AHRQ Annual Highlights, 2010 (continued)

 

Patient Safety Portfolio

Finding ways to eliminate medical errors and improve patient safety are an integral part of the Agency's agenda. AHRQ-funded projects and partnerships identify, develop, test, and implement patient safety and quality measures and solutions. During FY10, AHRQ funded $25 million in grants for demonstration projects on patient safety and medical liability reform and $34 million to expand projects to help prevent healthcare-associated infections. The Agency also released updates to its Common Formats for Patient Safety Organizations, Team Strategies and Tools to Enhance Performance and Patient Safety (TeamSTEPPS®), and its suite of patient safety culture surveys. Finally, AHRQ developed several tools to help engage patients and families as part of the health care team.

TalkingQuality Web site

In FY10, AHRQ released the newly revised TalkingQuality Web site, which is a resource for organizations that produce reports for consumers on the quality of care provided by hospitals, health plans, medical groups, nursing homes, and physicians. The TalkingQuality Web site offers resources that help organizations develop strategies for creating and disseminating reports to improve the quality of care that consumers receive. The site also offers ways to assess reports' effectiveness and to use the lessons learned.

Go to http://talkingquality.ahrq.gov for more information.

Patient Safety and Medical Liability Demonstration Projects

In the largest Federal investment connecting medical liability to quality, AHRQ awarded $25 million in funding to support efforts by States and health systems to implement and evaluate patient safety approaches and medical liability reform models. The demonstration and planning grants are part of the patient safety and medical liability initiative that President Obama announced during a September 9, 2009, address to a joint session of the United States Congress.

As part of his vision for a health care system that puts patient safety first and allows doctors to focus on practicing medicine, the President directed the Secretary of Health & Human Services (HHS) to help States and health care systems test models that put patient safety first and work to reduce preventable injuries; foster better communication between doctors and their patients; ensure that patients are compensated in a fair and timely manner for medical injuries, while also reducing the incidence of frivolous lawsuits; and reduce liability premiums.

Some of the reforms that are being planned and tested under HHS's initiative address limitations of the current medical liability system, such as costs, patient safety, and administrative burden for doctors. Grants support the creation of a judge-directed negotiation program, the development of "safe harbors" for State-endorsed evidence-based care guidelines, and early disclosure and offers of prompt compensation.

Of the $25 million, $23 million is allocated to grants and $2 million is allocated to a contract to evaluate improvements in both patient safety and medical liability systems. The evaluation is designed to develop the evidence base that will inform long-term solutions to the medical liability problem.

Blood Thinner DVD Wins Two Telly Awards

AHRQ's patient education video "Staying Active and Healthy with Blood Thinners" earned two Telly Awards, one in the "Health and Wellness" category and one for "How-To/Instructional Video." Telly Awards honor the best local and regional cable television commercials and programs, as well as the finest video and film productions. More than 13,000 entries were evaluated against a rigorous set of standards.

To view the video or the booklet, go to http://www.ahrq.gov/consumer/btpills.htm.

Preventing Healthcare-Associated Infections

Healthcare-associated infections (HAIs) are the most common complication of hospital care, resulting in 1.7 million infections and 99,000 deaths each year, according to a 2002 study by the Centers for Disease Control and Prevention. The added financial burden attributable to HAIs is estimated to be between $28 billion to $33 billion each year.

According to data from the 2007 Nationwide Inpatient Sample, adults who developed HAIs due to medical or surgical care while in the hospital in 2007 had to stay an average of 19 days longer than adults who didn't develop an infection, (24 days versus 5 days). For patients with an HAI, the rate of death in the hospital, on average, was 6 times as high as the rate for patients without an HAI (9 percent versus 1.5 percent). Also, on average, the hospital stay of an adult patient who developed an HAI was about $43,000 more expensive than the stay of a patient without an HAI ($52,096 versus $9,377).

To address this growing problem, AHRQ is contributing in multiple ways to the national effort to prevent HAIs. At the HHS level, AHRQ participates in coordinated, Department-wide activities directed to HAIs by:

  • Serving on the HHS HAI Steering Committee, which is chaired by the Deputy Assistant Secretary for Healthcare Quality and has representatives from the HHS agencies involved in HAI projects.
  • Taking an active part in the Steering Committee's development and updating of the National Action Plan to Prevent HAIs.
  • Chairing the Research Working Group, the subcommittee of the Steering committee charged with developing plans for future research investments to combat HAIs; and,
  • On behalf of the Steering Committee, directing the longitudinal evaluation of progress in achieving the National Action Plan's goals.

