Module 5: Improvement - Developing a Strategy for Diabetes Quality Improvement

Quality health care is a goal that all health care professionals and policymakers can achieve, yet many do not know where to begin.

The challenge of the health care system is to define what is "quality health care" and lead participants in the health care system to increase quality, reduce mistakes, and attain quality results for every patient every time. Some may view this as impossible. Others can point to the great strides that have been made in manufacturing and other services by applying the principles of quality improvement. And some can point to dramatic improvements in reducing deaths in U.S. hospitals from applying the principles of quality improvement (Gabor, 2004). Additionally, a number of States today can point to gains that they have made in diabetes outcomes for their citizens to confirm that quality improvement in health care is possible.

Module 1: Background provided an overview of diabetes and quality improvement. Module 2: Data provided a variety of data sources with State-specific data on diabetes quality of care. Module 3: Information helped State leaders understand how data must be examined to make comparisons and create information for guiding decisions and leading change. Module 4: Action offered a variety of national, Federal, State, and local approaches, resources, and best practices that can inform State quality improvement efforts.

This module aims to assist State leaders to develop diabetes quality improvement strategies suited to State contexts. Module 5: Improvement provides models for quality improvement, presents a case study of how one State — Wisconsin — undertook an ambitious quality improvement effort, and discusses how State leaders can begin to develop their own State-specific strategies to improve diabetes care quality.

Return to Contents

A Model for Quality Improvement

While local contexts differ, models of quality improvement give the common elements needed to stimulate change and improvement in any situation. As State leaders embark on new initiatives or revitalize existing ones, quality improvement models can inform those efforts. The key is to find a suitable model for an individual State and its partners, and then pick and choose the components that are most useful for a specific local context. Explained below is a model that may be useful for State leaders developing quality improvement strategies.

Plan-Do-Study-Act (PDSA) Model

 A time-tested quality improvement tool still useful today is the "Plan-Do-Check-Act" or the "Plan-Do-Study-Act" model for guiding quality enhancement projects of all types (go to Figure 5.1). The PDSA model conceptualizes the continuing cycle of improvement.

Its steps for effective quality improvement include:

• Plan - Set the goals of the quality improvement cycle— questions, predictions, data to be collected, and the who, what, when, where of the project.
• Do — Carry out the plan and document problems and unexpected observations.
• Study — Complete the analysis of the data, compare to predictions, and summarize lessons.
• Act — Determine changes to be made and decide what will happen in the next cycle (Langley, Nolan, Nolan, et al., 1996).

The PDSA cycle usually applies at the point of production, in this case to the front-line of health care at the point of care. The concept also can be applied to the quality improvement role of State leaders. Drawing on insights from State quality improvement activities around diabetes care, State leaders might consider a "Partner-Plan-Do-Study-Act" model.

Figure 5.1 The PDSA Quality Improvement Cycle

• Partner — Decide who are strategic partners of quality improvement and recruit them to the project — champions in health care production, stakeholders (e.g., consumer/patient groups, health care professionals, purchasers, health plans, and topic experts, among others), and key State leaders and agencies (e.g., visible champions, diabetes experts, program planning/evaluation staff and quality improvement experts). Is the group large enough to include key leaders and perspectives, yet small enough to be productive?

• Plan — The goals of a project will be broad in the context of statewide activities because many partners and processes will need to be involved. What does the group predict are the current obstacles to quality care? How will the goals be put into action? What data need to be collected to prove that the changes are indeed improvements?

• Do — Test the plan and document problems and unexpected observations as data are collected. Initial plans seldom produce desired results the first time. Pilot test the ideas of the group with front-line health care programs, providers, and consumers. Reconvene the partners and discuss successes and problems.

• Study — Complete the data analysis, compare the results to predictions, and summarize lessons learned. Do the test results convince the partners that full-scale implementation will be successful? Because the scope of activities may be broad and costs may be involved, the planned action should be based on reasonable data and results.

• Act — Determine the changes to be made. Implement the changes State- or district-wide. Continually assess those changes through data collection and analysis. Are the changes working? What will happen in the next cycle?

The PDSA is one model of quality improvement that has withstood the test of time. There are other tools and methodologies for quality improvement to suit the various stages at which States find themselves. Following are additional resources that States might want to use to facilitate quality improvement wherever they are along the continuum.

• The Quality Assurance Project, sponsored by the United States Agency for International Development (USAID), presents models based on quality improvement on an international scale. These models are useful and easily translated for States (information available at http://www.qaproject.org/resourcesintro.html).

• The IHI breakthrough series focuses on change at the provider level, but is an important approach that State leaders should understand for developing change agents (information available at http://www.ihi.org/ihi).

