Organizational- and individual-level correlates of posttreatment substance use: a multilevel analysis
Effectiveness bank home page. Opens new windowResearch analysis

This entry is our analysis of a study considered particularly relevant to improving outcomes from drug or alcohol interventions in the UK. The original study was not published by Findings; click Title to order a copy. Free reprints may be available from the authors – click prepared e-mail. The summary conveys the findings and views expressed in the study. Below is a commentary from Drug and Alcohol Findings.

Links to other documents. Hover over for notes. Click to highlight passage referred to. Unfold extra text Unfold supplementary text
Copy title and link | Comment/query |

Organizational- and individual-level correlates of posttreatment substance use: a multilevel analysis.

Ghose T.
Journal of Substance Abuse Treatment: 2008, 34(2), p. 249–262.
Unable to obtain a copy by clicking title? Try asking the author for a reprint by adapting this prepared e-mail or by writing to Dr Ghose at

Using advanced methods, this US study asked what makes for an effective treatment agency. Being constrained by funders in terms of services and ability to individualise treatments was the clearest negative factor, quality accreditation the clearest positive.

Summary The objective in this US study was to identify what it is about a service or about its patients which makes for good treatment outcomes. To do this it used sophisticated 'hierarchical' models to disentangle relationships between post-treatment substance misuse and features of the organisation and the individual patients. Such methods are able to correct for the fact that treatment services operate within a broader environment, and that individual patients are clustered within treatment services. Without these corrections, multiple 'nesting' can give rise to false associations. For example, some apparent links between individual characteristics and treatment success no longer held when nesting was accounted for. The analysis had available to it one of the world's richest and largest datasets It used data from the Alcohol and Drug Services Study (ADSS) 1996–1999, a national treatment study conducted by the US government. The dataset permits an examination of a more comprehensive set of organisational and individual-level variables than previous similar studies. The current analysis relied on a random sample of 155 residential and outpatient non-methadone treatment facilities. Records of 2670 clients eligible to be followed up were abstracted from these facilities to gather information on the clients and their treatments. Three years after leaving treatment, 1184 clients from 128 facilities were interviewed to establish long-term outcomes. documenting features of the treatment services, the patients, and their substance use three years after leaving treatment.

Different services in different areas saw on average different types of patients. But even after this had been taken in to account, the probability of post-treatment substance misuse Including any use of: crack/cocaine; heroin; cannabis; amphetamines; hallucinogens; non-prescribed methadone, opiates, sedatives or tranquillisers; and non-medical use of over-the-counter medications. varied significantly across services. The single most important factor was the degree to which a service was funded through managed care A US system in which non-medical administrators such as insurance companies control and limit the provision of medical procedures and medications and allied services. A primary aim is to controls costs by restricting patient choice and limiting clinical discretion, though 'levelling up' and quality improvement may also be objectives. arrangements intended to contain costs; the more it relied on these funding sources, the greater chance that its patients would use drugs after treatment. On the other hand, patients were much less likely to misuse substances if they had been treated by services with a recognised quality accreditation. Namely that offered by the Joint Commission on the Accreditation of Healthcare Organizations. Now known as The Joint Commission (see, this not-for-profit agency aims "To continuously improve the safety and quality of care provided to the public through the provision of health care accreditation and related services". In both cases the links were far stronger for residential than non-residential services.

Several characteristics of the individual patients (such as their ages and main drugs of choice) were also related to later substance misuse. After these and organisational factors had been taken in to account, longer treatment episodes and completing treatment remained strongly associated with a reduced risk of later substance misuse. However, the apparent benefits of length of stay were reduced in services more reliant on managed care funding, subject to monitoring by a parent organisation, whose workers had large caseloads, or which had higher proportions of highly qualified (doctorate level) staff.

Findings logo commentary The observation in Britain and elsewhere that broadly similar services differ greatly in outcomes has driven attempts Leading investigators include Dwayne Simpson and his team at the Institute of Behavioral Research ( in Texas, who are working with the English National Treatment Agency for Substance Misuse to improve performance by diagnosing and rectifying organisational processes. to establish what accounts for this variation and to rectify poor performance. Mechanisms like accreditation and managed care are major ways to achieve this objective. Findings from the featured study, probably the most reliable to emerge from the USA, suggest that on-site inspection and accreditation has substantially improved substance use outcomes, while an externally imposed value-for-money mandate motivated by cost-containment has even more substantially eroded them. In each case there are more or less distant parallels in the UK. Because impacts depend on how these mechanisms are implemented and the services involved, and because of the different national contexts, these US findings are best seen as indicative of potential impacts in the UK.

The positive implication is that having a relatively widely used quality accreditation process does improve treatment processes in ways which also improve outcomes. However, an alternative explanation is that the minority of services which sought accreditation were already implementing quality processes and achieving good outcomes, for which they sought recognition. One study which advanced this explanation found that different accreditation and licensing processes differed in the degree to which (if at all) they were associated with various indicators of quality. The featured accreditation agency based its awards largely on on-site inspections (now typically unannounced). Services had strong incentives In the US context, partly competitive and financial. to raise their games to meet the agency's standards, and doing so may have been a condition of funding.

In contrast, England's inspection process dedicated to substance misuse services is largely paper-based, involving on-site visits Other than for data collection. only to the 'worst' 10% of services. Within this process there are no minimum standards services must meet to receive public funding, good services do not receive public recognition, and competitive pressures are muted compared to those in the USA. Much closer to the model tested in the study is the inspection work of the Commission for Social Care Inspection in England and allied bodies elsewhere in the UK. They inspect registered care homes which include most residential rehabilitation houses dealing with substance users. Inspections are usually unannounced and the reports and quality ratings are made public on the commission's web site. Services are assessed against national minimum standards but inspectors also attempt to judge how far they are making a difference to the lives of the residents. However, non-residential drug/alcohol services and unregistered care homes generally fall outside the commission's ambit, and English service providers have agreed that the process "provides little oversight of the quality or appropriateness of the treatment programme itself".

