in two there was considerable doubt over whether the drinking reductions resulted from what is normally thought of as a brief intervention. Conducted in Finland, one of these studies recorded an unusually large impact when patients seriously injured (we can fairly assume) as a result of their heavy drinking were repeatedly counselled by a nurse and their doctors. In the third study, a positive contribution to the finding that brief interventions were effective would have been negative had another intervention variant tested in the study been selected for the analysis. In two of the studies, the degree of loss to follow-up, or its imbalance across intervention and control groups, raise concerns over the reliability of the findings or their generalisability to heavy drinking patients in general.
The finding of a significant impact on drinking at 12 months was based on a greater reduction between the pre-intervention assessment and the repeat assessment 12 months later across two US studies. In both, only a minority of risky drinking patients on the wards contributed follow-up data, raising questions over the generalisability of the findings to risky drinking patients in general. By far the most convincing of these studies was conducted on a ward dealing with serious injuries and among patients whose injuries warranted at least a 24-hour stay. Its finding of much greater drinking reductions after intervention than after assessment and usual care may have been biased by the fact that many more of the intervention group had elevated blood alcohol levels. In turn this means many more were likely to have been admitted with a clearly alcohol-related injury, giving them extra motivation regardless of the intervention. The second study, of general medical inpatients, found only minor and statistically insignificant extra reductions in drinking after intervention.
It may be no coincidence that the two studies which recorded the greatest impacts at six and 12 months respectively concerned patients admitted due to serious injuries rather than illnesses; As opposed to patients injured due to intoxication, it could be that for patients with chronic alcohol-related illnesses the dangers of their drinking are either less obvious and convincing, or if obvious, have previously been rehearsed several times by their doctors. In pointing out the dangers, or encouraging the patients to reflect on them, brief interventions might add little to what they already know or to the influences already brought to bear on them. in both it seems virtually certain that many had been injured as a result of fairly extreme drinking. The interventions in these studies were well placed to build on this salutary experience.
However, hindsight rationalisation of why some studies recorded an impact, but others none or a minor one, can only be speculative. The certain conclusion is that impacts on drinking can happen, but are inconsistent and often lacking. Given the nature of the studies, it is also unclear whether such impacts as there were would be replicated in normal practice. All these verdicts are based on overall drinking amounts; if patients had been prompted to manage their drinking better, alcohol-related harm or potential harm resulting from injuries while intoxicated might have been reduced, even if drinking amounts remained unchanged. In the three studies which took relevant measures, this was substantially Though in study 1 the differences were not statistically significant. the case in two (1 2), but not in the third.
Such results give little to reason to rely on hospital inpatient settings for reducing drinking and related problems through brief interventions. On the other hand, such interventions can reduce drinking, and in some studies substantially reduced risk or harm. Since very brief and unsophisticated interventions often work as well as longer ones, even if the pay-off is uncertain, the required investment is minimal. Also the study excluded a model Findings is grateful to James Morris of the AERC Alcohol Academy (www.alcoholacademy.net) for pointing this out. being implemented in some British hospitals which links risky drinking patients to services in the community rather than (or as well as) conducting a brief intervention on the ward. The few formal studies of this approach (1 2) have had at best mixed results, though on the ground (for example in Liverpool) some success has been noted.
Current alcohol screening policy in England and Scotland focuses more on primary care and accident and emergency departments than general hospitals, but the Department of Health's programme of improvement for alcohol misuse interventions saw hospitals as one of the sites for such work, particularly clinics dealing with complaints often related to drinking. In the Welsh national alcohol/drug strategy, accident and emergency departments are the only specific hospital units singled out as a venue for brief interventions. The equivalent document for Northern Ireland makes no mention of brief interventions in hospitals.
Though aimed at other medical settings, practical guidance on alcohol brief interventions is available from a UK web site developed by leading researchers and in an officially published US guide from the American College of Surgeons.
Thanks for their comments on this entry to Jean McQueen of the Ayr Clinic, Partnerships in Care, Scotland. Commentators bear no responsibility for the text including the interpretations and any remaining errors.
