If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Pediatrics, Nationwide Children's Hospital, Columbus, OhioCenter for Perinatal Research at Nationwide Children's Hospital, Columbus, Ohio
Cerebral palsy is the most common physical disability in childhood, and is mostly diagnosed after age 2 years. Delays in diagnosis can have negative long-term consequences for children and parents. New guidelines for early cerebral palsy diagnosis and intervention were recently published, after systematic review of the evidence by international multidisciplinary experts aiming to decrease age at diagnosis. The current study tested the feasibility of implementing these guidelines in an American clinical setting.
Methods
We designed a stepwise implementation process in a neonatal intensive care follow-up clinic. Efficacy was tested by comparing 10-month pre- and post-implementation periods. Clinic visit types, cerebral palsy diagnosis, provider competencies and perspectives, and balancing measures were analyzed.
Results
Changes to infrastructure, assessments, scheduling algorithms, documentation and supports in diagnosis or counseling were successfully implemented. Number of three- to four-month screening visits increased (255 to 499, P < 0.001); mean age at diagnosis decreased (18 to 13 months, P < 0.001). Clinic team awareness of early diagnosis and interventions increased (P < 0.001). There was no decrease in family satisfaction with overall clinic function. Opportunities for improvements included documentation for transitioning patients, generalizabilty across hospital clinics, systematic identification of high-risk status during hospitalization, and need for cerebral palsy care guidelines for infants under age 2 years.
Conclusions
We demonstrated for the first time in a US clinical setting the feasibility of implementation of international early diagnosis and treatment guidelines for cerebral palsy. This process is adaptable to other settings and underscores the necessity of future research on cerebral palsy treatments in infancy.
Prevalence of cerebral palsy, co-occurring autism spectrum disorders, and motor functioning—Autism and Developmental Disabilities Monitoring Network, USA, 2008.
Delays in diagnosis of cerebral palsy are associated with worse long-term function and participation, parental dissatisfaction, and higher rates of mental health conditions, including depression.
In response to these concerns, a large working group of international experts in cerebral palsy-related disciplines and parent stakeholders published guidelines for early detection and intervention for cerebral palsy,
The guidelines include recommendations for neuroimaging, neurological examinations, and motor assessments for infants below five months corrected age and between five and 24 months corrected age. They define the concept of “high-risk for cerebral palsy” at three to four months, a category in which infants do not yet receive a diagnosis, but due to a constellation of clinical factors, neuroimaging findings and standardized assessments, have a 95% probability of developing cerebral palsy in the following years. Best current evidence for counseling is based on published parent surveys and high-quality research. The guidelines can be translated into clinical practice by adapting core elements to suit individual clinical settings illustrated in Fig 1 (setting is a neonatal intensive care unit (NICU) follow-up program with applied examples of the guidelines).
FIGURE 1Translation of international guidelines into a clinical practice algorithm in a neonatal intensive care unit follow-up program. Infants referred due to newborn-attributable risks (e.g., neuroimaging findings consistent with neonatal encephalopathy) most frequently start surveillance according to the guidelines at the three- to four-month visits. If they are still hospitalized (e.g., extremely preterm infant with white matter injury and lung disease), the guideline mandated surveillance happens at the bedside with a HINE, a TIMP, and a GMA. Infants referred due to infant-attributable risks start at closest point on the pathway. For example, an infant referred to the program at nine months by their pediatrician due to inability to sit would start with a HINE and a Bayley Scales of Infant and Toddler Development. If an MRI has not yet been performed by the three- to four-month visits when neither sedation nor anesthesia is required, it is deferred until 12 months when anesthesia is no longer required by this particular setting's radiology protocols. During the program, infants are referred to services and subspecialists according to needs and established clinical algorithms. * Select examinations target developmental progression and represent best feasible evidence for specific concern (e.g., Gross Motor Function Measure-66 at 18 months and 30 months when infant walking to begin establishing trajectory, Infant Toddler Sensory Profile for sensory processing concerns, Quality of Upper Extremity Skills Test for arm evaluation in hemiplegia, Receptive Expressive Language Test-3 for speech assessment, etc.). ** Neurological examination after 2-year visit includes Amiel-Tison or other. Bayley III, Bayley Scales of Infant and Toddler Development Third Edition; CBCL, Child Behavior Checklist; CO, cerebral palsy; GMA, General Movements Assessment; HINE, Hammersmith Infant Neurological Examination; TIMP, Test of Infant Motor Performance. (The color version of this figure is available in the online edition.)
