How a regional collaborative network of NICUs worked together to reduce chronic lung disease rates by increasing use of delivery room CPAP.

By Christopher Chambers, RRT, James Chisholm, RRT, and Jan Schriefer, RN, DrPH

Bronchopulmonary dysplasia (BPD), also referred to as chronic lung disease (CLD), remains a challenging outcome for very preterm infants.1 About one-quarter of infants born < 1,500 g birth weight have BPD or CLD. The disorder results in prolonged stays in the neonatal intensive care unit (NICU) and increased use of health care services, carries a risk of neonatal and postneonatal mortality, and is associated with long-term neurodevelopmental issues and increased cost.

One of the primary therapies delivered in the NICU to support infants with BPD and CLD is respiratory care. Many different methods are used to deliver respiratory care, based on published studies, specific equipment present in each NICU, and local practices. The primary modes are intubation with positive pressure ventilation, noninvasive positive pressure ventilation (NIPPV), and continuous positive airway pressure (CPAP). The following paper describes a respiratory quality improvement learning collaborative among 11 NICUs in the United States.

Quality Improvement of Respiratory Care in the NICU

Payne et al2 developed a number of potentially better practices based on published evidence, provided a resource kit for the improvements described, and initiated research to define BPD and CLD better. They found that multi-institutional collaboration facilitated development of potentially better practices to reduce bronchopulmonary dysplasia.

This past collaborative served as a model for our work. Another potentially better practice was more frequent use of CPAP in the delivery room, a practice change described by O’Donnell.3 Our work focused on ways to increase the use of delivery room CPAP and reduce the rates of intubation in the delivery room. Our hope was that change in practice would result not only in a reduction in the rates of chronic lung disease, but also fewer chest x-rays, less need for blood gasses, and a decrease in the number of ventilator days.


  1. Reduce chronic lung disease by 20%;
  2. Increase the use of delivery room CPAP;
  3. Standardize respiratory care in the NICU and delivery room;
  4. Reduce unwanted variation in respiratory care in the NICU and delivery room; and
  5. Use meetings, surveys, and email distribution lists to gain consensus across many NICUs on respiratory care in areas such as delivery room CPAP and weaning protocols.


The quality improvement approach standardized respiratory care by collaborating with 11 neonatal intensive care units in the Vermont Oxford Network (VON). VON is a nonprofit organization dedicated to improving the quality of NICU care. Horbar et al4 established a system for multiple NICUs to work collaboratively in quality improvement efforts. The collaborative structure involved multidisciplinary teams from many NICUs who meet in person twice a year and meet via webinar once monthly.

Each of the 11 NICUs in this collaborative had teams of providers including respiratory therapists, physicians, neonatal nurse practitioners (NNPs), nurses, and parents of former NICU patients. We requested a summary of delivery room and NICU respiratory practices through a respiratory therapy email distribution list. The list was created for this purpose and represented 40 respiratory therapists as well as other practitioners.

Based on our findings from the protocols and order sets that were collected, we decided to create an online survey to obtain more details and specific How a regional collaborative network of NICUs worked together to reduce chronic lung disease rates by increasing use of delivery room/respiratory practices. The Survey Method system ( was used. We selected items by working with two respiratory therapists, a neonatologist, and an ICU nurse to outline common practice. We structured the survey so it would not take more than 10 minutes to complete. The only testing that was done on the survey was that of face and content validity by sharing it with the approximately 75 people in the workgroup prior to administration. The survey was sent only to the respiratory therapy email distribution list. The response rate was 17 out of 45 (38%). We organized and summarized our findings using some of the summary charts provided by Survey Method.

Once this was complete, we created a PowerPoint case study presentation. The scenario was a premature infant who went through all phases of our accepted respiratory care. Examples of the case presentation slides can be seen in Figure 1, which summarizes the content the providers would like to see for standardizing care. This was presented by several NICUs at our twice annual meeting, giving us an opportunity to disseminate information and discuss it with other NICUs in the collaborative. The dissemination occurred through the use of an online improvement story website called NICQpedia, during a poster learning fair at our face to face meetings, and during an Internet-based collaborative (iNICQ) that focused on respiratory care.

A draft Respiratory Care Delivery Room CPAP Bundle and Playbook was created after careful review of all protocols in use and review of the results of our survey. Table 1 is the display of the bundle contents.

Definition of Delivery Room CPAP

Delivery room CPAP became a focus when the collaborative was looking at ways to decrease chronic lung disease. Delivery room CPAP is the application of either nasal prongs or mask providing positive end expiratory pressure (PEEP), without the administration of positive pressure breaths, within 5 minutes of life. The CPAP equipment was based on what each NICU used. In addition to implementing Delivery room CPAP, because the infant was placed on CPAP prior to weighing, the weight of the equipment needed to be established. Another obstacle was determining the appropriate equipment and adapting it to transport the infant back to the NICU on CPAP. With an increased number of babies on CPAP, more patient care and serial assessments were required to prevent nasal septal injury12,13 and for airway care.12-14

Improvement Cycles for Respiratory Care

Schriefer and Leonard15 outlined the Model for Improvement approach used in pediatric quality improvement utilizing plan do study act (PDSA) cycles. The approach starts with setting an improvement aim. In this case, the aim for all the centers was to reduce chronic lung disease rates by increasing use of Delivery room CPAP. The team felt this was a realistic aim since many centers were above the VON network average for chronic lung disease and well above the top quartile centers.