Last revised 16 August 2009 Background notes Comment on this entry Give us your feedback on the site (one-minute survey) Back to contents list at top of page
Universal screening for alcohol problems in primary care fails in Denmark and no longer on UK agenda NUGGET 2008
Brief contact and written advice as effective as a longer talk for heavy drinking hospital patients NUGGET 2008
Brief alcohol interventions: can they deliver population-wide health gains? HOT TOPIC 2010
A&E units save health service resources by addressing drinking NUGGET 2006
Injury rate cut in heavy drinking accident and emergency patients NUGGET 2003
Heavily drinking emergency patients cut down after referral for counselling NUGGET 2005
Emergency patients benefit from minimal alcohol intervention NUGGET 2001
The effectiveness of brief alcohol interventions in primary care settings: a systematic review REVIEW 2009
Treating pregnant women dependent on opioids is not the same as treating pregnancy and opioid dependence: a knowledge synthesis for better treatment for women and neonates.Winklbaur B., Kopf N., Ebner N. et al. Request reprint
Addiction: 2008, 103, p. 1429–1440.
New guidance on managing pregnant women dependent on heroin and allied drugs emphasises that maintenance prescribing is the core treatment but holistic, individualised care is essential; its warnings about the dangers of detoxification are not universally accepted.
Abstract Unlike alcohol, cocaine and benzodiazepines, opioid use during pregnancy does not cause birth defects or damage foetal cells. Nevertheless, fluctuating levels of opioids in the mother's blood may lead to withdrawal symptoms or overdose in the foetus. Expectant mothers who are actively misusing substances are also at high risk of violence, poor nutrition, and inadequate obstetric care. Those continuing to inject are at heightened risk of medical complications such as infectious diseases, endocarditis, abscesses and sexually transmitted diseases. Pregnant women are typically excluded from clinical trials of medications, limiting the adequacy of the evidence base for their treatment, and the response to these women is often complicated by psychological problems and their use of other substances. To develop recommendations to improve patient management from conception to postnatal follow-up it was decided to synthesise available research Found by searching the US government medical database PubMed for articles published between 1998 and 2006, plus older key texts. Due to the limited number of randomised controlled designs, the review included follow-up studies without a control group and/or where the patients knew which treatments they were being given. with the authors' extensive clinical experience.
Broadly their advice was that in the short and longer term, mother and child do best if multi-disciplinary treatment is initiated as soon as possible, maintenance prescribing is permitted, and there is regular monitoring. Opioid maintenance therapy is the recommended approach during pregnancy. Treatment must take account of the full clinical picture, particularly the frequent complications arising from psychiatric problems and the use of other substances. Babies born to mothers maintained on opioid medication frequently suffer withdrawal symptoms which may require treatment. Selected further details below.
From whichever of the varied gateways women enter prenatal care, it is essential that the whole healthcare team from receptionist to physician offer assessment, triage, case coordination and referral services in a supportive, culturally sensitive, and non-judgemental environment. Case management is crucial to coordinate care of the women and their families. Ideally, a case conference and referral to appropriate services should be managed by one healthcare professional who oversees the multi-disciplinary team. This team should include an outreach worker to (if required) visit the woman at home.
A comprehensive examination The Addiction Severity Index is recommended because it assesses not just substance use but medical, psychological and social issues. of the multiple problems Key areas include drug/alcohol use, legal, family/social and psychiatric status, and a detailed health history. of this population and standardised documentation and evaluation of all prescribed therapies are very important. Special attention should be given to psychiatric problems, and appropriate treatment initiated. In particular, an opioid-maintained patient with untreated depression may relapse and could be difficult to stabilise. Treatment planning and management should include the woman's partner. If they too are opiate dependent, preferably the same treatment should be provided for both partners. Throughout the woman should be closely monitored, checking psychiatric condition, urinalysis for non-prescribed drugs, and uptake of nutritional supplements.
