Secondary Logo

Journal Logo

Original Article

Implementing a Robust Process Improvement Program in the Neonatal Intensive Care Unit to Reduce Harm

Nether, Klaus G.; Thomas, Eric J.; Khan, Amir; Ottosen, Madelene J.; Yager, Lauren

Author Information
doi: 10.1097/JHQ.0000000000000310
  • Open



Two decades after the publication of the Institute of Medicine report “To Err Is Human,” errors in health care delivery continue to lead to patient harm and death.1–4 Although some improvements have been made industry-wide, significant and sustainable breakthroughs in reducing patient harm remain elusive.5 Despite efforts to improve processes and reduce harm, critically ill neonates continue to be harmed in the neonatal intensive care unit (NICU) because of such events as central line blood stream infections, very low birth weight infant nutrition issues, or unplanned extubations.6–13 Neonatal Intensive Care Units are complex and relatively free-standing with staff and processes that are largely independent from the rest of the health system.13 The NICU delivers care to the most fragile of patients (infants) who are easily harmed.13 The infants in the NICU are often hospitalized for months, increasing their risk for harm events.13

In 2013, in their review of high reliability industries such as commercial airlines and nuclear power that have achieved high reliability despite complex operations and the potential for catastrophe, Chassin and Loeb proposed a framework for the transformation of health care to high reliability and zero harm.5 The High-Reliability Health Care Maturity Model presents three major and interrelated domains of change for health care organizations to pursue so that they can progress on their journey toward high reliability.5 These are: leadership commitment to zero harm, a culture of safety and robust process improvement (RPI).5 Robust process improvement refers to widespread use of process improvement tools within health care organizations, so that staff have the skills to identify and pursue sustainable improvements. Health care organizations also continue to have challenges sustaining improvements; this has had a detrimental effect on their ability to pursue high reliability.14,15 A full third of quality improvement projects are not sustained after a year and almost 70% are not sustained over time.16,17 Our understanding about what leads to sustainability is not yet well developed, but change management is likely a key factor.15,18 Managing the change for sustainability is specifically addressed by RPI through its blended approach of Lean, Six Sigma, and formal change mangement.5

In addition to building confidence and knowledge with RPI concepts and tools, other components such as leadership support, mentoring, and patient/family engagement in improvement initiatives are necessary to implement a RPI program.5,13,19 This goes beyond having patients and families provide input through committees or gathering feedback, but rather having them in the redesign of the processes that need improvement.5

Through funding from a grant, our institution partnered with colleagues at the university medical school and the children's hospital to implement an RPI program within a Patient Safety Learning Laboratory focused on reducing harm in the NICU. This paper describes how an NICU within a large health care system implemented an RPI program (“RPI Yellow Belt training”) including the provision of leadership support, the methods for training and mentoring of staff and parents, and the sustained outcomes to date for the improvement initiatives completed as part of the training.


Project Overview

This project was part of a grant to reduce harm in the NICU. It was designed to improve the confidence and knowledge in improvement concepts and tools of NICU staff and parents of the Parent Advisory Council (PAC) from the children's hospital by implementing an RPI program.13 Our institution's staff collected data from December 2016 through August 2019 on the RPI Yellow Belt training through surveys that were administered before and after training, assessing participants' confidence and knowledge in improvement concepts and tools. The staff also collected data on the outcomes including sustainability of the improvement initiatives that were completed as part of the training. Given that our focus was on implementing an RPI program, we do not report the details of each improvement initiative. This study was approved by our university's Committee for the Protection of Human Subjects as Non-Human Subjects Research (IRB#HSC-MS-16-0542).

Project Setting

This project was conducted in the NICU at the children's hospital.13 The children's hospital is one of 13 hospitals in the health system.20 One of the country's largest pediatric hospitals, the children's hospital is a 310-bed quaternary care women and children's hospital.13 Neonatal intensive care unit leadership and staff (frontline staff, medical staff, unit educators and unit leadership), university medical school research staff that worked or conducted improvement initiatives in the NICU, and members of the PAC participated in the RPI Yellow Belt training and improvement initiatives.13

Project Planning and Implementation

The RPI Yellow Belt training provides education so that process improvement concepts and tools are used in the daily work of staff and PAC members. Online content is complemented by face-to-face classes on change management and regular mentoring sessions led by our institution's process improvement experts. The RPI Yellow Belt training is an online and interactive course consisting of seven modules that lead a training participant through the different phases of Six Sigma called DMAIC (Define, Measure, Analyze, Improve, and Control) by answering the questions that need to be asked for each phase.21 The online training is four hours and can be stopped and started at any time, so that participants could receive training at their own pace. Integrated into the online training was a pretraining and post-training survey (0–10 scale) that participants had to take to be given credit for completing the training. Training participants then joined improvement initiative teams and used their RPI skills to address safety problems in the NICU.