In a second role, AHRQ supports research and demonstrations to generate knowledge about the best ways to prevent HAIs and accelerate the wide-scale adoption of evidence-based approaches in routine practice. In FY10, AHRQ awarded $34 million for grants and contracts that target HAI prevention. This represents a doubling of the $17 million that was available to support HAI projects in FY09. The funded projects address HAI reduction efforts in all three tiers of health care settings: Tier 1, acute care hospitals; Tier 2, ambulatory surgery centers and hemodialysis/end-stage rental disease facilities, and Tier 3, long-term care.

A major focus of AHRQ's HAI prevention work involves the implementation of the Comprehensive Unit-based Safety Program (CUSP), which is based on an Intensive Care Unit (ICU) Safety Reporting System developed by the Johns Hopkins University Quality and Safety Research Group, Baltimore, MD. The CUSP integrates communication, teamwork, and leadership to create and support a "harm-free" patient care culture. CUSP is implemented at the unit level and provides a scalable program that can be implemented throughout an organization.

The Agency announced in October 2009 that the CUSP, which has successfully reduced central line-associated blood stream infections in intensive care units, will expand to all 50 States, the District of Columbia, and Puerto Rico as well as additional hospitals in States that are already participating in the program. Over a 3-year period, 10 of these States will test the CUSP as a strategy for reducing central line-associated blood stream infections in 100 ICUs. Participating hospitals will implement a checklist to ensure compliance with safety practices, educate staff on evidence-based practices to reduce blood stream infections, educate staff on team training, provide feedback on infection rates to hospitals and hospital units, and implement monthly team meetings to assess progress.

This additional funding will also extend the CUSP to general medical and surgical hospital units in addition to ICUs. It will also broaden its focus to address other types of HAIs, such as bloodstream infections in patients undergoing hemodialysis and catheter-associated urinary tract infections.

For more information on AHRQ's projects to prevent HAIs, go to . A complete list of institutions and projects funded in FY10 is available at http://www.ahrq.gov/qual/haify10.htm.

AHRQ WebM&M

AHRQ WebM&M (Morbidity and Mortality Rounds on the Web) is the online journal and forum on patient safety and health care quality. This site features expert analysis of medical errors reported anonymously by our readers, interactive learning modules on patient safety ("Spotlight Cases"), and Perspectives on Safety. Continuing medical education and continuing education unit credits are available.

WebM&M can be found at http://www.webmm.ahrq.gov.

Patient Safety Organizations

Established by the Patient Safety and Quality Improvement Act of 2005 (the Patient Safety Act), Patient Safety Organizations (PSOs) collect and analyze information on the patient safety events that health care providers report and provide feedback to help clinicians and health care organizations improve health care quality. Strong confidentiality provisions are also key to this voluntary reporting. At the end of FY10, there were 88 listed PSOs in 30 States, the District of Columbia, and Puerto Rico.

As outlined in the Patient Safety Act, AHRQ administers provisions governing PSO operations. To allow health care providers to collect and submit standardized information regarding patient safety events, AHRQ coordinates development of Common Formats (i.e., common definitions and reporting formats) for reporting events to the PSOs. Common Formats optimize the opportunity for the public and private sectors to learn more about trends and patterns in patient safety, with the purpose of improving health care quality. AHRQ released the initial set of Common Formats for hospitals in FY09. In FY10, AHRQ released Common Formats (Version 1.1) that includes updated event descriptions, reports, and data collection forms, as well as technical specifications for software developers. Once the data have been made non-identifiable by the standards of the Patient Safety Rule, they will be transmitted to the Network of Patient Safety Databases in 2011 for aggregate analysis and generation of initial reports to be included in AHRQ's National Healthcare Quality and Disparities Reports.

Additional sets are being developed for nursing homes, ambulatory surgery centers, physician offices, and adverse events related to the use of medical or surgical devices or health information technology. For more information on the PSOs and Common Formats, go to http://www.pso.ahrq.gov.

