Population selection was from a network pool of 8,704 veterans identified as high cost in the prior year (≥$25,000), stratified by VA medical centers, and identified with chronic conditions such as CHF, COPD, HTN, and DM. Each care coordinator reviewed the list for appropriateness, made contact to establish willingness of veterans to participate and enrolled those who were willing and appropriate candidates. Seven hundred and ninety-one veterans were enrolled. The drop-out rate was very low (<10%); however, the lists used for enrollment had many exclusions due to death, inability to make contact, or institutionalization. A comparison population with clinically similar but nonenrolled veterans was also assembled. This group was randomly selected from a stratified sample similar in diagnosis, age, and gender. A comparison of their 1-year average health care utilization rate compared to the intervened group is attached (Table 1). It is important to remember, however, that the in Months · Pre 6 TABLE 2. INTERVENED GROUP: FIRST-YEAR COMMUNITY CARE COORDINATION SERVICE EVALUATION ER slops Hospital admission Hospital BDOC NH admission NH BLOC Clinic visits ER, emergency room; BDOC, bed days of care; NH, nursing home. VIRTUALLY HEALTHY Months TABLE 3. Comparison Group: Data on RESOURCE UTILIZATION FOR Same Time Frame as IntervENED GROUP Clinic visits ER stops Hospital admission Hospital BDOC NH admission NH BDOC The intervened group comprised 791 veterans enrolled in the CCCS program for 1 year. A comparison group of veterans were also analyzed (Table 3). The comparison group received usual care with no care coordination or technology. Results for the intervened group from the change in first year to second year data analysis showed a reduction in ER visits by 40%, hospital admissions by 63%, and hospital BDOC by 60% (Table 4). Clinic visits went up 14% in the first quarter postenrollment for the intervened group (Fig. 1). This trend was reviewed, and it was noted that care coordinators who had been empowered to make assessments had scheduled clinic appointments during the first few months of enrollment to ensure all clinical needs were met in a timely fashion. After the first 3 months, the number of clinic visits steadily declined. It is also noted that, although this group went up in clinic visits overall, the comparison group went up even more (40%). In addition to these outcomes, nursing home admissions and bed days of care were evaluated. It was believed by CCCS program lead ers that the veteran population targeted by the program was at high-risk for premature institutionalization and thus could be impacted by the care coordination process. Nursing home admissions declined by 64% and nursing home BDOC were reduced by 88%. In the comparison group, nursing home admissions increased by 106% (Table 4). An Odds Ratio analysis revealed that patients enrolled in the program were 77.7% less likely to be admitted to a nursing home care unit than those not enrolled in the program (Table 5). Quality of life and functional ability as measured by the SF 36V indicated significant improvements in the Role Physical (p < 0.003), Bodily Pain (p<0.000), Social Functioning (p<0.004), Role Emotional (p < 0.000), and the Mental Composite (p <0.011) scores. The other five domains remained the same, which is also significant in a frail elderly population with complex medical/chronic disease condi tions. Overall, when comparing the intervened group findings to the comparison group, it was found that the intervened group showed con TABLE 4. INTERVENED AND COMPARISON GROUPS: PERCENT CHANGE From Year 1 toO YEAR 2 five complex medical/chronic disease projects is included here. The immunization measures were in line with VHA performance standards. Other measures were developed by CCCS staff based on identified problem areas. VHA immunizations target goals were 78% for both influenza and pneumococcal measures. Eightythree percent of the CCCS veterans had a current flu shot, and 90% had a current pneumococcal vaccine. Medication compliance, which was chosen as a measure because it is often an issue with the chronically ill population, was 93%. The target goal was also 78%. Primary care providers responded positively to the role of the care coordinator, with an 85% outcome measure for appropriate and timely communication. Eighty-five percent was also the target goal for this measure. DISCUSSION Many aspects of chronic disease management must be carefully coordinated and monitored. CCCS leaders therefore believe that the model does improve clinical outcomes and reduces healthcare utilization. One of the core principles behind successful chronic disease management is effective self-management.5 VIRTUALLY HEALTHY The chronic disease dialogues used by the inhome messaging device not only provided daily, repetitive education on self-management principles, but also monitored a variety of symptom parameters including blood sugar, weight, blood pressure, and chest pain. Leider and Krizan postulated that, for a disease management model to be effective, it must employ three basic strategies: improving patient compliance and self-management behaviors, strong physician leadership, and rigorous monitoring of patients so that clinical outcomes can be improved. The CCCS model embodied these strategies, and staff members were able to effectively operationalize them in practice. Technologies were chosen that supported patient compliance and provided educational opportunities to enhance self-management. Special emphasis was put on keeping the technology simple and user friendly to allow for the broadest use regardless of the patient's technological expertise. The CCCS leaders strongly relied upon the collaboration of physician providers with care coordinators. Physician champions were sought to provide leadership at local project sites and to work directly with CCCS leaders to promote acceptance of the care coordinator role. Care coordinators themselves were chosen for their judgment skills and their effectiveness in managing patient needs across the healthcare continuum. CONCLUSION Based on the first-year findings, it is evident that the CCCS model has benefited many frail elderly, medically complex patients. It has helped them to maintain their independence, improved their functional status and deterred from costly hospitalizations and institutionalizations. It is strongly believed that the key to this success has been the carefully constructed role of the care coordinator, with clinical expertise to properly assess patient needs. This role in tandem with the right tools and the technology most adaptable to the needs of the patient and clinician have provided the means for early detection of patients at risk for further deterioration. Through the use of technology, efficiencies in process and practice, previously not pos sible, are achieved. This approach has given the patients a safer and more secure environment in their most preferred setting, the home. The first step in the process of inventing a proactive healthcare model that facilitates patient-oriented and cost-effective delivery of services is improving health and information access. The primary concept of integrating technology into care coordination has gone beyond that first step. The model has successfully evolved into an effective approach for managing patients with multiple chronic diseases. The CCCS is in the initial phase of identifying best practices for the strategic model. The intent is to draw upon the lessons learned and develop standards that can serve as the basic foundation for any population management program. The early successes have warranted expansion of the program to other populations. In 2001, two new demonstration programs were added. There will be a second request for proposals in 2002 to explore the effect of the concepts on other populations and new technologies not yet tested in this environment. In addition, VISN 8 is exploring accreditation opportunities in disease management to further validate and strengthen both the clinical and business applications of the concepts. The aging in place model has been the most notable success of this program. It is readily apparent that more veterans are stable, satisfied and able to manage their chronic health problems in their home environment. ACKNOWLEDGMENTS We would like to thank the following individuals: (1) Project staff from the following clinical demonstration medical program sites: Lake City, Gainesville, Ft. Myers, Miami, and San Juan. (2) Douglas D. Bradham, Dr.P.H., Associate Professor, Division of Healthcare Outcomes Research, Department of Epidemiology and Preventive Medicine, School of Medicine, University of Maryland. (3) Neale R. Chumbler, Ph.D., Associate Professor, Department of Health Policy and Epidemiology, University of Florida. |