Online Performance Appendix: Crosscutting Activities Related to Quality, Effectiveness, and Efficiency Research
Budget Estimates for Appropriations Committees, Fiscal Year 2010
In addition to our research portfolios, funds are provided in Health Costs, Quality, and Outcomes (HCQO) to support a variety of research projects that support all of our research portfolios. Projects that support all portfolios are kept with the Crosscutting Activities Related to Quality, Effectiveness, and Efficiency portfolio. In order to meet its outcome goals, AHRQ has developed a set of research contract and grant mechanisms that support the work of the portfolios. These activities include data collection and measurement, dissemination, rapid cycle research, research management and salary costs, training, and intramural and extramural research sponsored by multiple portfolios.
Examples of projects that help portfolios with measurement in health care include the Consumer Assessment of Healthcare Providers and Systems (CAHPS®), Healthcare Cost and Utilization Project (HCUP), Quality Indicators (QIs), and the National Healthcare Disparities and Quality Reports (NHDR/QR). Additional information about these activities is found in the next section.
Creation of new knowledge is critical to AHRQ's ability to answer questions related to improving the quality of health care. Portfolios rely on intramural and extramural research to accomplish strategic goals. The questions addressed are of interest and contribute to each of the portfolios. These types of research allow portfolios to generate knowledge and test hypotheses. Investigator-initiated research and training projects that have over-arching research topics—not specific to one portfolio—are kept within Crosscutting Activities. In addition, research portfolios use other activities to ensure that their research is being disseminated to the appropriate health care stakeholders and translated to usable information so health care is directly improved. Examples of activities that help with dissemination and translation are the Eisenberg Center, Evidence-Based Practice Centers (EPCs), marketing outreach activities, clearinghouses, and direct dissemination and knowledge transfer activities. Finally, crosscutting activities support Rapid Cycle Research and include Accelerating Change and Transformation in Organizations and Networks (ACTION), Centers for Education and Research on Therapeutics (CERTs), Primary Care Practice-Based Research Networks (PBRNs), and Developing Evidence to Inform Decisions about Effectiveness (DeCIDE Network). These Rapid Cycle Research Activities are found both in Crosscutting Activities and within our research portfolios, depending on the topic.
Research and Training Grants
AHRQ-supported grantees in this portfolio are working to answer questions about cost, organization, and socioeconomics; long-term care; pharmaceutical outcomes; training; quality of care; and system capacity and bioterrorism. AHRQ will highlight two grant programs related to Crosscutting Activities: CAHPS® and CERTs.
CAHPS®®. CAHPS® is a multi-year initiative of AHRQ. Originally, "CAHPS®" referred to AHRQ's "Consumer Assessment of Health Plans Study." However, in 2005, AHRQ changed this to "Consumer Assessment of Health Providers and Systems." This name better reflects the evolution of CAHPS® from its initial focus on enrollees' experiences with health plans to a broader focus on consumer experience with health care providers and facilities. AHRQ first launched the program in October 1995 in response to concerns about the lack of reliable information about the quality of health plans from the enrollees' perspective. The survey was adopted by CMS, U.S. Office of Personnel Management, and the National Committee for Quality Assurance for public reporting and accreditation purposes. As of 2007, 138,000,000 Americans are enrolled in health plans for which CAHPS® data are collected. Over time, the program has expanded beyond its original focus on health plans to address a range of health care services and to meet the various needs of health care consumers, purchasers, health plans, providers, and policymakers. In June 2007, AHRQ funded the third iteration of CAHPS® grants to two organizations: RAND and Yale School of Public Health. Though instrument development is a part of CAHPS® 3, there is a heavier emphasis on using CAHPS® data for quality improvement and expanding our knowledge of how to report quality data to consumers and other audiences. In FY 2009 and FY 2010, AHRQ support for CAHPS® grants totaled $2.9 million. Here are some highlights of the past year:
TalkingQuality. TalkingQualilty is a Web site developed by the CAHPS® consortium (AHRQ, the CAHPS® grantees, and the CAHPS® support contractor). This Web site assembles existing research and best practices about reporting quality information to consumers and other audiences. The intended users are people and organizations who design health care quality reports. In the past year, the team has begun a large-scale revision to this site, including updating of all information, designating priority content, improvements to site navigation and the possibility of developing new features, such as a 'wiki' type mechanism for linking users with reporting questions or problems. We hope to release an initial version of this improved site in early 2009.
