Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
8 Technical Assistance for Quality Improvement CHAPTER SUMMARY During the 7th and 8th scopes of work (SOWs), Quality Improve- ment Organizations (QIOs) offered technical assistance to provid- ers to help them improve their quality of care. This task (Task 1) was entitled Improving Beneficiary Safety and Health Through Clinical Quality Improvement in the 7th SOW and Assisting Pro- viders in Developing the Capacity for and Achieving Excellence in the 8th SOW. This chapter presents an overview of this task and reviews general policy issues, including how QIOs may choose the providers they will work with intensely (the "identified partici- pants," who work in an "identified participant group") and the modes of interaction. Next, the chapter discusses details of this task, as delineated in the contracts for the 7th and the 8th SOWs, including specific examples of projects and activities of the QIO Support Centers. Chapter 9 will discuss the impacts of these activi- ties on clinical outcomes and the transfer of knowledge. As technical assistants, the Quality Improvement Organizations (QIOs) use one-on-one consulting, collaborative activities, workshops, training ses- sions, root-cause analysis, and other techniques to assist providers with improving their health care processes and organizational systems. Budget constraints limit the degree to which QIOs can assist providers, as well as the number of organizations or individuals that they can assist within the state. Also, the presence of other quality improvement entities in the states can affect the demand for QIO technical assistance and the potential for partnering. These entities may include departments of health, state survey agencies, specialty societies, or private corporations. Although technical assistance can take many forms, the value of one methodology over another has not been determined, as will be discussed in Chapter 9. 192
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 193 DEFINING TECHNICAL ASSISTANCE "Technical assistance" can have different meanings. In general, it is the process by which QIOs work with providers, managed care organizations, and other stakeholders to improve patient outcomes. Fundamentally, QIOs provide technical assistance by the following means, among others: · detecting areas in need of improved performance; · helping identify the root causes of problems; · helping implement interventions and systems changes; · teaching process improvement methodologies and promoting best practices; · facilitating knowledge transfer; · reducing reporting burdens on providers; · collecting, aggregating, and analyzing data on performance mea- sures; and · working with stakeholders to coordinate quality improvement efforts. RECRUITMENT OF IDENTIFIED PARTICIPANTS Identified participants are the providers with whom the QIOs work intensely on quality improvement projects. Recruitment of identified par- ticipants is generally left to the discretion of each QIO. Provider participa- tion is voluntary, but in many tasks the Centers for Medicare and Medicaid Services (CMS) stipulates the percentage of each provider type that the QIOs must recruit (CMS, 2002, 2005c). CMS expects the QIOs to demonstrate significant improvement in the identified participant group and, in some cases, greater improvement compared with statewide gains. These gains are evaluated by calculating the reduction in failure rate1 (Jencks et al., 2003). Many QIOs look to their identified participant group to act as leaders for other providers in the state, especially because CMS evaluates the QIOs, in part, on the basis of statewide improvements. QIOs often recruit identified participants using a number of criteria, including readiness for change, pro- vider volume or size, current level of quality performance, and other demo- 1A reduction in failure rate, also known as relative improvement, is the change in perfor- mance between the baseline and the follow-up (absolute improvement) divided by the differ- ence between the performance at the baseline and perfect (100 percent) performance. The reduction in failure rate may be viewed as a crude measures of improvement, as it does not distinguish between difficulty of improving from 90 percent to 95 percent versus from 70 per- cent to 85 percent, both having a 50 percent reduction in failure rate.
194 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM graphics. The advantages and disadvantages of each of the methodologies related to these criteria are discussed in the next few sections. Readiness for Change When developing a strategy for the recruitment of identified partici- pants, some QIOs look to the five categories derived from Rogers' theory on the diffusion of innovations. By this theory, "innovators" initiate the process by embracing new ideas. The "early adopters" are often highly regarded as opinion leaders in their communities and convince the "early majority" to adopt the innovation. Those in the "late majority" follow with adoption of the innovation because of overwhelming peer pressure. Finally, Rogers identified a group that he called the "laggards," who are the last to adopt any innovative idea or process, skeptical, and resistant to any change. This model proposes that as innovators and early adopters embrace a new process or philosophy, the process of natural diffusion will spread ideas to the rest of the community. Because CMS evaluates QIOs, in part, on the basis of statewide improvements, QIOs may opt to target opinion leaders of the community, hoping that if these providers change their practice pat- terns, the rest of the community will follow, leading to greater widespread change over time. Some studies show the use of opinion leaders to be ef- fective in changing practice patterns for specific interventions, but other studies show mixed results (Thomson O'Brien et al., 2005; Davis, 1998; Soumerai et al., 1998). In telephone interviews, many QIO chief executive officers (CEOs) expressed the value of working with early adopters: "Early adopters/willing participants are a huge resource for massive education be- cause they have proven knowledge of how things can work." On the other hand, QIOs might theoretically focus on the laggards who need the extra push and individualized attention provided through one-on- one relationships with QIOs. Early adopters may alter their practices on their own or with minimal assistance through other programs offered in the state. Additionally, early adopters may already be involved in other state programs and so may feel no need to participate with QIOs. High-Volume Versus Low-Volume Providers One theoretical methodology for recruiting identified participants is to target either providers with large patient populations (high-volume provid- ers) or those with small patient populations (low-volume providers). By working with high-volume providers, QIOs may hope to achieve a greater impact because of the larger number of beneficiaries seen by these provid- ers. However, others may believe that low-volume providers should be as-
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 195 sisted first, as they are the most likely to lack the resources and staff exper- tise needed to investigate options, adopt change systems, and learn quality improvement techniques. In telephone interviews, one QIO CEO stated: "Critical access hospitals (small and rural) usually welcome us with open arms. In facilities with very few providers, they are often so overworked that they do not have anyone to do the quality documentation. Here, the QIO has to take more responsibility." High Performers Versus Low Performers QIOs may recruit participants by targeting either high performers or low performers, but this method is confounded by the difficulty in defining that distinction. In fact, in telephone interviews, many QIO CEOs expressed concern for how this distinction can be made. In theory, high performers may be either those who produce a consistent level of quality care or those who have demonstrated significant improvements from the baseline. Pay- for-performance programs or public reports may help to identify the low performers, which may help the QIOs to determine who would benefit the most from assistance. Because pay-for-performance programs provide fi- nancial incentives for improved quality, providers may be more willing to work with QIOs to improve their performance. In telephone interviews, 10 of 20 QIO CEOs independently proposed that a barrier to technical assis- tance is a lack of motivation of providers to work on quality. All 10 agreed that pay for performance or sanctions would be strong motivators for pro- viders to work with QIOs. Additionally, in site visits to 11 QIOs, four raised the issues of pay for performance and public reporting as potentially strong motivators for providers to work with QIOs. High Performers Some QIOs prefer to work with high performers. In telephone inter- views, many QIO CEOs expressed the idea that high performers are the key to the diffusion of best practices. For example, two CEOs commented, "Dif- fusion of quality comes from good providers spreading the word," and "When you include high performers, you get more diffusion to other pa- tients. You are also more likely to engage specialty providers." However, because CMS evaluates QIOs on the basis of the amount of increased improvement achieved (by calculation of the reduction in the fail- ure rate), QIOs may have difficulty reaching contract goals when they are working with very high performers. If a provider is already performing highly, it may be more difficult to improve upon the failure rate.