Health care providers need to be vigilant about drug interactions. For example, enzyme induction may affect levels of opioids in the blood, requiring dose adjustments to avoid destabilising the mother-to-be or her foetus. Screening (and if tests are negative, later rescreening) for HIV and hepatitis infections are recommended to minimise dangers for the foetus.
Detoxification or maintenance using opioid medications should follow established procedures. Detoxification should be undertaken only if the patient is stable and clearly committed to abstinence, and not after 32 weeks of pregnancy. The recommended procedure is a slow, medically supervised taper of a substitute opiate-type medication. Well-controlled medication is safer than continued use of street drugs and improves antenatal monitoring in cases otherwise difficult to engage. To avoid additional use of non-prescribed drugs, doses should be set individually rather than reduced to a set minimum. The 'gold standard' for treating pregnant opioid-dependent women is methadone, though buprenorphine seems equally effective and safe. Opioid antagonist treatment is contraindicated because of the risk to the foetus.
If required, detoxification from benzodiazepines should be gradual to reduce risk of pre-term labour or psychiatric symptoms. Psychosocial therapies are the first line option for mild depression; medication and inpatient care are indicated for major depression with suicide risk. If antidepressants are used, the case management team should be briefed to look for drug interactions, and neonates should be monitored carefully for complications after birth. Obstetricians should be alerted that the patient's maintenance opioid is insufficient for perinatal analgesia, and advised to select opioid analgesics with a view to avoiding precipitated withdrawal.
Mothers should stay in hospital for 5–8 days while the baby is monitored for signs of the neonatal abstinence syndrome. Even if they are maintained on methadone or buprenorphine, those who want to breastfeed should be encouraged to do so, provided they are seronegative for HIV, not misusing other substances, and there are no other contraindications. Where relapse is likely, opioid maintenance should continue after delivery. If the mother is on opioid medication, weaning the baby must be carried out under medical care to avoid withdrawal. It is important to discuss contraception to avoid early unplanned consecutive pregnancy. Finally, the team should focus on stabilising the home environment for the developing child and arranging health service assistance.
Around the same time as this team from the Medical University of Vienna published their recommendations, so too did a largely US team which included one of the authors from Vienna. Their more restricted focus was on the appropriate use of methadone and buprenorphine in the treatment of opioid-dependent pregnant women prior to and during delivery, based partly on their experiences in the first large-scale, controlled comparative trial of these medications with opioid-dependent pregnant women. Usefully they provide a North American perspective and much more detailed guidance on the opioid maintenance therapies which the featured review sees as the primary response. Guidelines are offered on induction, stabilisation, preventing and managing relapse, medication during labour and delivery, pain relief, breastfeeding, the interactions of methadone and buprenorphine with other medications, and managing psychiatric conditions. Throughout the emphasis is on a flexible, non-judgemental stance.
British addiction treatment guidelines agree with the reviewers that substitute prescribing is preferable to continued illicit substance use. By facilitating engagement with health services, this treatment also enables health and social needs to be identified and advice given to improve outcomes. Because of the risk of spontaneous abortion or foetal distress, detoxification is generally considered advisable only in the second trimester in the form of a gradual taper, and the guidelines emphasise that maintenance prescribing should be strongly encouraged if illicit drug use continues or recurs. Though buprenorphine is not licensed for this purpose, the guidelines say that pregnant women who are stable on buprenorphine may best continue on this medication rather than switch to methadone. The authors agree with the reviewers that research shows buprenorphine is as safe as methadone. Breastfeeding is encouraged even if the mother is on methadone (though the dose should be kept as low as possible) and/or continues to use drugs, except for cocaine or a very high dose of benzodiazepines.
The review's and the guidelines' clear preference for maintenance prescribing and, if undertaken, only gradual detoxification and then only in certain phases of pregnancy, is not universally accepted. Though early case reports of foetal distress associated with withdrawal raised alarm, more recent studies with larger samples have not found adverse foetal consequences. Experienced British clinicians argue (1 2 3) that the theoretical risk to the foetus is not borne out in practice. Their advice is that as long as the woman can tolerate withdrawal and is willing to try it, and it is unlikely to lead to uncontrolled substance use, withdrawal can occur at any speed and at any stage of pregnancy. In particular, experience in Britain suggests that given these conditions, there is no reason to discourage women who want to withdraw late in pregnancy in order to avoid their baby suffering withdrawal symptoms. Relevant research is summarised below.
