Infection-Control Report Cards — Securing Patient Safety
http://www.100md.com
《新英格兰医药杂志》
For many of us, the specter of report cards conjures up anxiety dreams. Nevertheless, public report cards have infiltrated many industries — airlines and banking, for instance — and various levels of government, and health care appears to be next. The belief that hospitals reporting lower infection rates are safer and that informed consumers will obtain safer care has driven many U.S. states to consider legislation requiring report cards on nosocomial infections.
Advocates of public reporting have been spurred on by the occurrence of nosocomial infections in 5 to 10 percent of hospitalized patients; increasing rates of antibiotic resistance; press coverage of cases of devastating nosocomial infection; and the view that many infections should be preventable. In little more than a year, 39 states introduced legislation and 6 states passed laws requiring disclosure of such infections to the state and, in most cases, to the public (see map). Although the movement is consumer-driven, health care providers share the goals of reducing infection rates and improving patient safety.1
Status of Legislation Requiring Public Disclosure of Rates of Nosocomial Infection, by State.
The public reporting of hospitals' performance is not new, although older systems paid little attention to nosocomial infections. For decades, several states have required that hospitals report death rates associated with cardiac surgery and other conditions. Although the effectiveness of these experiments has been mixed, reductions in the rate of death associated with coronary-artery bypass grafting in New York and Pennsylvania have been attributed, in part, to public reporting.
A key lesson from past reporting is the importance of "risk adjusting" the outcome data to account for essential differences in populations of patients. For example, surgical risk among 80-year-old patients with diabetes is greater than that among previously healthy 45-year-olds. We have also learned that we must select denominators carefully in order to avoid artificial inflation or deflation of rates; that sophisticated information technology is required; and that it can be difficult to define useful benchmarks, especially for small hospitals, so that reporting a trend for a particular hospital may provide more useful information than does comparing hospitals.
The Joint Commission on Accreditation of Healthcare Organizations and the Center for Medicare and Medicaid Services (CMS) have established Web-based public reports for participating hospitals. Both groups initially focused on process measures rather than outcome measures, a strategy that avoided the quagmire of risk adjustment. Neither group has yet addressed nosocomial infections, although the CMS is developing a Surgical Care Improvement Project that will probably report process and outcome measures. The Centers for Disease Control and Prevention (CDC), through its National Healthcare Safety Network, provides a mechanism for the confidential disclosure of nosocomial infections, allowing hospitals to compare their performance with that of others. Several states are considering statewide participation in the network as a reporting solution.
Current report cards focus on three types of common infections that are associated with high morbidity and mortality and that are likely to be controllable: infections associated with central venous catheters, surgical-site infections, and ventilator-associated pneumonia. We know that informing surgeons of their wound-infection rates can lead to reductions in those rates, presumably by reinforcing the use of sensible interventions (e.g., limiting the amount of movement in and out of operating rooms in order to lower bacterial loads). The most rigorous study of the impact of surveillance — the CDC's Study on the Efficacy of Nosocomial Infection Control, conducted in the 1970s — demonstrated a 32 percent decrease in infection rates in hospitals that implemented standardized surveillance methods and ongoing control measures and that had adequate infection-control staffing and expert physician epidemiologists.
Studies of infections related to devices, particularly vascular and urinary catheters and ventilators, have demonstrated the usefulness of key performance measures — such as site preparation and care and operator expertise for the control of venous catheter–related infections — and of "bundling" evidence-based prevention measures into comprehensive control programs.2 Studies have also highlighted certain difficulties involved in measuring some outcomes — for instance, the lack of an easily applied clinical definition of ventilator-associated pneumonia, the difficulty of tracking surgical-site infections in the community (now that the average postoperative stay is shorter than the incubation period for most wound infections), and the large confidence intervals around reported infection rates in smaller hospitals and for uncommon procedures.
In the light of these difficulties, will this type of public reporting result in the sort of improvements achieved by reporting wound-infection rates to surgeons? The answer is uncertain. Many experts recommend further study before states initiate costly, labor-intensive reporting programs, and a few states have passed laws that require such studies (see map). But more states have already embarked on the path of public reporting, and their legislators need advice urgently.
