Since the launch of the 100,000 Lives Campaign by the Institute for Healthcare Improvement (IHI), preventing medical adverse events to reduce avoidable mortality has emerged as a central focus for health care providers, institutions, regulators, insurance companies, and patients. Evidence-based interventions targeting the 6 interventions in the campaign have been associated with a reduction in preventable hospital deaths in the United States. The generalizability of the IHI’s campaign to the pediatric population is only partly applicable. Pediatric experiences with rapid response teams and preventing central-line infections parallel the published experience of adults, with promise to significantly reduce preventable pediatric mortality.
Key Points
- •
Mortality in children can be significantly reduced through the prevention of avoidable adverse events.
- •
The Institute for Healthcare Improvement’s 100,000 Lives Campaign to reduce adverse events associated with mortality in the adult population is only partly applicable to the pediatric population.
- •
Innovative research, such as reducing diagnostic errors and errors related to communication, shows early promise but remains largely understudied.
Since the publication of 2 landmark reports on medical errors and health care quality by the Institute of Medicine (IOM), preventing adverse events has become a national priority. In To Err is Human: Building a Safer Healthcare System , the IOM concluded that between 44,000 and 98,000 Americans die each year as a result of medical errors. For comparison, even when using the lower estimate, deaths attributable to medical errors exceed those from motor vehicle accidents (43,458), breast cancer (42,297), and illicit drug use (17,000). Subsequent reports with improved detection methodologies have suggested that the incidence of medically related adverse events may be an underestimation. These studies, and several others with similar findings, effectively launched the patient safety movement in the United States and around the world.
In response, the Institute for Healthcare Improvement (IHI) embarked on a nationwide venture called the “100,000 Lives Campaign” to significantly reduce mortality related to avoidable adverse events in American hospitals. The initiative articulated an agenda that uses an evidence-based approach to operationally standardize care in 6 clinical areas among the enrolled estimated 2300 of the United States’ 6000 hospitals. Although not without controversy regarding the campaign’s interpretation of results, the IHI announced that, after 18 months, the campaign had successfully reached its goal by nationally preventing more than 120,000 avoidable deaths in hospitals across the United States during the 18-month intervention time frame. However, important statistical and methodological uncertainty has obliged IHI leadership to be skeptical in attributing mortality reduction to campaign efforts alone.
Given the unique circumstantial and physiologic differences between pediatric and adult resuscitation and mortality, the “natural fit” of the IHI’s 6 interventions for pediatrics is uncertain. Nearly a decade after the results were published, the specific impact of the IHI’s 100,000 Lives Campaign on the pediatric population has not been systematically summarized or studied. This article analyzes the impact of the individual elements of the IHI’s 100,000 Lives Campaign as it relates to pediatric caregivers and patients, as well as discusses additional interventions with evidence of their potential to systematically reduce mortality related to adverse events in pediatrics ( Table 1 ).
| Intervention | Relative Strength for Mortality Reduction in Children | Relative Strength of Evidence in Children | |
|---|---|---|---|
| IHI 100,000 Lives Campaign | Rapid response team | Strong | Good evidence |
| ADE prevention | Intermediate | Good evidence | |
| CLABSI prevention | Strong | Good evidence | |
| Evidence-based care for acute myocardial infarction | Not applicable | Not applicable | |
| Surgical site infection prevention | Weak | Limited studies available | |
| Ventilator-associated pneumonia prevention | Weak | Limited studies available | |
| Other | CPOE & clinical decision support systems | Unknown | Limited studies available |
| Diagnosis error prevention | Unknown | Minimal evidence | |
| Standardized communication | Unknown | Minimal evidence | |
| Raising culture of safety | Unknown | Minimal evidence |
The 6 interventions of the campaign platform
The enrolled institutions committed to reduce the number of avoidable deaths by implementing one or more of the following 6 interventions that comprise the IHI’s campaign platform: (1) Deploy rapid response teams (RRTs) to bring skilled resources to the bedside of any patient at the first sign of decline and potentially facilitating transfer to an intensive care unit (ICU) where rapid resuscitation efforts are more likely to be successful. (2) Prevent adverse drug events (ADEs) by reconciling medications at all transitions in care. (3) Deliver reliable evidence-based care for acute myocardial infarction (given the pediatric focus of this article, this intervention will not be discussed). (4) Prevent central-line–associated bloodstream infections (CLABSIs) by applying a bundle of evidence-based practices based on guidelines issued by the Centers for Disease Control and Prevention (CDC). (5) Prevent surgical site infections (SSIs) by implementing a bundle of evidence-based practices issued by the CDC. (6) Prevent ventilator-associated pneumonia (VAP) by implementing a bundle of evidence-based practices including a 30° elevation of the head of the bed, daily sedation “vacations,” and daily readiness-to-wean assessments. These interventions were largely chosen based on wide acceptance and consensus in the medical community, strong evidentiary support in the adult medical literature, and relative ease of institutional implementation to address problems that are common and widespread.
