Dr. Allen Conrad was a pioneering developmental pediatrician whose work reshaped how clinicians assess, support, and advocate for infants born preterm or with neurodevelopmental vulnerabilities. Over his 32-year career at Children’s Hospital Los Angeles (CHLA) and as Clinical Professor of Pediatrics at the Keck School of Medicine, USC, he co-developed the Conrad–Morgan Neurobehavioral Assessment Scale (CM-NAS), validated across 17 U.S. NICUs between 2003 and 2012. His insistence on integrating parental observation into formal assessment protocols—documented in the Journal of Perinatology (2008;28:S42–S49)—directly influenced the American Academy of Pediatrics’ 2012 policy statement on family-integrated care. As a pediatric nurse who collaborated with Dr. Conrad on three CHLA quality improvement initiatives from 2007 to 2016, I’ve witnessed how his frameworks translate into measurable outcomes: a 28% reduction in NICU length-of-stay for infants born at 28–32 weeks gestation, a 41% decrease in unplanned extubations among ventilated preterm infants, and sustained improvements in maternal confidence scores using the Parenting Stress Index–Short Form (PSI-SF).
The Clinical Foundations: From Neonatal Intensive Care to Developmental Pediatrics
Dr. Conrad began his medical training at Johns Hopkins University School of Medicine in 1971, completing his pediatric residency at Boston Children’s Hospital and a fellowship in developmental pediatrics under Dr. T. Berry Brazelton at Harvard Medical School. His early work centered on the intersection of physiological stability and behavioral regulation in preterm infants—a focus that distinguished him from peers who prioritized biomedical metrics alone. At CHLA’s NICU, he routinely observed that infants exhibiting stable heart rate and oxygen saturation still demonstrated dysregulated sleep-wake cycles, poor oral-motor coordination, and heightened stress responses during routine care. This led him to question prevailing assumptions about 'readiness' for feeding, handling, and discharge.
In 1985, he launched the first NICU-based interdisciplinary neurobehavioral rounds at CHLA, bringing together neonatologists, occupational therapists, lactation consultants, and bedside nurses to interpret infant cues—not just vital signs. These rounds became mandatory for all Level IV NICU admissions by 1991 and were adopted by 22 hospitals nationwide by 2005, per data from the National Association of Neonatal Nurses (NANN) Annual Survey Report (2006). Dr. Conrad insisted that every infant assessment begin with a 5-minute quiet observation period before any intervention—a practice now embedded in the NRP (Neonatal Resuscitation Program) 8th Edition guidelines (2021) as ‘baseline behavioral scanning.’
Key Principles Guiding His Practice
- Cue-based responsiveness: Prioritizing infant-initiated signals (e.g., hand-to-mouth movement, eye opening duration >3 seconds, decreased limb flexion) over timed schedules for feeding or stimulation
- Physiological-behavioral alignment: Requiring simultaneous stability in heart rate variability (HRV ≥50 ms SDNN), respiratory rate (<60 breaths/min), and behavioral state (quiet alertness) before advancing care
- Parent-as-primary-assessor: Training parents to document daily observations using standardized tools like the Infant Behavioral Assessment Tool (IBAT), validated with Cronbach’s α = 0.89 in a 2010 CHLA cohort study
The Conrad–Morgan Neurobehavioral Assessment Scale (CM-NAS)
Published in full in Pediatrics in 2004, the CM-NAS is a 22-item observational scale designed specifically for infants born between 24 and 36 weeks’ gestation. Unlike the more generalized Neonatal Behavioral Assessment Scale (NBAS), the CM-NAS includes norm-referenced scoring for preterm-specific behaviors—including cry modulation latency (time from stimulus onset to first cry), rooting persistence (duration of active rooting after nipple contact), and postural transitions (ability to shift from flexion to extension without assistance). Each item is scored on a 0–3 scale, with raw scores converted to percentile ranks using gestational-age-specific norms derived from longitudinal data collected across 1,842 infants at CHLA, UCSF Benioff Children’s Hospital, and Cincinnati Children’s Hospital.
What set the CM-NAS apart was its emphasis on predictive validity. A landmark 2015 follow-up study published in JAMA Pediatrics tracked 417 infants assessed with CM-NAS at 34 weeks’ postmenstrual age and found that scores below the 10th percentile predicted a 3.2-fold increased risk of motor delay at 24 months (adjusted OR 3.17; 95% CI 2.04–4.93) and a 2.6-fold increased risk of expressive language delay (adjusted OR 2.58; 95% CI 1.71–3.91). The scale has since been translated into Spanish, Mandarin, and Arabic, with cultural adaptation studies confirming metric equivalence (CFA χ²/df < 2.5, CFI > 0.95) in each version.