Numerous tools are also available to further assist quality improvement projects. Quality improvement tools suited for policymakers are available on AHRQ's Quality Tools Web site at: http://www.innovations.ahrq.gov/channels/channel.aspx?mode=3&inc=browse.policy_makers.inc. For another quality toolbox, go to Tague (1995).

The PDSA model can be applied to the context of State leadership in quality improvement. The actual approaches and actions that States will take will be as varied as the examples that appear in Module 4: Action of this Resource Guide. One State's experience, in particular, can help illustrate how the PDSA model can be applied to an actual quality improvement project.

PDSA Case Study: Wisconsin Collaborative Diabetes Quality Improvement Project

Wisconsin's DPCP, part of the Wisconsin Department of Health and Family Services, received CDC funding in 1994 and other funding since then. Over time the DPCP developed an ambitious strategy to improve diabetes care quality for clients of managed care organizations. Wisconsin's diabetes quality improvement efforts in many ways mirror the stages of the PDSA model described above.

• Partner — In 1996, the DPCP formed the Diabetes Advisory Group comprised of over 50 diverse groups and organizations. Wisconsin's health maintenance organizations (HMOs) were key partners in the advisory group.

• Plan — One of the first plans developed by the Diabetes Advisory Group was the development of Diabetes Mellitus Care Guidelines and supporting documents for use by all health care providers in the State. Released in 1998, these guidelines were endorsed by the Advisory board members, and members promoted the use of the guidelines throughout the health care system blanketing the State with a common message about quality diabetes care. The DPCP used materials from other States and also worked with the University of Wisconsin-Madison to use data to assess the status of diabetes care in the State and adherence to the Diabetes Mellitus Care Guidelines.

• Do — Out of this successful effort, the Wisconsin Collaborative Diabetes Quality Improvement Project was established in 1999. The goal of the Diabetes Quality Improvement Project is to improve the quality of diabetes care for people who receive services through Wisconsin's HMOs and two other large health systems by:
  • Evaluating implementation of the Wisconsin Essential Diabetes Mellitus Care Guidelines.
  • Sharing data issues, strategies, initiatives and lessons learned.
  • Improving diabetes care through collaborative quality improvement initiatives.

Collaborators included university experts, Wisconsin's QIO, the State Medicaid program, and other health care industry partners. The department used a two-pronged approach to convince the HMOs to sign on. First, they leveraged the support of a well-connected spokesperson to discuss the guidelines and the possibility of forming the collaborative. Secondly, they participated in ongoing discussions about quality improvement in the private sector. They presented the collaboration as a potential win-win opportunity. HMOs would get value from the project through access to information, tools and ongoing support as well as receiving good media coverage for their work with the State. The DPCP would reach the 68 percent of the State's population served by the participating HMOs.

• Study — Collaborators used the Health Plan Employer Data and Information Set (HEDIS®) comprehensive diabetes care measures to track progress in improving diabetes care. The State's DPCP contracted with the University of Wisconsin to provide confidential analysis and reporting. Each HMO was given a confidential identifying number so it could see how its performance compared with other organizations. The project was careful to use these data results cooperatively, not competitively, with a goal of improving diabetes care in Wisconsin. Participants in the Collaborative continued to discuss issues and strategies such as registry development, data collection, and provider profiles.

The quality improvement plan included evaluative efforts to assess improvements in diabetes care. An evaluation of the HEDIS® data showed that the project:

• Increased eye exams for people with diabetes from 62 to 69 percent.
• Increased cholesterol screening and control from 72 to 78 percent and 45 to 51 percent, respectively.
• Increased kidney disease monitoring from 47 to 52 percent.

• Act — With data collection and reporting in place, the Wisconsin Collaborative Diabetes Quality Improvement Project took further action by focusing its quality improvement efforts. With HbA1c rates already at 90 percent in the State but eye examination rates much lower, the partners determined that the project should establish a statewide Diabetes Eye Care Initiative. In 2001, this project began with the goal of increasing eye examination rates and enhancing communication among specialists and primary care providers related to diabetes eye care.

In October 2003, the project released a compendium of the diabetes quality initiatives implemented in the 5 years since the project began. In addition to describing the interventions used, the compendium provided information on barriers, ongoing challenges, and lessons learned. Some of the lessons and strategies used to achieve the encouraging results of the project were:

• More inclusive quality improvement teams over time
• Increased use of diabetes care teams, champions, and case management services
• More in-depth barrier analysis and intervention evaluation
• Community collaboration
• More advanced information systems for developing, tracking, and feedback on "targeted" interventions
• Support for providers and clinics
• An increased focus on the role of the consumer
• Increased use of technology to enhance communication and outreach (Wisconsin Department of Health and Family Services, 2003).