Recent recommendations from the CDC suggest that states focus on a combination of linked process and outcome measures.1 We support the study of reporting when possible. For states that have passed laws requiring public reporting, we suggest measuring rates that can be compared meaningfully, that should be tracked anyway, and whose reporting is most likely to lead to improved care. Such process measures include assessments of the timely administration of perioperative antibiotic prophylaxis, vascular-catheter insertion practices, and hand hygiene. Outcome measures include the rate of infections in the intensive care unit associated with central vascular catheters and the rate of reoperation or rehospitalization for surgical-site infections. Other measures for special settings could include the rates of nosocomial influenza, respiratory syncytial virus, rotavirus infection, and cases of diarrhea associated with Clostridium difficile. Infections caused by multidrug-resistant pathogens, such as methicillin-resistant Staphylococcus aureus and vancomycin-resistant enterococci, are also important, but because of laboratory logistics and the difficulty of verifying an infection's nosocomial origin, meaningful reporting is not yet possible.
States that have involved experts in health care epidemiology early in the development of laws have produced the most useful legislation. In addition, we favor phasing in reporting requirements incorporating the process measures that are the most readily obtained and compared, in order to allow hospitals and health departments to develop, refine, and validate data-collection systems. States must also consider the cost of these programs and work with hospital associations to develop realistic plans for support and funding.
To understand other relevant concerns, states should review the reasons given by Governor Arnold Schwarzenegger for his recent veto of California's legislation, which included problems with auditing and validating data, the need to redirect resources from successful programs, and extant mandates from national organizations that already scrutinize infection control. Most important, states must work with experts in health care communications and consumer reporting to define the sorts of rates that will tell patients what they need to know.
Report cards assessing nosocomial infections are a reality. Their success will require interdisciplinary collaboration, a greater commitment of resources to infection-prevention practices, and conspicuous inclusion of these efforts in patient-safety programs. Research is needed to identify the most meaningful metrics, determine the best way to report them, and assess whether such reporting improves patient safety. These challenges present unprecedented opportunities to improve patient care, if we can only put our anxieties to rest and move forward.
Source Information
Dr. Weinstein is the chair of infectious diseases at John H. Stroger (Cook County) Hospital and a professor of medicine at Rush University Medical Center, Chicago; Dr. Siegel is a professor of pediatrics at the University of Texas Southwestern Medical Center and chair of the Infection Control Committee at Children's Medical Center Dallas; and Dr. Brennan is a professor of medicine and chief medical officer at the University of Pennsylvania School of Medicine and Health System, Philadelphia.
References
McKibben L, Horan TC, Tokars JI, et al. Guidance on public reporting of healthcare-associated infections: recommendations of the Healthcare Infection Control Practices Advisory Committee. Am J Infect Control 2005;33:217-226.
Institute for Healthcare Improvement: 100k Lives Campaign. (Accessed June 30, 2005, at http://www.ihi.org/IHI/Programs/Campaign.)(Robert A. Weinstein, M.D.)
Advocates of public reporting have been spurred on by the occurrence of nosocomial infections in 5 to 10 percent of hospitalized patients; increasing rates of antibiotic resistance; press coverage of cases of devastating nosocomial infection; and the view that many infections should be preventable. In little more than a year, 39 states introduced legislation and 6 states passed laws requiring disclosure of such infections to the state and, in most cases, to the public (see map). Although the movement is consumer-driven, health care providers share the goals of reducing infection rates and improving patient safety.1
Status of Legislation Requiring Public Disclosure of Rates of Nosocomial Infection, by State.
The public reporting of hospitals' performance is not new, although older systems paid little attention to nosocomial infections. For decades, several states have required that hospitals report death rates associated with cardiac surgery and other conditions. Although the effectiveness of these experiments has been mixed, reductions in the rate of death associated with coronary-artery bypass grafting in New York and Pennsylvania have been attributed, in part, to public reporting.
A key lesson from past reporting is the importance of "risk adjusting" the outcome data to account for essential differences in populations of patients. For example, surgical risk among 80-year-old patients with diabetes is greater than that among previously healthy 45-year-olds. We have also learned that we must select denominators carefully in order to avoid artificial inflation or deflation of rates; that sophisticated information technology is required; and that it can be difficult to define useful benchmarks, especially for small hospitals, so that reporting a trend for a particular hospital may provide more useful information than does comparing hospitals.
The Joint Commission on Accreditation of Healthcare Organizations and the Center for Medicare and Medicaid Services (CMS) have established Web-based public reports for participating hospitals. Both groups initially focused on process measures rather than outcome measures, a strategy that avoided the quagmire of risk adjustment. Neither group has yet addressed nosocomial infections, although the CMS is developing a Surgical Care Improvement Project that will probably report process and outcome measures. The Centers for Disease Control and Prevention (CDC), through its National Healthcare Safety Network, provides a mechanism for the confidential disclosure of nosocomial infections, allowing hospitals to compare their performance with that of others. Several states are considering statewide participation in the network as a reporting solution.