Rapid Response Teams
Survival rates of all pediatric inpatients after cardiopulmonary arrest are poor, with just 34% surviving 24 hours, 27% surviving to discharge, and 15% surviving 1 year. Surviving a code outside of the pediatric ICU is similarly improbable, with just 33% of pediatric patients with an arrest outside of the ICU surviving to discharge. The concept of an ICU-trained multidisciplinary team to respond to “prearresting” deteriorating patients not in the ICU was developed in response to research that revealed that both adult and pediatric patients often have evidence of physiologic decline several hours before cardiopulmonary arrest. Although the concept of and rationale for RRT programs are sound, the recommendation made by the IHI to implement RRTs as a strategy to decrease nationwide in-hospital mortality has not been without debate. Systematic reviews have cited mixed, contradictory, or inconclusive evidence as arguments for further research before decreeing RRTs standard care. To date, there continues to be controversy on the effectiveness of RRTs on adult hospital-wide mortality.
In pediatrics, there are a few studies that describe the effect of RRTs on mortality and other patient outcomes. The first studies initially did not demonstrate a meaningful decrease in total hospital mortality after implementation of an RRT. Both study designs were limited in their short time frame of the postintervention period (12 months and 8 months, respectively), potentially contributing to an underestimation of RRT programs on mortality by premature analysis. Subsequently, however, Sharek and colleagues, demonstrated in a cohort study that the implementation of an RRT was statistically associated with an 18% decrease in hospital-wide mortality rate, bringing the preintervention monthly mortality rate of 1.01 deaths per 100 discharges to 0.83 deaths per 100 discharges. In the same study, the rate of codes outside of the ICU per 1000 eligible patient-days decreased by 71.2% after RRT implementation. The study controlled for potential bias by secular trends by using longer time frames for preintervention and postintervention periods (18-month postintervention period) and demonstrated similarities in characteristics and case mix complexity between the control and intervention populations. Since then, ensuing studies have shown consistent findings. In a cohort study by Tibballs and colleagues, implementation of a medical emergency team in a free-standing children’s hospital was associated with a significant decline in total hospital deaths from 4.38 to 2.8 per 1000 admissions (risk ratio [RR], 0.65; 95% confidence interval [CI], 0.57–0.75; P <.0001). In the same study, survivability from cardiac arrest increased from 7 of 20 patients to 17 of 23 (RR, 2.11; 95% CI, 1.11–4.02; P = .01). Hanson and colleagues demonstrated a reduction in the rate of non-ICU cardiac arrests with a risk reduction of 0.35 (95% CI, 0–1.24; P = .125) associated with RRT implementation. In the first large multicenter study of RRTs in pediatrics, Kotsakis and colleagues developed, implemented, and examined a standardized rapid response system across 4 pediatric academic centers in Ontario, Canada. Using a prospective observational design, analysis of more than 110,000 hospital admissions and more than 14,000 pediatric ICU admissions during a 4-year study period revealed significant decreases in code blue events (defined as actual and near cardiopulmonary arrests) with a risk reduction ratio of 0.71 (95% CI, 0.61–0.83; P <.0001). The study also demonstrated a significant decrease in mortality rate in patients who were readmitted to the ICU with a risk reduction ratio of 0.43 (95% CI, 0.17–0.99; P <.05).
There are several potential reasons for discrepancies between outcomes when implementing RRTs in adult versus pediatric inpatient populations. First, the cause and pathophysiology of pediatric respiratory and cardiopulmonary arrests differ from that of adults. Consequently, the parameters for activating an RRT are likely to reflect the epidemiologic differences between adult and pediatric prearrest physiologies. Second, unique adult protocols and time-sensitive response teams for unique adult crisis situations such as coronary insufficiency syndromes, heart failure, and stroke may contribute to the dilution of a mortality benefit of adult RRTs. Finally, higher prevalence of do-not-resuscitate orders in adults compared with pediatrics may be an important confounder for interpreting adult RRT outcomes. With the advent of better methodologies to evaluate rapid response systems and improved sensitivity in early warning indicators, there is potential for RRTs to be increasingly valuable in preventing pediatric mortality.
