The NEONATE score predicts freedom from interstage mortality or transplant in a modern cohort.

BACKGROUND: Derived from the National Pediatric Cardiology Quality Improvement Collaborative registry, the NEONATE risk score predicted freedom from interstage mortality or heart transplant for patients with single ventricle CHD and aortic arch hypoplasia discharged home following Stage 1 palliation. OBJECTIVES: We sought to validate the score in an external, modern cohort. METHODS: This was a retrospective cohort analysis of single ventricle CHD and aortic arch hypoplasia patients enrolled in the National Pediatric Cardiology Quality Improvement Collaborative Phase II registry from 2016 to 2020, who were discharged home after Stage 1 palliation. Points were allocated per the NEONATE score (Norwood type-Norwood/Blalock-Taussig shunt: 3, Hybrid: 12; extracorporeal membrane oxygenation post-op: 9, Opiates at discharge: 6, No Digoxin at discharge: 9, Arch Obstruction on discharge echo: 9, Tricuspid regurgitation ≥ moderate on discharge echo: 12; Extra oxygen plus ≥ moderate tricuspid regurgitation: 28). The composite primary endpoint was interstage mortality or heart transplant. RESULTS: In total, 1026 patients met inclusion criteria; 61 (6%) met the primary outcome. Interstage mortality occurred in 44 (4.3%) patients at a median of 129 (IQR 62,195) days, and 17 (1.7%) were referred for heart transplant at a 167 (114,199) days of life. The median NEONATE score was 0(0,9) in those who survived to Stage 2 palliation compared to 9(0,15) in those who experienced interstage mortality or heart transplant (p < 0.001). Applying a NEONATE score cut-off of 17 points that separated patients into low- and high-risk groups in the learning cohort provided 91% specificity, negative predictive value of 95%, and overall accuracy of 87% (85.4-89.5%). CONCLUSION: In a modern cohort of patients with single ventricle CHD and aortic arch hypoplasia, the NEONATE score remains useful at discharge post-Stage 1 palliation to predict freedom from interstage mortality or heart transplant.

Digoxin Associated With Greater Transplant-Free Survival in High- vs Low-Risk Interstage Patients.

BACKGROUND: Digoxin has been associated with reduced interstage mortality for patients with functional single ventricles with aortic hypoplasia or ductal-dependent systemic circulation. The NEONATE (type of stage 1 palliation operation, postoperative extracorporeal membrane oxygenation, discharge with opiates, no digoxin at discharge, postoperative arch obstruction, moderate to severe tricuspid regurgitation without an oxygen requirement, and extra oxygen required at discharge in patients with moderate to severe tricuspid regurgitation) score can stratify patients by risk of death or transplantation (DTx) on the basis of clinical factors. The study investigators suspected a variable transplant-free survival benefit of digoxin in high-risk vs low-risk patients. METHODS: National Pediatric Cardiology Quality Improvement Collaborative patients discharged after stage 1 palliation with complete data were categorized as high- or low-risk on the basis of a modified NEONATE score. The primary outcome of DTx was evaluated. A mixed-effect regression evaluated associations between digoxin prescription and risk factors. RESULTS: A total of 1199 patients were included; 399 (33%) were high risk. Baseline demographics were similar between the cohorts. Blalock-Taussig shunt or a hybrid operation, postoperative extracorporeal membrane oxygenation, opiate prescription, and significant tricuspid regurgitation or arch obstruction were more common in high-risk patients. The odds of DTx were 65% lower in high-risk patients prescribed digoxin compared with patients who were not (P = .001). Digoxin prescription was associated with 60.8% lower DTx in the high-risk cohort (7.8% vs 19.9%; P = .001). There was no significant difference in the DTx rate according to digoxin prescription in the low-risk cohort (4.7% vs 5.7%; P = .46). Blalock-Taussig shunt, aortic arch obstruction, and significant tricuspid regurgitation were most strongly associated with deriving a benefit from digoxin. CONCLUSIONS: Digoxin use is associated with significant improvement in transplant-free survival in high-risk but not in low-risk interstage patients. A tailored approach to the use of digoxin in interstage patients may be warranted.

Characteristics of Interstage Death After Discharge from Stage I Palliation.

