Spotlight on Quality Improvement in Infection Prevention

Healthcare-associated infections (HAI) are at the center of patient safety initiatives and a focus of wide-spread attention from healthcare organizations, consumers, insurers, legislators, and the media at regional and national levels.  This focused attention has led to the creation of several national healthcare-associated infection prevention initiatives which offer support, guidance, recommendations, and strategies for reducing HAIs both to infection prevention professionals as well as to educate the public.
 
Infection prevention is a top priority in all Vermont hospitals.  Vermont hospitals have implemented evidence-based “best practices” and utilize national guidelines for reducing and preventing HAI.  Every Vermont hospital has implemented and adheres to the Institute for Healthcare Improvement’s (IHI) recommendations for prevention of central line-associated bloodstream infections.  Similarly, most hospitals have implemented all of the routine recommendations from the Centers for Disease Control (CDC) for preventing multi-drug resistant organisms; and twelve of the fourteen hospitals also have policies and methods in place to implement more intensive recommendations if needed.
 
Several hospitals have embarked on quality improvement activities focused on reducing and preventing HAIs.  One notable result has been a 62% reduction in central line-associated bloodstream infections in medical intensive care units.  
 
Section Outline [click on a section to jump ahead] 

• Why focus on healthcare-associated infections?
• What are the most common healthcare-associated infections?
• What are Vermont hospitals doing to prevent healthcare-associated infections?
• How do infection rates in Vermont compare?

Why focus on healthcare-associated infections? [back to top]
The Centers for Disease Control and Prevention estimates that, nationally, 5-10% of all hospitalized patients develop healthcare-associated infections (HAI) which equates to approximately $6 billion in attributable health care spending (an average of $20,000 per hospitalization)[1][2].  
 
HAIs are a major cause of morbidity and mortality and fall within the category of preventable medical errors.  The latter has placed healthcare-associated infections at the center of patient safety initiatives and made it the focus of wide-spread attention from healthcare organizations, consumers, insurers, legislators, and the media at regional and national levels.  This focused attention has lead to the creation of several national healthcare-associated infection prevention initiatives which offer support, guidance, recommendations, and strategies for reducing HAIs both to infection prevention professionals as well as to educate  the public.
 
What are the most common HAIs? [back to top]
A recent CDC study reports the most common healthcare-associated infections are urinary tract infections (32%), surgical site infections (22%), pneumonias (15%), and bloodstream infections (14%).  These estimates are based on three national data sources: 1) National Nosocomial Infections Surveillance (NNIS) system, a voluntary network of U.S. hospitals; 2) National Hospital Discharge Survey (NHDS), an annual survey of characteristics of inpatients discharged from U.S. hospitals; and 3) American Hospital Association (AHA) annual survey of hospitals and their characteristics[3]. 

Frequency of occurrence, however, does not directly translate to greater overall costs for treatment, to impairment, or to other consequences. In terms of severity of infection and having the greatest financial impact, the most burdensome infections are central line-associated bloodstream infections (CLABSI), surgical site infections (SSI), and ventilator-associated pneumonia (VAP).  Despite being the most common in occurrence, urinary tract infections (UTI) have relatively minor impact.  Conversely, MRSA (Methicillin Resistant Staphylococcus Aureus) infections caused from a type of bacteria that is resistant to certain antibiotics and Clostridium difficile (C.diff) infections, which can cause diarrhea and more serious intestinal conditions, occur much less frequently but the magnitude of the burden, both in terms of severity of illness and financial impact is far greater.

What are Vermont hospitals doing to prevent HAIs? [back to top]
Infection prevention has been a top priority in all Vermont hospitals for many years.  In 2003 the legislature passed legislation (Act 53) mandating that hospitals publish information on cost and quality measures and directed oversight of the reporting to the Department of Banking, Insurance, Securities, and Health Care Administration (BISHCA).  In 2006 the mandate on public reporting was amended to include the publication of HAI.  BISHCA contracts with VPQ to assist the hospitals with these reporting requirements.  As such, VPQ is a resource for the hospitals with regard to navigating the reporting system, NHSN (National Healthcare Safety Network), provides data analysis, and produces reports for publication for various infection rates.  Vermont hospitals also self-report adherence to IHI-recommended best practices for preventing central line infections and adherence to CDC-recommended best practices for prevention and control of multi-drug resistant organisms for which VPQ provides data collection, collation, and report preparation.      
 
Adherence to national guidelines and established best practices
Vermont hospitals have implemented evidence-based “best practices” and utilize national guidelines for reducing and preventing HAI.  These guidelines were developed by the Centers for Disease Control (CDC), Institute for Healthcare Improvement (IHI), and the Society for Healthcare Epidemiology of America (SHEA) and offer strategies for preventing, reducing, and controlling HAI[4].  Because healthcare-associated infection is broadly defined, the approach taken by a hospital to implement these guidelines will depend both on the hospital as well as the HAI.        
 
Central Line-Associated Bloodstream Infections
IHI-recommended best practices for preventing central line-associated bloodstream infections include five important prevention steps: 
1)      Hand hygiene
2)      Using barriers
3)      Using appropriate antiseptic
4)      Using the best site for the central line
5)      Reviewing the necessity of the central line every day
 
Every hospital in Vermont reports that they adhere to each of these five steps.
 
Multi-drug Resistant Organisms
CDC-recommended best practices for preventing multi-drug resistant organisms are complex.  Vermont hospitals publicly report the degree to which they have implemented CDC’s routine recommendations (Tier 1).  Additionally, they indicate whether they have policies and methods in place to implement more intensive recommendations, should a more aggressive approach be needed.
 