On the debit side in the featured study was the association between poor outcomes and funding through the cost-containment mechanism of managed care. This fetters the discretion of patients and providers to provide expensive Either in time, breadth, intensity or the cost of staff time and facilities. services, but also aims to eliminate unnecessary or less effective practices and to mandate good practice. Here again though, the possibility remains that reliance on managed care was not a causal factor but merely reflected features which, whatever the funding arrangements, would have led to poorer outcomes. If managed care was detrimental, how that came about is unclear. The researcher highlighted the pressure to de-individualise service provision, apparent in another analysis drawing on the same dataset which found that specialised treatment for HIV positive drug users was much less likely in services reliant on managed care. Though de-individualisation may be the typical result, it can be countered. Aware of this risk, in Oregon authorities paired managed care implementation with mandatory guidelines on the intensity of care required to match individual need and discharge criteria which took account of the patient's progress. The combination actually enhanced individualisation of treatment placement and discharge Though the average length of stay did not differ from that in a comparator state, there was significantly greater variability. and increased the use of more intensive (and expensive) outpatient options.

Possible implications of the managed care findings for the UK relate to the 'value for money' exercise being mounted by the National Treatment Agency for Substance Misuse A special health authority within the NHS, established by government in 2001 to improve the availability, capacity and effectiveness of treatment for drug misuse in England. (NTA). This attempts to link model treatment system components with the standard or typical costs of those components. The aim is to offer a benchmark to local drug action teams to help level up quality and identify possible economies. With falling per-patient funding for treatment, it also has the unintended potential to be used as a cost-containment tool to cap and de-individualise service provision. In all these respects, the exercise shares aims and possibilities with managed care, though it also differs in important ways. Notably the UK exercise does not directly force cost-capping or standardisation through funding mechanisms; instead pressure to maintain or increase patient numbers with in real terms less funding may have a similar effect.

The featured study indicates the potential for such initiatives to worsen outcomes, but this is not inevitable. Several other There are several studies which found no or a mixed impact. The study linked to next as an example is particularly useful for its mini-review of the literature. more limited studies did not find such a relationship. Managed care funders vary in which treatment processes they seek to control and how stringently, and in their focus on cost-containment versus quality improvement. Impacts of managed care also differ for different treatment modalities. In the featured study the apparently detrimental impact was much greater on residential than non-residential services, presumably because funders were more concerned to contain expensive residential provision. Similarly, a national US survey of youth substance misuse services found that, as expected, in residential services quality was impaired when funders were prepared to pay less for therapeutic programmes. But this was not the case for non-residential services, perhaps (the authors suspected) because even the highest levels of funding available to them were insufficient Relatedly, in Oregon researchers suspect that the advent of managed care was unrelated to outcomes because outcomes were already so poor that little detriment was possible. See: McFarland B.H. et al. "Outcomes for clients of public substance abuse treatment programs before and after Medicaid managed care." Journal of Substance Abuse Treatment: 2005, 28(2), p. 149–157, to support quality improvement initiatives.

Methadone services were excluded from the featured study. In another national US study, managed care was associated But not always to a statistically significant degree. with increased provision of psychosocial therapies and more drug/alcohol testing in methadone maintenance services and better discharge planning, but also with a lesser focus on reintegration through employment and housing or testing for infectious diseases. In respect of medical services, similar relationships were found among drug-free counselling agencies. The findings were broadly consistent with managed care resulting in a narrowing in on core services, implementing these more consistently but at the cost of broader medical, social and public health concerns.

One of the clearest illustrations of how managed care can diminish quality and outcomes comes from a US study of a specialist service for drug dependent pregnant women. Counselling time and with it therapeutic content were reduced and limits on methadone treatment durations imposed by health insurance companies forced mothers to drop out or seek alternative providers, fragmenting care. It is unclear whether these mechanisms resulted in the worsening in neonatal and child welfare outcomes.

Thanks for their comments on this entry in draft to independent consultant Richard Phillips. Commentators bear no responsibility for the text including the interpretations and any remaining errors.

Last revised 18 February 2009. First uploaded

Open Effectiveness Bank home page

Top 10 most closely related documents on this site. For more try a subject or free text search

STUDY 2010 Gender differences in client-provider relationship as active ingredient in substance abuse treatment

STUDY 2011 Performance-based contracting within a state substance abuse treatment system: a preliminary exploration of differences in client access and client outcomes

STUDY 2002 The grand design: lessons from DATOS

STUDY 2012 Advancing recovery: implementing evidence-based treatment for substance use disorders at the systems level

STUDY 2010 Does meeting the HEDIS substance abuse treatment engagement criterion predict patient outcomes?

DOCUMENT 2017 Drug misuse and dependence: UK guidelines on clinical management

STUDY 2008 Does following research-derived practice guidelines improve opiate-dependent patients’ outcomes under everyday practice conditions? Results of the Multisite Opiate Substitution Treatment study

DOCUMENT 2009 Guidelines for the psychosocially assisted pharmacological treatment of opioid dependence

MATRIX CELL 2018 Drug Treatment Matrix cell D2: Organisational functioning: Generic and cross-cutting issues

STUDY 2012 Randomized trial of standard methadone treatment compared to initiating methadone without counseling: 12-month findings