Early concerns were eased when in 1998 a study systematically monitored foetuses for signs of distress during the mothers' detoxification. There were none, and neither were there any miscarriages, despite the fact that some of the mothers were prescribed only clonidine to relieve their withdrawal symptoms and half were detoxified during the last third of pregnancy. One of the few studies to test different regimens took advantage of the switch at a specialist US centre for drug dependent pregnant women from a three-day to a seven-day methadone detoxification; in both phases, other women were prescribed methadone on a maintenance Some after detoxification. basis. The briefer detoxifications were conducted throughout pregnancy yet none of the 140 women suffered a miscarriage. There was one such event among 70 patients withdrawn more slowly and three among the 116 methadone-maintained patients. Measures of neonatal welfare revealed no adverse effects of withdrawal, or of brief withdrawal in particular. Where maintenance prescribing did win out was in retaining patients at the centre both for the births and for their addiction treatments, with possible longer term benefits. Even when conducted in the last third of pregnancy, hospital records in London revealed no elevated risk of miscarriage due to methadone detoxification, though in this study the reduction schedule was spread over three weeks in an inpatient unit. Similarly in Birmingham, a specialist mother and baby unit in a hospital addiction treatment centre noted no particular problems among the many opiate-dependent women who opted to slowly withdraw during pregnancy. In Glasgow, even abrupt 'cold turkey' withdrawal has not been found to cause elevated rates of miscarriage.
The recommendation for particularly slow withdrawal of benzodiazepines was based on the desire to avoid precipitating premature labour or exacerbating psychiatric symptoms. Set against this is the need to minimise foetal exposure to drugs which can cause birth defects and withdrawal symptoms in the neonate. Experience at a specialist unit in Glasgow is that inpatients can safely be withdrawn over about a week depending on the woman's ability to cope, though outpatient withdrawal may best be slower.
Opinions in Britain differ too on whether, as the featured review suggests, continued non-medical substance use means breastfeeding is inadvisable. Instead it has been argued that the vulnerability of the baby is such that breastfeeding should be encouraged, regardless of the type of drugs being used by the woman or their dosage (1 2). A major benefit is that by delivering small amounts of the drugs to which the foetus had been exposed during pregnancy, breastfeeding substantially reduces the risk of withdrawal symptoms.
A review for the Cochrane collaboration tentatively supported both opiates and phenobarbitone for the treatment of withdrawal in babies born to opiate-dependent mothers.
Thanks for their comments on this entry in draft to Mary Hepburn of the Glasgow Royal Maternity Hospital. Commentators bear no responsibility for the text including the interpretations and any remaining errors.
Last revised 18 August 2009 Comment on this entry Give us your feedback on the site (one-minute survey) Back to contents list at top of page
The Drug Treatment Outcomes Research Study (DTORS): final outcomes report STUDY 2009
Addressing medical and welfare needs improves treatment retention and outcomes NUGGET 2005
Prescription of heroin for the management of heroin dependence: current status REVIEW 2009
International review and UK guidance weigh merits of buprenorphine versus methadone maintenance NUGGET 2008
Traditional medicine in the treatment of drug addiction REVIEW 2009
Critical issues in the treatment of hepatitis C virus infection in methadone maintenance patients REVIEW 2008
Effective dissemination: a systematic review of implementation strategies for the AOD field.Bywood P.T., Lunnay B., Roche A.M.
Adelaide: National Centre for Education and Training on Addiction, 2008
Comprehensive Australian review garners the lessons from across health promotion and medical care on how best to improve practice by introducing research-based innovations, and evaluates their applicability to substance misuse.