Hospitals changed protocols for prevention of venous thromboembolism

Between September 2008 and May 2009, AHRQ hosted a series of seven Web conferences about its toolkit, Preventing Hospital-Acquired Venous Thromboembolism: A Guide for Effective Quality Improvement as part of an AHRQ Knowledge Transfer project. As a result of the Web conference series, several hospitals from New York changed their policies on caring for patients at risk for venous thromboembolism (VTE). The AHRQ toolkit is a comprehensive guide to help hospitals and clinicians implement processes to prevent dangerous blood clots.

Massena Memorial Hospital in Massena, New York, did not have a VTE prevention protocol in place at the beginning of the Web conference series. The hospital used the information to develop and implement a VTE protocol for the medical, surgical, intensive care, and obstetrics units. A baseline audit of Massena patient charts in November 2008 showed that 25 percent of patients received some form of VTE prevention. By September 2009, patients receiving some form of VTE prevention had increased to 90 percent.

St. Barnabas Hospital in New York City also did not have a VTE prevention protocol prior to participating in the Web conferences. The hospital developed a VTE order set for the medical unit that aligns with the toolkit's recommendations. St. Barnabas has provided in-service training for its residency programs on using the resources.

Catskill Regional Medical Center revised an existing VTE prevention protocol and switched from a points-based to a risk-category protocol and now recommends drugs to prevent blood clots for all patients at moderate risk for VTE. The new protocol also expanded the range of medications recommended for patients in this risk group. Catskill now conducts daily real-time reviews of VTE prevention rates. As a result of these efforts, the percent of patients at Catskill receiving some form of mechanical or medical VTE prevention increased from 30 percent in October 2008 to 80 percent in August 2009.

Lawrence Hospital Center also revised its VTE order set based on the information presented in the Web conference series. Originally using a points-based tool, the hospital now uses a protocol that groups patients into three clearly differentiated categories of risk. The protocol includes information about VTE prevention for each risk group and about contraindications.

Maimonides Medical Center made significant changes to the hospital's VTE prevention protocol after participating in the Web conferences. Maimonides streamlined what had been a fragmented paper-based form and put it online. The hospital also reduced the number of VTE risk categories from four to three and provided clearer prevention recommendations for each category.

Kingsbrook Jewish Medical Center used the information from the Web conferences to improve the risk assessment component of its existing protocol. Information about contraindications for certain approaches to VTE prevention was added electronically to the order set. The new computerized physician order entry VTE prevention protocol was implemented in the medical and surgical units in September 2009.

TeamSTEPPS®

TeamSTEPPS® is an evidence-based teamwork system designed for improving communication and other teamwork skills among health care professionals. Developed by the Department of Defense in collaboration with AHRQ, TeamSTEPPS includes a comprehensive set of ready-to-use materials and training curricula necessary to integrate teamwork principles successfully into a health care system. TeamSTEPPS is now a part of the Centers for Medicare & Medicaid Services (CMS) 9th Scope of Work for all Quality Improvement Organizations.

In FY10, 18 TeamSTEPPS 3-day master trainer courses were held at 5 Team Resource Centers: Duke University Medical Center, Durham, NC; Carilion Clinic, Roanoke, VA; University of Minnesota Fairview Medical Center, Minneapolis, MN; Creighton University Medical Center, Omaha, NE; and University of Washington Medicine, Seattle, WA. Additional training sessions for State hospital systems are underway. These 3- day training sessions will create a national network of master trainers, who will in turn offer TeamSTEPPS training to frontline providers in hospitals and other health care settings throughout the country.

Also, AHRQ released the TeamSTEPPS Teamwork Perception Questionnaire (T-TPQ). T-TPQ is a measurement tool that helps determine how an individual perceives the current state of teamwork within an organization. Measuring an individual's perception of collective teamwork offers a broader picture of an organization's team climate; thus, a measure of perception of overall teamwork serves as an additional measure of the effectiveness of TeamSTEPPS training. The T-TPQ has been designed to correlate with the AHRQ Hospital Survey on Patient Safety Culture.

More information on TeamSTEPPS® can be found at http://teamstepps.ahrq.gov.