BACKGROUND: Interstage mortality (IM) remains high for patients with single-ventricle congenital heart disease (SVCHD) in the period between Stage 1 Palliation (S1P) and Glenn operation. We sought to characterize IM. METHODS: This was a descriptive analysis of 2184 patients with SVCHD discharged home after S1P from 60 National Pediatric Cardiology Quality Improvement Collaborative sites between 2008 and 2015. Patients underwent S1P with right ventricle-pulmonary artery conduit (RVPAC), modified Blalock-Taussig-Thomas shunt (BTT), or Hybrid; transplants were excluded. RESULTS: IM occurred in 153 (7%) patients (median gestational age 38 weeks, 54% male, 77% white), at 88 (IQR 60,136) days of life, and 39 (IQR 17,84) days after hospital discharge; 13 (8.6%) occurred ≤ 30 days after S1P. The mortality rate for RVPAC was lower (5.2%; 59/1138) than BTT (9.1%; 65/712) and Hybrid (20.1%; 27/134). More than half of deaths occurred at home (20%) or in the emergency department (33%). The remainder occurred while inpatient at center of S1P (cardiac intensive care unit 36%, inpatient ward 5%) or at a different center (5%). Fussiness and breathing problems were most often cited as harbingers of death; distance to surgical center was the biggest barrier cited to seeking care. Cause of death was unknown in 44% of cases overall; in the subset of patients who underwent post-mortem autopsy, the cause of death remained unknown in 30% of patients, with the most common diagnosis being low cardiac output. CONCLUSIONS: Most IM occurred in the outpatient setting, with non-specific preceding symptoms and unknown cause of death. These data indicate the need for research to identify occult causes of death, including arrhythmia.

Development of a validated risk score for interstage death or transplant after stage I palliation for single-ventricle heart disease.

OBJECTIVE: To develop a risk score to predict mortality or transplant in the interstage period. BACKGROUND: The "interstage" period between the stage 1 and stage 2 palliation is a time of high morbidity and mortality for infants with single-ventricle congenital heart disease. METHODS: This was an analysis of patients with single-ventricle congenital heart disease requiring arch reconstruction who were enrolled in the National Pediatric Cardiology Quality Improvement Collaborative registry from 2008 to 2015. The primary composite endpoint was interstage mortality or transplant. Multivariable logistic regression and classification and regression tree analysis were performed on two-thirds of the patients ("learning cohort") to build a risk score for the composite endpoint, that was validated in the remaining patients ("validation cohort"). RESULTS: In the 2128 patients analyzed in the registry, the overall event rate was 9% (153 [7%] deaths, 42 [2%] transplants). In the learning cohort, factors independently associated with the composite endpoint were (1) type of Norwood; (2) postoperative ECMO; (3) discharge with Opiates; (4) No Digoxin at discharge; (5) postoperative Arch obstruction, (6) moderate-to-severe Tricuspid regurgitation without an oxygen requirement, and (7) Extra Oxygen required at discharge in patients with moderate-to-severe tricuspid regurgitation. This model was used to create a weighted risk score ("NEONATE" score; 0-76 points), with >75% accuracy in the learning and validation cohorts. In the validation cohort, the event rate in patients with a score >17 was nearly three times those with a score ≤17. CONCLUSIONS: We introduce a risk score that can be used post-stage 1 palliation to predict freedom from interstage mortality or transplant.

Statewide Quality Improvement Initiative to Reduce Early Elective Deliveries and Improve Birth Registry Accuracy.

OBJECTIVE: To evaluate the success of a quality improvement initiative to reduce early elective deliveries at less than 39 weeks of gestation and improve birth registry data accuracy rapidly and at scale in Ohio. METHODS: Between February 2013 and March 2014, participating hospitals were involved in a quality improvement initiative to reduce early elective deliveries at less than 39 weeks of gestation and improve birth registry data. This initiative was designed as a learning collaborative model (group webinars and a single face-to-face meeting) and included individual quality improvement coaching. It was implemented using a stepped wedge design with hospitals divided into three balanced groups (waves) participating in the initiative sequentially. Birth registry data were used to assess hospital rates of nonmedically indicated inductions at less than 39 weeks of gestation. Comparisons were made between groups participating and those not participating in the initiative at two time points. To measure birth registry accuracy, hospitals conducted monthly audits comparing birth registry data with the medical record. Associations were assessed using generalized linear repeated measures models accounting for time effects. RESULTS: Seventy of 72 (97%) eligible hospitals participated. Based on birth registry data, nonmedically indicated inductions at less than 39 weeks of gestation declined in all groups with implementation (wave 1: 6.2-3.2%, P<.001; wave 2: 4.2-2.5%, P=.04; wave 3: 6.8-3.7%, P=.002). When waves 1 and 2 were participating in the initiative, they saw significant decreases in rates of early elective deliveries as compared with wave 3 (control; P=.018). All waves had significant improvement in birth registry accuracy (wave 1: 80-90%, P=.017; wave 2: 80-100%, P=.002; wave 3: 75-100%, P<.001). CONCLUSIONS: A quality improvement initiative enabled statewide spread of change strategies to decrease early elective deliveries and improve birth registry accuracy over 14 months and could be used for rapid dissemination of other evidence-based obstetric care practices across states or hospital systems.