Current report cards focus on three types of common infections that are associated with high morbidity and mortality and that are likely to be controllable: infections associated with central venous catheters, surgical-site infections, and ventilator-associated pneumonia. We know that informing surgeons of their wound-infection rates can lead to reductions in those rates, presumably by reinforcing the use of sensible interventions (e.g., limiting the amount of movement in and out of operating rooms in order to lower bacterial loads). The most rigorous study of the impact of surveillance — the CDC's Study on the Efficacy of Nosocomial Infection Control, conducted in the 1970s — demonstrated a 32 percent decrease in infection rates in hospitals that implemented standardized surveillance methods and ongoing control measures and that had adequate infection-control staffing and expert physician epidemiologists.
Studies of infections related to devices, particularly vascular and urinary catheters and ventilators, have demonstrated the usefulness of key performance measures — such as site preparation and care and operator expertise for the control of venous catheter–related infections — and of "bundling" evidence-based prevention measures into comprehensive control programs.2 Studies have also highlighted certain difficulties involved in measuring some outcomes — for instance, the lack of an easily applied clinical definition of ventilator-associated pneumonia, the difficulty of tracking surgical-site infections in the community (now that the average postoperative stay is shorter than the incubation period for most wound infections), and the large confidence intervals around reported infection rates in smaller hospitals and for uncommon procedures.
In the light of these difficulties, will this type of public reporting result in the sort of improvements achieved by reporting wound-infection rates to surgeons? The answer is uncertain. Many experts recommend further study before states initiate costly, labor-intensive reporting programs, and a few states have passed laws that require such studies (see map). But more states have already embarked on the path of public reporting, and their legislators need advice urgently.
Recent recommendations from the CDC suggest that states focus on a combination of linked process and outcome measures.1 We support the study of reporting when possible. For states that have passed laws requiring public reporting, we suggest measuring rates that can be compared meaningfully, that should be tracked anyway, and whose reporting is most likely to lead to improved care. Such process measures include assessments of the timely administration of perioperative antibiotic prophylaxis, vascular-catheter insertion practices, and hand hygiene. Outcome measures include the rate of infections in the intensive care unit associated with central vascular catheters and the rate of reoperation or rehospitalization for surgical-site infections. Other measures for special settings could include the rates of nosocomial influenza, respiratory syncytial virus, rotavirus infection, and cases of diarrhea associated with Clostridium difficile. Infections caused by multidrug-resistant pathogens, such as methicillin-resistant Staphylococcus aureus and vancomycin-resistant enterococci, are also important, but because of laboratory logistics and the difficulty of verifying an infection's nosocomial origin, meaningful reporting is not yet possible.
States that have involved experts in health care epidemiology early in the development of laws have produced the most useful legislation. In addition, we favor phasing in reporting requirements incorporating the process measures that are the most readily obtained and compared, in order to allow hospitals and health departments to develop, refine, and validate data-collection systems. States must also consider the cost of these programs and work with hospital associations to develop realistic plans for support and funding.
To understand other relevant concerns, states should review the reasons given by Governor Arnold Schwarzenegger for his recent veto of California's legislation, which included problems with auditing and validating data, the need to redirect resources from successful programs, and extant mandates from national organizations that already scrutinize infection control. Most important, states must work with experts in health care communications and consumer reporting to define the sorts of rates that will tell patients what they need to know.
Report cards assessing nosocomial infections are a reality. Their success will require interdisciplinary collaboration, a greater commitment of resources to infection-prevention practices, and conspicuous inclusion of these efforts in patient-safety programs. Research is needed to identify the most meaningful metrics, determine the best way to report them, and assess whether such reporting improves patient safety. These challenges present unprecedented opportunities to improve patient care, if we can only put our anxieties to rest and move forward.
Source Information
Dr. Weinstein is the chair of infectious diseases at John H. Stroger (Cook County) Hospital and a professor of medicine at Rush University Medical Center, Chicago; Dr. Siegel is a professor of pediatrics at the University of Texas Southwestern Medical Center and chair of the Infection Control Committee at Children's Medical Center Dallas; and Dr. Brennan is a professor of medicine and chief medical officer at the University of Pennsylvania School of Medicine and Health System, Philadelphia.
References
McKibben L, Horan TC, Tokars JI, et al. Guidance on public reporting of healthcare-associated infections: recommendations of the Healthcare Infection Control Practices Advisory Committee. Am J Infect Control 2005;33:217-226.
Institute for Healthcare Improvement: 100k Lives Campaign. (Accessed June 30, 2005, at http://www.ihi.org/IHI/Programs/Campaign.)(Robert A. Weinstein, M.D.)