The improvement PDSA cycles were based on the evidence that suggested early CPAP is beneficial.6,16 The PDSA cycles started with Delivery room CPAP for larger infants, and once comfortable using delivery room CPAP with those babies, some of the NICUs lowered the age limit for initial trial of DR CPAP.

Education of Providers

The initial goal was to obtain consent and approval from all NICUs in the collaborative on the details of the Delivery room CPAP Playbook. This required agreement from all parties and education for all of the providers. As we presented the information from the playbook and survey, we utilized an interactive electronic audience participation device, a “Clicker” with a variety of questions (shown in Figures 2-4). This helped us to survey the practitioners and anonymously obtain opinions and topics for further discussion. The number of responders at our face to face meetings using the clicker technology was 65 NICU providers. Each person voted independently, and we did not check for concordance of responses from members of the same team. There were 11 different NICU teams represented in the 65 providers. The response was 100% since all 65 people present responded. The summary of the survey topics is presented in Figure 1.

Once the results were compiled, we determined that ventilator management and initial ventilator settings needed further discussion. From there, we separated into individual centers and discussed the desire to implement RT/RN respiratory-driven protocols. Many NICUs have different initial settings based on population, resources, and cultures. After discussion, it was determined that all centers could agree on common ventilator settings for initial stabilization. The agreement was reached through large group discussion. We did not use a more formal approach such as Delphi. Consensus setting was in accord with external guidance and opinion from our two expert faculty members and two clinical leaders.

Ventilator weaning protocols were not so eagerly met. Most agreed that having guidelines that address ventilator weaning strategies can be helpful. However, many thought that patients were too individualized to have one approach. The group did review the study by Hermeto and colleagues17 (which showed favorable results for RT-driven protocols) but was not ready to develop one weaning protocol between our 11 NICUs. Figure 3 shows the variation in opinions regarding weaning protocols. The concern with using an RT-driven weaning protocol is that not all NICUs have unit-based RTs and some centers felt the protocol should be patient driven, not RT, MD, or RN driven.

The next step for many facilities was to develop a strategy to address the patient’s readiness to wean. Some examples included discussion on daily rounds or designing specific questions pertaining to meeting specific weaning criteria. All 11 NICUs had nurse-driven oxygen weaning protocols to maintain oxygen saturations in target ranges as seen in Figure 4. Although ventilator weaning protocols have been used in many adult intensive care units, they have not been commonly used in NICUs.17

Statistical Analysis and Results

We have seen an increase in the number of infants receiving Delivery room CPAP since the implementation of our playbook and care bundle. This was a primary process measure we hoped to impact. The data from the 11 centers from 2010 and 2011 showed a 22% decrease in chronic lung disease. Also noted in this same data, DR endotracheal tube (ETT) was down 7%, ventilator after early CPAP was down 23%, O2 at 28 days was down 15%, and O2 on discharge home was down 21% (see Figure 5).

The real limitation of this analysis is the lack of a counterfactual comparison group and the problems with before-and-after comparisons of process and outcome measures. However, the VON network as a whole did not see the same. The network chronic lung disease rate dropped only 4%, ventilator after early CPAP dropped just 13%, O2 at 28 days dropped just 3%, O2 on discharge home dropped just 4%, and mortality dropped about 3%. In summary, the overall network relative percentage changes in these process and outcome measures were much smaller than for our 11 NICU homeroom group.

The patient populations between 2010 and 2011 for our 11 NICUs had a number of similar demographic characteristics, which suggests they are acceptable for comparison. The sample size was 1,145 VLBW infants in 2010 and 1,108 in 2011. The use of antenatal steroids was 82.4% for 2010 and 81.7% for 2011. The percentage with an admission temperature less than 36 degrees was 18.9% for 2010 and 17.3 for 2011. The percentage with an Apgar score at 5 minutes < 4 was 10.6% for 2010 and 9.5% for 2011, and the average NICU length of stay was 59 days in 2010 and 58 in 2011.


There were improvements in process and outcome measures. However, this was not a randomized controlled trial with a control group. Therefore, the causality that our quality improvement efforts are related to the changes cannot be assumed.

Based on our survey results, we determined that the interest in weaning protocols was important to the collaborative. However, we determined through the use of an electronic audience participation device that there were many concerns regarding this issue. Most importantly, people agreed that protocols can expedite weaning and extubation, but the concern should not overshadow patient variability. Most believed that each patient weans very differently and needed individualized plans. Perhaps our future work can develop a more structured analysis of the major themes concerning extubation and their relative importance.