Adverse Drug Events
The available data on the incidence of medication errors suggest that prescribing, dispensing, and administering medications are error-prone processes in pediatrics. The published estimates of medication errors causing harm (ADE rates) in pediatrics are few compared with those of adults. There is a demonstrable relationship between ADEs and mortality in adults, whereas the evidence of ADEs as an important contributor to pediatric mortality is sparse. Kaushal reported ADE rates in children in inpatient wards at 2 urban teaching hospitals to be 2.3 per 100 admissions (26 events), of which 5 (19%) were classified as preventable, with 2 (10%) categorized as fatal or life threatening. Holdsworth measured an ADE rate in pediatric inpatients of 6 per 100 admissions (76 events), with 46 (61%) classified as preventable and 8 (11%) classified as life threatening. Takata and colleagues, using the rigorous trigger tool methodology pioneered by the IHI, identified rates of 11.1 ADEs per 100 admissions in 14 children’s hospitals across the United States. Of the 107 ADEs identified in 960 total charts, none were classified as fatal or life threatening. Overall, at present, there is little evidence suggesting that ADEs are substantial contributors to mortality in the hospitalized pediatric population.
The prevention of ADEs and implementation of computerized order entry systems (CPOEs) have merged as parallel priorities for many institutions. Most evaluations of CPOEs and ADEs in adult and pediatric settings have focused on measuring intermediate outcomes (eg, prescribing errors, rule violations, and compliance) until Holdsworth demonstrated that the implementation of a CPOE was associated with a risk reduction of preventable ADEs of 0.56 (95% CI, 0.34–0.91) in the inpatient pediatric population. The impact on mortality was not studied. In the first study to demonstrate reductions in pediatric mortality rates with implementation of a CPOE, Longhurst and colleagues reported a 20% decrease (1.008–0.716 deaths per 100 discharges per month) in the mean monthly adjusted mortality rate after CPOE implementation (95% CI, 0.8%–40%; P = .03) during 18 months. In this report, introduction of the CPOE was shown to improve medication turnaround times by 19% and 5% in the solid organ transplant unit and pediatric ICU, respectively, a finding that other investigators have previously argued as the most proximate cause of CPOE-related mortality changes. The investigators noted that ADE rates were low before and after the CPOE implementation and likely not an important contributor to the shift in mortality.
Central-Line–Associated Bloodstream Infections
CLABSIs are common health care-associated infections and result in increases in length of stay, morbidity, and mortality in adult and pediatric patients. Each year in the United States, it is estimated that CLABSIs result in approximately 31,000 deaths. Wenzel and Edmond calculated that nosocomial bloodstream infections represented the eighth leading cause of death in the United States. In response, the CDC has issued evidence-based guidelines that articulate many essentials of care shown to reduce the risk of CLABSIs. Preventing CLABSIs has been the focus of substantial pediatric research and quality improvement efforts. In a New York quality improvement collaborative among 18 neonatal and pediatric ICUs, Schulman and colleagues demonstrated a 67% statewide decline in CLABSI rates (6.4–2.1/1000 central-line days, P <.0005) with the standardization of central-line insertion and maintenance practice. In a similar California collaborative of 13 regional neonatal ICUs, Wirtschafter and colleagues demonstrated a 25% reduction in CLABSIs from 4.32 to 3.22 per 1000 central-line days. Bizzarro and colleagues used the evidenced-based recommendations by the CDC to reduce CLABSIs from 8.4 to 1.7 per 1000 central-line days (adjusted rate ratio, 0.19 [95% CI, 0.08–0.45]) in a neonatal ICU. In 2010, the National Association of Children’s Hospitals and Related Institutions (NACHRI) structured a collaboration of 29 pediatric ICUs and described the reduction of CLABSIs by 43% (5.4–3.1/1000 central-line days) through the collective use of insertion and maintenance bundles. Finally, in an extended 3-year postintervention report, the NACHRI quality transformation effort realized a decrease in CLABSI rate of 56% or 5.2 to 2.3 CLABSIs per 1000 central-line days (rate ratio, 0.44 [95% CI, 0.37–0.53]; P <.0001). The authors estimated that greater than 900 CLABSIs were prevented, avoiding greater than 100 mortalities in 29 pediatric ICUs during the 3-year study period, and more than 65 pediatric ICUs have since joined the collaborative. CLABSIs have captured the nation as a preventable complication that contributes prominently to pediatric mortality. Future efforts should focus on continued implementation of the best-practice bundles of care in central-line management and further research to improve bundle elements that can prevent CLABSIs altogether.
