Appendix B: Summaries of PFQ Grantee Activities (continued)

PFQ Grant Summary: Partnering for Improved Primary Care Diabetes Management

Lead Organization: Lehigh Valley Hospital and Health Network (LVHHN)
Partner Team: LVHHN, Helwig Diabetes Center at LVHHN
Title: Partnering for Improved Primary Care Diabetes Management
Topic Area: Improve diabetes care in the primary care setting through intensive physician and patient education and consultations with specialists
Principal Investigators: Originally Dr. Mark Young, chair of Community Health & Health Studies at LVHHN & professor of Health Evaluation Sciences, Penn State University, College of Medicine (died April 2004); replaced by Dr. Kenneth D. Coburn, CEO of Health Quality Partners
AHRQ Project Officer: Margaret Coopey
Total Cumulative Award: $294,841
Funding Period: 10/2002–10/2004
Project Status: Terminated after 2 years

1. Project Description

Goals.  The project had two major goals: (1) to provide a packaged educational intervention to improve primary care physicians' (PCP) management of their diabetic patients in order to improve patient health status and (2) to devise a cost-efficient model of intensive intervention that could be delivered in primary care physician practices, which is where the majority of diabetes patients receive care. The project aimed to design, implement, and evaluate a diabetes management model that would deliver to diabetes patients (Type 2 only, excluding the very highest-risk patients) in primary care practices the same type of support (via referral to the regional diabetes center) received by high-risk diabetic patients.⁴

Activities and Progress. In the first year, diabetes educators from the Helwig Diabetes Center at LVHHN provided intensive team-based education with primary care physicians in four practices in two phases. In the first phase, called "intensive education," which lasted for three to six months, a Certified Diabetes Educator (CDE), nutritionist, and diabetes physician specialist conducted an initial assessment of the practice; recommended practice-specific process improvements; provided structured education for clinicians, other staff, and patients; and conducted biweekly case review. The CDE worked on site 16 to 24 hours per week. In the phase called "education reinforcement," the CDE was on site for eight hours per week for the next six to nine months, providing patient-specific problem solving and episodic consultation with an endocrinologist. Patient group visits, delivered by a team consisting of an educator, dietician, and support staff, were initiated in the four practices with 10 to 15 patients in each group. 

In the second year, the project introduced the same model in another six primary care practices but with a "refined model" that used Achievable Benchmarks of Care (ABC™) to motivate improved physician clinical performance and patient health outcomes. ABC sets a benchmark for care based on best practices of local or regional peers and, to motivate physicians, provides them with reports on how they compare to their peers. ABC reports, prepared by a Penn State College of Medicine biostatistician, were distributed to the six PCP practices, which received ongoing feedback on their progress.

2. Partnership Structure/Function

A project advisory committee was established to review project successes, barriers, data, and general operations and budget.  Members included the principal investigator, co-investigator, medical director of the Helwig Diabetes Center (Dr. Merkle), project director and project coordinator from Helwig, medical director of the Lehigh Valley Physician Hospital Organization, and two advisors from Penn State University: Pamela Short, Department of Health Policy Research, and Robert Gabbay, MD, College of Medicine. LVHHN's relationship to the primary care practices was primarily limited to providing technical assistance and clinical practice support. Neither PCPs nor patients appeared to have any input into program design, assessment, or modification. 

Table 1. Major Partner Organizations and Roles in the Project

	Organization	Role in Project
Lead Organization (grant recipient)	Lehigh Valley Hospital and Health Network	Project management, planning/development, and leadership; chair of Advisory Committee.  When Dr. Young died, Dr. Kenneth Coburn of Health Quality Partners assumed the administrative and leadership roles for the project, but for only four months.
Key Collaborators	Helwig Regional Diabetes Center at LVHHN Dr. Larry Merkle, Medical Director	Project director and project coordinator based at Helwig Diabetes Center staffed and coordinated delivery of diabetes interventions in PCPs, monitored progress, and helped collect data for evaluation Medical director and his staff provided endocrinologist consultation to PCPs
Target Organizations	Primary care practices in southeastern Pennsylvania St. Luke’s Health System and Sacred Heart Health Network	Ten primary care practices in southeastern Pennsylvania participated in the first two years; had the project continued, another eight PCPs were supposed to be added in years 3 and 4, and plans would have called for rolling out the project region-wide through the Physician Hospital Organization (PHO) affiliated with LVHHN Two other major hospital systems in southeastern Pennsylvania were to have been involved in the regional roll-out in years 3 and 4 had the project continued

3. Project Evaluation and Outcomes/Results

Structure/Process of Care. In February 2004 the project submitted data to the Agency for Healthcare Research and Quality showing promising improvements in the percent of physicians in the first four practices who were screening for glycosylated hemoglobin (HbA1c) and lipids, but not for micro-albuminiuria, per the time line set forth by the American Diabetes Association guidelines. On the Achievable Benchmarks of Care scores, physicians in the top-performing groups remained near the top while those in lower-performing groups showed improved scores. An initial assessment of the financial feasibility of providing group visits in private practice settings indicated that 12 patients per group provide income comparable to routine office visits, demonstrating that "a replicable and sustainable financial model has been developed."  

