Chapter III. What Did Grantees Seek To Do?

This chapter describes the PFQ grantees and their goals. Specifically, it discusses the intended grant focus; intervention strategies; the characteristics of the organizations awarded PFQ grants, proposed partners and their roles; and the expected outcomes and how they intended to measure their success.  We focus on grantees' initial intentions, based on the applications and interviews conducted with grantees. We defer to Chapter IV for our analysis of grantees' success in implementing these plans and the outcomes of their efforts.

A. PFQ Project Focus

The central focus of PFQ was to apply evidence-based practices to improve quality of health care. PFQ also provided grants to improve the health care system's readiness to address bioterrorism, although grants in this area were smaller than were the core grants focused on improved quality of health care. Of the 21 PFQ grants, 18 received funds for the first purpose and five for the second. The latter five included two (JCAHO and RTI) whose grants had both clinical quality improvement and bioterrorism preparedness components. 

The RFA allowed grantees substantial flexibility in choice of focus and approach, though it encouraged work in at least one of AHRQ's targeted priority health care settings, health conditions/issues,⁷ and/or populations.⁸  These priorities are broadly defined and so were the foci of PFQ grants. Appendix Table A.1 provides details on the specific focus of each grantee, but the themes across the projects are briefly described here.

Quality Improvement Grants. Of the 18 grants funded to encourage providers to better use evidence-based care to enhance its quality, 15 did so by working directly with providers, or through intermediaries that represented them, and 3 by attempting to leverage purchasing power to change incentives to reward providers that provide high quality care.  Of the 15 grants focused directly on changing provider behavior (Box 1), 5 worked to improve the quality or safety of hospital-based care, 4 with long-term care/home health providers, 5 with office-based physicians, and one with large integrated health delivery systems. 

Box 1. 15 Grants to Improve Process of Care and Clinical Outcomes by Changing Provider Behavior (grouped by setting)

Hospital American Hospital Association/Health Research and Education Trust (Original grantee: Institute for Healthy Communities) (J.R. Combes):  Increase and enhance hospital-based palliative care by creating learning center hospitals to host site visits from staff from other hospitals.  Association of California Nurse Leaders (N. Donaldson):  Decrease incidence of hospital-based falls and falls-with-injury by coaching nurse "linkers" to implement evidence-based interventions in medical-surgical hospital units. Catholic Healthcare Partners (D.S. Casey): Improve health care outcomes for patients with congestive heart failure using hospital-based approaches to encouraging consistent use of evidence-based guidelines for care. Child Health Corporation of America (P.J. Sharek): Work with a subset CHCA's member children's hospitals to integrate evidence-based practices on pain management, medication safety, and patient safety. JCAHO (J. Loeb): Identify whether the introduction of JCAHO's core performance measure sets for hospital care for patients with four conditions were perceived as valuable by hospitals, whether and how they influenced the process of care, and with what impact. (See separate bioterrorism component.) Long-term care and home health American Medical Directors Association Foundation (D. Polakoff): Create local long-term care partnerships and pilot test the use of clinical practice guideline implementation toolkits in nursing facilities in six states. International Severity Information Systems, Inc. (S. Horn): Incorporate findings from the National Pressure Ulcer Long Term Care Study into routine, evidence-based practice in long-term care facilities. New York State Department of Health (S. Broderick/B. Dichter):  Evaluate two alternative methods for disseminating evidence-based best practices in long-term care and adult care facilities. Visiting Nurse Service of New York (P.H. Feldman):  Establish a national learning collaborative for home health care agencies to improve care for elderly clients with diabetes. Physician office practice American Academy of Pediatrics (C. M. Lannon). Improve care for children with ADHD by using web-based tools and practice-based CME to encourage pediatrician's adherence to evidence-based guidelines, and if successful, extend the model to other conditions American College of Physicians (V.T. Snow).  Develop and test a team-oriented, practice-based continuing medical education strategy focused on improving care for patients with chronic disease and develop a business case to support its practical application American Medical Association (K.S. Kmetik):  Test two approaches for transferring clinical data to support large-scale improvement in ambulatory care for patients with chronic diseases—adult diabetes, coronary artery disease, and major depressive disorder—by promoting use of AMA's performance guidelines.  Lehigh Valley Hospital and Health Network (M. Young/K. Coburn): Develop and test a cost-efficient educational intervention to improve care for diabetes in primary care practice. Physicians Micro Systems Inc. (S. Ornstein): Expand availability and use of clinical indicators in physician offices for practice-based quality improvement in practices using one electronic medical record system.  Integrated delivery system RTI (L. Savitz):  Unlike other PFQ grants, this project focused generically on partnerships. It sought to leverage the experience of its health system partners in the Integrated Delivery System Research Network to improve quality, support more communication across partners, and study partnership issues in AHRQCoPs.