no clinical programs implemented the guidelines in their entirety as standard care. NICU follow-up programs routinely follow two infant populations with common identifiable factors for cerebral palsy: preterm infants and term infants with neonatal encephalopathy. Many children with less common risk factors for cerebral palsy also receive care in such programs due to medical complexity in the perinatal period. The goal of our study was therefore to test the effectiveness of an implementation process for clinical use of the early diagnosis guidelines
and apply them to a NICU follow-up clinic setting, where a large population of high-risk infants is routinely monitored. We hypothesized that comparison of quantitative metrics for clinic visits, diagnoses, and provider training competency would demonstrate improvements before and after implementation. We complemented our study with balancing measures of parent satisfaction and qualitative data on provider perceptions.
Methods
We analyzed data under an exempt determination from the Nationwide Children’s Hospital Institutional Review Board (IRB17-00409) due to the use of unidentified data.
Part one—baseline conditions
Setting and population
The NICU follow-up clinic in a large referral tertiary care center encompassing a network of nine NICUs including levels II, III, and IV units (a range from special care nurseries to intensive care to surgical referral centers)
and geographical region of approximately 200 mi2 was used as the study setting. Before the implementation of the guidelines, no specific early cerebral palsy program existed. Suspicion of cerebral palsy prompted referral at discharge from the clinic (age 2 years) to a developmental pediatrician or various specialty programs outside the NICU follow-up program for further evaluation. The exception to this pattern was for infants who received hypothermia, all of whom were followed by the pediatric neurologist within the NICU follow-up program. The clinic saw an average of 5000 high-risk visits per year, mostly referred from the NICU network, and also from community providers. Referrals to the clinic were based on established criteria (Supplemental Table S1).
Infrastructure
The electronic medical record was on an Epic platform, commonly used by large US health systems to access, organize, store, and share patient medical records. Therapist documentation of Bayley Scales of Infant and Toddler Development Third Edition
was standardized with flow sheets and checklists. Another feature of Epic was the MyChart option, an online secure portal, allowing parents to communicate with clinic providers. Pre-implementation, documentation of assessments was heterogeneous, reflecting personal choice and occasional practice in other clinics. Clinic scheduling templates were based on therapist scheduling availability and medical practitioners rarely had assigned patients.
Visit schedules, assessments, and referral
Pre-implementation, schedules followed very preterm infants at three to four months, nine to 12 months, and 22 to 26 months, with interim visits based on specific therapist- or dietitian-identified needs (Fig 2). Assessments were guided by participation in the Neonatal Research Network
as well as a strong evidence-based practice group in the therapy department. In addition, specialized providers examined those patients with high neurological complexity. Therapist assessments often determined referrals for therapy services, whereas medical providers made the referrals to subspecialty providers and programs (audiology, ophthalmology, physical medicine, developmental pediatrics, psychology, complex care, cerebral palsy) or for procedures based on their clinical judgment and a developmental algorithm for the two-year visit.
FIGURE 2Schedule of visits prior (A) and after (B) implementation. The journey in blue represents the schedule before guideline implementation, the red one after. The cognitive, motor, and language domains of the Bayley Scales of Infant and Toddler Development Third edition are administered. Standardized neurological examinations modified from the Amiel-Tison
are performed at 22 to 26 months on research patients by medical providers trained and certified on an annual basis by a gold-standard per NRN procedures.
Bayley III, Bayley Scales of Infant and Toddler Development Third Edition; CBCL, Child Behavior Checklist; GMA, General Movements Assessment; HINE, Hammersmith Infant Neurological Examination; NRN neuro, Neonatal Research Network neurological examination; OT, occupational therapist; PT, physical therapist; SLP, speech and language pathologist; TIMP, Test of Infant Motor Performance. (The color version of this figure is available in the online edition.)
The NICU follow-up team was multidisciplinary and diverse. Supplemental Table S2 lists roles and responsibilities of each provider in the NICU follow-up clinic pre- and post-implementation.
Research
Clinic personnel were involved in research through Neonatal Research Network studies requiring follow-up of two-year neurodevelopmental outcomes as a measure of efficacy of NICU-based interventions. However, interventional studies for infants were not yet integrated into the workflow of the clinic, making it difficult to incorporate new and proven interventions for cerebral palsy and to promote a culture of change into the setting.