Piloting a Statewide Home Visiting Quality Improvement Learning Collaborative.

Objective To pilot test a statewide quality improvement (QI) collaborative learning network of home visiting agencies. Methods Project timeline was June 2014-May 2015. Overall objectives of this 8-month initiative were to assess the use of collaborative QI to engage local home visiting agencies and to test the use of statewide home visiting data for QI. Outcome measures were mean time from referral to first home visit, percentage of families with at least three home visits per month, mean duration of participation, and exit rate among infants <6 months. Of 110 agencies, eight sites were selected based on volume, geography, and agency leadership. Our adapted Breakthrough Series model included monthly calls with performance feedback and cross-agency learning. A statewide data system was used to generate monthly run charts. Results Mean time from referral to first home visit was 16.7 days, and 9.4% of families received ≥3 visits per month. Mean participation was 11.7 months, and the exit rate among infants <6 months old was 6.1%. Agencies tested several strategies, including parent commitment agreements, expedited contact after referral, and Facebook forums. No shift in outcome measures was observed, but agencies tracked intermediate process changes using internal site-specific data. Agencies reported positive experiences from participation including more frequent and structured staff meetings. Conclusions for Practice Within a pilot QI learning network, agencies tested and measured changes using statewide and internal data. Potential next steps are to develop and test new metrics with current pilot sites and a larger collaborative.

Digoxin Use Is Associated With Reduced Interstage Mortality in Patients With No History of Arrhythmia After Stage I Palliation for Single Ventricle Heart Disease.

BACKGROUND: Interstage mortality (IM) remains significant after stage 1 palliation (S1P) for single-ventricle heart disease (SVD), with many deaths sudden and unexpected. We sought to determine whether digoxin use post-S1P is associated with reduced IM, utilizing the multicenter database of the National Pediatric Cardiology Quality Improvement Collaborative (NPCQIC). METHODS AND RESULTS: From June 2008 to July 2013, 816 infants discharged after S1P from 50 surgical sites completed the interstage to stage II palliation, transplant, or IM. Arrhythmia during S1P hospitalization or discharge on antiarrhythmic medications were exclusions (n=270); 2 patients were lost to follow-up. Two analyses were performed: (1) propensity-score adjusted logistic regression with IM as outcome and (2) retrospective cohort analysis for patients discharged on digoxin versus not, matched for surgical site and other established IM risk factors. Of 544 study patients, 119 (21.9%) were discharged on digoxin. Logistic regression analysis with propensity score, site-size group, and digoxin use as predictor variables showed an increased risk of IM in those not discharged on digoxin (odds ratio, 8.6; lower confidence limit, 1.9; upper confidence limit, 38.3; P<0.01). The retrospective cohort analysis for 60 patients on digoxin (matched for site of care, type of S1P, post-S1P ECMO use, genetic syndrome, discharge feeding route, ventricular function, tricuspid regurgitation, and aortic arch gradient) showed 0% IM in the digoxin at discharge group and an estimated IM difference between the 2 groups of 9% (P=0.04). CONCLUSIONS: Among SVD infants in the NPCQIC database discharged post-S1P with no history of arrhythmia, use of digoxin at discharge was associated with reduced IM.

Reliable implementation of evidence: a qualitative study of antenatal corticosteroid administration in Ohio hospitals.