Building an RPI program for the NICU required two phases of engagement: program building and training. Program building was the first step and began with training of NICU leaders since a transformation to continuous improvement requires leadership support. Training of NICU leaders started with formal change management (Facilitating Change Model).22 The training provided NICU leaders with change management concepts and tools to support improvement initiatives such as a stakeholder analysis, communication plan, priority payoff matrix, and helping/hindering to sustain the gains (SeeTable 1for Complete List of Concepts and Tools for Formal Change Management).22 In addition, NICU senior leaders built the infrastructure for a successful and sustainable program by examining roles and responsibilities, developing selection criteria for trainees, and selecting improvement initiatives and focus areas that aligned with the NICU's strategic initiative of reducing harm.

Table 1. - The Formal Change Management Curriculum (Facilitating Change Model): Face-To-Face Training With NICU Leaders22
Plan your project Inspire people Launch the initiative Support the change
Assess the culture
Define the change
Assemble a strategy
Engage the right people
Brainstorm barriers to success
Build the need for change
Paint a picture of the future state
Solicit support and involvement
Make it personal
Look for resistance
Lead change
Align operations and infrastructure
Get the word out
Permeate the culture
Monitor progress
Sustain the gains

The second phase, which was the RPI Yellow Belt training phase, started almost 5 months later and included NICU leadership and staff who were selected to lead improvement initiatives. Participants completed RPI Yellow Belt training (SeeTable 2for List of Concepts and Tools for RPI Yellow Belt training).21

Table 2. - The Online RPI Yellow Belt Training Curriculum: Interactive Online Training Consisting of 7 Modules Including Examples and Activities to Complement the Concepts and Tools21
Module 1—Welcome to yellow belt
 Problems occur, why?
 What is robust process improvement?
 Lean reduces waste
 Six sigma eliminate defects
 Change management involves people in creating and sustaining solutions
Module 2—What needs to be improved?
Tool list
 15 Words
 Threats versus opportunities
 Elevator speech
 Process map (current state)
 Project charter
Module 3—How do you know?
Tool list
 Voice of customer
 Pareto chart
 5 Whys
 Cause and effect diagram
 Spaghetti diagram
Module 4—How do you get others involved?
Tool list
 Cultural landscape map
 Stakeholder analysis
 Resistance analysis
 More of/Less of
 Communication plan
 Meeting facilitation
Module 5—How will you solve it?
Tool list
 Priority payoff matrix
 Process map (future state)
Module 6—How will you make it last?
Tool list
 Control plan
 Standard work
 Data display
Module 7—Training review and next steps:
 Taking action
 Challenging yourself
 Use robust process improvement everyday
 Next steps

One of the goals of the grant was to integrate and engage parents in improvement initiatives. The PAC was created as a component of the Patient Safety Learning Laboratory to engage former NICU parents as partners in the improvement process. Parent Advisory Council members were parents of NICU infants discharged from the hospital within the past one and a half to 5 years and who had experienced varied complications throughout their NICU stay. They were nominated by NICU clinicians, interviewed by a PAC steering committee, and selected based on their ability and motivation to serve on improvement initiatives in the unit. The 11-member council included 7 parents, 3 NICU nurse clinicians (manager, nurse practitioner and staff nurse), and a nurse researcher who facilitated implementation and evaluation of the council activities.

Parent Advisory Council members received training on patient and family-centered care, their role as members, and information about current improvement initiatives. In preparation to work on improvement initiatives, each member also received RPI Yellow Belt training.

Robust process improvement Yellow Belt training included a requirement for participants to apply the concepts and tools they learned in their training to an actual improvement initiative. For example, every team worked together to create a project charter to define the improvement initiative including the problem, business case, primary customer, scope, goals, timeline, and team. The team then got signoff from a project sponsor and champion that was identified for leadership support to move forward with the initiative. Some other tools that were used by multiple teams during their improvement initiative included the stakeholder analysis to gain support, process maps, communication plans, Pareto charts, cause and effect diagrams, control charts, design failure modes and effects analysis, standard work, and control plans. Neonatal intensive care unit leaders identified three improvement initiatives by focusing on the greatest opportunities to reduce harm in the NICU. These included reducing central line blood stream infections, improving very low birth weight (VLBW) infant nutrition, and reducing unplanned extubations.13