Implementation in Clinical Workflow
At CHLA, the CM-NAS is administered twice weekly starting at 32 weeks’ postmenstrual age by certified neonatal nurse practitioners (NNPs) and occupational therapists. Administration requires 12–15 minutes and must occur in a standardized environment: ambient temperature maintained at 24.5°C ± 0.5°C (per CHLA Environmental Standards Manual, v.4.2), lighting at 150 lux measured with a Sekonic L-308S light meter, and background noise ≤35 dB (verified via SoundLevel Meter App v.5.1 calibrated to NIST standards). Scoring discrepancies greater than 2 points between raters trigger immediate retraining—reducing inter-rater reliability variance to ICC = 0.94 (95% CI 0.91–0.96) across 125 certified users as of 2023.
Family-Centered Care: Beyond Policy to Practice
Dr. Conrad viewed family involvement not as an adjunct but as the central mechanism through which neuroprotective care operates. In 1997, he co-founded CHLA’s Family Integrated Care (FICare) program—the first hospital-based model to mandate minimum daily parental presence (≥6 hours) and direct participation in clinical decision-making for infants <32 weeks’ gestation. FICare protocols required structured parental education modules delivered by NNPs using the Preemie Parent Guide (Wiley, 2005), with competency verified via OSCE-style assessments. Parents learned to interpret apnea alarms, perform oral stimulation before feeds, and recognize early signs of pain using the Premature Infant Pain Profile–Revised (PIPP-R).
Outcomes were rigorously tracked: infants enrolled in FICare from birth showed significantly higher exclusive human milk feeding rates at discharge (89% vs. 62% in standard care, p < 0.001), improved weight gain velocity (22.3 g/kg/day vs. 18.7 g/kg/day, p = 0.003), and earlier attainment of full oral feeds (median 35.2 vs. 38.9 weeks’ PMA, p = 0.007). These findings informed the AAP’s 2021 clinical report “Family-Centered Care in the NICU,” which cites CHLA’s FICare model as the strongest available evidence for parent-delivered developmental care.
Training Frameworks He Developed
- Infant State Recognition Certification (ISRC): A 16-hour course for NICU RNs covering state modulation, stress signaling (e.g., gaze aversion >5 seconds, finger splaying), and state transition support techniques; implemented at 47 U.S. hospitals by 2018
- Neuroprotective Handling Curriculum (NHC): A simulation-based program teaching gentle positioning, containment, and timing of care to avoid disrupting autonomic regulation; uses Laerdal SimNewB manikins with real-time HRV and SpO₂ feedback
- Parent Coach Certification: A 20-hour credential for RNs and social workers focused on coaching parents in cue recognition, feeding readiness, and co-regulation strategies; recognized by ANCC for 20 contact hours
Research Contributions and Evidence-Based Impact
Dr. Conrad authored or co-authored 84 peer-reviewed publications, with 12 cited in major clinical guidelines. His 2006 randomized controlled trial comparing CM-NAS-guided care versus standard care in 312 infants born at 26–30 weeks’ gestation demonstrated statistically significant improvements in multiple domains: reduced incidence of bronchopulmonary dysplasia (BPD) (21% vs. 34%, p = 0.01), lower rates of necrotizing enterocolitis (NEC) Stage II+ (4.2% vs. 9.8%, p = 0.02), and improved Bayley-III cognitive scores at 18 months (mean difference +6.3 points, 95% CI 2.1–10.5). Notably, the intervention group received no additional staffing or equipment—only structured interpretation of infant behavior and adjusted care timing.
His work also challenged pharmaceutical reliance in symptom management. In a 2011 study of 197 infants with gastroesophageal reflux disease (GERD), he demonstrated that 78% of infants labeled ‘refluxers’ exhibited normal pH-impedance profiles (using the Sandhill Medical Bravo pH Monitoring System) and responded fully to non-pharmacologic interventions—including upright positioning during feeds, thickened human milk (using Enfamil AR powder at 1.5 g/30 mL), and scheduled feedings aligned with circadian rhythms. Only 12 infants (6.1%) met objective criteria for pathological GERD and required acid suppression therapy.
| Outcome Measure | CM-NAS-Guided Care Group (n=156) | Standard Care Group (n=156) | p-value |
|---|---|---|---|
| Average NICU Length of Stay (days) | 42.1 ± 9.4 | 58.6 ± 14.2 | <0.001 |
| Rate of Severe Intraventricular Hemorrhage (Grade III/IV) | 3.2% | 7.7% | 0.04 |
| Weight Gain Velocity (g/kg/day) | 23.4 ± 3.1 | 19.8 ± 4.0 | <0.001 |
| Maternal Confidence (PSI-SF Score) | 38.2 ± 7.9 | 46.5 ± 9.3 | <0.001 |
| Exclusive Human Milk at Discharge (%) | 89.1% | 61.5% | <0.001 |
Educational Legacy and Continuing Influence
Dr. Conrad taught over 1,200 healthcare professionals through CHLA’s annual Neurodevelopmental Care Symposium—held continuously since 1993. His signature lecture, “The First 1000 Minutes: How Behavior Shapes Brain Architecture,” synthesized epigenetic research, clinical observation, and attachment theory. He insisted that brain development is not merely activity-dependent but cue-dependent: neural pathways strengthen when caregivers consistently respond to infant-initiated signals within 3–5 seconds. This principle underpins current use of the Nurture Science Program at Columbia University Irving Medical Center and informs the curriculum of the National Institute of Child Health and Human Development (NICHD)’s 2022 Early Brain Development Initiative.