On the reverse side, the respiratory therapists agreed that having experienced NICU RT staff may be the most important contribution for successful protocols. The consensus was that NICU care teams should develop a daily weaning plan and daily rounding that includes respiratory care planning.


The collaborative approach was effective for developing guidelines and strategies to assist centers in the management of respiratory care of the neonate. By collaboratively compiling resources and information, through email distribution lists, webinars, and twice annual meetings, we were able to develop a “Respiratory Playbook” that could be used by NICUs to help guide the respiratory care provided to their neonates.

The team will focus next on sharing the playbook with referral hospitals so respiratory care can be more standardized for the outborn infants. The team observed improvements in clinical outcomes and process measures as well as a drop in NICU length of stay of 1 day for 1,108 infants.


Christopher Chambers, RRT, is supervisor, respiratory care, and James Chisholm, RRT, is pediatric clinical educator at Fletcher Allen Health Care Vermont Children’s Hospital, Burlington, Vt. Jan Schriefer, RN, DrPH, is assistant professor of pediatrics, Division of Pediatric Hospital Medicine, University of Rochester, and director of quality improvement at Golisano Children’s Hospital at University of Rochester Medical Center, Rochester, NY. For further information, contact [email protected].


The authors want to thank the NICUs that participated in this collaborative: Advocate Lutheran General, Vermont Children’s at Fletcher Allen, University of Minnesota Amplatz Children’s Hospital, Rockford Memorial Hospital, Meriter Hospital, Geisinger Medical Center, Inova Fairfax Hospital for Children, Women’s Hospital, Benefis Hospital, Brenner Children’s Hospital, and CHOC at Mission Hospital.


  1. Van Marter LJ. Epidemiology of bronchopulmonary dysplasia. Semin Fetal Neonatal Med. 2009;14(6):358-66.
  2. Payne NR, LaCorte M, Sun S, Karna P, Lewis-Hunstiger M, Goldsmith JP. Evaluation and development of potentially better practices to reduce bronchopulmonary dysplasia in very low birth weight infants. Pediatrics. 2006;118(suppl 2):S65-72.
  3. O’Donnell CP. Turn and face the strange— to neonatal resuscitation guidelines in the past decade. J Paediatr Child Health. 2012;48(9):735-9.
  4. Horbar JD, Plsek PE, Schriefer JA, Leahy K. Evidence-based quality improvement in neonatal and perinatal medicine: the Neonatal Intensive Care Quality Improvement Collaborative experience. Pediatrics. 2006;118(suppl 2):S57-S64.
  5. de Klerk R. Care of the Infant of NCPAP Vermont Oxford Network Delivery Room Management Clinical Trial. Burlington, Vt: Vermont Oxford Network; 2002.
  6. Dawson JA, Kamlin CO, Vento M, et al. Defining the reference range for oxygen saturation for infants after birth. Pediatrics. 2010;125(6):e1340-e1347.
  7. Levesque B, Kalish L, LaPierre J, Welch M, Porter V. Impact of implementing 5 potentially better respiratory practices on neonatal outcomes and costs. Pediatrics. 2011;128(1):e218-6.
  8. Verder H, Albertsen P, Ebbesen F, et al. Nasal continuous positive airway pressure and early surfactant therapy for respiratory distress syndrome in newborns of less than 30 weeks’ gestation. Pediatrics. 1999;103(2):E24.
  9. Polin RA, Sahni R. Newer experience with CPAP. Semin Neonatol. 2002;7:379–389.
  10. Dani C, Corsini I, Bertini G, Fontanelli G, Pratesi S, Rubaltelli FF. The INSURE method in preterm infants of less than 30 weeks’ gestation. J Matern Fetal Neonatal Med. 2010;23(9):1024-9.
  11. Stevens TP, Harrington EW, Blennow M, Soll RF. Early surfactant administration with brief ventilation vs selective surfactant and continued mechanical ventilation for preterm infants with or at risk for respiratory distress syndrome. Cochrane Database Syst Rev. 2007;4:CD003063.
  12. Blondin J, Schriefer J, Shinozaki T, et al. The Quality Cup winner: Fletcher Allen Health Care’s early extubation team. Qual Manag Health Care. 1996;4(2):42-54.
  13. Continuous Positive Airway Pressure (CPAP). Newborn Services Clinical Guidelines. 2004.
  14. de Klerk R. Nasal Prong Continuous Positive Airway Pressure: A Practical Guide. New York: Children’s Hospital of New York; 2004.
  15. Schriefer J, Leonard MS. Quality improvement—an overview. Pediatr Rev. 2012;33(8):1-8.
  16. Bhandari V. Nasal intermittent positive pressure ventilation in the newborn: review of literature and evidence-based guidelines. J Perinatol. 2010;30(8):505-12. Epub 2009 Oct 22.
  17. Hermeto F, Bottino MN, Vaillancourt K, Sant’Anna GM. Implementation of a respiratory therapist-driven protocol for neonatal ventilation: impact on the premature population. Pediatrics. 2009;123(5):e907-e916.