CAHPS® Hospital Survey (HCAPS). This survey, developed at the request of CMS and jointly funded by CMS and AHRQ, is a standardized survey of the experiences of adult inpatients concerning care and services they received while hospital patients. CMS began voluntary national implementation of the CAHPS® Hospital Survey in fall 2006 and publicly reported survey results via the HospitalCompare Web site for the first time in March 2008. In the week before HCAHPS® data were added to HospitalCompare, CMS reports that there were 161,000 page views; in the week after, page views increased to 1.4 million.
CAHPS® Clinician and Group Survey. This survey, which we released in spring 2007, asks patients about their recent experiences with physicians and other office staff. We are currently working with the American Board of Medical Specialties (ABMS) to develop a version of this survey, which ABMS will use as part of their Maintenance of Certification process. ABMS will use survey results to improve physician performance and will ultimately release these data to consumers.
CAHPS® Home Health Care Survey. We are finishing work on this survey, which asks for patients' assessment of services they received from home health agencies (HHAs). These services include nursing, physical therapy, occupational therapy, or other medical care, as well as personal assistance. The field test involved 34 HHAs in 15 States. The questionnaire is expected to be refined, completed, and sent to CMS by fall of 2008, at which time it will be available to the public free of charge. CMS anticipates submitting the survey for endorsement to the National Quality Forum later this year.
CAHPS® Nursing Home Survey (Family Members). AHRQ plans to officially transmit the CAHPS® Nursing Home Survey for family members to CMS this summer and post the survey and related development documents to the CAHPS® Web site. The Technical Expert Panel met in February 2008 to review the final report and comment on the survey and administration protocol.
Surveys modules under development include: Health Literacy, Cultural Competence, and Health Information Technology.
The long-term goals of CAHPS® are to ensure that consumers/patients have accurate and timely information about health care providers and facilities to inform their selection decisions, and providers and health care facilities have accurate information from their patients to use as a basis for quality improvement efforts. CAHPS® has set a program performance goal of ensuring that CAHPS® data will be more easily available to the user community and that the number of consumers who have accessed CAHPS® information to make health care choices will increase by over 50 percent from the FY 2002 baseline of 100 million. By moving to create surveys for a range of providers beyond the widely used CAHPS® health plan surveys, including clinicians, hospitals, nursing homes, and dialysis facilities, CAHPS® is rapidly expanding the capacity to collect data that can be utilized to make more informed choices by the purchasers who contract with and the consumers who visit these providers. In FY 2007, CAHPS® met the performance target (see performance measure # 1.3.23) to increase 40 percent over the baseline of the user community. In FY 2007, AHRQ increased this usage to 41 percent over the baseline of 100 million users—141 million users of CAHPS® information—and maintained this performance level in FY 2008.
CERTs. The Centers for Education and Research on Therapeutics (CERTs) demonstration program is a national initiative to conduct research and provide education that advances the optimal use of therapeutics (i.e., drugs, medical devices, and biological products). The program consists of 14 research centers and a Coordinating Center and is funded and run as a cooperative agreement by AHRQ in consultation with the U.S. Food and Drug Administration (FDA). The CERTs receive funds from both public and private sources, with AHRQ providing core financial support—$11.5 million in both FY 2009 and FY 2010. The research conducted by the CERTs program has three major aims:
- To increase awareness of both the uses and risks of new drugs and drug combinations, biological products, and devices, as well as of mechanisms to improve their safe and effective use.
- To provide clinical information to patients and consumers; health care providers; pharmacists, pharmacy benefit managers, and purchasers; health maintenance organizations and health care delivery systems; insurers; and government agencies.
- To improve quality while reducing cost of care by increasing the appropriate use of drugs, biological products, and devices and by preventing their adverse effects and consequences of these effects (such as unnecessary hospitalizations).
Upper GI Bleeding: Measures 4.4.3 and 4.4.4
Results show that from FY 2005 through FY 2007, the actual rate of hospitalizations for upper gastrointestinal (GI) bleeding due to adverse effects of medication or inappropriate treatment of peptic ulcer disease in those between 65 and 85 years of age have consistently met or slightly exceeded the targets. In FY 2004, baselines rates were established (55/10,000). In FY 2006, the target was a 1.1-percent drop and the actual result was a 2-percent drop (54.38/10,000). In FY 2007, the target was a 2-percent drop and the actual result was a 5.2-percent drop (51.56/10,000).