Most hospitals have implemented all of the routine recommendations from CDC for preventing multi-drug resistant organisms; and twelve of the fourteen hospitals also have policies and methods in place to implement more intensive recommendations if needed.  
 
Quality improvement activities related to HAI
Several hospitals have implemented strategies for preventing and controlling HAI.  Infection prevention specialists at several hospitals have developed curriculums to educate and train bedside nurses on best practices in infection prevention.  One example of targeted HAI quality improvement has resulted in a 64% reduction in central line-associated bloodstream infections in the medical ICU unit at Fletcher Allen Health Care.  This was accomplished primarily through the implementation of a “checklist” for following best practice for inserting central lines as well as infection prevention staff participation in rounds with new residents.  
 
How do infection rates in Vermont compare? [back to top]
As mentioned above, the legislature expanded the public reporting mandate on Vermont hospitals to include HAI in 2006.  Currently hospitals use NHSN to report infection rates on Central Line-Associated Bloodstream Infections, abdominal hysterectomy surgeries, knee replacement surgeries, and hip replacement surgeries.  NHSN provides standardized reporting criteria and definitions, risk adjustment, and comparison infection rates to a national NHSN average.  Reports providing hospital-level infection rates are published in the Act 53 Hospital Report Cards[5]. 

Central Line Associated Bloodstream Infections (CLABSI)
Vermont began reporting CLABSI rates in November 2006 in three types of Intensive Care Units (Medical ICU, Surgical ICU, and Combined Medical/Surgical ICU).  Over this cumulative time period, the CLABSI rate in Vermont’s medical ICU was significantly lower (p<.05) than the National NHSN rate (1.4 infections per 1,000 central line days vs. 2.4 infections per 1,000 central line days).  Vermont’s CLABSI rate in combined medical/surgical ICUs was also significantly lower (p<.05) than the National NHSN rate (0.3 infections per 1,000 central line days vs. 1.5 infections per 1,000 central line days).   The CLABSI rate for Vermont’s surgical ICU over this time period was 1.9 infections per 1,000 central line days and not statistically different than other surgical ICUs reporting to NHSN (the National NHSN rate for surgical ICUs was 2.3 infections per 1,000 central line days).
 
The following graph (Figure 1) presents one-year snapshots of CLABSI rates reported at quarterly intervals for three types of ICUs in Vermont.  As illustrated in the graph below, there has been a 62% reduction in CLABSIs in the medical ICU.  This reduction coincides and is largely attributable to targeted quality improvement efforts by Fletcher Allen Health Care.
 
Figure 1: Central Line-Associated Bloodstream Infections in three types of ICUs in Vermont

 
Infection after surgery
Vermont began reporting infection rates for abdominal hysterectomy in October 2007 and added knee replacements and hip replacements in April 2008.  Infections for these surgeries may have been identified during the initial hospital stay, or if patients returned for inpatient, emergency room, or observation care to the same hospital, a different Vermont hospital, or any reporting out-of-state hospital.  For each of the surgical procedures below, infection rates are presented for three risk categories.  The degree of risk for developing an infection is determined by three major factors: length of surgery; contamination of the wound; and an assessment of the overall physical status of the patient (using a standardized 1 to 5 scale).  
 
Abdominal Hysterectomy
Over the past year and a half, Vermont hospitals have performed 802 abdominal hysterectomy surgeries and have reported thirteen infections.  The measure is reported as the percentage of women who underwent abdominal hysterectomies who developed infections in their surgical wounds within 30 days of the surgery.  As figure 2 illustrates, infection rates for abdominal hysterectomy in Vermont are similar to other hospitals reporting to NHSN.
 
Figure 2: Infections Rates for Abdominal Hysterectomy surgeries in Vermont, October 2007 - March 2009

 
Hip Replacement
In the past year, Vermont hospitals have reported 872 hip replacement surgeries to the NHSN system; eleven of which have developed infection.  The measure reported as the percentage of people who underwent hip replacement surgeries who developed infections in their surgical wounds within one year of the surgery.  All hip replacement surgeries are included in this measure including total hip replacements, hip replacement revisions, and partial hip replacements.  As figure 3 illustrates, infection rates for hip replacements in Vermont are similar to other hospitals reporting to NHSN for these surgeries at low to moderate risk for developing an infection.  There have been 76 surgeries considered to be at higher risk for developing an infection and there have been no reportable infections to date.  On average, NHSN reporting hospitals report infections in three percent of surgeries considered to be higher risk.   It should be noted that hip replacement surgeries have a one year follow-up period during which a reportable infection may develop.  Many of the surgeries included in this section have an incomplete follow-up period so if a reportable infection develops the infection rates may change.
 
Figure 3: Infections Rates for Hip Replacement surgeries in Vermont, April 2008 - March 2009

 
Knee Replacement
In the past year, Vermont hospitals have reported 1,215 knee replacement surgeries to the NHSN system; six of which have developed infection.  The measure reported as the percentage of people who underwent knee replacement surgeries who developed infections in their surgical wounds within one year of the surgery.  All knee replacement surgeries are included in this measure including total knee replacements, knee replacement revisions, and partial knee replacements.  As figure 4 illustrates, infection rates for knee replacements in Vermont are similar to other hospitals reporting to NHSN for these surgeries.  There have been 477 surgeries considered to be at low risk for developing an infection and there have been no reportable infections to date.  It should be noted that knee replacement surgeries have a one year follow-up period during which a reportable infection may develop.  Many of the surgeries included in this section have an incomplete follow-up period so if a reportable infection develops the infection rates may change.
 
Figure 4: Infections Rates for Hip Replacement surgeries in Vermont, April 2008 - March 2009