Each grantee defined its target group in different ways. Of the three projects whose quality improvement strategies focused on purchasers, shown in Box 2, one focused on office-based physician care (HealthFront), one on rewarding higher quality hospitals (The Leapfrog Group), and one on creating general measures of performance by the health care system (Pacific Business Group on Health, whose project terminated prematurely, and is therefore not described in the report after this point).

Box 2. Three Grants to Modify Purchaser Incentives to Promote Quality

Health Front (M. Callahan): Develop nationally recognized measures of provider performance and use them to support purchaser value-based decision making on the part of health plans. The Leapfrog Group (S.F. Delbanco): Leverage payer and purchaser groups in select communities involved in Leapfrog's "Regional Roll Out" to pilot test financial reward and incentive programs targeting hospital and consumer groups. Pacific Business Group on Health (D. Hopkins): Develop comparative performance data on physicians using Medicare claims. Project was terminated early when access to the necessary data could not be negotiated.

In addition to provider type and health care setting, most grantees also focused their efforts by health condition or population group. The most common priority health issues and conditions addressed by the PFQ grants awarded include diabetes (five), long-term care (three), heart disease (four), mental health (three), and child health (two). Fire projects targeted two or more conditions. The most common priority populations targeted by grantees included: the elderly (six projects), special needs populations, including those with disabilities, chronic care, or end of life care (six projects), and children (two projects).

Bioterrorism Preparedness Grants. Projects addressing bioterrorism and emergency preparedness often defined their target audience more broadly than did grantees seeking to improve quality. The three grants funded exclusively to focus on bioterrorism preparedness pursued goals related to increasing health providers' ability to respond to bioterrorism or other disasters.

Both grantees with dual-purpose funding (JCAHO, RTI)—to improve both quality and bioterrorism preparedness—built on strong hospital links and sought to bring in other community groups as appropriate. JCAHO's bioterrorism grant sought to assess the existence and effectiveness of linkages for community-wide bioterrorism preparedness among health care, public health, public safety, and government agencies. JCAHO also planned to compare preparedness for communities with and without disaster experience and identify exemplary practices. RTI hoped to develop and use the same infrastructure used for a previous AHRQ-funded project with its integrated health system partners that used evidence-based research to improve quality, and to facilitate communication that would also address bioterrorism preparedness in the health systems. 

Box 3. Three Grants to Improve Bioterrorism and Emergency Response Preparedness by Health Providers

Altarum Institute (PI G. Miller). This project focused on developing simulation models to project demand for medical care within communities in response to a bioterrorist attack or acute outbreak of infectious disease. The intent was to test the utility of these models in planning with an urban and a rural healthcare network. Connecticut Department of Public Health (PI L. Dembry). The project focused on developing, providing, and evaluating the effectiveness of Web-based bioterrorism preparedness and response training for "front line" practitioners in Connecticut. Texas A&M University Health Sciences Center PI (J. Williams). The bioterrorism preparedness component of this grant, which ultimately became its exclusive focus, focused on analyzing bioterrorism readiness among provider systems in counties in and around San Antonio and Dallas/Fort Worth, Texas.

Return to Contents

B. Intervention Models and Strategies

1. Models

PFQ projects were expected to design their interventions to include three major types of activities:  1) designing, supporting and facilitating evidence-based improvements in health care security, safety, and quality; 2) sustaining these improvements by making them part of the ongoing practice of health care providers and clinicians; and 3) disseminating improvements beyond targeted selected population groups. The AHRQ grant solicitation instructed grantees to design their interventions using one of the following models:

1. Short-term; single, relatively limited target.
2. Complex plan of multiple targets requiring a sequence of interventions over a longer period.
3. Expand over time, adding additional targets or partners in a planned sequence over the period of time

PFQ ultimately included few short-term grants (type 1), with the vast majority of grants funded for at least three years and designed to fit models (2) or (3). An example of the first model is the Connecticut Department of Health Grant that developed a bioterrorism preparedness training program for physicians.  An example of the second model is the American Medical Directors Association Foundation grant that focused on nursing facilities in six states to determine the effectiveness of an approach for training nursing home staff to implement clinical practice guidelines and to evaluate nursing homes' experiences and lessons in using implementation toolkits. An example of the third model is the American Hospital Association-HRET grant that planned to expand the number of palliative care learning centers from the three Pennsylvania-based units in Phase I to an additional four national facilities in Phase II.