196 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM Low Performers Alternatively, QIOs may choose to recruit low performers on the basis of the opinion that low performers, by definition, are the ones who need the most immediate help. Greater absolute and relative gains in quality may be achieved by bringing the low performers up to the level of the majority of providers. QIOs may perceive that high performers are already doing well and that their limited resources are most effectively spent in the areas of greatest need. Alternatively, some QIOs encourage participation in their quality interventions by all providers and include any providers who agree to the conditions of the program. This enables QIOs to meet their participa- tion level requirements and offer assistance to all who are willing and able to participate. In telephone interviews, all QIO CEOs said they preferred working with providers with a mix of performance levels. Telephone Interviews: Working with Low Performers When the QIO CEOs reacted to how a mandate to work only with the worst-performing providers would affect how they operate and what the likely outcomes would be, they stated that it would require more resources and would affect the diffusion of their quality improvement efforts state- wide. The CEOs thought that providers might perceive the focus of the QIO program to be a return to weeding out "bad apples" rather than pro- moting quality. They also questioned how to define "worst performers" and how CMS would evaluate the QIOs. Eighteen of 20 CEOs thought that a focus on the worst providers was not workable and has many disadvantages, such as the possibility of losing champions, diminishing diffusion of ideas, and increased investment in time and money. Only two CEOs thought that the focus on the worst- performing providers would not have much of an effect on their QIOs. However, they did qualify that by saying that the QIO might have to do more handholding of the poorer performers. Sixteen of the 20 CEOs thought that diffusion to other providers would be negatively affected if there was a focus on the worst-performing providers. Eighteen of the 20 CEOs raised the issue that poor performers barely have sufficient infra- structure for day-to-day survival, let alone quality improvement systems. All CEOs said that the worst performers would require more financial resources per site than the other providers. They require more intensive interventions (e.g., one-on-one onsite assistance and longer periods of inter- vention), and they need support for data and communications systems. Some providers do not even have the components of a basic communications infrastructure, like e-mail. Funds from CMS for the direct provision of tech-
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 197 nology or the support of technology acquisition by poor performers may be necessary. Random Selection Another option for the recruitment of identified participants is random selection, which, to date, the QIOs have not used. This method would be useful for evaluation purposes because it eliminates selection bias (partici- pation by the most highly motivated providers) and allows the greatest range of providers to be involved with the QIO program. Such an approach, com- bined with the random allocation of providers to interventions, would per- mit an accurate evaluation of the impact of the quality interventions. How- ever, because participation with QIOs is voluntary, it would be impossible to enforce participation by unwilling providers who are chosen randomly. One could, however, sample with replacement, in which those providers who choose not to participate would be replaced by other randomly se- lected providers who agree to participate. In multiple interviews and visits, the QIO staff expressed the opinion that the willingness of the provider to participate is an important part of the success of their technical assistance work. In telephone interviews, the QIO CEOs echoed this sentiment: "Re- alizing that the QI [quality improvement] process is voluntary, the issue is really whether the poorest performers want to make change. If the provider has no desire to improve quality, the QIO's hands are tied unless the perfor- mance is so egregious that it requires sanctioning." INTERACTION WITH PROVIDERS One-on-One Consulting Versus Collaboratives QIOs generally interact with providers through (1) individual consulta- tion and (2) community, statewide, and national collaboratives. One-on- one consultation provides direct, specialized attention. By receiving techni- cal assistance tailored to their needs, providers may be more likely to adopt changes. However, the development of multiple individualized relationships can be labor-intensive and cost prohibitive. This relationship can be espe- cially complicated when many parties are involved, such as when organiza- tions that hold contracts in more than one state work with integrated deliv- ery networks or large nursing home chains. Even if this method leads to big changes in quality improvement, it would be in a small subset of the pro- vider population because current budget restraints do not allow individual- ized assistance to every provider. On the other hand, one-on-one consulting
198 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM with key providers, such as those who would act as opinion leaders, could help disseminate information as effectively. Collaboratives can foster relationships among providers and allow the sharing of best practices, but their impacts on quality measures and health outcomes are unclear (see Chapter 9). The success of this type of interaction depends to some extent on the topic chosen, the enthusiasm of the partici- pants, and the organization of the collaborative (Ovretveit et al., 2002). Successful collaboratives must not only have strong leadership, but the par- ticipants also must be effective leaders when they return to their practice settings. The literature shows that effective leadership is the key to organi- zational change (Joshi, 2000; Shortell et al., 2004; Bradley et al., 2005). The Institute for Healthcare Improvement (IHI) developed a model for a Breakthrough Series collaborative that has been widely adapted by many organizations, including the QIO program, and has been used for many topic areas. The Breakthrough Series model creates a structured collabora- tive that lasts 6 to 15 months, during which the collaborative teams partici- pate in three learning sessions (meetings in which all participants gather together to learn and share experiences), followed by three action periods (implementation of changes, tailored to local settings). Further support is provided through conference calls, Internet-based conferences, and site vis- its (Institute for Healthcare Improvement, 2003). Box 8.1 presents a de- scription of the IHI's Spread Initiative, a typical example of an IHI collabo- rative. A detailed evaluation of the impacts of these and other methods is presented in Chapter 9. Telephone Interviews: Collaboratives In telephone interviews, 13 QIO CEOs talked about collaboratives. Four of the 13 mentioned the IHI model specifically, with 2 saying that they modify it because of the costs and time associated with the IHI model. Overall, QIOs widely use collaboratives, regardless of the model. One QIO CEO stated, "We have a large state geographically. We deploy collabora- tives in our work in all settings. Identified participants work with col- laboratives for 12 to 18 months. They come together in follow-up confer- ences. There is lots of learning and sharing of collected data." Another indicated, "We use collaboratives as much as possible, especially through the hospital association. We have about 60 percent voluntary participation. We use an IHI model, though [it is] less intense; IHI requires so much meet- ing time that we did not think that would work for our providers." Nationally Defined Projects Versus Local Needs As described in Chapter 2, the QIO program has undergone an evolu- tion in which the core contract contains an increased emphasis on standard-
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 199 BOX 8.1 IHI's Spread Initiative Overview "IHI's Spread Initiative is a collaborative improvement project to help organizations establish processes and practices to spread successful change throughout their health care systems." How It Works "During the one-year membership in The Spread Initiative, partici- pants will meet three times for Learning Sessions. During these ses- sions, IHI's expert faculty will guide participants in the development of spread goals and outcomes measures, and lead discussions about proven methods for successful spread of improvement throughout the organization. Between Learning Sessions, participants will work with IHI faculty and each other--through conference calls, a list serve, and email--and with colleagues at their own organizations to test and imple- ment recommended changes. Throughout the process, IHI faculty will guide organizations through the necessary steps, and will formally assess their progress on a monthly basis, providing customized feedback and guidance based on results." SOURCE: Institute for Healthcare Improvement (2005). ized national projects and less emphasis on projects related to local needs. On the site visits to 11 QIOs, 5 specifically expressed frustration with the lack of flexibility in their contracts to address local needs. Of those 5, 3 suggested the need to return to a more balanced mix of local and national projects. They suggested the possibility of substituting a local project for a national task if the QIO has successfully achieved a high level of perfor- mance on that task in previous scopes of work (SOWs), but there is concern that the high level of performance will regress if performance on the waived national task is not actively monitored. During the 8th SOW, requirements to work with underserved local populations are incorporated into part of the Physician Office Task. TECHNICAL ASSISTANCE DURING THE 7TH AND 8TH SOWS Over the last 35 years, the QIO program's priorities have evolved along with the environment of health care (see Chapter 1). Predecessor organiza-
200 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM tions attempted to control wasteful and ineffective practices by identifying outliers at the local level. As national standards of care gained acceptance, the QIO program developed national projects. In the 6th SOW, QIOs per- formed some standardized work in the hospital setting, but other projects were local in nature and differed from state to state (CMS, 1999). In the 7th SOW, all quality improvement projects in each state were standardized to achieve the same goals by use of the same measures. Projects expanded to include provider settings, in addition to the hospital: nursing homes, physi- cians' offices, and home health agencies (CMS, 2002). As the program pro- gressed from a focus on cost containment to a focus on improved quality, the SOWs changed to reflect those priorities. Today, the QIO program stresses broader quality improvement in a shorter period of time. The 8th SOW looks to achieve transformational cultural and systems changes rather than the incremental changes more characteristic of the 7th SOW (CMS, 2002, 2005c). Technical assistance activities relate to five dimensions of performance defined by CMS: · Improvement in performance measure results (such as increased rate of mammography screening in the physician's office setting), · Improvement in clinical performance measurement and reporting (such as attaining a 25 percent level of self-reporting of expanded measures in the hospital setting), · Systems adoption and use (such as the implementation of an elec- tronic health record), · Implementation of key process changes (such as implementation of an immunization assessment survey by home health agencies), and · Changes in organizational culture (such as data collection by nurs- ing homes on satisfaction of residents and staff). Overall, the QIOs believe that they are extremely capable in their role as technical assistants. On the site visits to 11 QIOs, the QIO staff attributed their successes in the provision of technical assistance to positive relation- ships in the community (11 QIOs), internal experience and skills (9 QIOs), and a dedicated staff and a culture of quality (9 QIOs). When the QIOs were asked about challenges or threats to their technical assistance activi- ties, 2 QIOs voiced a need for more knowledge sharing, 3 related a need for more help from Quality Improvement Organization Support Centers (QIOSCs), and 6 discussed the difficulty of engaging providers. Other is- sues mentioned included a lack of information and a lack of experience with communications technology, the need for funding for start-up work before the contract begins, and individual staffing problems. When the QIOs were questioned about opportunities for new areas in which the QIOs could offer technical assistance, 2 QIOs mentioned information and communica-
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 201 tions technology and 4 identified pay for performance and public reporting. One QIO recommended more crosscutting initiatives. NURSING HOMES 7th SOW With the release of public reports on the performance of nursing homes in November 2002 (CMS, 2004), CMS added Task 1a to the 7th SOW for QIOs to work with nursing homes. Details of the work of the QIOs on public reporting initiatives are discussed in Chapter 11. For technical assis- tance, CMS charged the QIOs to work with nursing homes on quality im- provement projects to help improve performance on selected measures cho- sen from the Minimum Data Set, which was developed to assess the quality of care for both long-term and short-stay residents. The QIOs chose to work on between three and five of the following publicly reported quality of care measures: · Decrease in pain, · Decrease in infections, · Decrease in pressure sores, · Decrease in use of restraints, · Decrease in delirium, · Improvement in ability to ambulate, and · Improvement in ability to perform basic daily tasks (see Table A.5 in Appendix A). The measures were examined both at the statewide and at the individual nursing home levels. CMS expected the QIOs to partner with stakeholders such as state nursing home associations, professional societies, or local chapters of the AARP. CMS also expected the QIOs to interact with the Nursing Home QIOSC and offer quality improvement information to all nursing homes in the state. The QIOs had to enlist at least 10 percent of the nursing homes in the state to serve as members of the identified participant group (CMS, 2002). On average, the QIOs actually worked with 15 percent of the facilities in their individual states, with participa- tion ranging from 10 percent to 100 percent (Rollow, 2005). QIO activities included one-on-one consultations between the QIO's medical director and the nursing home's medical director, the provision of manuals on the use of restraints and on fall prevention to all directors of nursing in the state, the development of e-mail listserves with all nursing homes to share ideas, and the management of workshops with continuing education units (CMS, 2004). Several QIOs initiated innovative programs.