More information on the Wisconsin Collaborative Diabetes Quality Improvement Project, is available at: http://dhs.wisconsin.gov/health/diabetes/hmo.htm.

As this case study demonstrates, quality improvement can take many years and iterations before actual change and quality improvement can be documented. However, the reward is that once the partnerships and processes are in place there is the opportunity to see measurable advances in care quality and in health outcomes.

Return to Contents

Developing a State Strategy for Improving Diabetes Care Quality

The PDSA quality improvement model described above can be combined with previous modules of this Resource Guide to build a strategy for improving diabetes care quality. AHRQ has also developed a companion Workbook that can assist State leaders through a step-by-step process for using the data, information and resources from this Resource Guide to develop the case for diabetes quality improvement in a particular State, examine strategic areas for improvement, and develop a detailed strategy.

 Described below are tools that can help State leaders develop a State quality improvement strategy. These tools can be used in conjunction with the Workbook exercises. The first tool is the State Diabetes Quality Improvement Worksheet (go to Figure 5.2) that can assist State leaders with assembling the data about their State and diabetes. Another tool is the PDSA checklist that provides the common steps State leaders need to take to build a quality improvement strategy. In working through these tools, State leaders are advised to work closely with State DPCP officials to plan and develop their State's diabetes quality improvement strategy.

Building the Case for Diabetes Quality Improvement

One step in the process of developing a quality improvement strategy is for State leaders to gather information about diabetes in their State. The worksheet below helps State leaders to assemble State specific information on diabetes prevalence, cost and quality of care to assess opportunities for improvement. This worksheet information combined with the inventory of programmatic activities related to diabetes assembled at the end of Module 4: Action allows State leaders to assess the current condition of diabetes care and public policy in their State. Using this information, they can make the case that diabetes quality improvement is important for their State.

Putting the PDSA Model to Work

The adapted PDSA model is a general model intended to capture the most important components of quality improvement; but State leaders may wonder how to put it to use. Provided below is a checklist of PDSA quality improvement steps. This checklist outlines the common steps that State leaders need to take to develop a quality improvement strategy. Using the checklist as a framework, State leaders can fill in the State and local background, data, information, public policy approaches and other resources to develop a strategy suited to their particular context.

As State leaders do this, one of the most important factors to bear in mind is the cyclical nature of quality improvement. Improving health care quality is not a one-time activity but must be ongoing. Sustained improvement usually occurs over many years. Thus, the most effective action plans will include not only short-term goals but long-range ones as well.

Return to Contents

Integrating Quality Improvement Activities Across Conditions

Diabetes is one of several chronic illnesses with demonstrated opportunities for quality of care improvements. Care for asthma, cancer, heart disease, and other common chronic conditions affecting millions of Americans too often falls short of what research has indicated to be the most effective treatments. Diabetes and other chronic conditions combined account for 78 percent of all health care spending and 7 out of 10 deaths.

State leaders can use this information to help decide how broadly or narrowly to focus their quality improvement efforts. Diabetes may be just one of several costly health care conditions that are appropriate areas to invest in quality improvement efforts in a given State. In addition, there may be advantages to integrating quality improvement efforts across several conditions, such as stretching scarce resources by using economies of scale across programs, minimizing investment costs in infrastructure, and maximizing the effect of systemic changes in health care delivery.

Thus, some States may choose to expand on existing diabetes quality improvement efforts while other States may want to establish comprehensive quality improvement efforts that target several diseases at once. For example, several States, including Wisconsin, have expanded their efforts with diabetes to address heart disease since the two are related. On the other hand, Vermont has initiated a broader chronic care initiative to improve the quality of care for all chronic diseases but has chosen diabetes as the first focus area for the initiative. (Go to the Vermont Chronic Care Initiative Web site at http://healthvermont.gov/blueprint.aspx for more information.)

Return to Contents

The Importance of Evaluation

Evaluation is essential to understand whether a quality improvement activity is accomplishing planned goals, whether goals and actions are ultimately improving the health outcomes of the population, and what adjustments are necessary. Evaluation in quality improvement can be done quickly, as often look at longer term, underlying components of the program. One program that can serve as a resource for State leaders in developing an evaluation plan for diabetes quality improvement efforts is the CDC's accountability efforts for State DPCPs.

In 1999, the CDC began addressing the need for more systematic State-level programmatic evaluation and accountability in the National Diabetes Prevention and Control Program. The CDC devised seven national objectives for diabetes care, including increasing the percentages of people with diabetes receiving HbA1c testing, eye exam, foot exam, and influenza vaccination. In addition, the national objectives include reducing health disparities and establishing linkages to other wellness programs, such as: physical activity or smoking cessation programs for people with diabetes.