Part two—implementation process
The process followed a predetermined timeline (Fig 3). The planning phase involved the design of a clinical schedule based on the guidelines (Fig 1), also respecting the primary mandate of the NICU follow-up program to follow all high-risk infants, not only those with cerebral palsy, and to promote effective transition to existing services after leaving the program. The preparation phase initiated provider culture change through a series of small moderated workgroups, each charged with guiding individual aspects of optimal developmental surveillance for all patients. Training in the General Movements Assessment (GMA) was accomplished by Prechtl's Trust certified trainers. Clinical competency in the Hammersmith Infant Neurological Examination (HINE) was a novel implementation process published recently.
Training of medical providers in scope of practice of various therapy disciplines, available services, and best evidence for occupational, physical, speech and feeding therapies in infancy was accomplished by a series of self-directed modules designed by clinical therapy leaders. Dietitians utilized first World Health Organization curves for infants,
based on Gross Motor Function Classification System levels. Documentation in Epic was modified to match the new visit schedule, driven by providers practicing regularly in the clinic. Clinic templates were switched to a provider-based system, allowing continuity of medical care and neurological trajectories as well as longitudinal perspective from the team leaders who could request specific therapy assessments based on new knowledge gained in the modules. Visits before three months were scheduled for those with the greatest risk factors and complexity. Research was integrated through additional workspace, training of coordinators, improved communication with nursing and social work leaders, and visibility of rehabilitation study materials.
FIGURE 3Implementation timeline and phases. The initial three phases of the implementation process occurred over the course of 12 months. A consolidation phase with reassessment of strengths and weaknesses offered the opportunity of a new development phase, planning for priorities of the next implementation cycle.
Parent support was addressed through counseling with the elements recommended in published literature and the guideline mandates: “face-to-face, with both caregivers present if appropriate, private, honest, jargon-free, with empathic communication tailored to the family.”
Identification of strengths and goals was included, and questions were fostered by allowing the diagnosing provider to return at conclusion of the visit to reassess the emotional context and answer additional questions. MyChart use was explained and encouraged. Referral to services was complemented by a set of written materials (therapist handouts and parent educational booklet).
Implementation occurred over one month, during which process improvement guides were available to observe, troubleshoot, record challenges, and provide support if necessary. Written feedback was also solicited.
Consolidation involved further observation and support in areas identified as challenges such as knowledge verification, training gaps, hand-off problems, template and workflow, scheduling difficulties, and referral difficulties. Development was driven by group identification of new priorities for the next cycle in clinical, operational, and research domains during regular provider meetings. Intra-hospital and community outreach were added as new priorities.
Part three—evaluation metrics
Quantitative metrics (Table 1) were obtained in the 10-month periods from July 1, 2014 to May 30, 2015 and July 1, 2016 to May 30, 2017 to prevent the possibility of crossover effects from early phases of implementation. Balancing measures were obtained using standard anonymous feedback clinic questionnaires at the end of each visit from any parent in the clinic, not only those of children with cerebral palsy. The goal of these measures was to determine whether implementation of the program negatively impacted parent satisfaction through three questions about overall perception of clinic function. Provider perspectives on cerebral palsy recognition and intervention for those involved in the care of infants 12 months and under were obtained through a survey scored on a five-item ranked scale.
TABLE 1Clinic Visit Numbers and Diagnosis Metrics for Implementation
Includes patients either discharged from clinic before a diagnosis was made or patients referred to other clinics where diagnosis was subsequently made.
28 (SD 7)
16 (SD 7)
Total number of patients at NCH with a diagnosis of cerebral palsy under 3 years
Does not reflect all patients with cerebral palsy at NCH, only those under age 3 years. Incidence unchanged between periods.
70
175
Abbreviations:
NCH = Nationwide Children's Hospital
SD = standard deviation
* Number of visits over a 10-month period in all NICU follow-up clinics with developmental testing.
† Includes patients either discharged from clinic before a diagnosis was made or patients referred to other clinics where diagnosis was subsequently made.
‡ Does not reflect all patients with cerebral palsy at NCH, only those under age 3 years. Incidence unchanged between periods.