Process improvement teams formed to address each of the three improvement initiatives. Each multidisciplinary team including one PAC member was led by a physician and clinician such as a nurse or respiratory therapist, depending on the improvement initiative.13 Parent Advisory Council members were asked to participate on teams that they had personal experience with based on their infant's NICU journey. For example, parents of infants who had been intubated for several months participated on the unplanned extubation team. In addition, each of the teams had mentors from our institution who were experts in RPI.13 Mentoring with our institution staff on the RPI tools and improvement initiatives occurred every 2 weeks as the trained RPI leaders guided their teams through the different discovery and problem-solving phases. The mentoring calls lasted for about 60 minutes and included a status update on work to date, any challenges and barriers the team was facing, next steps the team was planning including concepts and tools being used, strategy for next steps, and a work plan outlining the next steps called a WWW (What, Who, When) plan.


Statistical Analysis of the Robust Process Improvement Yellow Belt Training

To evaluate the RPI Yellow Belt Training, paired sample T-tests were performed to test for significant improvement. The paired sample T-Tests were conducted on the pretraining and post-training survey results for training participants who completed the RPI Yellow Belt training. Two-sample proportions tests were conducted on the improvement initiative outcomes. All statistical tests were conducted at the 0.05 level of significance. The data were analyzed using Minitab software, version 19.

Robust Process Improvement Yellow Belt Training Results

Sixty-seven of 88 participants (76%) completed the training. The participants included NICU leadership and staff (frontline staff, medical staff, unit educators, and unit leadership), university medical school research staff that worked or conducted improvement initiatives in the NICU and members of the PAC. Training scores (0–10 scale) improved from an average of 4.45–7.60 for confidence in leading process improvement work, 2.36 to 7.49 for RPI knowledge, and 2.19 to 7.30 for confidence in using RPI tools; relative improvement of 71%, 217%, and 233% respectively (Figure 1). All three indicators showed a statistically significant (p < .001) improvement and less variability in post-training scores. We also analyzed the data of staff and PAC members separately and found similar results (Figures 2 and 3).

Figure 1.:
Box plot on the pre- and post- robust process improvement (RPI) Yellow Belt Training Survey Results (zero-10 scale) overall (N = 67) on Confidence in Leading Process Improvement Projects, RPI (Lean, Six Sigma and formal change management) Knowledge and Confidence in using RPI tools (all three indicators had a p < .001).
Figure 2.:
Box plot on the pre- and post- robust process improvement (RPI) Yellow Belt Training Survey Results for Staff (zero-10 scale) staff (N = 62) on Confidence in Leading Process Improvement Projects, RPI (Lean, Six Sigma and formal change management) Knowledge and Confidence in using RPI tools (all three indicators had a p < .001).
Figure 3.:
Box plot on the pre- and post- robust process improvement (robust process improvement) Yellow Belt Training Survey Results for PAC members (zero-10 scale) parents (N = 5) on Confidence in Leading Process Improvement Projects, RPI (Lean, Six Sigma and formal change management) Knowledge and Confidence in Using RPI Tools (the three indicators had a p = .030, p = .008 and p = .004 respectively).

Improvement Initiative Results

The heart of implementing an RPI program was the application of process improvement concepts and tools learned by participants during training to reduce harm in the NICU. The teams were formed from staff and PAC members that completed the RPI Yellow Belt training. The improvement initiatives started once RPI Yellow Belt training was completed and focused on reducing central line blood stream infections, improving nutrition of VLBW infants, and reducing unplanned extubations.

The results from these initiatives showed significant and sustainable improvements. The central line blood stream infections rates showed a relative reduction of 67.1% from baseline to the improve phase that has been furthered reduced and sustained for 17 months with a relative reduction of 77.9% from baseline to the control phase (p = .001). The unplanned extubations rate showed a relative reduction of 22.7% from baseline to the improve phase that has also been furthered reduced and sustained for 20 months with a relative reduction of 54.3% from baseline to the control phase (p = .001). Nutrition of VLBW infants improved. Calories and protein orders that met daily goals showed a relative improvement of 63% and 35%, respectively from the baseline to the improve phase (p < .001). Two of the three projects showed sustainability beyond 1 year as we continue to monitor the outcomes. One of the projects was entering the phase for sustainability so at this time we cannot validate the sustainability of those improvements made. The children's hospital's definition for sustainability is that improvements are sustained forever unless the improvement is no longer relevant, or the process is eliminated.