The most recent results from FY 2008 also met the corresponding target. In FY 2008, the target was a 1.8-percent drop and the actual result was a 3.5-percent drop (49.75/10,000). Although FY 2007 and FY 2008 had approximately double the targeted decrease in hospitalizations for GI bleeding, we retained the previously modeled FY 2009 target of a 3-percent decrease pending a planned evaluation in FY 2009 as described above under 4.4.2. AHRQ did not revise this target because of an ongoing external evaluation that is currently gathering information on multiple factors that might explain why the rate of GI bleeding hospitalizations is exceeding targeted declines. The evaluation will assess the precision of the annual HCUP measurement of GI bleeding hospitalizations and its ability to discern meaningful changes in annual rates, so AHRQ can determine whether the observed annual changes are sufficiently robust to re-project a new and more ambitious trend. The evaluation will also examine potential contributions from non-pharmaceutical factors (including but not limited to changes in health care systems, treatment methods, and population lifestyle factors such as alcohol and tobacco use) to hospitalizations for GI bleeding, so that changes due to pharmaceuticals can be appropriately attributed and projected.
Results show that from FY 2005 through FY 2007, the number of admissions for GI bleeding have generated a per year drop in per capita charges for GI bleeding and our targets have consistently been met. In FY 2004, baselines rates were established ($96.54 per capita). In FY2006, the target was a 3-percent drop and the actual result was a 3.2-percent drop ($93.36 per capita). In FY 2007, the target was a 4-percent drop and the actual result was a 4.9-percent drop ($91.81 per capita).
The most recent results from FY 2008 also met the corresponding target. In FY 2008, the target was a 5-percent drop and the actual result was a 5.1-percent drop ($87.10 per capita). Given the past trend, we believe it is reasonable to expect that hospitalization for upper GI bleeding due to adverse events of medication or inappropriate treatment of peptic ulcer disease in those between 65 and 85 years of age will decrease, and the decreased number of admissions will continue to generate an annual drop in per capita charges for GI bleeding. The target selected for FY 2009 is a 6-percent drop. The target selected for FY 2010 is a 7-percent drop. In FY2009, the program will assess the ambitiousness of current targets.
Many external factors could have affected this performance trend. For example, upper GI bleeding is common in people taking drugs such as anticoagulants, medications affecting platelet functions, and those affecting gastrointestinal mucosal defenses. Increased or more appropriate monitoring of these drugs could have affected the number of hospitalizations for upper GI bleeding due to adverse events of medication. An increased use of pharmacologic agents such as proton pump inhibitors to prevent gastric irritation in patients could also have affected this performance trend.
The CERTs program initiated a warfarin interaction study to better define the relative safety of commonly used antibiotics and antifungals when co-administered with warfarin. The safety outcome will be major bleeding complications of warfarin, as confirmed by medical record review. This study will test the hypothesis that in a cohort of warfarin users, the risk for major GI bleeding complications differs among the specific study antimicrobials. At present, clinicians cannot make evidence-based choices when prescribing antibiotics and antifungals with warfarin, because the overall quality of interaction literature for warfarin is poor. These data on the relative safety of antimicrobials would inform clinical decisions for this vulnerable population. This research, once complete, will have a direct impact on AHRQ's performance measure 4.4.3: reduce hospitalization for upper GI bleeding due to the adverse effects of medication or inappropriate treatment of peptic ulcer disease in those between 65 and 85 years of age by implementing the research findings.
CERTs, as part of the now obsolete Pharmaceutical Outcomes program, underwent a program assessment in 2004. The program received a Moderately Effective rating. The assessment cited research to be conducted by AHRQ's CERTS program to reduce inappropriate antibiotic use in children, congestive heart failure hospital readmission rates, and hospitalizations for upper GI bleeding due to the adverse effects of medication or inappropriate treatment of peptic ulcer disease. As a result of the program assessment, the CERTs program is taking actions to: (1) analyze trends to determine if targets for measures need to be adjusted; and (2) produce reports on best practices in observational methods research.
Research Contracts and IAAs
Examples of types of research contracts and IAAs AHRQ has supported related to Crosscutting Activities includes the following:
Contracts and IAAs Support the Development and Release of the Annual National Healthcare Quality Report and Its Companion Document, the National Healthcare Disparities Report. These reportsmeasure quality and disparities in four key areas of health care: effectiveness, patient safety, timeliness, and patient centeredness. In addition, AHRQ provides a State Snapshots Web tool that was launched in 2005. It is an application that helps State health leaders, researchers, consumers, and others understand the status of health care quality in individual States, including each State's strengths and weaknesses. The 51 State Snapshots—every State plus Washington, DC—are based on 129 quality measures, each of which evaluates a different segment of health care performance. While the measures are the products of complex statistical formulas, they are expressed on the Web site as simple, five-color "performance meter" illustrations. Support for these contracts and IAAs totals $2.9 million in both FY 2009 and FY 2010.