202 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM Health Services Advisory Group (Arizona's QIO) focused on the high per- formers among its participating nursing homes to determine common fea- tures and to try to replicate their successes in other locations (CMS, 2004). MetaStar (Wisconsin's QIO) evaluated innovative nursing home models around the country to better understand the implementation of resident- based models of care (CMS, 2004). The National Nursing Home Collabo- rative operated on a larger level by the Process Improvement QIOSC con- tract was held by Qualis Health through its QIO work for Washington state. In this project, 43 QIOs each worked with a subset of the participat- ing nursing homes in their home states to improve pressure ulcer manage- ment (CMS, 2004; Eloranta, 2005). Qualis Health conducted this collabo- rative in the style of IHI to learn how to identify, measure, monitor, and treat pressure ulcers. The impacts of the various quality initiatives are dis- cussed in Chapter 9. 8th SOW In the 8th SOW, QIOs work with two groups of identified participants, in addition to their work on statewide nursing home activities (CMS, 2005c). In this subtask, QIOs focus on improving clinical performance on specific measures (as reported on the Nursing Home Compare website), setting improvement targets, and analyzing resident and staff satisfaction, which includes monitoring of workforce turnover. Statewide Statewide, QIOs provide assistance to any nursing home that requests assistance with their performance on clinical measures. QIOs also set state- wide targets for decreasing the frequency of pressure ulcers in high-risk patients, decreasing the frequency of use of physical restraints, and helping all nursing homes set their own annual targets for these measures (and oth- ers, if desired). CMS does not define the requirements for these targets-- those are left to the discretion of each nursing home. QIOs may opt to work with a subset of nursing home providers on the documentation of specific processes of care (CMS, 2005c), including: · Skin inspection and pressure ulcer risk assessment, · Screening and treatment for depression, · Evaluation of physical restraint requirements or alternatives, and · Pain assessment and treatment. If a QIO chooses this activity, the providers must document information on 50 percent of their new admissions. For the QIO to receive credit for this
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 203 TABLE 8.1 Required Minimums of Data Transmission for Optional Statewide Work Number of Nursing Homes Minimum Number of Nursing Homes Transmitting in State or Jurisdiction Process-of-Care Data for QIO to Receive Credit Up to 30 5 31150 8 151300 15 301500 25 More than 500 40 SOURCE: CMS (2005c). activity, a minimum number of nursing homes (Table 8.1) must transmit data on a monthly basis to the QIO Data Warehouse using a Nursing Home Improvement Feedback Tool (provided free by CMS) or any other compat- ible tool. The nursing homes must submit data for at least 14 of the months between January 2006 and September 2007, but the monthly submissions need not be consecutive (CMS, 2005c). Identified Participant Groups As stated above, in the 8th SOW, QIOs are working with two identified participant groups. For both groups, QIOs administer satisfaction surveys to nursing home residents and staff (CMS, 2005c). These surveys must be completed annually; therefore, successful performance includes the comple- tion of three annual surveys of both residents and staff by at least 90 per- cent of the identified participants. Additionally, for both groups, QIOs col- lect and monitor data on the retention of certified nursing assistants and aides for at least 90 percent of the identified participants. For the first group of identified participants, QIOs work with providers to improve upon the clinical quality of care for nursing home residents. Specifically, they strive to improve upon measures related to pressure ulcers among high-risk patients, the use of physical restraints, depression manage- ment, and pain management (CMS, 2005c). For the second group of identi- fied participants, the QIOs work only on measures related to physical re- straints and pressure ulcers. New in the 8th SOW, CMS set specific criteria for recruitment for the two identified participant groups. Members of the two identified partici- pant groups may not overlap. For the first group, the selection criteria in- clude consideration of the total number of nursing homes in the state or jurisdiction (Table 8.2). CMS will ensure that the identified participants in this group are distributed across the state or jurisdiction (including rural
204 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM TABLE 8.2 Required Numbers of Participants for First Identified Participant Group in the Nursing Home Setting Number of Nursing Homes in State or Minimum Number of Maximum Number of Jurisdiction Identified Participants Identified Participants 30 All nursing homes in state or All nursing homes in state or jurisdiction (excluding providers jurisdiction (excluding providers in in second identified participant second identified participant group) group) 31300 30 total nursing homes 45 >300 10 percent of nursing homes in 10 percent of nursing homes in state state or jurisdiction or jurisdiction SOURCE: CMS (2005c). areas). Additionally, this group must have an even distribution across the state, including rural areas. Finally, QIOs must consult with the state sur- vey agency and local stakeholders for the selection of group participants. Confidentiality is maintained unless the provider chooses to reveal its participation. For recruitment of the second identified participant group, a QIO must work with its respective state survey agency to identify poorly performing nursing home providers. Although CMS requires a minimum number of participants for this group (Table 8.3), it encourages QIOs to work with as many of these providers as possible. TABLE 8.3 Minimum Numbers of Participants for Second Identified Participant Group in the Nursing Home Setting Number of Nursing Homes Minimum Number of in State or Jurisdiction Identified Participants <30 1 31300 2 More than 300 3 SOURCE: CMS (2005c).
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 205 HOME HEALTH 7th SOW Under Task 1b, the QIOs provided technical assistance to home health agencies (CMS, 2002). The QIOs directed their efforts toward the imple- mentation of processes for the continuous improvement of home health care quality by focusing on outcomes, a methodology known as the Outcome-Based Quality Improvement (OBQI) system. As a contractor to CMS, the University of Colorado Center for Health Services Research de- veloped the Outcome and Assessment Information Set (OASIS) as a pri- mary tool for the collection of outcomes data in the home health care set- ting. OASIS includes publicly reported quality measures related to the demographics, the physical and mental health, and the health care utiliza- tion of each Medicare patient receiving home health care (see Table A.5 in Appendix A). In 1999, CMS mandated that all Medicare-certified home health agencies start OASIS data collection and transmission (CMS, 2004). Quality improvement work in the home health setting started as a five- state pilot project, led by the Delmarva Foundation for Medical Care, in April 2000 to see if the QIOs could work with home health agencies on the OBQI system (CMS, 2004). In the pilot, 68 percent of the agencies in the five states agreed to be trained on the OBQI system, and 76 percent of those that were trained submitted a subsequent plan of action (CMS, 2004). Plans of action outline best practices, implementation schemes, and the specific activities to be changed or monitored. CMS declared the pilot successful and added the home health care setting to the 7th SOW. Because the home health care setting was a new provider setting for most QIOs, all QIOs had to have staff trained on OBQI techniques. QIOs then offered similar training to home health agencies in their own states. The next challenge was to get 30 percent of the trained home health agen- cies to select one or two outcome measures to improve upon and develop a plan of action for each measure. QIOs developed relationships with key stakeholders (such as state trade associations, OASIS education coordina- tors, and state survey and certification agencies), provided communications support (through listserves, teleconferences, and newsletters), and coordi- nated seminars and workshops (CMS, 2002, 2004). In the 7th SOW, QIOs trained about three-quarters of the agencies in their states and actively worked with about 55 percent of the agencies (Rollow, 2005). Box 8.2 pre- sents a story about an action plan for one home health agency that worked with Health Care Excel (Indiana's QIO) on intractable pain.
206 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM BOX 8.2 Clarifying the Definition for Intractable Pain Makes the Difference "A small hospital-based home health agency (HHA) in urban Indiana successfully used the Outcome-Based Quality Improvement (OBQI) pro- cess to improve its target outcome. After implementing a Plan of Action (POA) for Improvement in Pain Interfering with Activity, the agency achieved a rate of 32.5 percent, a statistically significant increase over its adjusted prior rate of 26.8 percent. A standard definition for intractable pain, consistent pain assessment, and appropriate interventions were important elements of the agency's quality improvement success. "The OBQI team consisted of staff from various components of the agency, including a clinical manager, direct care staff including a physi- cal therapist (PT), a home care coordinator, and a registered nurse (RN) who was a computer specialist. Their activities were regularly communi- cated to staff via staff meetings. "The OBQI team believed that the following clinical actions were criti- cal to their success: · Adopting a definition for intractable pain and having all staff consis- tently use the same definition · Ensuring that therapists ask patients to take their pain medication prior to exercising · Educating staff to perform consistent pain assessment and pain rating with appropriate interventions "The best practices implemented by the agency included: · Patient education related to correct pain medication administration for optimal benefit · Consistent pain assessment with correct rating by clinicians · When the patient is in pain, the physician is contacted, within 24 hours, for a change pain medication orders · Consistent follow-up, within 72 hours, of pain management interventions "The OBQI team used staff training and supervision as the interven- tion actions to implement the clinical best practices. The team monitored the clinical documentation for a month to ensure that the staff imple- mented the Best Practices on the Plan of Action. Further monitoring ac- tivities included the case manager making two admission supervisory visits per month for three months to compare data gathered by the admit- ting discipline. They also conducted quarterly reviews of ten charts for one year." SOURCE: Jones (2003).