Some projects followed a model that combined these strategies.  For example, over the course of the grant period, Physician Micro Systems, Inc. in collaboration with the Medical University of South Carolina, aimed both to increase the clinical indicators tracked from 22 to over 70 and the number of participating physician practices from 40 to 100 (model 3), and to involve a sequence of interventions, including quarterly reports, site visits, and annual network meetings (model 2). While most of the grantees planned to expand their targets, interventions, and/or partners over the course of the grant periods, some ran into hurdles, such as recruiting issues and staff turnover that delayed and/or inhibited their progress (discussed further in Chapter IV).

Because the PFQ solicitation required that the proposed interventions use successful care models, most of the PFQ projects built on work already underway. One grantee noted that the PFQ program "offered an opportunity to continue what we had already started and what we wanted to do." PFQ funding allowed organizations to expand upon their prior quality improvement or bioterrorism preparedness work and/or accelerate their efforts. Several used the funding to strengthen operational and/or infrastructure support to more comprehensively carry out their work. In addition, a few of the grantees transformed concepts from proposals rejected by other funders into projects that were more in line with the aims of the PFQ program. 

Though the RFA encouraged applicants to build their proposed interventions on published evidence of effectiveness, the evidence base is stronger in some areas than others. Bioterrorism projects, in particular, were challenged to address topics where a strong base of evidence and knowledge of how to proceed is just now developing and has many gaps.

2. Intervention Strategies

To achieve their quality improvement goals, PFQ grantees intended to implement a variety of changes in health care systems, organizations, and clinical practices.  Projects seeking direct improvements in clinical care primarily utilized training, education, or technical assistance to implement organizational and/or operational process changes in target organizations. Projects seeking to utilize purchaser power to leverage change focused on mechanisms for implementing policy/reimbursement changes. Some bioterrorism preparedness projects also included training and technical assistance, and some studied or developed emergency preparedness planning processes and tools. The effectiveness of these strategies will be examined further in Chapter V.

Changes in Provider Practices and Operations. Of the 15 grants focused directly on changing provider behavior, 12 planned to conduct some form of training, education, or technical assistance to increase use of clinical guidelines in daily practice. This involved staff training on guidelines and/or working with staff to change workflow, the documentation of care processes, or organizational policies to increase adherence to clinical guidelines. Most of these grantees also planned to offer follow-up support to providers.

The majority of these 12 grantees combined the three strategies to maximize providers' adoption of clinical guidelines.  For example, the American College of Physicians developed a practice-based continuing medical education course, based on the Institute for Health Improvement (IHI) rapid cycle quality improvement model, to train teams of doctors, nurses, and office administrators on how to improve quality of care and outcomes for patients with chronic diseases.  They also developed a toolkit to help the teams implement clinical, administrative, and patient education techniques to be incorporated into daily workflow, and planned to follow up in between training sessions via conference calls to help providers deal with problems putting the tools into practice.

A few grantees provided intensive on-site training/technical assistance to their targets. For example, project leaders from the Medical University of South Carolina made site visits to some of the groups participating in the practice partner research network (PPRNet) in PMSI's project. During these visits, PPRNet staff or consultants would meet with all members of the practice for about a half day to assess the practice's performance, highlight what was working well and explore opportunities for improvement.

In addition to ACP's project, two grantees incorporated IHI's rapid-cycle quality improvement approach as the basis for their interventions. CHCA adopted this approach in the last two years of its project, to bring more rigor and consistency to its quality improvement efforts in pediatric hospitals. It launched two rapid-cycle improvement projects, each with different sets of hospitals. The hospitals sent teams to learning sessions and received intensive coaching on change implementation in conference calls between sessions. Like ACP, CHCA also created and tested toolkits for implementing patient safety best practices in hospitals. VNSNY also used the IHI rapid-cycle improvement model to design and implement diabetes care improvements in the eight participating home health agencies.