Association of periOperative Registered Nurses develops toolkit based on TeamSTEPPS®

As part of its Patient Safety First program, the Association of periOperative Registered Nurses (AORN) developed a patient handoff toolkit based on Team Strategies and Tools to Enhance Performance and Patient Safety (TeamSTEPPS®). Because problems in communication cause the majority of avoidable adverse events, AORN customized TeamSTEPPS materials for the perioperative environment to improve communication among surgical team members by standardizing handoffs. The perioperative toolkit includes a summary of the evidence-based research on patient handoffs, sample patient handoff templates, and a PowerPoint® presentation on standardizing handoffs. The patient handoff toolkit is available free of charge on the AORN Web site at http://www.aorn.org/PracticeResources/ToolKits/PatientHandOffToolKit.

Patient Safety Culture Surveys

In its 1999 landmark report, To Err Is Human: Building a Safer Health System, the Institute of Medicine cited studies that found that at as many as 98,000 people die in U.S. hospitals each year as a result of preventable medical errors. As part of its goal to support a culture of patient safety and quality improvement in the Nation's health care system, AHRQ sponsors the development of patient safety culture assessment tools for hospitals, nursing homes, and medical offices. A survey for pharmacies also is under development. Health care organizations use these survey assessment tools to assess their patient safety culture, track changes in patient safety over time, and evaluate the impact of specific patient safety interventions. The three survey tools are: Hospital Survey on Patient Safety Culture, Medical Office Survey on Patient Safety Culture, and Nursing Home Survey on Patient Safety Culture.

AHRQ Patient Safety Network (PSNet)

The AHRQ PSNet (http://www.psnet.ahrq.gov) is a national Web-based resource featuring the latest news and essential resources on patient safety. The site offers weekly updates of patient safety literature, news, tools, and meetings ("What's New"), and a vast set of carefully annotated links to important research and other information on patient safety ("The Collection"). Supported by a robust patient safety taxonomy and Web architecture, the AHRQ PSNet provides powerful searching and browsing capability, as well as the ability for diverse users to customize the site around their interests (My PSNet). It also is tightly coupled with AHRQ's WebM&M, the popular monthly journal that features user-submitted cases of medical errors, expert commentaries, and perspectives on patient safety.

Hospital Survey on Patient Safety Culture: 2010 Comparative Database Report

In FY10, AHRQ released the Hospital Survey on Patient Safety Culture 2010 Comparative Database Report. This report contains more data than any previous report, providing results from nearly 350,000 hospital staff in 885 hospitals. The report presents statistics on the patient safety culture areas or composites assessed in the survey, as well as the survey items. The appendixes present breakouts of the data by hospital characteristics (bed size, teaching status, ownership and control, region), respondent characteristics (hospital work area/unit, staff position, interaction with patients), and trends over time for the hospitals that administered the survey and submitted data more than once.

Hospitals implement AHRQ's Door-to-Doc patient safety toolkit

Bon Secours St. Mary's Hospital, a 320-bed nonprofit hospital in Richmond, VA, and Carilion Roanoke Memorial Hospital, an 825-bed, nonprofit teaching hospital in Roanoke, VA, implemented AHRQ's Door-to-Doc patient safety toolkit after attending an AHRQ-sponsored training session in September 2008.

With the Door-to-Doc model, patient flow is split into "less sick" and "sicker" patient groups. This determination is based on a brief analysis rather than a full triage. The system has the advantage of keeping the vast majority of patients—those who are less sick—moving during busy times, rather than waiting in the lobby or the ED. A key advantage with the Door-to-Doc model is that ED beds are reserved for sicker patients who truly need them. Less sick patients, who tend to be ambulatory, are not assigned beds; instead, they move among treatment areas as they would in a clinic setting. These patients remain dressed and mobile as much as possible and wait for lab and other test results away from the flow of other patients.

As a result of implementing Door-to-Doc, St. Mary's Hospital ED was able to decrease by nearly 50 percent the time patients must wait to see an ED physician—from more than 30 minutes to just 16 minutes. St. Mary's was also able to accommodate an increase in patient volume without experiencing any increase in the rate of patients leaving without being treated. Patient satisfaction scores increased from the 70th to the 90th percentile.

Carilion Roanoke Memorial Hospital had experienced a left-without-being-seen rate consistently over 7 percent, which had a negative impact on both patients and providers. Since implementing Door-to-Doc, Carilion reduced the number of patients leaving the ED without treatment to 2 percent, enabling the ED to see more patients and increase patient volume from 190 patients per day to 205 patients per day. This increase in patient volume has resulted in a $156,000 increase in hospital revenue for every 1 percent reduction in the rate of patients leaving without treatment. In addition, using the toolkit allowed Carilion to reduce the wait times that less-sick patients spend in the ED from 228 minutes pre-Door-to- Doc to 186 minutes after implementation—an 18 percent improvement.