Outcomes of Care. Data on HbAlC levels, lipids, and blood pressure were monitored at baseline and then at 6 and 12 months after the intensive education phase of activities in the primary care practices. In February 2004, the data showed an increase in patient adherence to guidelines and statistically significant improvement in all the core clinical measures: blood pressure, lipid levels, cholesterol, triglycerides, and hemoglobin. In the absence of a control group, the project "corrected for the regression to the mean." 

4. Major Products

• Presentation on the project delivered at the American College of Physicians, spring 2005.  
• Najarian et al., Improving Outcomes for Diabetic Patients Undergoing Vascular Surgery. Diabetes Spectrum 18:53-60, 2005.

5. Potential for Sustainability/Expansion after PFQ Grant Ends

Project representatives report that the intervention remains in place in the 10 participating primary care practices. The project's financial sustainability study showed that group visits by patients to receive diabetes education are billable services and can generate enough revenue that primary care practices can sustain the model.  The project demonstrated a model of providing chronic care to diabetes patients that could be replicated by other specialty diabetes centers working in conjunction with primary care practices; however, project representatives were not aware of any other centers that had done so.

---

^4. The projected was terminated shortly after the end of the second year of the grant, eight months after the principal investigator died. Had the project continued into the third and fourth years of the grant (after December 2004), it would have addressed several additional goals: (1) to evaluate the sustainability of models of care for improving primary care diabetes management, (2) to disseminate the model to other systems in southeastern Pennsylvania (16 practices and over 3,000 individuals in conjunction with the LVHHN Physician Hospital Organization), and (3) to disseminate the lessons learned to a national audience.

---

Return to Appendix B Contents

PFQ Grant Summary: Different Approaches to Information Dissemination

Lead Organization: New York State Department of Health (NYSDOH) (through Health Research Inc.)
Partner Team: Research Division of the Hebrew Home for the Aged at Riverdale (RDHHAR), Columbia University Stroud Center, New York State Psychiatric Institute, American Health Care Association (AHCA), Association of Health Facilities Survey Agencies (AHFSA), Institute for the Future of Aging Services, and The Commonwealth Fund
Title: Different Approaches to Information Dissemination
Topic Area: Implementation of evidence-based long-term care practices in nursing homes and adult care facilities in New York State
Principal Investigators: Beth Dichter, PhD, NYSDOH (formerly Suzanne Broderick); with coprincipal investigators from RDHHAR: Douglas Holmes, PhD, and Jeanne Teresi, EdD, PhD
AHRQ Project Officer: Margaret Coopey
Total Cumulative Award: $1,161,932
Funding Period: 9/2002–9/2006
Project Status: Grantee has a no-cost extension through September 29, 2007, to conduct and complete data analysis

1. Project Description

Goals. The project aims to evaluate two methods for disseminating best practices to nursing homes and adult care facilities.  The research design is quasi-experimental with two intervention groups and a comparison group.  Each group includes 15 nursing homes and 7 adult care facilities (ACFs), for a total of 45 nursing homes and 21 ACFs.  The first intervention group received special training modules provided to facility in-service educators.  The second intervention group received the same special training modules while the state surveyors responsible for quality assurance in the facilities also underwent training on the modules. The comparison group conducted its own training as required by state regulations, on topics selected by each facility. The project will make pre- and post-training comparisons of staff knowledge of accident/fall prevention and conditions (e.g., vision disorder, affective and behavioral states) that may increase the risk of accidents/ falls as well as comparisons between control and experimental groups (see below). 

Researchers hypothesized that training modules provided to nursing homes and ACFs in the experimental groups, as compared to the control group, would enhance quality of life for residents as measured by the reduction in indicators such as accidents/falls and by secondary quality indicators, including behavior and affect.  The primary outcome was reduction in accidents/falls.