Several grantees planned to collect data on provider performance and report back to them on their progress in following clinical guidelines or meeting performance standards.  Lehigh Valley Hospital and Health Network, for example, used a system called Achievable Benchmarks of Care (ABC™), which sets a benchmark for care based on best practices of regional peers and reports to physicians on how they compare to their peers.

Changes in Payment Policies to Reward Quality.  The two purchaser-focused PFQ grants used different strategies for creating or aligning payment incentives to promote quality care.  The Leapfrog Group recruited payer and purchaser groups to pilot test financial incentive and reward programs that utilized their recommended hospital patient safety practices in six health care markets around the country.  One of the pilots was led by the Boeing Company, which worked with consultants secured by the Leapfrog Group to implement a program for employees enrolled in the company's PPO, which offered a discount on care provided in hospitals that met Leapfrog's quality and patient safety practices. In another pilot project, Leapfrog arranged for technical assistance to the Maine Health Management Coalition to help design and implement a bonus pool for high performing hospitals. HealthFront, which led the other purchaser-focused project, studied the current status of pay-for-performance and public reporting in two health care markets, to identify the degree of alignment among insurers and payers in their use of provider incentive programs. HealthFront reported its findings to the purchasers to prompt discussions about how to make the incentives more consistent. The project also conducted surveys of medical group managers in Minnesota and physicians in Colorado to determine their awarenessof and response to different types of incentive programs.

Study of Providers' Bioterrorism/Emergency Response Preparedness. While the five bioterrorism/emergency response preparedness grantees all sought to improve the capacity of the health care delivery system to respond to crises, they did so in different ways. Connecticut Department of Health, in partnership with Yale New Haven Health System, proposed to create and evaluate the effectiveness of a training program for front-line clinical staff. JCAHO assessed the linkages between the health care system and public health infrastructure through the use of a survey of hospitals and community health centers.  Altarum Institute modeled the surge capacity of health care systems in the event of a bioterrorism event, under varying assumptions regarding the public health response.

Return to Contents

C. PFQ Grantee Organizations and Partnerships

1. Lead Grantee Organizations

The PFQ solicitation encouraged applicants with the capacity to influence health care organization and delivery and the ability to evaluate the impact of their efforts (Chapter I). Specifically, the solicitation targeted applicants from health care professional organizations, accrediting agencies, practice networks, employer coalitions, and health insurers. Twelve of the 20 PFQ grants were awarded to organizations falling within these categories: five were awarded to provider groups, five to health care professional organizations, one to an accrediting/certifying body, and one to an employer coalition/purchaser collaborative. Of the remaining eight grants, four were awarded to research organizations, two to state government agencies/departments, one to a university, and one to a private company. The organizational types of the PFQ grantees are shown in Box 4.

Seeking to fund a "different kind" of project, AHRQ's RFA solicitation excluded universities from being eligible for PFQ grants, though academically-based individuals were not precluded from being involved in the grants. In fact, Principal Investigators affiliated with academic institutions led 6 of the 20 PFQ grants. Of these six, only one of the academic institutions was the actual grant recipient (Texas A&M University). This grant was also the only one of the six that focused on bioterrorism preparedness, which we believe may have been the reason for this exception. The remaining five university-affiliated Principal Investigators applied to the PFQ program through other organizations, whose responsibilities included an administrative/fund disbursement role.⁹

Box 4. PFQ Grantees by Organizational Type

Organizational Type	PFQ Grantees
Provider organizations(PFQ grant usually housed in the research division)	American Hospital Association/HRET Catholic Healthcare Partners, Child Health Corporation of America, Lehigh Valley Hospital and Health Network, Visiting Nurse Service of New York
Health professional associations	American Academy of Pediatrics, American College of Physicians, American Medical Association, American Medical Directors Association Foundation, Association of California Nurse Leaders
Health care accrediting/certifying body	Joint Commission on the Accreditation of Healthcare Organizations (JCAHO)
Employer/purchaser collaborative	The Leapfrog Group
Independent research organizations	Altarum Institute, HealthFront, International Severity Information Systems, RTI
State health departments	New York State Department of Health, Connecticut Department of Health
University	Texas A & M University System
Private company	Physician Micro Systems, Inc.