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 207 8th SOW For the home health setting in the 8th SOW, as with nursing homes, CMS defined criteria for the makeup of the identified participant groups more specific than those in the 7th SOW and presented details for the re- quired activities more intricate than those in the 7th SOW (CMS, 2002, 2005c). The QIOs work with two groups of identified participants, in addition to addressing statewide performance. The membership of the groups--the Clinical Performance group and the Systems Improvement and Organizational Culture Change group--may overlap. Pediatric agencies and agencies with less than 10 episodes of care are excluded. Both at the statewide level and with the Clinical Performance group, QIOs use OBQI methods to reduce the failure rate on the measure related to hospitalization for acute care as well as additional publicly reported OASIS measures (as determined by CMS) (see Table A.3b in Appendix A) (CMS, 2005c). QIOs work with home health agencies to set targets. CMS defined intricate criteria for choosing OASIS measures and for activities related to the rate of hospitalization for acute care, based on the provider's previous performance on those measures. Statewide, the QIOs must also work with home health agencies to in- clude influenza and pneumococcal immunizations in the comprehensive patient assessment (CMS, 2005c). The QIOs use a CMS tool to survey all the home health agencies (with a required minimum response rate of 50 per- cent) to determine if immunizations and follow-up activities are included in patient assessments. The QIO is charged to reach either a 50 percent rela- tive improvement over the baseline or 80 percent performance on the inclu- sion of immunizations in the agencies' patient assessments. QIOs work with the Systems Improvement and Organizational Culture Change group to focus on telehealth and culture change (CMS, 2005c). First, the QIOs help providers implement or use some form of telehealth to reduce the rates of hospitalization for acute care. Second, the QIOs admin- ister a culture change survey related to organizational practices, including teamwork, communication, leadership, quality improvement, and patient- centeredness (CMS, 2005c). The QIOs then help the identified participants implement plans of action on the basis of the survey results. The contract for the 8th SOW delineates selection standards for each identified participant group on the basis of the number of home health agencies in the state, not including pediatric agencies (Table 8.4). Again, the members of the two groups may overlap. In addition, for the Clinical Performance group, selection must include specific levels of representation of small, medium, and large agencies (Table 8.5). The size designations are based on the numbers of episodes of care that each agency provides. Ten percent of the identified participants
208 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM TABLE 8.4 Required Numbers of Participants for Identified Participant Groups in the Home Health Setting Number of Home Number for Clinical Number for System Improvement and Health Agencies Performance Group Organizational Culture Change Group <14 6 2 1525 8 3 2645 10 4 4665 14 5 6690 16 6 >91 20 percent of all agencies 8 percent of all agencies SOURCE: CMS (2005c). TABLE 8.5 Required Sizes of Agencies in the Clinical Performance Group Small agencies <90 episodes of care annually Medium agencies 91350 episodes of care annually Large agencies >351 episodes of care annually SOURCE: CMS (2005c). must come from small agencies, 10 percent must come from medium agen- cies, and 15 percent must come from large agencies (CMS, 2005c). The rest of the participants may be chosen without regard to size. No similar re- quirement exists for the Systems Improvement and Organizational Culture Change group. The QIOs may choose up to eight additional agencies to work with on these group projects (or 8 percent of the total number of agencies for states with more than 100 agencies). These agencies may act as a substitute for one of the identified participants in the evaluation process if one of the original participants goes out of business or changes ownership. HOSPITALS 7th SOW The provision of QIO technical support to hospitals began in the 4th SOW (1993 to 1996). Thus, the QIOs may have long-standing relation- ships with hospitals and related stakeholders in their states. In the 7th SOW, the QIOs had to show only statewide improvement, as their evaluations did not involve an identified participant group. The QIOs interacted with acute care and critical access hospitals. Interventions were designed to prevent
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 209 surgical infections and reduce systems failures for hospitalized patients with acute myocardial infarction, heart failure, and pneumonia. The QIOs used measures derived from preexisting clinical guidelines and scientific evidence (see Table A.3c in Appendix A). CMS chose hospital topics on the basis of the incidence of hospitalization, the rates of morbidity and mortality, and annual payments related to those diseases (CMS, 2004). Another factor was evidence showing a link between care practices and improved outcomes. In the 7th SOW, of the approximately 6,000 hospitals across the country, QIOs worked with approximately 2,400 hospitals on cardiac care, 2,000 hospitals on pneumonia, and 1,500 to 2,000 hospitals on surgical infection prevention (Rollow, 2005). As in other provider settings, the QIOs helped providers implement quality improvement plans, provided written materials and guidelines, and gave individualized feedback. They provided much technical assistance to help hospitals collect and report data (CMS, 2004). Additionally, the QIOs facilitated collaboratives for hospitals working on the same performance measures. One example of a national collaborative is the National Surgical Infection Prevention Collaborative. Sponsored by CMS in 2003, this pro- gram allowed each QIO to select one or two motivated hospitals in the state to work on surgical site infection reduction. During the 13-month collabo- rative, the participating providers reduced the incidence of surgical site in- fections by 27 percent (CMS, 2004). 8th SOW In the 8th SOW, QIO work in the hospital setting continues in much the same manner that it did in the 7th SOW, but the work has been divided into two subtasks: (1) all prospective payment system hospitals (Task 1c1) and (2) critical access and rural prospective payment system hospitals (Task 1c2) (CMS, 2005c). 8th SOW: Hospitals (Task 1c1) For Task 1c1, CMS defined four strategies that can be used to improve quality of care in the hospital setting: improving performance on clinical performance measures, increasing clinical performance measurement and reporting, process improvement, and systems improvement and organiza- tional culture change (CMS, 2005c). As in the 7th SOW, the QIOs must demonstrate improvement statewide. However, under the contract for the 8th SOW, Task 1c1 requires QIOs to also work with three groups of iden- tified participants related to the defined strategies: the Appropriate Care Measure group, which focuses on clinical performance measurement; the Surgical Care Improvement Project group, which focuses on process im-
210 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM provement; and the Systems Improvement and Organizational Culture Change group (CMS, 2005c). Statewide Performance Statewide, the QIOs help prospective payment system hospitals report on the expanded measures set, which includes measures required by the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 (P.L. 108-173), to receive the Annual Payment Update plus other measures collected by the QIOs in the 7th SOW (CMS, 2005c) (see Table A.3c in Appendix A). Reporting on the additional measures remains voluntary for hospitals; but to achieve success on this subtask, the QIOs must demon- strate that 25 percent of the hospitals statewide are reporting on this ex- panded set. The QIOs are available to all prospective payment system hos- pitals for assistance with data collection and validation. Additionally, the QIOs assist both prospective payment system hospitals and critical access hospitals to improve the validity, timeliness, and completeness of the data that they submit to the QIO Clinical Data Warehouse. Finally, the QIOs also work with hospitals statewide to improve upon the appropriate care measure and measures related to the Surgical Care Improvement Program (CMS, 2005c). Identified Participants The identified participants in these groups should have broad represen- tation by size, geography, and performance level. Each group must include 15 percent of all prospective payment system hospitals and stay within a range of 6 to 36 participants (CMS, 2005c). Exceptions exist for states or jurisdictions with an inadequate number of hospitals. Additionally, Public Health Service hospitals and hospitals owned by Indian tribes may also be included under specific conditions. All identified participants in all groups (except for critical access hospitals of the Surgical Care Improvement Project group) must report on the measures required by the Medicare Prescription Drug, Improvement, and Modernization Act of 2003. The members of the three identified participant groups may overlap. Appropriate Care Measure Identified Participant Group The QIOs work with the Appropriate Care Measure group to improve performance on the clinical measures required by the Medicare Prescription Drug, Improve- ment, and Modernization Act of 2003 that address acute myocardial in- farction, heart failure, and pneumonia care in the hospital setting (see Table A.3c in Appendix A). Identified participants must report on these measures to the QIO Clinical Data Warehouse. The appropriate care mea-
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 211 sure is a composite score based on how often a patient receives the care specified by all the measures. Participants must represent both hospitals that cover the distribution of urban and rural prospective payment system hospitals2 and hospitals that cover the range of performance at the baseline on the appropriate care measure in the state (CMS, 2005c). The QIOs may do extra-credit work on this task. (See Chapter 10 for discussion of evalua- tion of QIO performance.) Surgical Care Improvement Project Identified Participant Group The QIOs assist the Surgical Care Improvement Project group to standardize processes for the following conditions (see Table A.3c for measures related to some of these): · Surgical site infections, · Venous thromboembolism, · Ventilator-associated pneumonia, · Cardiovascular complications, and · Fistula use in hemodialysis (vascular access). The QIOs assist hospitals with the collection of related measures for all these topic areas, but CMS will evaluate each QIO only on a subset of the measures for its contract performance evaluation. The QIOs work in con- junction with the American College of Surgeons on these activities. The work in this group is part of the project's larger national effort, and thus, the QIO must coordinate activities with the local chapter of the American College of Surgeons (if a chapter is present in the state). Hospitals must demonstrate a caseload of at least 300 annual surgical procedures to par- ticipate in this group, and critical access hospitals may count toward the required 15 percent recruitment. Additionally, the QIOs will survey End- Stage Renal Disease Networks to evaluate their satisfaction with QIO assis- tance in the Fistula First program3 (CMS, 2005c). Extra-credit work is also available. 2The definition of "urban" comes from the U.S. Census Bureau's determination of Metro- politan Statistical Areas as areas with a single city of at least 50,000 residents and at least 100,000 total residents (or 75,000 total residents in New England). All other areas are consid- ered rural. 3The Fistula First program is a national effort supported by CMS, the 18 End-Stage Renal Disease Networks, the Institute for Healthcare Improvement, and other key stakeholders to promote arteriovenous fistula use in hemodialysis.