Surgical Site Infections
In a 1999 report by the National Nosocomial Infections Surveillance (NNIS) system, SSIs were considered the third most frequently reported nosocomial infections accounting for 14% to 16% of all nosocomial infections among hospitalized adults and children. Moreover, they were the most common nosocomial infection among surgical patients and deaths among patients with SSIs; 77% were related to infection, and the majority (93%) were serious infections involving organs or spaces accessed during surgery. However, most of the epidemiologic reviews from broad national surveillance systems are more than a decade old and do not capture the impact resulting from improved care practices since 2002. There are only a few US studies published that specifically address the problem of SSIs in children and even fewer pediatric studies that attempt to quantify the relationship between pediatric SSI and mortality. Published SSI rates after sternotomy for pediatric cardiac surgery range from 2.3% to 5%. In a prospective multicenter study by Horwitz and colleagues, the incidence of wound infection in a general pediatric surgical population was 4.4% in the 30 postoperative days among 846 patients within 3 pediatric institutions in Texas. Ryckman and colleagues demonstrated reductions in pediatric SSI rates from 1.5 to 0.54 per 100 procedure-days (64% reduction) in a single center study using established adult evidence-based bundles with pediatric-based modifications. However, influence on clinical outcome, including mortality, was not analyzed. Since then, many studies have better characterized pediatric SSIs but are limited in design with only a few controlled studies. Most studies use a comparative before–after structure with inadequate control for secular trends. In short, it remains unclear whether successful SSI prevention efforts in the pediatric inpatient population has and/or will result in significant impact on mortality.
Ventilator-Associated Pneumonias
In adults, VAP is a frequent hospital acquired infection that is considered an important quality-of-care indicator based on multiple reports that link it to attributable mortality. Although the accuracy of the association has been controversial, systematic reviews have estimated the attributable mortality rate to range between 3% and 17%. Because VAP is considered the most common nosocomial infection in critically ill adults, preventing it has emerged as an important component to the IHI’s campaign. VAP is considered the second most common nosocomial infection in pediatric ICUs in the United States. In 2004, the NNIS system of the CDC reported a mean VAP rate of 2.9/1000 ventilator days for 52 participating US pediatric ICUs. Unfortunately, the characteristics, associated risk factors, and outcomes of pediatric VAP are less established compared with the adult population. In a 2009 prospective, observational study, Srinivasan and colleagues reported VAP to be significantly associated with a greater need for mechanical ventilation, longer intensive care length of stay, and higher hospital costs. The investigators reported an association with increased absolute hospital mortality (10.5% vs 2.4%, P = .56), but given the limitation of the small sample size, they failed to demonstrate statistical significance. Apisarnthanarak and coworkers reported high rates of VAP demonstrated prospectively in a cohort study of neonates with birth weight less than 2000 g (6.5/1000 ventilator days for patients with an estimated gestational age [EGA] <28 weeks and 4/1000 ventilator days for EGA ≥28 weeks). In this study, VAP was an independent predictor of mortality (adjusted odds ratio, 3.4; 95% CI, 1.2–12.3) and prominent in extremely preterm neonates who stayed in the neonatal ICU for at least 30 days (RR, 8.0; 95% CI, 1.9–35; P <.001). Because of the paucity of epidemiologic and outcome studies on pediatric VAP, there are no reports, to date, that describe effective shifts in VAP rates or outcome using evidence-based guidelines issued by the CDC, the American Thoracic Society, and/or the Infectious Diseases Society of America.