BACKGROUND: Antenatal corticosteroids (ANCS) reduce complications of preterm birth; however, not all eligible women receive them. Many hospitals and providers do not have the right processes and conditions to enable ANCS administration with high reliability. The objective of this study was to understand conditions that enable delivery of ANCS with high reliability among hospitals participating in an Ohio Perinatal Quality Collaborative (OPQC) ANCS project. METHODS: We conducted focus groups and semistructured interviews with members of the OPQC project team (n=27) and other care providers (n=70) using a purposeful sample of 6 sites involved in the OPQC ANCS project. Participants including nurses (n=57), attending obstetricians (n=17), physician trainees (n=21) and certified nurse midwives (n=2) were asked to reflect on their experiences and to identify factors contributing to optimal use of ANCS. Focus groups and interviews were transcribed verbatim and were analysed by a multidisciplinary team using an iterative approach that combined inductive and deductive methods to identify and categorise themes. RESULTS: Six major themes supporting reliable implementation of ANCS at these hospitals emerged including: (1) presence of a high reliability culture, (2) processes that emphasise high reliability, (3) timely and efficient administration process, (4) multiple disciplines are involved, (5) evidence of benefit supports ANCS use and (6) benefit is recognised at all levels of the care team. CONCLUSIONS: Our findings identify the key processes and supports needed to ensure delivery of ASCS with high reliability and are reinforced by implementation and reliability science. They are useful for foundation of the successful implementation of other evidence-based practices at high levels of reliability.

Transparency in a Pediatric Quality Improvement Collaborative: A Passionate Journey by NPC-QIC Clinicians and Parents.

Transparency-sharing data or information about outcomes, processes, protocols, and practices-may be the most powerful driver of health care improvement. In this special article, the development and growth of transparency within the National Pediatric Cardiology Quality Improvement Collaborative is described. The National Pediatric Cardiology Quality Improvement Collaborative transparency journey is guided by equal numbers of clinicians and parents of children with congenital heart disease working together in a Transparency Work Group. Activities are organized around four interrelated levels of transparency (individual, organizational, collaborative, and system), each with a specified purpose and aim. A number of Transparency Work Group recommendations have been operationalized. Aggregate collaborative performance is now reported on the public-facing web site. Specific information that the Transparency Work Group recommends centers provide to parents has been developed and published. Almost half of National Pediatric Cardiology Quality Improvement Collaborative centers participated in a pilot of transparently sharing their outcomes achieved with one another. Individual centers have also begun successfully implementing recommended transparency activities. Despite progress, barriers to full transparency persist, including health care organization concerns about potential negative effects of disclosure on reputation and finances, and lack of reliable definitions, data, and reporting standards for fair comparisons of centers. The National Pediatric Cardiology Quality Improvement Collaborative's transparency efforts have been a journey that continues, not a single goal or destination. Balanced participation of clinicians and parents has been a critical element of the collaborative's success on this issue. Plans are in place to guide implementation of additional transparency recommendations across all four levels, including extension of the activities beyond the collaborative to support transparency efforts in national cardiology and cardiac surgery societies.

Pediatric collaborative networks for quality improvement and research.

Despite efforts of individual clinicians, pediatric practices, and institutions to remedy continuing deficiencies in pediatric safety and health care quality, multiple gaps and disparities exist. Most pediatric diseases are rare; thus, few practices or centers care for sufficient numbers of children, particularly in subspecialties, to achieve large and representative sample sizes, and substantial between-site variation in care and outcomes persists. Pediatric collaborative improvement networks are multi-site clinical networks that allow practice-based teams to learn from one another, test changes to improve quality, and use their collective experience and data to understand, implement, and spread what works in practice. The model was initially developed in 2002 by an American Board of Pediatrics Workgroup to accelerate the translation of evidence into practice, improve care and outcomes for children, and to serve as the gold standard for the performance in practice component of Maintenance of Certification requirements. Many features of an improvement network derive from the Institute for Healthcare Improvement's collaborative improvement model Breakthrough Series, including focus on a high-impact condition or topic; providing support from clinical content and quality improvement experts; using the Model for Improvement to set aims, use data for feedback, and test changes iteratively; providing infrastructure support for data collection, analysis and reporting, and quality improvement coaching; activities to enhance collaboration; and participation of multidisciplinary teams from multiple sites. In addition, they typically include a population registry of the children receiving care for the improvement topic of interest. These registries provide large and representative study samples with high-quality data that can be used to generate information and evidence, as well as to inform clinical decision making. In addition to quality improvement, networks serve as large-scale health system laboratories, providing the social, scientific, and technical infrastructure and data for multiple types of research. Statewide, regional, and national pediatric collaborative networks have demonstrated improvements in primary care practice as well as care for chronic pediatric diseases (eg, asthma, cystic fibrosis, inflammatory bowel disease, congenital heart disease), perinatal care, and patient safety (eg, central line-associated blood stream infections, adverse medication events, surgical site infections); many have documented improved outcomes. Challenges to spreading the improvement network model exist, including the need for the identification of stable funding sources. However, these barriers can be overcome, allowing the benefits of improved care and outcomes to spread to additional clinical and safety topics and care processes for the nation's children.