212 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM Systems Improvement and Organizational Culture Change Identified Par- ticipant Group Finally, through the Systems Improvement and Organiza- tional Culture Change group, the QIOs engage the senior leadership of hospitals to consider the adoption of health information and communica- tion technology, such as computerized provider order entry, bar-coding, and telehealth technologies. The QIOs act to advise, strategize, and ulti- mately, help hospital leadership implement plans for the use of any or all of these technologies (CMS, 2005c). 8th SOW: Critical Access and Rural Hospitals (Task 1c2) In the 8th SOW, CMS designated a subtask for the QIOs to specifically interact with both critical access and rural prospective payment system hos- pitals4 (CMS, 2005c). This is a significant change from the 7th SOW when QIOs could choose to work with rural providers under Task 1e, but were not required or incentivized to work with this specific population. Under Task 1c2, the QIOs work both at the statewide level and with an identified participant group. The QIOs in states with less than two critical access hospitals are exempt from this task. Additionally, a QIO's state must have at least six critical access or rural prospective payment system hospitals for the QIO to work with an identified participant group. If this is not the case, the QIO must get approval from the Project Officer and Government Task Leader to perform this task. Statewide, the QIOs work with all critical access hospitals that are re- porting on Hospital Quality Alliance measures to improve performance through process redesign for at least one measure (chosen by the hospital and the QIO). QIOs also help nonreporting critical access hospitals start reporting data to the QIO Clinical Data Warehouse on at least one topic for two consecutive quarters. QIOs may perform extra-credit work related to new acute myocardial infarction transfer measures or emergency depart- ment transfer measures, or both, if they become available during the 8th SOW. QIOs also work with a Rural Organizational Safety Culture Change identified participant group. Participants may include both critical access and rural prospective payment system hospitals, with a minimum of six hospitals participating (unless a different minimum number is otherwise approved). CMS expects the QIOs to use a Rural Organizational Safety Culture Change toolkit to work with senior leaders to determine the culture of safety, including the use of the Hospital Survey on Patient Safety Culture 4For the purposes of this task, a "rural" hospital is one that is not in a Metropolitan Statis- tical Area (as defined earlier in this chapter).
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 213 (developed by the Agency for Healthcare Research and Quality). The QIOs will then help these providers individually to analyze the survey results. Extra-credit work is available by recruiting at least one critical access hospi- tal to work on certain health information technology activities. PHYSICIANS' OFFICES AND PRACTICES 7th SOW Under Task 1d of the 7th SOW, QIOs expanded upon work begun in the 6th SOW with the implementation of quality improvement projects fo- cused on the physician's office. In the 7th SOW, the QIOs were required to work with at least 5 percent of the physicians in the state as identified par- ticipants. On average, the QIOs actually worked with about 7.5 percent of the eligible practitioners (Rollow, 2005). Specific topic areas were care for chronic disease (diabetes) and preventive services, including cancer screen- ing (mammography) and adult immunizations (see Table A.3d in Appendix A) (CMS, 2002). Box 8.3 gives an example of how the Oklahoma Foundation for Medi- cal Quality (OFMQ; Oklahoma's QIO) worked with a single physician's office to improve the rates of mammography. In this story, the Clinton Medical Clinic in Oklahoma, headed by Sharad Swami, worked with Oklahoma's QIO to increase mammography rates by 40 percent. QIOs also supported collaboratives for Quality Assessment and Perfor- mance Improvement projects required of Medicare+Choice organizations (see the discussion of managed care later in this chapter) (CMS, 2002). The work performed in Quality Assessment and Performance Improvement projects could also fulfill the requirements for working with underserved populations (see the discussion of underserved populations later in this chap- ter). QIOs received individual state-level analyses of data, including county- specific and provider-level information from the Outpatient Data QIOSC, operated by the Iowa Foundation for Medical Care (Iowa's QIO). Data were not provided at the practitioner level (CMS, 2002). 8th SOW In the 8th SOW, CMS has divided efforts with physicians' practices into three distinct subtasks: · Physician practices (Task 1d1), · Physician practices: underserved populations (Task 1d2), and · Physician practice and pharmacy: Part D prescription drug benefit (Task 1d3).
214 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM BOX 8.3 Improvement Story: Mammography Scheduling Block Improves Rates "Dr. Swami found that patients were not scheduling mammograms on their own, even with his recommendation to do so. So, his staff called the local hospital and asked for a block of time in which they could sched- ule his patients. The hospital radiology center was very agreeable. This program works well not just for Dr. Swami and his patients, but also for the radiology center where patients receive mammograms. They do not have to worry about scheduling these time slots since the medical center is doing it for them. "In the physician office, the program works well because everyone is responsible for making sure that if a patient needs a mammogram, it is scheduled. Before the patient even sees the doctor, she is asked about when her last mammogram was. If it has been more than one year, one is immediately scheduled for her. "Intervention used: Reserved mammography time slots at the hos- pital, made appointment for appropriate patients before they left the of- fice, made reminder calls to patients the day before mammogram ap- pointment. "Barriers: Manual reminders to patient don't work, uninformed pa- tients are less compliant. "Strategies to Overcome Barriers: Worked with OFMQ to imple- ment changes and find what works, educated staff on processes and rationale. "Lessons learned: This intervention is reproducible in other settings--for example it could be used for eye exams for diabetic pa- tients or to schedule a patient's lab work. Educating all staff on processes is key." SOURCE: Oklahoma Foundation for Medical Quality (2005). These subtasks combine Tasks 1d and 1e of the 7th SOW, as well as add activities related to the Part D prescription drug benefit under Medi- care. The term "physician office" was changed to "physician practice" to incorporate multiple types of settings, including offices with single practi- tioners as well as practices with multiple physicians at multiple sites (CMS, 2005c). As with other settings, the evaluation formulas and the criteria for the identified participant groups are more detailed and complex than those in the 7th SOW.
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 215 8th SOW: Physician Practices (Task 1d1) Under Task 1d1, the QIOs work statewide as well as with an identified participant group. The statewide work focuses on the promotion of quality initiatives, whereas the work with the identified participants focuses on the reliability of preventive care delivery and the effective management of chronic conditions. Additionally, the identified participants work on im- proving clinical performance through the use of health information and communications technologies and process redesign. Statewide Performance Statewide, the QIOs work with physicians' offices to improve upon clinical performance measures through the support of ini- tiatives such as the Physician Voluntary Reporting Program5 (CMS, 2005c). The QIOs also coordinate with state agencies working on process improve- ment, such as the Welcome to Medicare Visit (CMS, 2005c). As in the 7th SOW, Medicare Advantage organizations (previously known as Med- icare+Choice organizations) must be included in statewide activities, and the QIOs must work with these organizations on their quality projects. In this task, the QIOs may also work with End-Stage Renal Disease Networks (upon their request) to help physicians' practices improve their rates of fis- tula use and immunization. Finally, QIOs must collaborate with the Medi- care Management Demonstration Project as required by the Medicare Pre- scription Drug, Improvement, and Modernization Act of 2003 (CMS, 2005c). Identified Participants The QIOs assist identified participants with the use of electronic clinical information, the design of care processes for preven- tive care and chronic conditions (including self-management), and report- ing of and improvement upon quality measures (CMS, 2005c). This area of assistance began as a four-state pilot project in the 7th SOW, known as the Doctor's Office QualityInformation Technology program, to recruit doc- tors' offices to adopt electronic health records (iHealthBeat, 2005). To par- ticipate in this group, a physician's practice must complete a readiness as- sessment form indicating its request for assistance, and the QIO must accept the form. Participants may be at different stages of technology adoption, but no more than 25 percent of the identified participants can already have a full electronic health record system in place (with some exceptions). The 5Under this CMS program, physicians will voluntarily report on quality data, receive feed- back on their performance, and suggest improvements to streamline reporting requirements. As of this writing, CMS plans a January 2006 launch.