The only limitation of this study is that we cannot be certain which solutions (training, leadership support, mentoring or parent engagement) implemented had the greatest impact on the improvement initiatives.


It is well known that frontline staff, and patients/families have highly variable levels of knowledge about and experience with process improvement concepts and tools. In addition, health care organizations continue to struggle with sustaining improvement gained through project work.14–18 We found that implementing an RPI program not only addressed these gaps by providing the training to increase confidence and knowledge for improvement work but led to sustainable improvements.

Our experience also suggests that several additional factors were associated with the success and sustainability of the improvement initiatives. These factors were validated through the continuous monitoring of the improvement initiative outcomes. They are also equally important to the success and sustainability of an RPI program. It starts with the leadership support necessary to implement a program. Neonatal intensive care unit leaders provided dedicated time for staff to attend training, conduct process improvement initiatives during work time and provide the support needed for the improvement teams to be successful. The leadership support for the RPI program created a common language for improvement work for NICU staff and PAC members including an approach for that work. Common language allowed exchange of ideas, clarity of timelines, and more confidence. The common language empowered members from the PAC to be more involved in improvement initiatives. The PAC members were part of the team providing feedback on what the team identified as root causes and participated in the identification of solutions addressing those root causes. Even PAC members felt confident to lead an improvement initiative in the future. Having access to and participating in the same type of improvement training leveled the field, making everyone feel like they had an equal stake in the improvement initiatives especially the PAC members. This was achieved through training and validated by the pre- and post-training surveys. Sustainability was also designed into the RPI program. Leaders of the NICU were trained in RPI tools and coached in developing and leading a program resulting in a culture for improvement because NICU staff and PAC members are actively seeking out opportunities for improvement, engaging in improvement initiatives and sustaining improvements that were made. Increased capability and capacity for process improvement was achieved through the RPI training which created a common language and skillset that became embedded into daily work.

The second factor that was key to the success and sustainability of the improvement initiatives was the mentoring provided from our institution experts in improvement work. This is critical to the application of the concepts and tools learned by the staff and PAC members during the training. Mentoring was essential to help team members remember to use tools, focus on the right problem and to maintain accountability of milestones or outcomes.

The final factor was the involvement of the PAC members in the actual improvement initiatives. Because training of members from the PAC was delivered through this project, parents became active participants in improvement initiatives. Parent Advisory Council member participation added a patient-centered dimension. One member was part of each team working on an improvement initiative. Parent Advisory Council members were assigned to each team based on the completion of the RPI training and on the experiences of their own infant in the NICU. Members were able to share stories of their infant's NICU journey, which shed light on important quality and safety aspects of care delivery. Their input to the improvement initiatives were invaluable and assisted in the implementation of solutions that were not only successful but also sustainable.


Patients continue to experience death and harm during the course of receiving care despite the best efforts of health care organizations to improve processes and reduce harm. Attaining significant and sustainable improvements will require training of all staff and patients/families to improve care processes. However, improvement training programs designed to meet these goals are uncommon. This project focused on implementing an RPI program by building confidence and knowledge in RPI concepts and tools. The RPI Yellow Belt training including leadership support, mentoring and parent engagement in improvement initiatives not only led to improvements in confidence and knowledge for process improvement, but also resulted in significant and sustainable improvements in the initiatives that were completed as part of the training.


Although improvements have been made in health care, the magnitude of those improvements have not significantly moved the needle when it comes to sustainment and high reliability. Having an effective RPI program addresses the issues with improvement initiatives which is essential for achieving significant and sustainable improvements. The implementation of an RPI program helps to create a culture of continuous improvement as the children's hospital progresses on its journey towards high reliability and zero harm.

Author's Biographies

Klaus G. Nether, DHSc, MMI, CSSMBB, is the executive director for high reliability product delivery at the Joint Commission Center for Transforming Healthcare. Dr. Nether has expertise in high reliability science, patient safety and process improvement.

Eric J. Thomas, MD, MPH, is a professor of medicine at the McGovern Medical School at The University of Texas Health Science Center at Houston (UTHealth), and director of the UTHealth—Memorial Hermann Center for Healthcare Quality and Safety.

Amir Khan, MD, is the Medical Director of the Neonatal Intensive Care Unit at Children's Memorial Hermann Hospital. After completing his fellowship in Neonatal-Perinatal Medicine, Dr. Khan joined the faculty at UTHealth, McGovern Medical School. Over the years, he has assumed roles and is now the Director of the Division of Neonatal-Perinatal Medicine.