The CDC asked States to devise their own State objectives for improved diabetes population health in order to address the uniqueness of each State's population. A critical objective that States usually include is to establish measurement procedures to track progress. The CDC focus on evaluation emphasizes the importance of measurement. The State DPCP has become a catalyst for statewide improvements through partnering and accountability on various operational levels.

CDC Model for Program Evaluation

 The CDC employs a model of evaluation that includes four groupings of standards for program evaluation, and six repeating steps in the evaluation process as illustrated in Figure 5.3.

The CDC provides 30 standards under the four subgroups of utility, feasibility, propriety, and accuracy. These standards are guidelines for conducting sound and fair evaluations and may be briefly described as follows:

• Utility ensures that the evaluation serves the needs of intended users.
• Feasibility results in evaluations that are realistic and sensible.
• Propriety ensures ethical integrity in the conduct of the evaluation.
• Accuracy leads to information that is technically sound.

Steps in the Evaluation Process

The six steps in the evaluation process may vary as to when they are carried out, though one step usually lays a foundation for the next. Steps will be repeated as results become clear and new issues arise. Each step serves to ensure the effectiveness of the evaluation.

• Engaging stakeholders is essential to ensure that the evaluation addresses the important elements of the program and that the evaluation is used. Stakeholders include those served by the program, those planning and directing the program, and those involved in program operations.

• Describing the program serves two functions. First, it lays out in detail the program's goals and strategies so that everyone involved understands them. Second, it provides an opportunity for consensus building around the goals and strategies.

• Focusing the evaluation design addresses the greatest issues of concern. This step includes identifying the purpose of the evaluation; defining the users and usefulness of the evaluation; listing stakeholders' questions that need to be addressed; establishing methods to ascertain information upon which the evaluation will be based; and developing consensus around particular roles and responsibilities pertaining to the evaluation.

• Gathering credible evidence contributes to the robustness of the evaluation. Developing credible evidence involves defining appropriate indicators, identifying legitimate sources of information; ensuring the quality of data gathered; and aligning the infrastructure for collecting evidence with the environment (and individuals) from which the information is gathered.

• Justifying conclusions is important to ensure that the evaluation will be used. When consensus is reached regarding the goals and strategies of the program, when the values of the evaluation are aligned, and when the evidence gathered is credible, then conclusions will naturally be justified. At this point, conclusions and recommendation can be made.

• Ensuring use and sharing lessons learned includes designing mechanisms for feedback and dissemination of the information gained in the evaluation.

Employing an evaluation specialists or, at the least, assembling an evaluation team with a designated leader will help facilitate the process. Issues regarding internal bias and external influences must be addressed to ensure integrity of the analytic work and a trusted evaluation of a program or project.

To be effective, however, evaluation strategies must be timely and useful. They should be considered at the beginning of the project and they should have a reasonable deadline for completion. Including an experienced evaluator on the quality improvement team can help ensure that the evaluation will be sound, useful, and timely. The evaluation should feed back to the quality improvement cycle and direct future actions.

Return to Contents

Summary and Synthesis

Module 5 has provided a model of quality improvement, a case study of one State's innovative quality improvement strategy and some tools and considerations to help State leaders develop their own quality improvement strategy. This module does not provide States with one approach or answer. Instead, decisions about the kind of quality improvement strategies that a State should pursue are the responsibility of State leaders and their partners, who are positioned to know what is best suited for their State context.

There are common elements to quality improvement that can inform the development of State strategies. The PDSA model adapted for the policymaking context is one approach that can assist State leaders. It is also important that State innovators examine the current condition of diabetes care and what diabetes programs are underway. State quality improvement efforts can then build on and fill in the gaps to develop a more comprehensive, coordinated approach to improving care. Because quality improvement occurs over a long time frame, evaluation is crucial to determine what effects the strategy has had and to justify continued efforts and resources over time.

Return to Contents

Resources for Further Reading

Langley G, Nolan K, Nolan T, Norman C, Provost L. The Improvement Guide: A Practical Approach to Enhancing Organizational Performance. San Francisco: Jossey-Bass Publishers, 1996.

Tague NR. The Quality Toolbox. Milwaukee, WI: ASQC Quality Press, 1995.

Return to Contents

Associated Appendix for Use With This Module

Appendix F: NHQR Quality Measures for All Conditions by State

Appendix F provides quality measures for all conditions and topics in the NHQR. It includes the national estimate and an indicator for whether or not the State estimate (not shown due to space limitation) is statistically greater, lower, or no different from the national average. This resource can help States identify which diseases and their treatments may be in need of attention.