The 6 interventions of the campaign platform
The enrolled institutions committed to reduce the number of avoidable deaths by implementing one or more of the following 6 interventions that comprise the IHI’s campaign platform: (1) Deploy rapid response teams (RRTs) to bring skilled resources to the bedside of any patient at the first sign of decline and potentially facilitating transfer to an intensive care unit (ICU) where rapid resuscitation efforts are more likely to be successful. (2) Prevent adverse drug events (ADEs) by reconciling medications at all transitions in care. (3) Deliver reliable evidence-based care for acute myocardial infarction (given the pediatric focus of this article, this intervention will not be discussed). (4) Prevent central-line–associated bloodstream infections (CLABSIs) by applying a bundle of evidence-based practices based on guidelines issued by the Centers for Disease Control and Prevention (CDC). (5) Prevent surgical site infections (SSIs) by implementing a bundle of evidence-based practices issued by the CDC. (6) Prevent ventilator-associated pneumonia (VAP) by implementing a bundle of evidence-based practices including a 30° elevation of the head of the bed, daily sedation “vacations,” and daily readiness-to-wean assessments. These interventions were largely chosen based on wide acceptance and consensus in the medical community, strong evidentiary support in the adult medical literature, and relative ease of institutional implementation to address problems that are common and widespread.
Rapid Response Teams
Survival rates of all pediatric inpatients after cardiopulmonary arrest are poor, with just 34% surviving 24 hours, 27% surviving to discharge, and 15% surviving 1 year. Surviving a code outside of the pediatric ICU is similarly improbable, with just 33% of pediatric patients with an arrest outside of the ICU surviving to discharge. The concept of an ICU-trained multidisciplinary team to respond to “prearresting” deteriorating patients not in the ICU was developed in response to research that revealed that both adult and pediatric patients often have evidence of physiologic decline several hours before cardiopulmonary arrest. Although the concept of and rationale for RRT programs are sound, the recommendation made by the IHI to implement RRTs as a strategy to decrease nationwide in-hospital mortality has not been without debate. Systematic reviews have cited mixed, contradictory, or inconclusive evidence as arguments for further research before decreeing RRTs standard care. To date, there continues to be controversy on the effectiveness of RRTs on adult hospital-wide mortality.
In pediatrics, there are a few studies that describe the effect of RRTs on mortality and other patient outcomes. The first studies initially did not demonstrate a meaningful decrease in total hospital mortality after implementation of an RRT. Both study designs were limited in their short time frame of the postintervention period (12 months and 8 months, respectively), potentially contributing to an underestimation of RRT programs on mortality by premature analysis. Subsequently, however, Sharek and colleagues, demonstrated in a cohort study that the implementation of an RRT was statistically associated with an 18% decrease in hospital-wide mortality rate, bringing the preintervention monthly mortality rate of 1.01 deaths per 100 discharges to 0.83 deaths per 100 discharges. In the same study, the rate of codes outside of the ICU per 1000 eligible patient-days decreased by 71.2% after RRT implementation. The study controlled for potential bias by secular trends by using longer time frames for preintervention and postintervention periods (18-month postintervention period) and demonstrated similarities in characteristics and case mix complexity between the control and intervention populations. Since then, ensuing studies have shown consistent findings. In a cohort study by Tibballs and colleagues, implementation of a medical emergency team in a free-standing children’s hospital was associated with a significant decline in total hospital deaths from 4.38 to 2.8 per 1000 admissions (risk ratio [RR], 0.65; 95% confidence interval [CI], 0.57–0.75; P <.0001). In the same study, survivability from cardiac arrest increased from 7 of 20 patients to 17 of 23 (RR, 2.11; 95% CI, 1.11–4.02; P = .01). Hanson and colleagues demonstrated a reduction in the rate of non-ICU cardiac arrests with a risk reduction of 0.35 (95% CI, 0–1.24; P = .125) associated with RRT implementation. In the first large multicenter study of RRTs in pediatrics, Kotsakis and colleagues developed, implemented, and examined a standardized rapid response system across 4 pediatric academic centers in Ontario, Canada. Using a prospective observational design, analysis of more than 110,000 hospital admissions and more than 14,000 pediatric ICU admissions during a 4-year study period revealed significant decreases in code blue events (defined as actual and near cardiopulmonary arrests) with a risk reduction ratio of 0.71 (95% CI, 0.61–0.83; P <.0001). The study also demonstrated a significant decrease in mortality rate in patients who were readmitted to the ICU with a risk reduction ratio of 0.43 (95% CI, 0.17–0.99; P <.05).