The proportion of preterm infants seen in the clinic was not significantly different between the two periods (85.7% versus 76.6%, P = 0.9). The incidence of cerebral palsy was unchanged pre- and post-implementation of the guidelines in the hospital system.
(Table 1). Number of three- to four-month visits increased compared with other types of visits (due to the decreased clinic load of new patients and increased focus on three- to four-month visits), resulting in more children having GMA and HINE assessments performed at this time point. The number of children in the clinic diagnosed by 12 months increased, with the age at diagnosis decreasing significantly. In addition, changes in the clinic were reflected by a decreased mean age at diagnosis in the overall population of children with diagnoses of cerebral palsy at the hospital under age 3 years.
Provider metrics
Provider metrics (Table 2) demonstrated significant improvement of awareness of best practices around early diagnosis and early interventions for infants less than 12 months. When providers were asked to describe what they had learned post-implementation, five major themes emerged: optimizing developmental outcomes for children, improvement of support to parents, awareness of assessment tools, effective rehabilitation interventions, and importance of early detection. Balancing measures of parent satisfaction with clinic processes remained high and unchanged (Supplemental Table S3).
TABLE 2Survey Results for Provider Awareness
n
2014
2017
P Value
How aware that cerebral palsy could be frequently diagnosed by 12 months
26
1
3
<0.001
How aware of research interventions targeted to infants ≤12 months with cerebral palsy
22
0
3
<0.001
Therapist and provider: awareness of GMA and HINE use
11
0
4
<0.001
Nurse: awareness of sleep importance for infants with cerebral palsy
5
0
2
0.05
Dietitian: awareness of cerebral palsy specific nutrition
Continuous variables were analyzed using two-tailed t tests. Distributions of visit types pre-implementation versus post-implementation were compared using McNemar test with alpha = 0.05. Scores on scales using ranked items were analyzed using two-tailed Wilcoxon matched-pairs signed-rank test, with with alpha = 0.05. All calculations were performed using SPSS software (IBM Corp. Released 2013, IBM SPSS Statistics for MacIntosh, Version 22.0, Armonk, NY)
Discussion
This study represents the first successful attempt in the United States to implement the international guidelines for early detection and intervention for cerebral palsy. Average age at diagnosis decreased in the clinic program with an increase in provider knowledge base. Overall, parents were not dissatisfied by the process changes, and qualitative reports indicated satisfaction from providers and patients on the new model.
The NICU follow-up program presented a unique opportunity for success in implementation due to the strong institution-wide culture of commitment to excellence in clinical care, research, and quality improvement. Institutional support at the operational level allowed business managers and clinical leaders to direct the change in vision. For example, the therapist group already had a funded evidence-based practice group. The commitment of the NICU follow-up program team to their high-risk population made a change in practice relatively easy, as it did not represent a culture change.
In the consolidation phase, several immediate opportunities for process improvement were identified. In particular, as previously reported,
various providers had different comfort levels with giving the diagnosis of high-risk for or of cerebral palsy. Establishing confidence in administering the HINE or reading GMAs was sometimes difficult. For all these issues, a team-based support system was established so that providers could always rely on a more advanced or experienced provider for assistance. For diagnosis, if the provider did not feel comfortable or assessed that the family was not ready (emotional state, lack of support system, other parent not present, child acutely ill at visit) to receive a diagnosis, they explained that findings of an assessment were not typical. They then scheduled a return visit with a neurodevelopmental-trained specialist in the clinic. Similarly, for the GMA, therapists with less experience or without certification recorded the GMA on a secure tablet, explained that results would be reviewed by an experienced provider and communicated as soon as available. Experienced providers were available either in person or remotely. Provider confidence in neurological examinations improved.
On average, providers currently perform five to six HINE examinations per day, leading to increasing clinical competency and confidence. A year after the initial training program, a repeat observation of clinical examinations prevented drift and clarified questions. When trained providers left and new providers were incorporated, a systematic training process and team-support system helped overcome this new challenge.
The documentation implementation was overall a success, with streamlined notes and flow sheets routinely used for the GMA and HINE. However, a noted failure was lack of compliance with an additional documentation form used in the Comprehensive Cerebral Palsy program for older children. The goal of filling out this flow sheet in addition to regular Epic documentation was to provide data for the Learn from Every Patient project
and facilitate transitions. However, the added burden of repetitive documentation proved unsustainable and use of the forms was paused while new processes were developed to capture the needed information without duplication of effort. A streamlined referral process was devised with the indication of “transition to the cerebral palsy program,” placed at 30 to 33 months to ensure continuity between the early and later cerebral palsy programs.
