The New York State Office of Alcoholism and Substance Abuse Services (OASAS) envisions a performance-driven system of care, one that is adaptable to the changing needs of individuals with substance use disorder (SUD) and their families. A learning system that is integrated with healthcare, person-centered, utilizes the best addiction technology available and can prove its benefits to patients. This will require us to be able to track outcomes at an individual, program, network and system level so that we can target interventions and resources. In this article, I will give an overview of several of these projects with a broad update on each and the overall research plan.
OASAS and the Center on Addiction worked extensively with the providers and payers to develop the LOCADTR 3.0 application. The LOCADTR is an algorithm for determining a recommendation for level of care for individuals in need of treatment based on a clinical assessment of current risks associated with use and the resources available to the individual in her or his community. The collective wisdom of everyone who participated in workgroups and testing of the tool has helped us to make a practical clinical tool that will have tremendous potential to inform policy, identify service gaps and identify factors impacting outcomes. To date we have a data base of over 830,000 completed LOCADTR assessments. We can connect 80% of individuals who had a LOCADTR to data in our statewide registry of treatment episodes (Client Data System) so we can study system performance, including capacity gaps in service needs and patient outcomes.
We recently published an article in The Journal of Behavioral Health Services and Research describing the development of the tool and its reliability. The paper was comprised of two studies: one in which 139 clinicians were tested for interrater reliability and the second study looked at the characteristics of system overrides to the initial LOCADTR recommendation. In the first study, we found acceptable interrater reliability (IRR = 0.57-0.59) in the administration of the LOCADTR. In the second study of nearly 390,000 LOCADTR administrations we found that clinicians changed the clinical algorithm determination in only 10% of the administrations. We will continue to study the LOCADTR data and make adjustments to the tool and/or our system of care informed by our analyses.
This year we rolled out a concurrent review tool to complement the LOCADTR using the same process with providers and payers to develop the tool. The concurrent review includes questions to determine the appropriateness of continued stay in the current level of care as well as guide clinicians through a process of documenting provision of care in accordance with the clinical practice guidelines.
The LOCADTR and Continuing Review tools provide a rich trove of data for examining treatment system gaps as well as performance. They provide data on an individual level that include: SUD symptoms, primary substance of focus, risks including medical and psychiatric, hazardous use patterns, interpersonal and intrapersonal patterns that interfere with recovery goals, and social, interpersonal, community and personal resources available to support recovery goals. Together with the admission and discharge information gathered, that includes data on admission and discharge disposition, criminal justice, mental health, medical, substance use and employment, we will be able to identify correlations between factors that relate to successful engagement and outcomes and support, and modify algorithms within the tool.
Our team at OASAS has been working closely with the Department of Health to establish a practical yet adequate set of metrics that will be included in forthcoming value-based purchasing (VBP) contracts. Metrics for VBP need to fit certain criteria: feasibility, parsimony, consistency with other healthcare metrics, and clinical meaningfulness. The challenge we face is that there are few national measures of effectiveness or quality of substance use treatment that have been validated. Given the current changes in the health care delivery system and the crisis facing the state due to the opioid crisis, we recognized the urgent need to develop and implement metrics. OASAS and the Center on Addiction worked with the Medicaid Clinical Advisory Group, Managed Care Plans, Providers and the Department of Health to more fully develop and implement several measures to support Value Based Purchasing arrangements. Some of these new measures assess whether patients continue in care after inpatient discharge, use of FDA approved medications, and retention in care. OASAS expects to extend their use to monitoring provider performance across payers. These measures have support in the literature and are process indicators that are correlated with outcomes. We recognize that work in this area is still very early and that metrics that are more focused on actual outcomes associated with treatment are needed.
Opioid Treatment CASCADE
Drawing from our work on VBP metrics and inspired by an article published by Williams and colleagues “To Battle the Opioid Overdose Epidemic, Deploy The ‘Cascade of Care’ Model” (Arthur Williams, Edward Nunes, Mark Olfson, 2017), OASAS has developed a monitoring tool for tracking our treatment system progress in addressing the needs of those afflicted by opioid use disorders. The tool is based on a similar one that was developed for addressing the HIV/AIDS epidemic. The management tool shows performance on treatment indicators across different components of effective interventions, from identification of the disorder to retention in care for six months or longer. We will be able to use this tool to highlight gaps in care as well as track progress over time and for different regions and subgroups of people.
Treatment Progress Assessment Tool
The measures workgroup recommended that SUD treatment needed a tool to measure treatment progress that could inform clinical decision making, and also to establish a baseline and subsequent scores to measure individual and program level performance in treatment improvement. There is no Hemoglobin A1C measure for substance use. A toxicology report can determine if someone has used substances within a few days to weeks, individuals can report perception of care, but we lack a practical, quick measure of response to treatment that is sensitive enough to capture changes that are indicative of progress.
OASAS and the Center have been working with several providers to pilot a tool that we developed focused on eight self-reported measures of SUD. It includes question related to symptoms such as cravings, use pattern and problems relating to use and questions related to response to the treatment process including hopefulness and bonding with addiction program staff. Initial testing of this tool show promise in the tool’s sensitivity and correlation with other signs of progress including completion of treatment.
To achieve our reform goals for New York, OASAS is encouraging providers to become learning systems of care. Metrics alone are not sufficient for creating a performance driven system. Organizations need to have the capacity and protocols for improving clinical practice in response to feedback from measures. OASAS together with the Center on Addiction will be providing technical assistance for providers to develop this capacity for clinical practice change. The methods are drawn from those used in other industries for improving quality.