Risk adjustment of heroin treatment outcomes for comparative performance assessment in England

Abstract

There is some exceptional variability in the apparent effectiveness of the English treatment system for substance use disorders. It is important to determine the source of this variability in order to inform drug treatment delivery and its evaluation both in England and overseas.

Effectiveness of treatment for opioid use disorder: A national, five-year, prospective, observational study in England

“…A fixed-effects approach was used to determine the relative effectiveness of each local treatment system because random intercepts could mask real variation in achieving the outcome (Marsden et al, 2012). For each treatment system, predicted outcome probabilities were summed across patients and divided by the number of patients treated in the area.…”

“…Unlike other comprehensive administrative datasets (Sahker et al, 2015;Alterman et al, 2001;Stahler et al, 2016), NDTMS has patient-level identifiers that enable cross-referencing with subsequent treatment admissions. This utility provides not only an objective summative measure of the sustainability of recovery in this population, it also enables national administrative systems to objectively capture whether patients had previously accessed treatment services, which is an important negative predictor of treatment outcome (Siguel and Spillane, 1978;Marsden et al, 2012).…”

“…At the national level, the impact of this is twofold: it is not possible to objectively assess whether an individual has previously engaged in treatment (an indicator of patient-level complexity (Marsden et al, 2012;Siguel and Spillane, 1978). It is also not possible to determine whether a patient's successful completion status is enduring.…”

Is treatment for alcohol use disorder associated with reductions in criminal offending? A national data linkage cohort study in England

“…NDTMS includes almost all specialist services for alcohol and drug use disorders. It reports annually on the characteristics of patients treated (see Public Health England, 2014a, 2014b and provides national outcome monitoring of treatment episodes and performance benchmarking for local treatment systems (see Marsden et al 2009;Marsden et al, 2012).…”

“…We followed an general evaluation approach used by our group in which an outcome measure is adjusted by pre-treatment offending; patient level demographics; local area deprivation; clinical indicators of clinical disorder severity and complexity; and summary measures of treatment exposure (see Marsden et al 2012;Marsden et al, 2014).…”