216 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM identified participant group must include, at a minimum, 5 percent of the physicians' practices in that state or jurisdiction. Additionally, CMS re- quires the group to be divided as follows: · At least 40 percent of the participants must come from small prac- tices (one to three physicians), · No more than 60 percent of the participants may come from me- dium practices (four to eight physicians), and · No more than 20 percent of the participants may come from large practices (nine or more physicians). QIOs will continue to recruit participants throughout the 8th SOW. All identified participants must complete office systems surveys at the baseline and undergo a remeasurement. To achieve success, the QIOs must help the identified participants achieve improvements in reporting, implementation of new care processes, and adoption of clinical information systems. 8th SOW: Physician Practices: Underserved Populations (Task 1d2) In the 8th SOW, CMS incorporated Task 1e of the 7th SOW (under- served and rural populations) into the physician practice setting task. Under Task 1d2, the QIOs work statewide as well as with two identified partici- pant groups whose members may overlap. Statewide, QIOs work to im- prove performance on measures of diabetes care, mammography, and adult immunizations for one of the following populations (CMS, 2005c): · African Americans, · Asians-Pacific Islanders, · American Indians-Alaska Natives, and · Hispanics and Latinos. Unlike the 7th SOW, rural and dual-eligible populations do not qualify for this task (see the discussion of underserved and rural populations later in this chapter). To work on this task, at least 3 percent of the state's benefi- ciary population must fall into one of the categories listed above. The QIOs in the U.S. Virgin Islands and Puerto Rico are automatically excluded from this task. QIOs work with a Task 1d1 underserved identified participant group on the same activities described for Task 1d1. The practices in this group must proportionately represent the underserved population in that state or jurisdiction. With the second group, QIOs promote Culturally and Linguis- tically Appropriate Services (CLAS) standards and cultural competency. This group is known as the CLAS/Cultural Competency identified partici-
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 217 pant group. For this group, the QIOs promote the adoption of cultural standards and requirements at both the practice level and the practitioner level using a tool from the Office of Minority Health of the U.S. Depart- ment of Health and Human Services and self-assessments. This group must include 5 percent of the total number of primary care practices, within boundaries of between 20 and 50 participants. At the practitioner level, the group must include at least 10 percent of the practitioners from participat- ing practices, within an accepted range of 20 to 100 practitioners (CMS, 2005c). 8th SOW: Physician Practice and Pharmacy: Medicare Part D Prescription Drug Benefit (Task 1d3) As a result of enactment of the Medicare Prescription Drug, Improve- ment, and Modernization Act of 2003, CMS added Part D (the Medicare prescription drug benefit) as a new topic area for the QIOs in the 8th SOW. CMS plans to work with the QIOs on developing methods for improving the dissemination of information and the implementation of registries. The QIOs will work with identified participants (physicians' practices or phar- macies) to improve safety in prescription delivery. Their services may range from providing information to physicians to modify their practices to help- ing with the implementation of electronic prescribing systems. The QIOs will partner with prescription drug plans on this task, including Medicare Advantage prescription drug plans. Section 109(b) of the Medicare Pre- scription Drug, Improvement, and Modernization Act authorizes QIOs to offer assistance regarding improving the quality of prescription drug therapy for all Medicare providers, Medicare Advantage organizations offering pre- scription drug plans under Part C, and organizations offering prescription drug plans under Part D (CMS, 2005c). The Part D benefit is scheduled to start on January 1, 2006, and as of this writing, the QIOs will begin quality improvement projects the following August. These projects will include baseline assessments of performance, implementation of an intervention, identification of targets, and follow-up assessments. CMS will identify ap- propriate measures for this task on the basis of evidence-based guidelines and collaborations with multiple partners. The contract for the 8th SOW outlines four preliminary options for QIO activities in this task. The QIO must select two options--either Op- tion 1 or Option 2 and either Option 3 or Option 4 (see below)--for which the QIO will submit concept papers to CMS (CMS, 2005c). If CMS deter- mines that the concept papers from all QIOs are varied enough, the QIOs will then submit project proposals. The QIOs will partner with prescription drug plans for all of these activities.
218 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM Option 1: Improve prescribing using Part D data This option focuses on electronic prescribing. The measures will likely address drugs whose use by elderly populations should be avoided, drug interactions, the use of generic drugs, preferred drugs, and polypharmacy. Option 2: Improving patient self-management through medication therapy management services This option focuses on the delivery side of drug therapy. Measures will likely address process measures (such as the identifi- cation of appropriate patients) and outcomes measures (such as those re- lated to patient experience and rehospitalization rates). As of this writing, the QIO will have to recruit a group of participants representing at least 5 percent of the total number of pharmacies working with a prescription drug plan. Option 3: Improving disease-specific therapy using integrated Medicare Part A, B, and D data This option focuses on physicians' practices that use electronic health records or electronic prescribing tools. It requires work- ing with the identified participants for Task 1d1 who are using these tech- nologies, as well as with others who are using the technologies but who are not working in the Task 1d1 identified participant group. Measures will likely address drugdisease interactions and therapeutic monitoring. Option 4: QIO-directed project This option requires approval of the Project Officer and the Task 1d3 Government Task Leader. UNDERSERVED AND RURAL BENEFICIARIES 7th SOW Under Task 1e of the 7th SOW, each QIO worked to reduce a health disparity in its state. Work with underserved populations began in the 6th SOW, and in the 7th SOW, the QIOs could continue the same project or start a new one (CMS, 2002, 2004). The disparity had to exist between a medically underserved population and a reference group from the general population of all Medicare beneficiaries. The following is a list of the ac- ceptable populations for this subtask during the 7th SOW (the numbers of QIOs that chose each of the populations to work with are given in paren- theses) (CMS, 2004): · African Americans (22), · American Indians-Alaska Natives (3), · Asians-Pacific Islanders (1), · Hispanics (4),
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 219 · Dually enrolled (3), and · Rural beneficiaries (19). The reference group for nonrural projects was white, non-dually eligible Medicare beneficiaries residing or receiving care in the same geographic area as the underserved population. The reference group for rural projects was Medicare beneficiaries residing or receiving care in all urban counties of the state. CMS granted a waiver to the QIO for the U.S. Virgin Islands because the entire population was considered "underserved" (CMS, 2002, 2004). The QIOs targeted a subset of the underserved population in their states, called the intervention group, which had to be at least 25 percent of the underserved population's entire size. The QIO addressed one quality measure used in either Task 1c or Task 1d for its underserved population. If the QIO was continuing a project from the 6th SOW, the QIO was encour- aged to increase the size of its intervention group. In this subtask, the QIOs also supported state Quality Assessment and Performance Improvement projects to reduce health disparities (see the discussion of managed care later in this chapter). The story presented in Box 8.4 demonstrates a successful intervention by Florida Medical Quality Assurance, Inc. (Florida's QIO), which used multiple techniques, including beneficiary education, the use of opinion leaders, partnering with key stakeholders, communications tools, use of QIOSC materials, and individualized assistance to providers to improve the rate of hemoglobin A1c testing in the African-American population (CMS, 2002, 2004). Telephone Interviews: Challenges In telephone interviews, many QIO CEOs mentioned difficulties with the design of interventions for the underserved population, including access versus quality; the resource-poor state of some providers; and the ability to track changes in populations whose providers bundled charges for tests into visits, such as the Indian Health Service. They expressed concern that access issues are not addressed by the QIOs as much because there is a tendency to look at the care of people who already have access. They also related logis- tical difficulties with the study of underserved populations, including suffi- cient sample sizes, the increased use of services by all populations, and spe- cial geographical needs. The following are comments of some of the CEOs: · "I don't think the issue is the quality of care delivered in the area, but it is the issue of access instead. For example, people in rural areas or even midcity often do not have transportation to care."
220 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM BOX 8.4 Glycosylated Hemoglobin A1c Testing Among African Americans in Florida "The Florida Medical Quality Assurance, Inc. (FMQAI) sought to re- duce the disparity in glycosylated hemoglobin A1c (HbA1c) use between African American Medicare beneficiaries and non-African American Medi- care beneficiaries with diabetes in the state. "FMQAI used previous statement of work (scope of work) research to identify barriers, such as beneficiary knowledge regarding the impor- tance of healthcare screening techniques for diabetes. FMQAI also found a need for providers to learn about the healthcare disparity, existent bar- riers to care, and providers systemic changes that could address pre- scribing HbA1c for African Americans such as diabetic beneficiary identi- fication and follow up by the provider. "The QIO identified existing networks, used them to enhance com- munication, and employed African Americans from the community to lead the project team and build partnerships with community champions. Out- reach tools were designed with target audience involvement and included mailings, radio public service announcements, and press releases. The QIO also worked with stakeholders to develop a culturally sensitive edu- cation program. FMQAI made presentations at public meetings with local health departments and church groups and developed an Internet web- site. Intervention tools included key chains inscribed with the phrase `Dia- · "In a rural area there might only be five or six Medicare admissions a month, so it is hard to break down rural facilities on an individual basis; we need 10 to 12 facilities to have adequate data." · "Trying to meet the needs of lower-scoring rural hospitals really adds to the QIOs' costs not only because of their needs but because of the dis- tance required to go to serve their needs." 8th SOW In the 8th SOW, CMS integrated efforts to take a more community- based approach to improve beneficiary health by incorporating underserved populations into Tasks 1a to 1d, by requiring the adequate representation of providers to underserved populations and, in many cases, in the selection of identified participants. Underserved populations are specifically ad- dressed in Task 1d2 (CMS, 2005b).
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 221 betes HbA1c <7,' a train-the-trainer module, and a diabetes education module for beneficiaries. Another beneficiary intervention involved face- to-face interaction and educational presentations, seminars, and involve- ment at trade shows. FMQAI implemented the Front Porch initiative, which involved a `Closing the Gap' grant, the SHARE program, the Frenchtown initiative, and diabetic educators. "The QIO provider intervention targeted physicians' offices and in- cluded an adaptation of the Oklahoma Foundation for Medical Quality, Inc. methodology. FMQAI visited `best practice' offices and recruited those physicians to become FMQAI consultants, who conducted other provider office visits. The QIO also made direct mailings, which contained recruitment materials and a `project-in-a-box' with tools for system changes. Other efforts involved physician-to-physician mailings, QIO partner articles, presentations, and face-to-face contact with providers by teleconferencing. Diabetic educators and pharmacists also assisted the QIO in working with physicians. "The HbA1c-testing rate for the target population showed an abso- lute improvement of 14.6 percent. There was a reduction in the disparity of HbA1c use between African American Medicare beneficiaries and non- African American Medicare beneficiaries with diabetes of 3.1 percent." SOURCE: QSource: The Center for Healthcare Quality (2005). MANAGED CARE By law, the QIOs must also review the services provided to beneficia- ries in managed care plans. Independent of the QIO program, all Medicare managed care organizations must execute one national Quality Assessment and Performance Improvement (QAPI) project to improve health outcomes and beneficiary satisfaction. CMS chooses the clinical topic for the national project each year. Past topics included diabetes, community-acquired pneu- monia, congestive heart failure, and clinical health care disparities or cul- turally and linguistically appropriate services (CMS, 2005a). In 2005, CMS did not assign a specific topic because of the overwhelming work that re- sulted from implementation of the Medicare Prescription Drug, Improve- ment, and Modernization Act (P.L. 108-173). Instead, organizations per- formed a task of their choosing based on local needs (Moreno, 2004). In years with an identified clinical topic, the organizations could initiate addi- tional projects on topics in response to local needs.