There are several potential reasons for discrepancies between outcomes when implementing RRTs in adult versus pediatric inpatient populations. First, the cause and pathophysiology of pediatric respiratory and cardiopulmonary arrests differ from that of adults. Consequently, the parameters for activating an RRT are likely to reflect the epidemiologic differences between adult and pediatric prearrest physiologies. Second, unique adult protocols and time-sensitive response teams for unique adult crisis situations such as coronary insufficiency syndromes, heart failure, and stroke may contribute to the dilution of a mortality benefit of adult RRTs. Finally, higher prevalence of do-not-resuscitate orders in adults compared with pediatrics may be an important confounder for interpreting adult RRT outcomes. With the advent of better methodologies to evaluate rapid response systems and improved sensitivity in early warning indicators, there is potential for RRTs to be increasingly valuable in preventing pediatric mortality.
Adverse Drug Events
The available data on the incidence of medication errors suggest that prescribing, dispensing, and administering medications are error-prone processes in pediatrics. The published estimates of medication errors causing harm (ADE rates) in pediatrics are few compared with those of adults. There is a demonstrable relationship between ADEs and mortality in adults, whereas the evidence of ADEs as an important contributor to pediatric mortality is sparse. Kaushal reported ADE rates in children in inpatient wards at 2 urban teaching hospitals to be 2.3 per 100 admissions (26 events), of which 5 (19%) were classified as preventable, with 2 (10%) categorized as fatal or life threatening. Holdsworth measured an ADE rate in pediatric inpatients of 6 per 100 admissions (76 events), with 46 (61%) classified as preventable and 8 (11%) classified as life threatening. Takata and colleagues, using the rigorous trigger tool methodology pioneered by the IHI, identified rates of 11.1 ADEs per 100 admissions in 14 children’s hospitals across the United States. Of the 107 ADEs identified in 960 total charts, none were classified as fatal or life threatening. Overall, at present, there is little evidence suggesting that ADEs are substantial contributors to mortality in the hospitalized pediatric population.
The prevention of ADEs and implementation of computerized order entry systems (CPOEs) have merged as parallel priorities for many institutions. Most evaluations of CPOEs and ADEs in adult and pediatric settings have focused on measuring intermediate outcomes (eg, prescribing errors, rule violations, and compliance) until Holdsworth demonstrated that the implementation of a CPOE was associated with a risk reduction of preventable ADEs of 0.56 (95% CI, 0.34–0.91) in the inpatient pediatric population. The impact on mortality was not studied. In the first study to demonstrate reductions in pediatric mortality rates with implementation of a CPOE, Longhurst and colleagues reported a 20% decrease (1.008–0.716 deaths per 100 discharges per month) in the mean monthly adjusted mortality rate after CPOE implementation (95% CI, 0.8%–40%; P = .03) during 18 months. In this report, introduction of the CPOE was shown to improve medication turnaround times by 19% and 5% in the solid organ transplant unit and pediatric ICU, respectively, a finding that other investigators have previously argued as the most proximate cause of CPOE-related mortality changes. The investigators noted that ADE rates were low before and after the CPOE implementation and likely not an important contributor to the shift in mortality.
Central-Line–Associated Bloodstream Infections
CLABSIs are common health care-associated infections and result in increases in length of stay, morbidity, and mortality in adult and pediatric patients. Each year in the United States, it is estimated that CLABSIs result in approximately 31,000 deaths. Wenzel and Edmond calculated that nosocomial bloodstream infections represented the eighth leading cause of death in the United States. In response, the CDC has issued evidence-based guidelines that articulate many essentials of care shown to reduce the risk of CLABSIs. Preventing CLABSIs has been the focus of substantial pediatric research and quality improvement efforts. In a New York quality improvement collaborative among 18 neonatal and pediatric ICUs, Schulman and colleagues demonstrated a 67% statewide decline in CLABSI rates (6.4–2.1/1000 central-line days, P <.0005) with the standardization of central-line insertion and maintenance practice. In a similar California collaborative of 13 regional neonatal ICUs, Wirtschafter and colleagues demonstrated a 25% reduction in CLABSIs from 4.32 to 3.22 per 1000 central-line days. Bizzarro and colleagues used the evidenced-based recommendations by the CDC to reduce CLABSIs from 8.4 to 1.7 per 1000 central-line days (adjusted rate ratio, 0.19 [95% CI, 0.08–0.45]) in a neonatal ICU. In 2010, the National Association of Children’s Hospitals and Related Institutions (NACHRI) structured a collaboration of 29 pediatric ICUs and described the reduction of CLABSIs by 43% (5.4–3.1/1000 central-line days) through the collective use of insertion and maintenance bundles. Finally, in an extended 3-year postintervention report, the NACHRI quality transformation effort realized a decrease in CLABSI rate of 56% or 5.2 to 2.3 CLABSIs per 1000 central-line days (rate ratio, 0.44 [95% CI, 0.37–0.53]; P <.0001). The authors estimated that greater than 900 CLABSIs were prevented, avoiding greater than 100 mortalities in 29 pediatric ICUs during the 3-year study period, and more than 65 pediatric ICUs have since joined the collaborative. CLABSIs have captured the nation as a preventable complication that contributes prominently to pediatric mortality. Future efforts should focus on continued implementation of the best-practice bundles of care in central-line management and further research to improve bundle elements that can prevent CLABSIs altogether.