Madelene J. Ottosen, PhD, RN, teaches and conducts research about strategies to improve patient safety with particular emphasis to create partnerships with patients and families as part of our healthcare culture. Dr. Ottosen has expertise in qualitative research methods, quality improvement, and development of patient and family advisory councils.

Lauren Yager, MBA, BSN, RNC-NIC, is the clinical director for the Neonatal ICU and Children's Transport Teams at Children's Memorial Hermann Hospital. She has spent over 13 years as a nurse at CMHH, and the last 7 years in leadership.


1. Kohn LT, Corrigan JM, Donaldson MS. To err is human (book review). J Altern Complement Med. 2001;7(1):98.
2. James JT. A new, evidence-based estimate of patient harms associated with hospital care. J Patient Saf. 2013;9(3):122-128.
3. Giraldo P, Sato L, Sala M, Comas M, Dywer K, Castells X. A retrospective review of medical errors adjudicated in court between 2002 and 2012 in Spain. Int J Qual Health Care. 2016;28(1):33-39.
4. Wong BM, Dyal S, Etchells EE, et al. Application of a trigger tool in near real time to inform quality improvement activities: A prospective study in a general medicine ward. BMJ Qual Saf. 2015;24(4):272-281.
5. Chassin MR, Loeb JM. High-reliability health care: Getting there from here. Milbank Q. 2013;91(3):459-490.
6. Sharek PJ, Horbar JD, Mason W, et al. Adverse events in the neonatal intensive care unit: Development, testing, and findings of an NICU-focused trigger tool to identify harm in north American NICUs. Pediatrics. 2006;118(4):1332-1340.
7. Lanzillotti LS, De Seta MH, de Andrade CL, Mendes Junior WV. Adverse events and other incidents in neonatal intensive care units. Ciencia Saude Coletiva.2015;20(3):937-946.
8. Samra HA, McGrath JM, Rollins W. Patient safety in the NICU: A comprehensive review. J Perinatal Neonatal Nurs. 2011;25(2):123-132.
9. Hsu HE, Mathew R, Wang R, et al. Health care–associated infections among critically ill children in the US, 2013-2018. JAMA Pediatr. 2020;174(12):1176-1183.
10. Nkwata A, Soe M, Li Q, Godfrey-Johnson D, Edwards J, Dudeck MA. Incidence trends of central-line–associated bloodstream infections in neonatal intensive care units, NHSN, 2009–2018. Sixth decennial international conference on healthcare-associated infections. Infect Control Hosp Epidemiol. 2020;41(S1):s68.
11. Morii C. Prevention strategies for unplanned extubation in NICU—A literature review. J Neonatal Nurs. 2016;22(3):91-102.
12. Crezeé KL, DiGeronimo RJ, Rigby MJ, Carter RC, Patel S. Reducing unplanned extubations in the NICU following implementation of a standardized approach. Respir Care. 2017;62(8):1030-1035.
13. Sedlock EW, Ottosen M, Nether K, et al. Creating a comprehensive, unit-based approach to detecting and preventing harm in the neonatal intensive care unit. J Patient Saf Risk Manag. 2018;23(4):167-175.
14. Scheirer MA, Dearing JW. An agenda for research on the sustainability of public health programs. Am J Public Health. 2011;101(11):2059-2067.
15. Santos P, Joglekar A, Faughnan K, et al. Sustaining and spreading quality improvement: Decreasing intrapartum malpractice risk. J Healthc Risk Management. 2019;38(3):42-50.
16. Beer M, Nohria N. Cracking the code of change. Harv Business Rev. 2000;78(3):133.
17. Silver SA, McQuillan R, Harel Z, et al. How to sustain change and support continuous quality improvement. Clin J Am Soc Nephrology. 2016;11(5):916-924.
18. Wiltsey Stirman S, Kimberly J, Cook N, Calloway A, Castro F, Charns M. The sustainability of new programs and innovations: A review of the empirical literature and recommendations for future research. Implementation Science. 2012;7:17.
19. Millman EA, Pronovost PJ, Makary MA, Wu AW. Patient-assisted incident reporting: Including the patient in patient safety. J Patient Saf. 2011;7(2):106-108.
20. Children's Memorial Hermann Hospital. About Children's Memorial Hermann Hospital.
21. Joint Commission Center for Transforming Healthcare. RPI Yellow Belt [Web-Based Course]. The Joint Commission Blackboard. 2016.
22. Joint Commission Center for Transforming Healthcare. Leaders Facilitating Change [Course Material]. 2013.

lean; six sigma; patient safety; robust process improvement; neonatal intensive care unit

© 2021 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the National Association for Healthcare Quality.