Contracts and IAAs to Support the National Quality Measures Clearinghouse™ (NQMC) and Its Companion the National Guideline Clearinghouse™ (NGC). The NQMC and the NGC provide open access to thousands of quality measures and clinical practice guidelines to clinicians and health care providers. The NQMC and NGC receive close to two million visits each month. They can be found at http://www.qualitymeasures.ahrq.gov and http://www.guideline.gov. Support for these two clearinghouses total $7.0 million in FY 2010.
Contract Support for HCUP and the AHRQ Quality Indicators (QIs). Efforts to improve the quality, safety, effectiveness, and efficiency of health care and reduce disparities in the United States require detailed knowledge about how the health care delivery system works now and how different organizational and financial arrangements affect this performance. Improving health care requires easy access to detailed information and data on costs, access to health care, quality, and outcomes that can be used for research and policymaking at the national, State, and local levels. It also requires tools to measure and track progress in these areas. The Healthcare Cost and Utilization Project (HCUP) provides the necessary data through a long-standing partnership with State data organizations, hospital associations, and private data organizations. HCUP is a family of health care databases and related software tools and products that support the mission of AHRQ. HCUP includes the largest collection of all-payer, encounter-level data in the United States, beginning in 1988. It includes detailed information on 90 percent of all inpatient stays in the country—including information about the diagnosis, the procedures, the cost, and who paid for the care, as well as encrypted non-identifiable demographic information. For over 25 States, it also includes ambulatory surgery and emergency department data. Support for the HCUP contract totals $4.1 million in FY 2009 and 2010. For more information, go to http://www.hcup-us.ahrq.gov/overview.jsp.
One widely used HCUP tool is the AHRQ Quality Indicators (QIs), a set of quality measures developed from HCUP data. Support for QIs total $0.4 million in FY 2010. This measure set is organized into four modules—Prevention, Inpatient, Patient Safety, and Pediatrics. The Prevention Quality Indictors (PQIs) focus on ambulatory care-sensitive conditions that identify adult hospital admissions that evidence suggests could have been avoided, at least in part, through high-quality outpatient care. Inpatient Quality Indicators (IQIs) reflect quality of care for adults inside hospitals and include: inpatient mortality for medical conditions; inpatient mortality for surgical procedures; utilization of procedures for which there are questions of overuse, underuse, or misuse; and volume of procedures for which there is evidence that a higher volume of procedures may be associated with lower mortality. Patient Safety Indicators (PSIs) also reflect quality of care for adults inside hospitals, but focus on potentially avoidable complications and iatrogenic events. Pediatric Quality Indicators (PDIs) reflect quality of care for children below the age of 18 and neonates inside hospitals and identify potentially avoidable hospitalizations among children. These measures are free and made publicly available as part of an AHRQ-supported software package.
The AHRQ QIs are based upon a few guiding principles which make them unique. They:
- Were developed using readily available administrative data (HCUP).
- Use a transparent methodology.
- Are risk-adjusted and use a readily available, familiar methodology.
- Are constantly refined based on user input.
- Are updated and maintained by a trusted source.
- Have documentation and program software in the public domain.
The HCUP/QI family of data and products supports the achievements of a number of AHRQ objectives including two major goals:
- Expand and improve data and tools
- Expand use of HCUP and the AHRQ QIs by policymakers and others.
Expand and Improve Data and Tools
The HCUP databases have been a powerful resource for the development of tools that can be applied to other similar databases by health services researchers and decisionmakers. The expanded data and tools can then be translated to inform decisionmaking and improve health care delivery. A major achievement in 2008 and 2009 was creation and release of the largest all-payer emergency department database in the United States.The first Nationwide Emergency Department Sample (NEDS) database was created to enable national analyses of emergency department (ED) utilization patterns and support public health professionals, administrators, policymakers, and clinicians in their decisionmaking regarding this critical source of care. The NEDS contains clinical and non-clinical information on patients, regardless of payer—including those covered by Medicare, Medicaid, private insurance, and the uninsured. The ED serves a dual role in the U.S. health care system infrastructure as a point of entry for approximately 50percent of inpatient hospital admissions and as a setting for treat-and-release outpatient visits.
In FY 2008, AHRQ also met our performance target (see performance table 1.3.15) to increase the number of partners contributing outpatient data to the HCUP databases. AHRQ added data from Maine for a total of 39 statewide data organizations participating in HCUP. The number of State Ambulatory Surgery Databases increased by three partners (California, Maine, and Oklahoma) and the number of State Emergency Department Databases increased by four partners (California, Maine, New York, and Rhode Island). They were selected based on the diversity— in terms of terms of geographic representation and population ethnicity—they bring to the project, along with data quality performance and their ability to facilitate timely processing of data. This outcome exceeded the goal by adding seven new Partner databases instead of four. Progress has already been made on FY 2009 goals with the addition of a 40th partner and two new outpatient databases.