The success of the program in early identification prompted troubleshooting questions currently not addressed in the care of cerebral palsy patients. Partnering with physical and occupational therapists, orthopedic surgeons and orthotists specializing in the care of very young children and infants allowed a reframing of expectations and ordering patterns for upper and lower extremity orthotics. Evidence-base for truncal supports is also being reviewed, and guidelines for prescription in the clinic are still in development. Another area of critical importance to be addressed was hip surveillance in infants 12 months and under, in circumstances where the severity of cerebral palsy is uncertain or unstable at best
The development phase highlighted opportunities for future improvements. Early identification in the NICU of all infants at risk for cerebral palsy in addition to those already meeting criteria for follow-up became a priority. Patients referred to the NICU for reasons that are not commonly associated with neurological insults are neither routinely screened using neuroimaging or referred to the follow-up program (surgical intervention, routine preterm care from 32 to 37 weeks, respiratory distress in term newborns, hyperbilirubinemia or sepsis evaluations). Routine use of the GMA before discharge from the NICU can help screen for infants meeting the “infant-attributable risk” pathway of the international guidelines.
Results of a one-year experience using the GMA in the NICU for this purpose are forthcoming.
In addition, several other clinics at the hospital that follow children at high risk for cerebral palsy have not implemented the guidelines. Many of these children are referred when pediatricians identify the need for a specialist after age 3 years, whereas others were never cared for in a NICU. Opportunities and willingness currently exist to adapt the guidelines to these settings to promote more consistent early care for high-risk infants. Community outreach and education to support pediatric providers through education materials and care coordination is ongoing.
Finally, this single site experience in implementation processes prompted interest at the national level for implementation of the international guidelines. The Cerebral Palsy Foundation therefore initiated the creation of a structured implementation network across six major US sites in July 2017
to be expanded over three years in a widening US-specific network. Knowledge translation, toolkits, training, and infrastructure will be provided to all participants, many derived from this single site experience.
Conclusions
This work demonstrates the first successful attempt at implementing the international early diagnosis and treatment guidelines for cerebral palsy in a clinical setting in the United States. The process feasibly lends itself to adaptation to other settings. The current experience highlights the need for new clinical care guidelines and novel intervention research for infants with cerebral palsy diagnosed under age 2 years.
Authors' Contributions
NM conceptualized the implementation process and the research study. RB contributed to the research design and wrote the first draft of the manuscript. GN contributed to the implementation process and draft of the manuscript. All authors in the NCH Early Developmental Group designed and participated in components of the implementation process and contributed to the manuscript. All authors read and approved the final manuscript.
We would like to thank all our patients and their families. In addition, we would like to thank Ms. Rebecca Lam and Ms. Olena Chorna for their help in preparing this manuscript and obtaining all regulatory materials. NCH Early Developmental Group includes Rukhshana Ahmed, Erin Anderson, Nancy Batterson, Lawrence Baylis, Brooke Bernstein, Teresa Borghese, Regina Bray, Chelsea Britton, Marie Cain, Ilene Crabtree, Caitlin Cummings, Kelli Dilver, Jaclyn Gray, Jennifer Haase, Andala Hardy, Megan Harrison, Alyssa Huddleston, Cherie Isler, Cheryl Kokensparger, Beth Kosch, Joanna Kinner, Jessica Lewis, Ryan Nicoll, Brianne Palay, Audrey Shifflett, Laurel Slaughter, Michelle Smith, Annie Temple, Kristy Theller, Lindsey Tobias, Jill Tonneman, Elizabeth Wagner, Kylie Weber.
Appendix. Supplementary data
The following is the supplementary data to this article:
Prevalence of cerebral palsy, co-occurring autism spectrum disorders, and motor functioning—Autism and Developmental Disabilities Monitoring Network, USA, 2008.
Conflicts of interest: All the authors declare that they have no competing interests. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
30782-8&id=pnu9207-fig-0001.jpg){kind=link}
30782-8&id=pnu9207-fig-0002.jpg){kind=link}
30782-8&id=pnu9207-fig-0003.jpg){kind=link}