2. Partners and Other Affiliates

Number of Partners. In contrast to traditional research grants, the PFQ program encouraged grantees to form partnerships with a variety of types of organizations and individuals that could help reach target providers. The numbers of partners involved in PFQ grant activities varied tremendously across the projects. Some had few partners, while others had as many as 20 or more partners with varying levels of involvement. A full list of partner organizations is shown in Appendix Table A.2, which displays the partners associated with each project according to organizational type.

Grantees structured relationships and communication among partners differently, depending on the scope and focus of their projects. The projects led by the AMA, ISIS, Lehigh Valley Health and Hospital System, and VNSNY intended to collaborate with a dozen or fewer provider organizations as working partners, usually because their interventions were more time-intensive, either for the lead agency or the provider organizations. Other projects, such as those led by the American Academy of Pediatrics, American College of Physicians, Child Health Corporation, and PMSI, planned to engage between 35 and 180 provider organizations, and in these cases group training sessions, quarterly reporting and occasional teleconferences were used to interact with a larger number of target organizations.

Types of Partner Organizations. The four most common types of partner organizations affiliated with PFQ grantees included:

1. Research organizations or university-based researchers, typically responsible for leading the projects' research and evaluation design and implementation.
2. National or state health care professional organizations led 5of the 20 projects as noted earlier, but were involved in several other projects as partners to help promote QI approaches or recruit their members to participate.
3. Provider organizations or practices, which were often the targets of QI tools and methods.
4. State or local public health agencies, one of which led a project (NYS-DOH) and involved as partners in bioterrorism and emergency preparedness projects.

Type of Role. Partners played different roles with the grantee team.  In some cases, partners were expected to work very closely with the lead grantee on overall leadership for the project. They could be involved in any or all of the following: grants management, research design, quality improvement training, data collection and analysis, and marketing/dissemination of the project results. Instead, or in addition to being part of the leadership team, some partners were asked to perform the following roles:

• Intermediaries, sometimes referred to as key collaborators, who recruited, trained or provided technical assistance to the target organizations, and served as a critical link between leadership and targets.  Those filling the intermediary role included a variety of health care professional organizations, providers, or quality improvement organizations (QIOs).
• Targets,who included the health care organizations or providers on whom the quality improvement intervention was focused, as discussed earlier. 
• Advisors, who provided expert input to project leaders in their areas of clinical, health services research, and health delivery expertise.

Types of Partnerships. While the way in which each grantee worked with its partners differed greatly among the projects, there were two major types of partnerships, which differed by how the grantee organization related to the target organizations:

• In one model, used largely by the projects that focused on bioterrorism and emergency preparedness, grantees largely involved target organizations as advisors or as study participants.
  
• In the second model, used by the 14 projects that targeted providers for quality improvement efforts, and 2 focused on purchasers, grantees forged direct working relationships with the target organizations to design, implement, and assess the success of efforts to translate research into quality improvements.  Virtually all of the projects adopting this model also involved advisors as partners, but the advisors usually had little or no interaction with target organizations.

We describe in more detail in Chapter VI how these partnerships actually worked—how the partnerships functioned, how partners communicated and made decisions, and how they involved staff in target organizations. Chapter VI also assesses how partnership structure and function contributed to the success of individual projects and to the overall goals of the PFQ program.

---

⁷ The RFA stated that grants could focus on priority health conditions, including: cancer, diabetes, heart disease, chronic kidney disease, or respiratory disease, as well as priority health issues, including maternal and child health, mental health, long-term care, and bioterrorism. Some of these priority conditions and issues were expected to fall within the categories to be addressed by AHRQ's National Health Care Quality Report, under development when the RFA was released. The 2005 National Healthcare Quality Report identified nine clinical conditions or care settings: cancer, diabetes, end stage renal disease, heart disease, HIV/AIDS, maternal and child health, mental health and substance abuse, respiratory disease, and nursing home and home health care.
⁸ The RFA stated that PFQ applications should address priority populations identified in AHRQ's authorizing legislation: inner-city areas and rural areas (including frontier areas); low-income groups; minority groups; women; children; the elderly; and individuals with special health care needs, including individuals with disabilities and individuals who need chronic care or end-of-life health care.
⁹ The remaining five university-affiliated Principal Investigators led projects for two professional organizations, one provider group, one state government department, and one private company.

---

Return to Contents
Proceed to Next Section