Health care-associated infections (HAIs) have been recognized as a problem at least since the time of Ignac Semmelweis, an obstetrician who associated puerperal fever with lack of hand-washing among physicians and medical students in a Vienna General Hospital maternity ward in 1847.1 Despite improved understanding of risk factors for HAIs in the 21st century, it has been estimated that there were 1.7 million HAIs and 99 000 deaths from HAIs in US hospitals in 2002.2 These numbers may be underestimates because of underreporting, incomplete postdischarge surveillance, and other factors. In addition, the recent increase in Clostridium difficile infections (CDIs)3 and the ongoing emergence of infections caused by multidrug-resistant organisms4 might cause current rates to be even higher.

In recent years, there has been increasing interest in combating HAIs, beginning with placing them under public health surveillance. In 2006, only one state mandated HAI reporting; in 2012, a total of 28 states and the District of Columbia mandated reporting of at least 1 type of HAI.5 Almost all states with HAI reporting programs conduct HAI surveillance using the Centers for Disease Control and Prevention's National Healthcare Safety Network,6 which combines surveillance functions with analysis functions to assist facilities in focusing their quality improvement efforts to reduce HAIs. The Centers for Medicare & Medicaid Services also encourages reporting of certain HAIs via the National Healthcare Safety Network through its Inpatient Prospective Payment System and Value-Based Purchasing program.7,8 Since 2011, acute care hospitals have been reporting central line-associated blood stream infections (CLABSIs) in intensive care units (ICUs) to the Centers for Medicare & Medicaid Services, and since 2012, they have been reporting catheter-associated urinary tract infections, colon surgical site infections (SSIs), and abdominal hysterectomy SSIs. Reporting of CDIs and methicillin-resistant Staphylococcus aureus bacteremia from hospitals to the Centers for Medicare & Medicaid Services begins in 2013. One-time funding for state HAI programs was available through the American Recovery and Reinvestment Act of 2009, and funding for HAI surveillance infrastructure and prevention activities has been available through the Patient Protection and Affordable Care Act.6

In this issue of the Journal, Stricof and colleagues9 describe the implementation and findings of the New York State Department of Health's (NYSDOH) mandatory hospital-acquired infection public reporting program. From 2007 to 2010, the program documented a 34% reduction in adult/pediatric ICU CLABSIs, a 42% reduction in neonatal ICU CLABSIs, a 12% reduction in colon SSIs, and a 13% reduction in coronary artery bypass graft chest site SSIs. These reductions resulted in substantial cost savings estimated at $7.3 million to $29.4 million for CLABSIs and $7.9 million to $23.1 million for all reported SSI types.10

The value of HAI surveillance ultimately lies in its ability to reduce HAI rates. Although public reporting would not be expected to directly reduce HAI rates, it may prompt improvements in prevention practices that do. In New York State, HAI program staff develop relationships with hospital personnel in their geographic regions and provide technical assistance for both reporting procedures and prevention practices. When performed by HAI staff who are part of a cohesive HAI program, who are committed to reducing HAI rates, and who interact with health care facility infection preventionists regularly, audits of hospital reporting offer opportunities beyond validating reported rates. As noted by Stricof and colleagues, they may expose areas in need of improvement and stimulate discussions of prevention practices. In addition, New York State funds HAI prevention projects for collaborative groups of health care facilities. Projects have included methicillin-resistant S aureus screening policies in hospitals and CLABSI reduction practices in non-ICU settings, in peripherally inserted central catheters, and with chlorhexidine bathing.

One strength of the New York State program is its flexibility with regard to hospital reporting indicators. Reporting indicators are reviewed and selected by the NYSDOH in consultation with an external Technical Advisory Workgroup and may be changed annually in response to current priorities without changing legislation. In 2010, hospitalized CDIs were added to the list of mandated reporting indicators; 2011 hospital-onset rates increased by 3%.11 Although part of the increase might be explained by the increased use of more sensitive testing methods, the lack of improvement has stimulated further interest in CDI prevention activities throughout the state.

The NYSDOH hospital-acquired infection reporting program has recently improved access to its data by posting all publicly reportable data from 2008 to 2011 on the NYSDOH Web site (https://apps.nyhealth.gov/METRIX/main.action). Future activities may include development of criteria, such as infection burden, ease of surveillance, and cost, to determine which HAIs should be added to or removed from the public reporting program's list of selected indicators.

The NYSDOH hospital-acquired infection reporting program has several limitations. Although HAIs occur throughout the health care delivery system, the New York State program, like that of many states, focuses exclusively on acute care hospitals, and in the case of CLABSIs, only on ICUs. In New York State, the reasons for this emphasis include both legal and logistical issues, such as the fact that New York's statute calls only for reporting from hospitals, the availability of appropriate National Healthcare Safety Network modules, and the lack of infection preventionist staff in nonhospital settings.

In addition, the New York State program, as with most HAI reporting programs, is designed to monitor endemic rates rather than to detect outbreaks. The reporting lag is approximately 2 months, and the types of HAIs reported are limited to CLABSIs, SSIs, and CDIs. There is a general requirement in New York State for health care providers to report any infectious disease outbreak,12 regardless of etiology, whether the outbreak is HAI related, and whether it affects inpatients or outpatients. An electronic reporting system is available to report outbreaks in hospitals and long-term care facilities directly to the state.

Recent HAI-related multistate outbreaks affecting primarily outpatients have drawn attention to the importance of HAI surveillance and outbreak detection in diverse settings. For example, an outbreak of hepatitis C infections in New Hampshire alleged to have been caused by narcotic diversion was discovered when the involved hospital noticed an unusual cluster of 4 hepatitis C cases and reported the cluster to the New Hampshire Department of Health and Human Services.13 Similarly, an outbreak of fungal infections associated with methylprednisolone injections was detected after a clinician alerted the Tennessee Department of Health about a patient with a history of epidural steroid injection and no other explanation for the patient's fungal meningitis.14 The Tennessee Department of Health quickly initiated active surveillance and identified other patients with meningitis with the same exposure,15 leading to an investigation that revealed an outbreak with several hundred cases in multiple states.16

These examples illustrate the critical importance of collaboration between clinicians and public health authorities. Although neither of these outbreaks would have been detected through traditional HAI reporting programs, which are appropriately focused on common HAIs, such programs may foster stronger relationships between the medical and public health communities that ultimately result in more rapid identification of HAI clusters. Although the primary rationale behind HAI reporting programs is to define endemic rates of common HAIs as a first step to decreasing those rates, the indirect benefits of such programs should not be underestimated. State HAI programs should continue to develop and expand to reach their full prevention potential, and adequate funding should be allocated to these programs so that their value may be realized.

REFERENCES