222 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM Under Task 1f of the 7th SOW, the QIOs supported Medicare+Choice organizations' (M+COs') performance of QAPI projects. The QIOs encour- aged consistent practice patterns for beneficiaries, regardless of the type of plan in which they were enrolled. CMS required the QIOs to invite all Medicare+Choice organizations in the state to participate in any projects related to Tasks 1a to 1e. If possible, a QIO was supposed to offer technical assistance to Medicare+Choice organizations for any quality improvement activities not specifically related to QAPI project requirements. This work continued collaborations initiated during the 6th SOW. No specific set of measures existed for this task, but the QIOs reported on their activities to CMS on a quarterly basis (CMS, 2002). Box 8.5 gives an example of how BOX 8.5 Cultural Competency Organizational Assessment (CCOA) Pilot "CMRI (now known as Lumetra) sponsored a pilot project utilizing the Organizational Self-Assessment tool and protocol developed by Dennis Andrulis, PhD, MPH, a research professor at the department of Preventive Medicine & Community Health at the State University of New York, Downstate Medical Center. "In consultation with Dr. Andrulis, the original self-assessment tool was modified to meet the needs of the managed care community. Five California M+COs participated in the pilot project. The participating plans contributed to the adaptation of the tool and shared their experiences with other M+COs at a meeting in December 2002 in Oakland, California. "The M+COs who participated in the CCOA Pilot have been given permission by CMS to use that experience as the basis for their 2003 QAPI Project. More importantly, the participants contributed to the devel- opment of an organizational self-assessment tool that can be used by all types of managed care organizations nationwide. Dr. Andrulis is consult- ing with each participant plan confidentially about their scoring results and `next steps.' Although all participants note that the self-assessment took a lot of hard work, they enthusiastically acknowledge how valuable the experience has been to their organization. They report increased awareness of organizational resources and improved inter-departmental communications. The CCOA pilot results help their organizations to develop a multi-year action plan to address issues related to cultural competency." SOURCE: Lumetra (2005).
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 223 Lumetra (California's QIO) interacted with managed care plans in that state on its national project for 2003. In the 8th SOW, CMS did not define a separate task for the Medicare Advantage beneficiary population, but the QIOs must include Medi- care Advantage organizations in all of their activities at a level that is equiva- lent to their representation in the state. QIO SUPPORT CENTERS As described in Chapter 7, a QIOSC is a QIO that acts as a central resource on a specific task or crosscutting topic. QIOSCs conduct analyses, develop materials, and share information. In interviews with four QIOs (representing seven QIOSCs), all stated that they had various levels of inter- action with the QIOs. They provided information when they were asked, but they could not intervene in a QIO's activities unless they were asked to do so. They agreed that individual QIOs needed different levels of help, depending on their own skills. All QIOSCs saw their role as assistants to QIOs in their activities and as sources for the sharing of knowledge. The following sections present specific examples of QIOSC activities related to the role of QIOs in offering technical assistance to providers. Nursing Home QIOSC During the 7th SOW, Quality Partners of Rhode Island (Rhode Island's QIO) served as the Nursing Home QIOSC. In addition to providing general support, it provided technical information and reports, training, and imple- mentation materials to the QIOs. The QIOSC served a convening function for QIOs to communicate among themselves through the establishment of a community of practice, a group that comprised staff working on this spe- cific task in each QIO. Communities of practice regularly engaged in roundtables by telephone and communicated through a listserve. Addition- ally, the QIOSC helped develop and maintain a Nursing Home Information Clearinghouse, an Internet-based database of best practices, change con- cepts, interventions, and guidelines available to the QIOs and nursing homes. The data included findings from the literature, as well as the ex- periences of QIOs and nursing homes (CMS, 2004). Quality Partners of Rhode Island continues these activities as the Nursing Home QIOSC in the 8th SOW. Box 8.6 represents part of a document developed by the Nursing Home QIOSC in the 7th SOW to serve as a resource guide for the QIOs when they are working on delirium.
224 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM BOX 8.6 Delirium: Tools and Web Links Resources for Creating Your Own Delirium Relief Resource Manual "Assessment: 10-Point Clock Test Screens http://www.psychiatrictimes.com/p981049.html A method of using clock-drawing tests to identify delirium. This tool is accompanied by an article, which describes the administration of the 10-Point Clock Test, as well as the interpretation of the score for cogni- tive impairment. Source: Psychiatric Times Confusion Assessment Method http://www.hartfordign.org/publications/trythis/issue13.pdf The Confusion Assessment tool has two parts. Part one is an as- sessment instrument that screens for overall cognitive impairment. Part two includes only those four features that were found to have the great- est ability to distinguish delirium from other types of cognitive impairment. Source: The Hartford Institute for Geriatric Nursing [adapted from Inouye, S., van Dyck, C., Alessi, C., Balkin, S., Siegal, A. & Horwitz, R. (1990). Clarifying confusion: the confusion assessment method. Annals of Internal Medicine, 113(12), 941948.] Mini-Mental State Examination (MMSE) http://www.minimental.com Home Health QIOSC In the 7th SOW, the Maryland-based QIO of the Delmarva Foundation for Medical Care served as the Home Health QIOSC as a result of a pilot study that it led during the 6th SOW. As with the Nursing Home QIOSC, the Home Health QIOSC offered general assistance as well as technical information, reports, and implementation materials to the QIOs. Because this was a new setting for QIOs, the QIOSC worked to orient the QIOs to the home health setting in general. They also provided OBQI system train- the-trainer programs to all the QIOs. The 3-day training included lectures on the state of the home health industry and OASIS (CMS, 2004). The Home Health QIOSC operated in a fashion similar to that described above for the Nursing Home QIOSC and maintained an information clearing-
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 225 A practical method for grading the cognitive state of patients for the clinician. Source: Marshal F. Folstein, MD; Susan E. Folstein, MD; Paul R. McHugh, MD Guidelines: Acute confusion and delirium http://www.guideline.gov/VIEWS/summary.asp?guideline=000536 &summary_type=brief_summary&view=brief_summary&sSearch_string =delirium Major recommendations including the assessment and management of delirium. Source: Research Dissemination Core. Acute confusion/delirium. Iowa City (IA): University of Iowa Gerontological Nursing Interventions Research Center; 1998. 41 p. [81 references] Practice Guideline For the Treatment of Patients with Delirium http://www.psych.org/clin_res/pg_delirium.cfm This practice guideline seeks to summarize data regarding the care of patients with delirium. It begins at the point where the psychiatrist has diagnosed a patient as suffering from delirium according to the DSM-IV criteria for the disorder. The purpose of this guideline is to assist the psychiatrist in caring for a patient with delirium. Source: American Psychiatric Association" SOURCE: Quality Partners of Rhode Island (2005). house. In the 8th SOW, the West Virginia Medical Institute (West Virginia's QIO) acts as the Home Health QIOSC. Hospital QIOSCs In the 7th SOW, the Colorado Foundation for Medical Care (Colo- rado's QIO) served as the Heart Failure QIOSC and focused on hospital- based measures of acute myocardial infarction and heart failure, whereas the Oklahoma Foundation for Medical Quality (Oklahoma's QIO) served as the Infectious Disease QIOSC and supported hospital tasks related to pneumonia and the prevention of surgical infections. These QIOSCs oper- ated in the same fashion described above for the other QIOSCs. For ex-