Abstract Australia's National Centre for Education and Training on Addiction acts as the country's centre for the development of the workforce responding to substance use problems. To help identify the best ways to narrow the gap between research-based knowledge and its application in practice, the centre undertook a systematic literature review of the most common strategies to enhance uptake of innovations in professional practice. Though most of the evidence derived from non-substance specific health and medical fields, the review sought to determine its applicability to substance use.
A literature search found 25 relevant systematic reviews plus 85 additional primary studies. These were used to assess 16 dissemination and implementation strategies in terms of changing professional practice and the impact of these changes on welfare and substance use. As well as rarely being focused on substance use, the evidence was typically of average quality and most studies were prone to some degree of bias. Nevertheless, findings from better quality studies indicated that some strategies can improve professional practice in sectors including:
• Preventive care; for example in the substance misuse sector, advice on smoking cessation or substance use in pregnancy or on the risks of alcohol or drug use more generally, and screening for actual or potential problems;
• Treatment; for example, methadone or nicotine replacement and psychosocial interventions like brief advice to risky drinkers, motivational interviewing and cognitive-behavioural therapy;
• Disease management; for example, management of patients on pharmacotherapies, of co-occurring psychological problems, or of chronic illness related to substance use such as hepatitis C and HIV infection and cirrhosis of the liver;
• Rehabilitation; for example, management of relapse in dependent clients; and
• Palliative care; for example, for clients with terminal illnesses related to substance use.
Of the 16 evaluated strategies, the four most successful across these clinical sectors are described below.
Interactive educational meetings are often conducted under the umbrella of continuing medical education. Examples include educational conferences, meetings, seminars, workshops, lectures and symposia. These resulted in small to moderate improvements in practice, particularly when participants were starting from a relatively low base, and sometimes (but inconsistently) were also shown to have improved patient outcomes. Interactive or personalised formats worked best. Improvements were most likely when meetings were followed up for example by telephone consultations, and participants were incentivised through mechanisms such as feedback on performance and medical education credits.
Educational outreach visits, also known as 'academic detailing', involve a trained health educator or specialist (typically a physician or pharmacist) visiting health care providers in their own workplaces to inform them about the evidence for a specific intervention or practice guideline with a view to encouraging its uptake. In general, these have improved practice in a range of settings and when they did so, the impacts were moderate to large. The few studies which investigated this generally found no impact on patient outcomes. Again, an interactive format and follow-up contacts helped.
Prompts and reminders (computerised or manual) prompt the practitioner to take certain actions or record key information when research suggests these are warranted by the situation and will improve patient management. They may be single-action systems such as prompts to deliver a brief intervention when screens indicate risky drinking, or may be elaborated in to decision support systems recommending a series of actions from a number of alternatives. Studies recorded mixed effects in respect of both practice improvements and patient outcomes. Practice improvements were commonly noted in preventive care, prescribing and disease management, but less so in diagnostic practice, while the few studies of patient outcomes rarely demonstrated significant positive effects. Systems worked best when the clinician could not suppress the prompt/reminder and was obliged to respond, even if only to acknowledge it.
Audit and feedback involves gathering information For example, from medical records, databases, patients or by direct observation. on a practitioner's performance, comparing this against a benchmark such as their peers' performance or clinical guidelines, and feeding back the results to the practitioner with or without recommendations. The rationale is that practitioners who are made aware of suboptimal behaviour will seek to improve. Research shows that commonly they do improve (though to a highly variable degree), but the few studies which measured patient outcomes found little or no impact. Impacts on practice were greatest among practitioners whose initial performance was least satisfactory, and when feedback was relatively intensive. Intensive feedback might, for example, entail feeding back verbally to each individual or prolonged feedback from a senior colleague. Non-intensive feedback might entail just written feedback, feedback to groups, from a less experienced colleague, or lacking personal incentives for improvement.