Understanding treatment decision making in juvenile idiopathic arthritis: a qualitative assessment.

BACKGROUND: The increase in therapeutic options for juvenile idiopathic arthritis (JIA) has added complexity to treatment decisions. Shared decision making has the potential to help providers and families work together to choose the best possible option for each patient from the array of choices. As part of a needs assessment, prior to design and implementation of shared decision making interventions, we conducted a qualitative assessment of clinicians' current approaches to treatment decision making in JIA. METHODS: Pediatric rheumatology clinicians were recruited from 2 academic children's hospitals affiliated with a quality improvement learning network, using purposive and snowball sampling. Semi-structured interviews elicited how clinicians with prescribing authority (n = 10) interact with families to make treatment decisions. Interviews were audio-recorded and transcribed verbatim. A multi-disciplinary research team used content analysis to analyze the interview data.To validate data from individual interviews and enrich our understanding, we presented the interview results to pediatric rheumatology clinicians attending a learning network meeting (n = 24 from 12 children's hospitals). We then asked the clinicians questions to further identify and discuss areas of variation in the decision-making processes. RESULTS: Clinicians described a decision-making process in which they, rather than the family or other care team members, consistently initiated treatment decisions. Initial treatment options presented to families generally reflected the clinician's preferred treatment approaches, which differed across clinicians. Clinicians used various methods to inform families about treatment options and tailor information according to perceptions of a family's information needs, level of comprehension or mood (e.g. anxiety). The attributes of medication presented to families fell into 4 categories: benefits, risks, logistics and family preferences. Clinicians typically included family members in the decision to initiate JIA treatment after limiting the options to fit the clinical situation and the clinician's own preferences. Family members' preferences were seen as more integral in the decision to stop treatment after symptom remission. CONCLUSIONS: Decision making about initial JIA treatment appears to be largely driven by clinician preferences. Family preferences are more likely to be considered for treatment discontinuation. Opportunities exist to develop, test, and implement tools to facilitate shared decision making in pediatric rheumatology.

Pediatric collaborative improvement networks: background and overview.

Multiple gaps exist in health care quality and outcomes for children, who receive <50% of recommended care. The American Board of Pediatrics has worked to develop an improvement network model for pediatric subspecialties as the optimal means to improve child health outcomes and to allow subspecialists to meet the performance in practice component of Maintenance of Certification requirements. By using successful subspecialty initiatives as exemplars, and features of the Institute for Healthcare Improvement's Breakthrough Series model, currently 9 of 14 pediatric subspecialties have implemented collaborative network improvement efforts. Key components include a common aim to improve care; national multicenter prospective collaborative improvement efforts; reducing unnecessary variation by identifying, adopting, and testing best practices; use of shared, valid, high-quality real-time data; infrastructure support to apply improvement science; and public sharing of outcomes. As a key distinguisher from time-limited collaboratives, ongoing pediatric collaborative improvement networks begin with a plan to persist until aims are achieved and improvement is sustained. Additional evidence from within and external to health care has accrued to support the model since its proposal in 2002, including the Institute of Medicine's vision for a Learning Healthcare System. Required network infrastructure systems and capabilities have been delineated and can be used to accelerate the spread of the model. Pediatric collaborative improvement networks can serve to close the quality gap, engage patients and caregivers in shared learning, and act as laboratories for accelerated translation of research into practice and new knowledge discovery, resulting in improved care and outcomes for children.

Exemplar pediatric collaborative improvement networks: achieving results.