Activities and Progress

Year 1.  Delays in the release of AHRQ grant funds delayed the start of project activities by about six months. By March 2003, the project had convened an Advisory Group comprising representatives of project partners and other stakeholder organizations.  Project staff conducted an exhaustive search for evidence-based best practices in long-term care.  Through careful screening and scoring on criteria such as cost, whether the module was indeed evidence-based (as determined by results reported in peer-reviewed journals, at conferences and meetings, and so forth), relevance to nursing home and ACF residents, and so forth, the project identified several possible candidate best practices for the evaluation. The Advisory Group further reviewed and scored the training modules and recommended a subset for use in the project.  Initially, the project intended to implement six to eight evidence-based best practices in the experimental nursing homes and ACFs.  During a meeting on September 10, 2003, convened by NYSDOH, the Advisory Group recommended limiting the number of practices to two for each facility; the group believed that nursing homes and ACFs would not be able to implement more than two practices successfully at one time.  After selection of the modules, project staff finalized the outcome measures for evaluating the effectiveness of the interventions.  The project randomly selected samples of nursing homes and ACFs from three regions in New York State and began recruiting facilities to participate in the study.  

Year 2. With guidance from the Advisory Group as described above, project staff selected three evidence-based best practices with associated training modules and worked with the developers of the modules to adapt the materials and training process to meet the specific needs of New York State facilities. The three training programs were (1) Bathing without a Battle, which focused on person-centered bathing of individuals with dementia; (2) Vision Awareness, which promoted a low-cost intervention that increases staff knowledge of visual impairments; and (3) Staff Training in Assisted Living Residences (STAR), which helped staff understand and deal more effectively with difficult behavior problems among residents with dementia.  Bathing without a Battle and Vision Awareness were selected for nursing homes and Vision Awareness and STAR for ACFs based on appropriateness for the target populations.

The project then recruited facilities: 15 nursing homes and 7 ACFs for each of the training programs. Training sessions for nursing homes and ACFs in the two experimental groups on all three modules began in the second year.  For experimental group one, the project trained one or two staff members of the facility.  In nursing homes, the trainee was usually the nurse educator.   In ACFs, the trainee was usually the administrator or case manager.  All trainees then returned to their facilities and trained other facility staff.  For experimental group two, the project also trained the state surveyors responsible for quality assurance.  Research staff collected baseline data on ACF residents by using a version of the Comprehensive Assessment and Referral Evaluation (CARE) and the Extended Interview, both of which are comprehensive assessment tools used extensively by RDHHAR in studies of comparable populations. As locally collected Minimum Data Set (MDS) data were to be used for nursing home residents, raw data collection for nursing home residents was not necessary. The first wave of data collection in ACFs, which also included interviews with staff and administrators and an environmental assessment, was completed for the control group and began for the experimental groups.

Year 3.  Training continued for both nursing homes and ACFs. Implementation forms were collected from participating facilities to monitor their progress with training and implementation. The project completed the first wave of data collection at ACFs in the experimental groups early in the grant year and began follow-up data collection at the facilities that had implemented training modules earlier in the year and at ACFs in the control group toward the end of the grant year.

Year 4.  During the fourth year, the project continued to provide training and implementation consultation to facilities.  Due to staff turnover, 10 facilities experienced difficulty in continuing staff training such that the project had to deliver new "train-the-trainer" sessions.  Retraining was conducted by the developer of the Vision module but not for STAR or Bathing without a Battle because of limited resources and the lack of available trainers.  

As of the last project report, which covers the period from September 30, 2005, through September 29, 2006, the project completed collection of follow-up data for ACFs (using the RDHHAR tools) and was in the process of extracting MDS data for the nursing homes.  Preliminary data analysis has begun, and final data analysis will begin once all data are compiled. 

2. Partnership Structure/Function

NYSDOH/Health Research Inc. contracted with the Research Division at the Hebrew Home for the Aged at Riverdale to serve as the research partner for the project. RDHHAR developed and implemented the project's research design, collected resident data from ACFs, and provided support to participating facilities in completing implementation tracking logs and other data collection forms.  Project staff from NYSDOH and RDHHAR met or held conference calls at least monthly throughout the project.  The two organizations consulted with experts at Columbia University and Advisory Group members to identify proven or effective evidence-based long-term care practices.  They also identified ways in which the training should be delivered or adapted to meet the needs of staff in nursing homes and adult care facilities or to comply with New York State rules and regulations.

The expectation is that the three national organizations (AHCA, AAHSA, and AHFSA) represented on the Advisory Group will help disseminate and promote adoption of the evidence-based practice programs and training approaches through their national conferences and education vehicles.  Project staff also sent updates to at least 40 "interested parties"–educators, researchers, trade association representatives, and regulators who offered to provide occasional advice or assistance. 