Cutting across the different types of dissemination strategies are some common features both of the strategies and of the contexts in which they operate which may facilitate or inhibit implementation. The most successful strategies provide clear and succinct messages, with simple, focused objectives, are reinforced by additional materials and support, and require small, practical changes. Effective messages are presented by reliable and credible sources (backed by research-based information) in an appealing and persuasive interactive format which encourages participation, are tailored to the local setting, and relevant to the needs of practitioners and clients. Systems or procedures are most likely to be implemented if they are accessible, demand little effort to comply, but require the practitioners to respond. Successful strategies are based on an assessment of the barriers to change, address change at multiple levels, including the individual practitioner, organisational structure and culture, and health system policy, and are sustainable over a prolonged period.
Contextual factors which may enhance effectiveness include identifying the need for change, communicating this to the target audience in a way which motivates change, providing resources and staffing to integrate change into existing systems, and ongoing monitoring of how well an innovation is being implemented to identify and correct 'drift'.
However, the reviewers cautioned that the evidence base was of only moderate quality, few studies assessed impacts on clients, and those which did found little or no benefits. This may have been because studies were typically conducted over a short time period; longer follow-up and reinforcement of change may be needed to detect such benefits. Since no single strategy proved effective in all situations, careful selection is required to ensure the best match with the topic or behaviour in question and the people and institutions targeted for change.
The featured review was comprehensive and seems to have truly deserved the term 'systematic' in its title, but was hampered by the quality of the evidence and its lack in the substance misuse sector. Even more hampered was an accompanying report which extracted cost-effectiveness data from the same set of studies reviewed in the featured report. It found that of the 25 reviews and 85 additional studies, just two reviews and 14 studies reported costs, and generally the methodological quality of the economic analyses was poor. Nevertheless it was concluded that educational meetings under the rubric of continuing medical education were generally effective and cost-effective, while educational materials alone were relatively cheap but also relatively ineffective.
The review focused on formally evaluated strategies, which have mainly been implemented at the level of the service or individual practitioner. As it acknowledged, beyond these are more macro health system and administrative structures which can themselves be agents for change. These and other factors also critically influence whether the reviewed strategies have a chance of making a difference. Some examples below.
Outside the review's scope were more macro, system-wide change efforts driven for example by national priorities and targets, backed by incentives or sanctions to promote those policies. Examples in Britain have included the numbers-in-treatment and waiting time targets which have improved treatment access. Similarly, in the US state of Delaware, state treatment recruitment and engagement targets promoted through financial sanctions and incentives applied to services resulted in an 87% increase between 2001 to 2006 in the average number of patients in treatment.
The Delaware study is an example of the harnessing of market mechanisms to public priorities. Other market mechanisms which contain costs but also require the demonstration of value and quality also have an impact. Most researched have been US managed care cost-containment mechanisms, and US quality-assurance accreditation systems, which seem to have had opposing effects on the quality of treatment for substance use problems.
Change strategies of any kind can only work well if the organisations and people at which they are targeted are willing and able to change and to sustain that change. Researchers at the Institute of Behavioral Research in Texas have been conducting a determined attempt to measure and improve organisational 'health' including ability to adopt new practices. In England they found that client participation in treatment was greater in services characterised by team working and mutual trust among staff, and which encouraged discussion and implementation of new ideas and procedures. One reason may be that staff in such agencies are personally most able to implement new learning gained in training. Across behavioural health services in general, agencies which prioritise achievement, individualism and self-actualisation rather than security are most likely to be open to change. However, more prosaic issues like high staff turnover and the 'churn' in organisations due to market forces and commissioning cycles have severely limited the sector's capacity for accumulating and implementing learning (see for example 1 2 3). For more on these issues click this link to trigger a search on the Findings site for studies related to organisational functioning.