A number of pediatric collaborative improvement networks have demonstrated improved care and outcomes for children. Regionally, Cincinnati Children's Hospital Medical Center Physician Hospital Organization has sustained key asthma processes, substantially increased the percentage of their asthma population receiving "perfect care," and implemented an innovative pay-for-performance program with a large commercial payor based on asthma performance measures. The California Perinatal Quality Care Collaborative uses its outcomes database to improve care for infants in California NICUs. It has achieved reductions in central line-associated blood stream infections (CLABSI), increased breast-milk feeding rates at hospital discharge, and is now working to improve delivery room management. Solutions for Patient Safety (SPS) has achieved significant improvements in adverse drug events and surgical site infections across all 8 Ohio children's hospitals, with 7700 fewer children harmed and >$11.8 million in avoided costs. SPS is now expanding nationally, aiming to eliminate all events of serious harm at children's hospitals. National collaborative networks include ImproveCareNow, which aims to improve care and outcomes for children with inflammatory bowel disease. Reliable adherence to Model Care Guidelines has produced improved remission rates without using new medications and a significant increase in the proportion of Crohn disease patients not taking prednisone. Data-driven collaboratives of the Children's Hospital Association Quality Transformation Network initially focused on CLABSI in PICUs. By September 2011, they had prevented an estimated 2964 CLABSI, saving 355 lives and $103,722,423. Subsequent improvement efforts include CLABSI reductions in additional settings and populations.

Impact of drug label changes on propofol use in pediatrics for moderate conscious sedation.

BACKGROUND: Drug product labeling is a critical component of communication regarding the appropriate use of medications. The information contained in a drug label is often complex, including contraindications and warnings that may be difficult to understand. In an attempt to further examine this issue, this article looks at one such difficult-to-understand label concerning anesthetic propofol and its use in the pediatric population. OBJECTIVE: The objective of this study was to describe the use of propofol for moderate conscious sedation (MCS) in pediatric patients (0-17 years) after drug warnings were disseminated. METHODS: This study was a retrospective, observational study from January 2001 to December 2007 that used data from the Premier Perspective Comparative Hospital database. This database includes approximately 425 hospitals with a broad range of hospital types and contains a weighting scheme that allows for the generation of national estimates in the United States. The main outcome measure was use of propofol during hospitalization. RESULTS: The study included 307,779 discharges in which MCS was used. Both the number of discharges for MCS and the percent of discharges using propofol increased from 2001 to 2007. After multivariable adjustment, there was more than a 3-fold increase in the odds of receiving propofol between 2001 and 2007 (odds ratio [OR] = 3.32; 95% CI, 2.96-3.72) for MCS. CONCLUSIONS: The results of this study suggest that the label changes and a "Dear Doctor" letter did not affect propofol utilization. A more cohesive approach to the assessment of safety and the dissemination of label change information to practitioners is needed.

Center variation in hospital costs for patients undergoing congenital heart surgery.

BACKGROUND: Congenital heart disease consumes significant health care resources; however, there are limited data regarding factors affecting resource utilization. The purpose of this study was to evaluate variation between centers in total hospital costs for 4 congenital heart operations of varying complexity and associated factors. METHODS AND RESULTS: The Premier Database was used to evaluate total cost in children undergoing isolated atrial septal defect (ASD) repair, ventricular septal defect (VSD) repair, tetralogy of Fallot (TOF) repair, or arterial switch operation (ASO) from 2001 to 2007. Mixed models were used to evaluate the impact of center on total hospital costs adjusting for patient and center characteristics and length of stay. A total of 2124 patients were included: 719 ASD (19 centers), 792 VSD (20 centers), 420 TOF (17 centers), and 193 ASO (13 centers). Total cost increased with complexity of operation from median $12 761 (ASD repair) to $55 430 (ASO). In multivariable analysis, models that accounted for center effects versus those that did not performed significantly better for all 4 surgeries (all P≤0.01). The proportion of total cost variation explained by center was 19% (ASD repair), 11% (VSD repair), 6% (TOF repair), and 3% (ASO). Higher-volume centers had significantly lower hospital costs for ASD and VSD repair but not for TOF repair and ASO. CONCLUSIONS: Total hospital costs varied significantly by center for all congenital heart surgeries evaluated, even after adjustment for patient and center characteristics and length of stay. Differences among centers were most prominent for lower complexity procedures.

Special challenges in comparative effectiveness research on children's and adolescents' health.

The United States is undertaking a major expansion of comparative effectiveness research, with the potential to achieve systemwide improvements in health care quality, outcomes, and resource allocation. However, to achieve these improvements in children's health and health care, comparative effectiveness research needs to be targeted, designed, conducted, and reported in ways that are responsive to the unique circumstances of children and adolescents. These include clinically important differences in the type and course of disease in children; demographic differences between the overall child and adult population in the United States, such as racial and ethnic makeup; and methodological issues involving study design. Our overarching point is that the base of evidence in pediatrics must not fall even further behind that for the adult population in an era of rapid advancement and funding of comparative effectiveness research.