Expand use of HCUP and the AHRQ QIs by Policymakers and Others
The AHRQ QIs are widely used for quality improvement and public reporting initiatives. We saw several major successes in FY 2008: The National Quality Forum (NQF) endorsed 41 of our QIs for public reporting, and there are a growing number of organizations who are using them for public reporting. There are currently over 2,000 subscribers to the AHRQ QI listerv and approximately 150 inquiries are received monthly.
AHRQ has fully met its FY 2008 performance target (Performance table 1.3.22): "3 new organizations use HCUP/QIs to assess potential areas of quality improvement, and at least 2 of them will develop and implement an intervention based on the QIs. Impact will be observed in 1 new organization after the development and implementation of an intervention based on the QIs."
As the result of NQF endorsement, a growing number of States are using the QIs for public reporting of hospital quality. Most recently, New Jersey and California became the 13th and 14th States to use the AHRQ QIs in a hospital level public report card. Nevada will begin public reporting using the QIs by the end of April 2009. With the addition of Nevada, 15 States, covering more than half the U.S. population, will be publicly reporting on hospital quality using AHRQ's QIs. A new Quality Indicators Learning Institute helps these States use the indicators effectively, and provides technical assistance to new States or communities as they plan their public reporting efforts.
In addition, CMS has incorporated 9 AHRQ Patient Safety Indicators in its 2009 IPPS Rule (acute hospital inpatient prospective payment system). The CMS has held a national "dry run" of the measures with its hospitals and is planning on releasing the measures by hospital on its Hospital Compare Web site in FY 2010.
HCUP and QI analyses and reports based on these tools have been greatly expanded through statistical briefs, peer-reviewed publications, and Web-based reports. For example, HCUP provides critical information on the U.S. health care system, such as:
- From 1997 to 2006, the number of uninsured hospitalizations increased by 34 percent, which far exceeds the 14 percent overall increase in hospital stays. Relative to all hospital stays, uninsured stays began in the emergency department (ED) much more frequently with nearly 60 percent of these stays originating in the ED compared to 44 percent of hospital stays overall.
- Uninsured patients accounted for 22.0 percent of tuberculosis (TB) stays, though they made up only 5.8 percent of all non-maternal, non-neonatal hospitalizations. Medicaid covered 24.4 percent of all TB stays, though it accounted for only 12.3 percent of all non-maternal, non-neonatal stays.
- In 2006, there were 503,300 total hospital stays with pressure ulcers (a potentially preventable complication) noted as a diagnosis—an increase of nearly 80 percent since 1993. Adult stays totaled $11 billion in hospital costs in 2006.
- One out of every 5 hospital stays (21.3 percent) had either a principal or secondary diagnosis of a mental health condition. Medicare and Medicaid were the expected payers for 60 percent of mental health stays.
- Potentially preventable hospital stays for chronic conditions were 42 percent higher among Hispanic adults than among non-Hispanic white adults. Disparities between Hispanics and non-Hispanic whites were greater for diabetes (37 versus 17 hospitalizations per 10,000 population, respectively).
- Almost two-thirds of the national bill for hospital care was billed to two government payers, Medicare ($444 billion) and Medicaid ($135 billion), while slightly less than one-third ($287 billion) was billed to private insurance and about 5 percent ($43 billion) was billed to uninsured individuals.
HCUP and the Quality Indicators projects also began development of a new AHRQ tool, My Own Network AHRQ (MonAHRQ)—a Web site builder that would allow any organization or Agency to input their data and then output a Web site. It is being developed to be used by anyone with access to hospital discharge data and will allow users to generate quality, cost, and utilization statistics for Web sites that will be hosted on local servers by individual organizations. These Web sites will provide information in a uniform way using uniform measures at whatever level the host user chooses (e.g., county-level, hospital-level) to various audiences (e.g., patients/consumers, constituent hospitals, public health officials). After testing is completed in the spring of 2009, AHRQ anticipates launching the tool for public use in summer 2009. The final Web site will be an interactive querying tool that users can navigate to learn about hospital care in their area. These efforts, along with others to speed up the production of HCUP databases, increase data representativeness, examine data linkages, facilitate the inclusion of clinical information in administrative data, and begin development of the new Web-based tool all combine to ensure future program performance and support of the Agency's portfolios.