There is also the possibly crucial but under-investigated issue of recruiting people whose personal qualities and 'training' in life as much as in their professions equip them for the tasks of relating (in prevention) to young people on highly sensitive issues, or (in treatment and harm reduction) to society's most stigmatised, damaged and disadvantaged members. Research has typically sought to to purge the data of the influence of the person(s) delivering the intervention, in order to isolate the impact of the intervention itself, rather than to investigate them as an active ingredient. Nevertheless, examples of their importance include studies of motivational interviewing, probably in the UK the most influential therapeutic approach for substance use problems. In one such study, client engagement was unrelated to 'trainable' features such as the frequency with which the therapist made statements seen as compatible (such as open questions) or incompatible (such as warnings) with the approach. However, engagement was strongly related to embodying the overall spirit of motivational interviewing and to more general social attributes including empathy, warmth, supporting the client's autonomy, and coming across as 'genuine'. Within (and only within) the kind of empathic, caring context socially skilled therapists were able to create, 'breaking the rules' by making interventions supposedly incompatible with motivational interviewing actually deepened the client's engagement, perhaps because the therapist came across as more genuine. Another study of motivational interviewing training found that initial gains in skills had waned two months later. However, this was not the case for the addiction and mental health clinicians who, even before training, had been more proficient than the other trainees would be after training. Not only did these 'natural experts' start from a higher level, they went on to absorb and retain more of what they had learnt. In contrast, within months much of the training was wasted when it fell on less fertile human ground. Among this set of trainees, given a choice between choosing the 'right' people who had not been trained, and the 'wrong' people who had, the right people would have been the better choice. A further example from Britain concerned brief interventions based on motivational interviewing versus simple advice and information, intended to prompt college students to curb their cannabis use. In this context, some interventionists were markedly more successful with motivational interviewing than simple advice, while others did slightly better with the more straightforward and familiar advice option. In contrast there was little difference in outcome between the approaches overall, normally the only finding to be reported in such studies. What made the difference it seemed was not the approach itself, but the combination of approach and individual.
In the treatment sector, many of these therapist qualities are among the 'common factors' which cut across specific therapies and seem responsible for much of the benefits of psychosocial therapies in general. Qualities such as optimism, empathy, and respect for the patient or client are not easily susceptible to change via the strategies featured in the review. In contrast, specific therapies can be codified and disseminated via these strategies, but are unlikely to have major impacts on patients; even when expertly implemented in well resourced research projects, generally they make little difference to outcomes. For more on these issues see for example the Findings analysis of a review of cognitive-behavioural therapies; click this link to trigger a search on the Findings site for other studies of the qualities of the therapist.
Factors such as institutional and personal receptivity to change, and personal and organisational qualities not easily susceptible to formal change strategies, may have been one reason why the review found that the impacts of these strategies – especially on patients and clients – were inconsistent and often minor. Nevertheless, innovations of the kind recommended in the featured review – sharply targeted, highly tailored and feasible in practice – can be integrated in to practice and make substantial differences to, for example, treatment uptake and aftercare adherence. One clear example was the system of simple, inexpensive prompts and patient motivators introduced at a US inpatient treatment centre which substantially improved aftercare attendance and helped sustain progress made during initial treatment. Another was the introduction in Philadelphia of a computerised system to link assessed client needs to local services addressing those needs. Not only did this improve 'wrap-around' service provision, it more than doubled the completion rate of the core treatment programmes. Innovations such as these to improve quasi-administrative procedures hold at least as much promise as the introduction of evidence-based therapies.
Thanks for their comments on this entry in draft to Ann Roche of the National Centre for Education and Training on Addiction at Flinders University in Australia. Commentators bear no responsibility for the text including the interpretations and any remaining errors.
Last revised 10 August 2009 Comment on this entry Give us your feedback on the site (one-minute survey) Back to contents list at top of page
Matching resources to needs is key to achieving 'wrap-around' care objectives NUGGET 2006
The motivational hallo THEMATIC REVIEW 2005
The power of the welcoming reminder THEMATIC REVIEW 2004
Wet day centres in Britain SERIES OF ARTICLES 2005
Wet day centres in Britain part 2: Care Control Challenge IN PRACTICE 2005
Oiling the wheels IN PRACTICE 2000
Evidence-based practice? The National Probation Service's work with alcohol-misusing offenders STUDY 2009
Guide to implementing family skills training programmes for drug abuse prevention REVIEW 2009
General practices can be trained to help families cope with drinkers and drugtakers NUGGET 2001
