Presentation
Variation in the Implementation of Hospital Care Transition Strategies: Institutional and Community Factors
AcademyHealth Annual Research Meeting
(2016)
Abstract
Research Objective
More than 21 million Americans experience a hospital stay each year, but too often the transition from inpatient to home and community settings is uncoordinated, disruptive, costly and incomplete. In recent years, the federal government has sponsored a series of demonstration programs and quality improvement initiatives designed to improve care transitions and reduce readmissions, and several research-tested models for managing care transitions have emerged. In 2015 PCORI commissioned Project ACHIEVE (Achieving Patient-Centered Care and Optimized Health In Care Transitions by Evaluating the Value of Evidence), a national comparative effectiveness study designed to identify the care transition outcomes that matter most to patients and caregivers, and to determine which combinations of care transition strategies produce these desired outcomes. As part of this study, this analysis uses data from a national survey of hospitals to (1) quantify variation across hospitals in the combination of transitional care (TC) strategies that are implemented; and (2) identify institutional, economic, and community factors associated with TC implementation.
Study Design:
We developed a cross-sectional survey administered to hospital staff who have responsibility for implementing TC strategies. Respondents reported on 17 TC strategies that may be used by their institution, including information on when each strategy was first implemented and what types of patients are targeted. Factor analysis and cluster analysis methods were used to identify combinations of TC strategies are commonly implemented together, and to classify hospitals into homogeneous groups based on the combinations of TC strategies they offer. Bayesian latent class estimation and multivariate hierarchical regression models were used to identify hospital, market, and community factors associated with patterns of TC implementation.
Population Studied:
We recruited a national selection of hospitals to complete the survey from lists of hospitals participating in one or more federal TC initiatives sponsored by CMS. A total of 417 hospitals completed the survey.
Principal Findings:
Hospitals varied widely in the specific combinations of 17 TC strategies implemented. The prevalence of individual strategies ranged from a high of 82% of hospitals reporting sharing inpatient data with outpatient providers to a low of 22% of hospitals reporting screening all patients to identify post-discharge care needs. Across all responding hospitals, a total of 302 unique combinations of TC strategies were identified in practice, which map into 17 clusters of related TC strategies based on factor analysis results. Multi-hospital systems and hospitals operating in more competitive markets were significantly more likely to implement broad-spectrum TC clusters that comprise 9 or more individual strategies, including risk-stratification, medication reconciliation and teach-back interventions. Broad-spectrum TC clusters were significantly less likely to be implemented in rural communities and areas with lower health insurance coverage, but were unrelated to patient case mix.
Conclusions:
TC strategies vary widely across hospitals and appear sensitive to resource availability and competitive pressures.
Implications for Policy or Practice:
Factors other than patient needs and preferences appear to drive variation in TC strategies, underscoring the need for clearer evidence on which strategies benefit which patients.
Disciplines
Publication Date
Summer June 27, 2016
Location
Boston, MA
Citation Information
Glen P. Mays, Arnold J. Stromberg, Jing Guo and Mark V. Williams. "Variation in the Implementation of Hospital Care Transition Strategies: Institutional and Community Factors" AcademyHealth Annual Research Meeting (2016) Available at: http://works.bepress.com/glen_mays/253/