Table 1. Major Partner Organizations and Roles in the Project

	Organization	Role in Project
Lead Organization (grant recipient)	New York State Department of Health, Division of Home and Community-Based Care (through Health Research Inc., an affiliated private organization)	Manage and coordinate project activities.  Convene and obtain input from Advisory Group.  Develop facility sample and recruit facilities to participate in project.  Ensure participation from surveyors.  Provide consultation to facilities as they trained staff and implemented best practices.  Extract MDS data and provide them to RDHHAR.
Key Collaborators	Research Division of the Hebrew Home for the Aged at Riverdale Consultants and Advisory Group members	Co-principal investigators (Douglas Holmes and Hebrew Home for the Aged Jeanne Teresi) responsible for performing evidence-based review of potential modules, evaluation design, data collection, technical assistance to participating facilities, and analysis of project outcomes. Identify and recommend evidence-based training programs, packages, or modules; review training approaches to ensure nursing facilities and ACFs can effectively implement them; and help disseminate or promote use of the training programs more broadly: American Association of Homes and Services for the Aging (AAHSA)—Institute for the Future of Aging Services. American Health Care Association (AHCA). Association of Health Facility Survey Agencies (AHFSA). Columbia University Stroud Center. New York State Psychiatric Institute. The Commonwealth Fund.
Target Organizations	45 nursing homes and 21 adult care facilities in three regions in New York State	Those assigned to the experimental groups participated in special training programs offered by the state, trained other staff in their facilities in evidence-based practices, and provided data on implementation of the practices. Those assigned to the control groups provided their usual training programs

3. Project Evaluation and Outcomes/Results

The project will evaluate process data collected with respect to each module.  To determine impact at the staff level, the project intends to look at the number of facility staff trained in the target facilities, assess how thoroughly best practices have been implemented, and compare pre- and post- training knowledge among staff.  The project will also make resident-level comparisons between control and experimental groups. The project will analyze the impact and significance of the project once all the data have been compiled and will include the analysis in a final report.   

After training was completed at the experimental sites, the project asked each facility to submit implementation forms that reported the number of staff trained as well as the fidelity of the particular intervention in that facility, i.e., how many vision logs were completed by those trained to assess vision, or how many "ABC" cards were filled out by those trained to address behavioral problems of patients with dementia.  As of June 2006, among the nursing home sample, 10 of 15 facilities in the first experimental group trained staff in at least one of the modules; in the second experimental group (with surveyor training in addition to staff training), 14 of 15 facilities completed training in at least one of the modules.  It is expected that the latter two numbers may increase somewhat after facilities are contacted and revisited in order to obtain final implementation data. Among ACFs, 6 of 7 in each of the two experimental arms completed one or both training modules.  In total, staff from 28 facilities received vision training, staff from 6 facilities received STAR training, and staff from 22 facilities received bathing training.  Several nursing homes and ACFs have neither trained staff nor implemented the modules. The two primary reasons facility administrators provided for inaction were (1) the need to address higher-priority issues and (2) attrition in staff trained at initial train-the-trainer sessions.

Some facilities participating in the experimental groups found the training to be useful. For example, some administrators say that, as a result of the bathing training, they have made some structural changes in the facility to improve residents' bathing experience.  One of the facilities' interviewed indicated that it uses the training it received through the project in nurse aide classes, and another interviewee mentioned that the facility has integrated some practices into its standard procedures.  Some facilities, however, mentioned that the time needed for training and/or completion of implementation monitoring logs and quality assurance forms was a significant burden. Others noted that turnover in directors of nursing often meant the loss of support for training programs while turnover in aides meant that the training had to be provided to all new aides if it were to be integrated into ongoing practice. 

With insufficient funding, the project was not designed to assess directly via interview the impact of training on state nursing facility surveyors' attitudes or understanding about what qualifies as an avoidable adverse outcome. However, the project will analyze staff training and implementation and resident indicators for the two experimental groups (one of which included state surveyors in the training program) to see if there were any differences in outcomes.

4. Major Products

• Presentation at the Gerontological Society of America Annual Meeting 2005—AHRQ Partnerships for Quality: Different Approaches to Information Dissemination.
• Planned preparation of a manuscript outlining the process used to determine the strength of the evidence base of available off-the-shelf training modules.

5. Potential for Sustainability/Expansion after PFQ Grant Ends

Some facilities indicated that a few project activities will continue in the future. For example, some aspects of the training will be provided to new staff, and some best practices have been integrated into standard procedures, e.g., asking new residents, upon admission, about their bathing preferences. The continued use of training programs depends on the availability of a trained "trainer" and the availability of off-the-shelf and easy-to-implement training modules, as facility education staff otherwise have difficulty in providing the training. 

The New York State Department of Health plans to use the project results to decide which types of training programs to support with the recurring funds available through its Dementia Grants Program. Pending the project's favorable outcome, the department may also require or recommend the inclusion of elements of evidence-based training programs in state-mandated certified nurse aide training.

Return to Appendix B Contents
Return to Contents
Proceed to Next Section