226 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM ample, the Infectious Disease QIOSC brought together representatives of the Infectious Diseases Society of America and the American Thoracic Soci- ety to develop joint guidelines for the treatment of community-acquired pneumonia. Examples of fact sheets include Antibiotic Timing and Selec- tion (developed by the Infectious Disease QIOSC) and Successful Inpatient Intervention Factors (developed by the Heart Failure QIOSC) (CMS, 2004). In the 8th SOW, the Iowa Foundation for Medical Care (Iowa's QIO) acts as the newly designed Hospital Reporting QIOSC. This QIOSC will provide data support for hospital reporting initiatives, help with the CMS Abstraction and Reporting Tool (see Chapter 13), and support validation for hospital-generated data and will perform overall data management (Qualis Health, 2005). The Oklahoma Foundation for Medical Quality (Oklahoma's QIO) operates the newly designed Hospital Interventions QIOSC to provide support for all Task 1c1 and Task 1c2 activities. Physician Office QIOSC In the 7th SOW, the Virginia Health Quality Center (Virginia's QIO) acted as the Physician Office QIOSC and thereby supported the QIOs in the same manner described above. Monthly calls included discussions of na- tional topics, as well as topics designed to target smaller groups of QIOs with specific demographic challenges. The QIOSC also supported the Medi- care Quality Improvement Community website through coordination with the Interventions QIOSC (CMS, 2004). The Virginia Health Quality Center acts as the Physician Office QIOSC in the 8th SOW. Additionally, it has three subcontractors: (1) Lumetra (California's QIO) for Medicare Advan- tage, (2) Lumetra (under a second subcontract) for Electronic Health Record Vendor Relations Updates, and (3) the Northeast Health Care Quality Foundation (New Hampshire's QIO) for Office System Survey (Qualis Health, 2005). Underserved QIOSC In the 7th SOW, the Center for Healthcare Quality (Tennessee's QIO) served as the Underserved/Rural QIOSC. This QIOSC provided support to the QIOs in a manner similar to that described above and collected a large scientific evidence base on disparities in health care quality. Specific efforts included participation in the Healthy People 2010 Partnership for Heart Disease and Stroke and the Southeast Health Disparities Collaborative. The Underserved/Rural QIOSC also conducted a needs assessment in January 2003 (5 months after the first round of the 7th SOW started) in which it surveyed the QIO community on how the QIOSC could best serve its needs. The survey found that QIOs had a strong preference to learn from other
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 227 QIOs and other experts. As a result of this and other findings, the QIOSC held conference calls on intervention strategies, convened a 1-day confer- ence on rural health for hospitals and outpatient rural projects, led web- based training sessions, and provided written educational materials (CMS, 2004). The Center for Healthcare Quality continues its activities as the Underserved QIOSC in the 8th SOW. Medicare Advantage QIOSC In the 7th SOW, Lumetra (California's QIO) acted as the Medicare +Choice QIOSC. The QIOSC provided support in a way similar to that described above for other QIOSCs. In the 8th SOW, no QIOSC is dedicated solely to Medicare Advantage issues. Instead, Lumetra acts as a supporting contractor to the Physician Office QIOSC for Medicare Advantage issues. SUMMARY This chapter has discussed issues related to the technical assistance ac- tivities of the QIO program. The following are some of the main themes of this chapter, which are reflected in the finding and conclusions presented in Chapter 2: · The activities involved under the broad term of technical assistance vary widely and include the implementation of interventions, the provision of support with public reporting, the provision of assistance with data col- lection and manipulation, and collaboration with stakeholders. · Recruitment of voluntary identified participants is largely left to the discretion of the QIO (aside from certain specific numeric or demographic requirements). The QIOs largely favor working with those showing an ea- gerness and readiness for change. · The QIOs have experience with many methods for interacting with providers, including collaboratives, one-on-one consulting, teleconferences, local or regional conferences, newsletters and other printed materials, and web-based tools. · The QIOs favor an increased ability to tailor interventions to local needs. · Because of the history of the QIO program, QIOs have the longest- standing relationships with hospitals. For some providers, particularly phy- sicians in ambulatory care, their interaction with QIOs has occurred over a much shorter length of time, and so many of those relationships are not fully developed. · The 8th SOW has many more detailed requirements for technical assistance activities than the 7th SOW did. The identified participant groups
228 MEDICARE'S QUALITY IMPROVEMENT ORGANIZATION PROGRAM are intricately described, including how they may be chosen, and more com- plex evaluation criteria are used to determine the QIOs' success on these tasks. The 8th SOW also includes many new challenging areas to be ad- dressed, such as the Part D prescription drug benefit and a program to work with physicians' offices on the adoption of information and communication technology. · Task-specific QIOSCs exist for each of the provider settings. These QIOSCs are very active in producing materials, disseminating information, and otherwise supporting the QIOs in their technical assistance activities. REFERENCES Bradley EH, Carlson MDA, Gallo WT, Scinto J, Campbell MK, Krumholz HM. 2005. From adversary to partner: Have quality improvement organizations made the transition? Health Services Research 40(2):459476. CMS (Centers for Medicare and Medicaid Services). 1999. 6th Statement of Work (SOW). [Online]. Available: http://www.cms.hhs.gov/qio [accessed April 9, 2005]. CMS. 2002. 7th Statement of Work (SOW). [Online]. Available: http://www.cms.hhs.gov/qio [accessed April 9, 2005]. CMS. 2004. The Quality Improvement Organization Program: CMS Briefing for IOM Staff. [Online]. Available: http://www.medqic.org/dcs/ContentServer?cid=1105558772835 &pagename=Medqic percent2FMQGeneralPage percent2FGeneralPageTemplate&c =MQGeneralPage [accessed December 26, 2005]. CMS. 2005a. Quality in Managed Care, National QAPI Project Information. [Online]. Avail- able: www.cms.hhs.gov/healthplans/quality [accessed May 25, 2005]. CMS. 2005b. 8th Statement of Work (SOW). [Online]. Available: http://www.cms.hhs.gov/ qio [accessed April 9, 2005]. CMS. 2005c. 8th Statement of Work (SOW), Version 080105-1. [Online]. Available: http:// www.cms.hhs.gov/quality improvementorgs/downloads/8thsow.pdf [accessed November 4, 2005]. Davis D. 1998. Does CME Work? An analysis of the effect of educational activities on physi- cian performance or health care outcomes. International Journal of Psychiatry in Medi- cine 28:2139. Eloranta S. 2005. Brief Report: National Nursing Home Improvement Collaborative Out- comes Congress. [Online]. Available: http://www.ihi.org/IHI/Topics/Improvement/ ImprovementMethods/ImprovementStories/BriefreportNationalNursingHome ImprovementCollaborativeOutcomesCongress.htm [accessed May 11, 2005]. iHealthBeat. 2005. CMS Expands DOQ-IT Program Nationwide. [Online]. Available: http:// www.ihealthbeat.org/index.cfm?Action=dspItem&itemID=114658 [accessed September 13, 2005]. Institute for Healthcare Improvement. 2003. The Breakthrough Series: IHI's Collaborative Model for Achieving Breakthrough Improvement. Cambridge, MA: Institute for Health- care Improvement. Institute for Healthcare Improvement. 2005. The Spread Initiative. [Online]. Available: http:/ /www.ihi.org/IHI/Programs/CollaborativeLearning/TheSpreadInitiative.htm [accessed April 21, 2005]. Jencks SF, Huff ED, Cuerdon T. 2003. Change in the quality of care delivered to medicare beneficiaries, 19981999 to 20002001. Journal of the American Medical Association 289(3):305312.
TECHNICAL ASSISTANCE FOR QUALITY IMPROVEMENT 229 Jones J. 2003. Improvement Story: Clarifying the Definition for Intractable Pain Makes the Difference. [Online]. Available: http://www.medqic.org/dcs/ContentServer?cid= 1110810401913&pagename=Medqic percent2FMQStories percent2FImprovement StoryTemplate&c=MQStories [accessed April 21, 2005]. Joshi M. 2000. Effecting and leading change in health care organizations. Journal of Quality Improvement 26(7):388399. Lumetra. 2005. QAPI--Cultural Competency. [Online]. Available: http://www.lumetra.com/ healthplans/culturalcompetency/index.asp [accessed May 24, 2005]. Moreno C (Director, Health Plan Benefits Group). 2004, March 3. Memo: Selection of 2005 QAPI Project Focus. [Online]. Available: http://www.cms.hhs.gov/healthplans/quality/ 2005memo2.pdf [accessed May 24, 2005]. Oklahoma Foundation for Medical Quality. 2005, March 7. Improvement Story: Mammogra- phy Scheduling Block Improves Rate. [Online]. Available: http://www.medqic.org/dcs/ ContentServer?cid=1110810342250&pagename=Medqic percent2FMQStories percent 2FImprovementStoryTemplate&c=MQStories [accessed September 14, 2005]. Ovretveit J, Bate P, Cleary P, Cretin S, Gustafson D, et al. 2002. Quality collaboratives: Les- sons from research. Quality and Safety in Health Care 11(4):345351. QSource: The Center for Healthcare Quality. 2005. Diabetes: African American Population. [Online]. Available: http://www.qsource.org/HDS/African percent20American percent20 Diabetes percent20-v1.pdf [accessed May 11, 2005]. Qualis Health. 2005. QIO Guidebook to QIOSC Resources. Unpublished. Seattle, WA: Qualis Health. Quality Partners of Rhode Island. 2005. Tools and Weblinks Related to Delirium. [Online]. Available: http://www.medqic.org/dcs/ContentServer?cid=1110810477361&pagename =Medqicpercent2FMQTools percent2FToolTemplate&c=MQTools [accessed May 1, 2005]. Rollow WC. 2005. The Medicare Quality Improvement (QIO) Program 7th SOW and Re- sults. PowerPoint Presentation to the Committee on Redesigning Health Insurance, June 13, Washington, DC. Shortell SM, Marsteller JA, Lin M, Pearson ML, Wu SY, et al. 2004. The role of perceived team effectiveness in improving chronic illness care. Medical Care 42(11):10401048. Soumerai S, McLaughlin TJ, Gurwitz JH, Guadagnoli E, Hauptman PJ, et al. 1998. Effect of local medical opinion leaders on quality of care for acute myocardial infarction: A ran- domized controlled trial. Journal of the American Medical Association 279:13581363. Thomson O'Brien M, Oxman A, Haynes R, Davis D, Freemantle N, Harvey E. 2005. Local opinion leaders. Cochrane Database of Systematic Reviews 2005(2).