A Case and Care Management Program to Reduce Use of Acute Care by Clients With Substance Use Disorders

In times of economic constriction, systems look for ways to make the most of available resources to address the needs of the populations they serve. For persons with substance use disorders, a strategy gaining favor in recent years is introduction of what have historically been considered case management services (1–4). These services were developed to meet the multiple needs of individuals with mental illnesses in the community after deinstitutionalization (5). Recent efforts have explored the utility of case management for individuals who do not derive benefit from existing substance abuse treatments. These individuals may encounter obstacles to making effective use of treatment, such as a lack of financial and instrumental resources, a history of trauma, child care or transportation issues, and comorbid conditions, among others (1). Although systems may want to offer additional services to these individuals—who often are high-cost service users—they also find it difficult to free up the resources required to do so.

This brief report describes the efforts by the State of Connecticut’s Department of Mental Health and Addiction Service to meet the needs of this population by reallocating acute care resources to the development of a targeted case and care management program aimed at engaging individuals who are frequent users of acute care in less costly nonacute services. Data are presented from five consecutive years after the introduction of this program in 2003. Service use patterns and outcomes over time were examined both for persons participating in the program and for the system as a whole.

Methods

The case and care management program focuses on clients identified through a centralized utilization management function as having four or more admissions to detoxification or acute inpatient services within a six-month period. Once identified, eligible clients are assigned a “recovery specialist” who arranges to meet with them while they are still in the acute setting. Recovery specialists have administrative support from a staff member of the state agency to ensure that they can access needed treatment and supports in a timely fashion to improve the likelihood of the client’s entering and sustaining recovery. In addition, the recovery specialists work to engage clients in treatment, improve their linkage from detoxification to another level of care, promote longer stays in treatment, link them to other entitlements for services and support that may enhance recovery, and assist them with returning to employment and taking steps to move off public assistance. The other primary function of the recovery specialists is to decrease clients’ repetitive need for and use of costly acute services, thus freeing up capacity in the system for new clients and creating opportunities for reallocation of resources. This combination of outreach staff (the recovery specialists) with centralized, statewide care management helps to ensure that no one “falls through the cracks” and that available resources are used more efficiently.

To assess the impact of the program, participants’ outcomes and service use were compared for the six- or 12-month period before they entered the program with the six- or 12-month period after enrollment. Additional outcome, service use, and cost data were derived from the management information system of the state agency’s administrative services organization and were tracked for five years (2004–2008) after introduction of the case and care management program. These data were compared across the five years; cost data were also compared with 2003 data.

Results

The analysis found a reduction in problematic discharges among the 636 clients who were hospitalized during this period. Problematic discharges were defined as episodes of clinical care that terminate prematurely in one of the following ways: client is administratively discharged from a program, client is discharged against medical advice, client refuses care, or client does not attend the program and is unresponsive to attempts at reengagement. The number of such discharges experienced by participants decreased 56% during the first six months after enrollment in the case and care management program compared with the six months before enrollment (327 problematic discharges before enrollment, compared with 145 discharges in the first six months postenrollment).

In addition, in the year after enrollment, the number of acute care episodes among participants was 56% lower than in the year before enrollment (1,727 episodes, compared with 759 episodes), with a 19% increase in engagement in nonacute care services (1,303 participants were engaged in nonacute services in the year after enrollment, compared with 1,092 in the year before enrollment). This shift from acute to nonacute care was even more pronounced in terms of units of service received. The number of paid units of acute care fell by 62% (from 14,452 units to 5,555 units), with a corresponding 63% increase in paid units of nonacute care (from 12,964 units to 21,144 units).

Given the cost differential between acute and nonacute care, this shift generated considerable cost savings. The cost for all levels of care was approximately 46% lower in the first 12 months of enrollment compared with the 12 months before enrollment ($9,393,657 compared with $5,047,371). The cost reductions were due to a 57% reduction in use of intensive residential services and a 53% reduction in use of inpatient psychiatric beds. These reductions were offset by a 40% increase in use of observation and respite beds, a 35% increase in general outpatient care, and a 24% increase in methadone maintenance services.

From the perspective of client outcomes, steady increases were found in the connect-to-care rate for participants leaving acute care settings. This rate reflects the percentage of persons leaving acute care settings who present for care in nonacute settings after discharge. Before the case and care management program, rates hovered around 35% for three consecutive years. This rate rose over 10 percentage points between 2004 and 2005 and has continued to increase, although at a slower rate, which confirms the finding that participants used less acute care as they became more engaged in nonacute services.

What was perhaps most surprising was that the reduction in use of acute care was sustained over time, with the number of high service users decreasing from almost 1,100 during 2004 to less than 400 in 2008 (Figure 1). As service use patterns among the high service users shifted from recidivism in acute care settings to use of nonacute services, a new and equivalent cohort of high service users did not replace them. Rather, the gains made in reducing acute care costs were sustained and decreased even further over time, as demand for acute care services decreased. This led to a reduction in mean service costs per client in the overall public assistance program, from approximately $2,300 per person in 2004 to just over $1,900 in 2008 (these amounts were not adjusted for inflation). Finally, when per-person rates before and after introduction of the program were compared, the decreasing service costs per person generated considerable savings, allowing the state system to extend care to more people in need each year. The unduplicated number of persons served increased from 23,500 in 2004 to 32,800 in 2008, with an average annual increase of approximately 8%.

Discussion and conclusions

The findings presented here did not emerge from a controlled trial nor were they subjected to inferential statistical analysis. Caution must therefore be used in concluding that these changes were due entirely or directly to the care and case management program. Within these limitations, the results suggest that investment in recovery-oriented care and case management initiatives may create opportunities for a cycle of reallocation from acute to nonacute care that can be self-sustaining. This outcome has the added benefit of less payer spending per person per year. As such, this model may be replicated as both a recovery-oriented and a cost-effective intervention, especially suitable for these difficult financial times.