For more information on Door-to-Doc and other patient safety tools, go to http://www.ahrq.gov/qual/pips/.

Consumer Guides

During FY10, AHRQ developed a wide range of tools to help patients and their families become more engaged in their health care. Bilingual brochures, videos, advice columns, podcasts, and other materials are available in English and Spanish at http://www.ahrq.gov/consumer.

In addition, AHRQ synthesized information from its patient safety research findings to develop a new easy-to read guide called Taking Care of Myself: A Guide for When I Leave the Hospital. After hospital staff or patients fill in the information, the guide requests that patients track their medication schedules, upcoming appointments, and important phone numbers. The guide is adapted from the AHRQ-funded Re-Engineered Hospital Discharge project, known as Project RED. AHRQ Partnerships in Implementing Patient Safety grantee Brian Jack, M.D., Associate Professor of Family Medicine at Boston University, developed Project RED, which showed that preparing patients to care for themselves when they leave the hospital can improve patient safety and reduce rehospitalization rates by 30 percent. The guide is available in both English and Spanish at http://www.ahrq.gov/qual/goinghomeguide.htm.

Implementing Re-Engineered Hospital Discharge (Project RED)

This Knowledge Transfer project aims to help hospitals implement the Re-Engineered Hospital Discharge (Project RED) intervention to improve the discharge process and reduce hospital readmissions. The ongoing project features development of Web-based training modules, and technical assistance for participants implementing the toolkit. Activities with each hospital/health system are customized based on unique needs. Below are highlights of the accomplishments of the project to date:

  • Trained nearly 50 hospitals, including 2 complete health systems, to implement Project RED.
  • Developed a comprehensive training plan for educating hospital staff during the discharge program implementation process.
  • Developed a tool for participating sites' to use when measuring the potential impact of the Project RED intervention program. Components of the tool include readmission rates, length of stay, patient experience, frontline staff and primary care physician opinion about discharge process, timeliness of RED intervention, and completeness of after hospital care plan provided to targeted patients.

Recent Research Findings on Quality and Patient Safety

AHRQ-funded researchers published numerous studies in FY10, abstracts of which can be found on AHRQ's Patient Safety Network Web site at http://www.psnet.ahrq.gov. Among the highlights:

  • Emergency departments (EDs) are a vulnerable area for patient care errors, especially during shift changes when physicians and nurses "hand off" patients. In researching how to make handoffs safer, the study found that the number of handoffs can be reduced by discharging patients near the time of shift changes. Another approach is to have outgoing and incoming ED personnel gather in a quiet, dedicated space where they account for every patient and provide succinct overviews of patients' statuses, including having lab and other study reports available. Researchers also recommend that receiving physicians ask questions and discuss outstanding tasks during these handoff sessions (Annals of Emergency Medicine, February 2010).
  • Drug labels that use simplified language, in some cases with patient-tested icons, can improve patients' ability to understand warning labels affixed to prescription drug containers. Researchers found that the rate of correct interpretation of drug warnings was lowest (80.3 percent) among patients who were shown standard warnings, higher (90.6 percent) for those shown simplified warning text, and highest (92.1 percent) for patients shown simplified text with icons. Patients with low literacy (below 7th grade reading level) were 35 percent less likely to correctly interpret standard drug warning labels than those who read at the 9th grade level or higher. Patients with marginal (7th and 8th grade reading levels) or low literacy were two to three times more likely to correctly understand warnings with both simplified text and icons than those with simplified text alone (Archives of Internal Medicine, January 2010).
  • Work hour restrictions allow medical residents to work up to 80 hours per week for no more than 30 consecutive hours. These restrictions have decreased the amount of time residents spend teaching medical students; however, residents report feeling less exhausted and more satisfied with the level of care they deliver. Residents completed a survey designed to measure various aspects of their working situations, including time spent teaching, number of hours worked, satisfaction with patient care, and level of exhaustion. Nearly a quarter (24 percent) of residents reported spending less time teaching, which was associated with working less than 80 hours a week, being a second-or third-year resident, and spending more time on administrative tasks. Residents with reduced teaching schedules reported feeling less emotionally exhausted and more satisfied with the care they provided patients (Journal of Hospital Medicine, October 2009).
  • Failing to order tests, report results to patients, and follow up with abnormal test findings are leading causes of diagnostic errors, according to a survey of U.S. primary care and specialist physicians. Responding to a 6-item survey, nearly 300 physicians from 22 hospitals reported 583 cases of diagnostic error—the largest report ever published on diagnostic errors. The most commonly missed or delayed diagnoses included pulmonary embolism, drug reactions or overdose, lung cancer, colorectal cancer, acute coronary syndrome (including heart attack), breast cancer, and stroke. Diagnostic errors occurred most often in the testing phase (failure to order, report, and follow up laboratory results, 44 percent), followed by clinician assessment errors (failure to consider and overweighing competing diagnoses, 32 percent; inadequate history taking, 10 percent; incomplete physical examination, 10 percent); and referral or consultation errors and delays, 3 percent. Overall, 28 percent of the 583 diagnostic errors were rated as major, resulting in patient death, permanent disability, or a near-life-threatening event. Another 41 percent resulted in moderate adverse outcomes that caused the patient short-term illness, a prolonged hospital stay, an invasive procedure, or more intense care; 31 percent of diagnostic errors were minor or insignificant (Archives of Internal Medicine, November 2009).
  • According to an AHRQ-funded research paper, health care organizations should disclose medical mistakes that affect multiple patients even if patients were not harmed by the event. Medical mistakes that affect multiple patients, known as large-scale adverse events (LSAEs) to researchers, are incidents or series of related incidents that harm or could potentially harm multiple patients. These events, which can include incompletely sterilized surgical equipment, poor laboratory quality control, and equipment malfunctions, are often identified after care has been provided and can affect thousands of patients. Researchers weighed ethical considerations of whether to disclose such events. For instance, is disclosure ethical if patients were unlikely to have been physically harmed by the event but could be harmed psychologically by the disclosure? The authors reviewed instances in which health care institutions disclosed an LSAE and analyzed the method of disclosure and existing disclosure policies. They concluded that, in most cases, these events should be disclosed and recommended that organizations develop an institutional policy for managing and planning the disclosure process, communicating with the public, and patient followup (New England Journal of Medicine, September 2010).

QIO Learning Network helps health care providers improve quality of care

The Quality Improvement Organization (QIO) Learning Network was established in January 2010 as a Knowledge Transfer project. Its primary focus has been to train QIOs and their providers on two AHRQ-supported tools: Preventing Hospital-Acquired Venous Thromboembolisms: A Guide for Effective Quality Improvement and the MATCH (medication reconciliation) toolkit. As part of this project, AHRQ is working with QIOs in the following States: Idaho, Indiana, Kentucky, Missouri, Nebraska, New Jersey, New York, North Carolina, South Carolina, Texas, and Washington.

The purpose of this learning network is to create an ongoing collaborative relationship between AHRQ and the QIOs. The goal is to provide training on AHRQ tools that will help the QIOs and their providers improve health care quality.

The QIO Learning Network has trained 3 QIOs and staff from 31 hospitals in 7 States on Preventing Hospital-Acquired Venous Thromboembolisms: A Guide for Effective Quality Improvement. As a result of the training, staff from all 31 hospitals have revised or developed a new venous thromboembolism (VTE) protocol based on information shared in the Guide. Staff from 3 hospitals are piloting the new or revised VTE protocol, and staff from 28 hospitals have implemented a new or revised VTE protocol.

In addition, 5 QIOs and staff from 58 hospitals, 11 home health agencies, 8 nursing homes and 1 pharmacy have been trained on the MATCH toolkit. The MATCH toolkit provides a comprehensive approach to improving medication reconciliation in the hospital and other health care settings. While this project is ongoing, of those trained, staff from 59 provider settings have used the toolkit to make changes to their medication reconciliation process. For example, staff from 34 hospitals have designed a document, called the "One Source of Truth" to list all medications a patient is taking, as recommended by the toolkit.

Return to Contents
Proceed to Next Section

Current as of February 2011
Internet Citation: AHRQ Annual Highlights, 2010 (continued). February 2011. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/news/newsroom/highlights/highlt10c.html