Surgical Site Infections
In a 1999 report by the National Nosocomial Infections Surveillance (NNIS) system, SSIs were considered the third most frequently reported nosocomial infections accounting for 14% to 16% of all nosocomial infections among hospitalized adults and children. Moreover, they were the most common nosocomial infection among surgical patients and deaths among patients with SSIs; 77% were related to infection, and the majority (93%) were serious infections involving organs or spaces accessed during surgery. However, most of the epidemiologic reviews from broad national surveillance systems are more than a decade old and do not capture the impact resulting from improved care practices since 2002. There are only a few US studies published that specifically address the problem of SSIs in children and even fewer pediatric studies that attempt to quantify the relationship between pediatric SSI and mortality. Published SSI rates after sternotomy for pediatric cardiac surgery range from 2.3% to 5%. In a prospective multicenter study by Horwitz and colleagues, the incidence of wound infection in a general pediatric surgical population was 4.4% in the 30 postoperative days among 846 patients within 3 pediatric institutions in Texas. Ryckman and colleagues demonstrated reductions in pediatric SSI rates from 1.5 to 0.54 per 100 procedure-days (64% reduction) in a single center study using established adult evidence-based bundles with pediatric-based modifications. However, influence on clinical outcome, including mortality, was not analyzed. Since then, many studies have better characterized pediatric SSIs but are limited in design with only a few controlled studies. Most studies use a comparative before–after structure with inadequate control for secular trends. In short, it remains unclear whether successful SSI prevention efforts in the pediatric inpatient population has and/or will result in significant impact on mortality.
Ventilator-Associated Pneumonias
In adults, VAP is a frequent hospital acquired infection that is considered an important quality-of-care indicator based on multiple reports that link it to attributable mortality. Although the accuracy of the association has been controversial, systematic reviews have estimated the attributable mortality rate to range between 3% and 17%. Because VAP is considered the most common nosocomial infection in critically ill adults, preventing it has emerged as an important component to the IHI’s campaign. VAP is considered the second most common nosocomial infection in pediatric ICUs in the United States. In 2004, the NNIS system of the CDC reported a mean VAP rate of 2.9/1000 ventilator days for 52 participating US pediatric ICUs. Unfortunately, the characteristics, associated risk factors, and outcomes of pediatric VAP are less established compared with the adult population. In a 2009 prospective, observational study, Srinivasan and colleagues reported VAP to be significantly associated with a greater need for mechanical ventilation, longer intensive care length of stay, and higher hospital costs. The investigators reported an association with increased absolute hospital mortality (10.5% vs 2.4%, P = .56), but given the limitation of the small sample size, they failed to demonstrate statistical significance. Apisarnthanarak and coworkers reported high rates of VAP demonstrated prospectively in a cohort study of neonates with birth weight less than 2000 g (6.5/1000 ventilator days for patients with an estimated gestational age [EGA] <28 weeks and 4/1000 ventilator days for EGA ≥28 weeks). In this study, VAP was an independent predictor of mortality (adjusted odds ratio, 3.4; 95% CI, 1.2–12.3) and prominent in extremely preterm neonates who stayed in the neonatal ICU for at least 30 days (RR, 8.0; 95% CI, 1.9–35; P <.001). Because of the paucity of epidemiologic and outcome studies on pediatric VAP, there are no reports, to date, that describe effective shifts in VAP rates or outcome using evidence-based guidelines issued by the CDC, the American Thoracic Society, and/or the Infectious Diseases Society of America.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