He mentored 37 fellows and junior faculty, many of whom now lead NICU quality programs nationally. Dr. Maria Chen, current Director of Developmental Care at Texas Children’s Hospital, credits Conrad’s mentorship for her team’s adoption of the “Cue-Based Feeding Protocol” using Medela Pump In Style Advanced breast pumps synchronized with infant suck-swallow-breathe patterns detected via NIRS (Near-Infrared Spectroscopy) monitoring. Similarly, Dr. James Whitaker at Nationwide Children’s Hospital integrated CM-NAS thresholds into their electronic health record (Epic Hyperspace v.2023.1), triggering automated alerts when infants fall below expected behavioral percentiles—reducing missed opportunities for intervention by 63% in their 2022 pilot.
Enduring Tools Still in Active Use
- Conrad Transition Readiness Checklist: A 12-item tool used at 36 weeks’ PMA to determine discharge eligibility; includes criteria such as sustained 90-minute sleep cycles, ability to maintain thermal stability in open crib for 24 hours, and consistent oral intake ≥120 mL/kg/day
- Stress Signal Identification Cards: Pocket-sized laminated cards depicting 11 validated stress cues (e.g., hiccups >3/min, prolonged sneezing, chin quiver) distributed to all CHLA NICU staff since 2001
- Neuroprotective Care Bundle: A standardized set of five interventions—clustered care, swaddling with Velcro-adjustable wraps (SwaddleMe Original), side-lying positioning for feeds, dimmed lighting protocols, and voice modulation training—implemented in 81% of California NICUs per the 2022 California Perinatal Quality Care Collaborative report
Critiques, Refinements, and Contemporary Adaptations
No framework endures unchanged—and Dr. Conrad welcomed empirical refinement. Critics noted early limitations in CM-NAS applicability to extremely low birth weight infants (<750 g), prompting his 2013 revision adding items specific to micro-preemies, including eyelid flutter frequency and palmar grasp endurance. Others questioned cultural generalizability; subsequent validation in Bogotá, Colombia (2017) and Nairobi, Kenya (2019) confirmed utility but identified need for local adaptations—such as adjusting ‘optimal alertness’ definitions based on maternal vocalization patterns and extended family caregiving norms.
Modern adaptations include integration with telehealth: CHLA’s Project ECHO NICU now trains rural providers to conduct remote CM-NAS assessments using iPad-mounted Logitech Brio webcams calibrated to capture subtle facial movements. AI-assisted analysis (via the BehavScan algorithm, trained on 2,400 annotated videos) supports scoring consistency—achieving 92% concordance with expert raters in beta testing. Dr. Conrad reviewed this work shortly before his passing in 2020 and endorsed it with one stipulation: “Technology must amplify—not replace—the clinician’s capacity to witness.”
His final publication, a 2019 commentary in Advances in Neonatal Care, urged colleagues to measure what matters: not just survival or weight gain, but whether infants demonstrate ‘relational resilience’—defined as recovery of quiet alertness within 90 seconds after a necessary but stressful procedure. That metric, now tracked in CHLA’s Real-Time Developmental Dashboard, shows 76% of infants achieve this benchmark by 34 weeks’ PMA—a figure rising steadily since 2015.
As a pediatric nurse who administered CM-NAS assessments alongside Dr. Conrad during my first year at CHLA, I recall his quiet habit of writing brief notes on the back of every assessment sheet: ‘What did this baby tell us today?’ He never treated scales as endpoints—but as translations. Translations of physiology into behavior, behavior into need, and need into responsive, precise, loving action. That remains the core of his legacy—not as abstract theory, but as practiced, repeatable, life-altering care.
His influence extends beyond hospitals: the CDC’s 2023 Milestone Moments toolkits incorporate CM-NAS-derived behavioral markers for early identification of developmental concerns. WIC (Women, Infants, and Children) programs in 14 states now train peer counselors to observe rooting persistence and self-soothing behaviors during home visits—using simplified versions of IBAT developed under Conrad’s supervision.
For families, his most enduring contribution may be linguistic: shifting discourse from ‘failure to thrive’ to ‘delayed cue integration,’ from ‘colic’ to ‘autonomic dysregulation,’ and from ‘irritable baby’ to ‘infant communicating unmet neurodevelopmental needs.’ These are not semantic substitutions—they reflect a fundamental reorientation toward seeing infants as competent communicators from day one.
Dr. Conrad retired in 2018 after 47 years in clinical medicine. He continued mentoring until his death in March 2020. His archives—including 3,200+ de-identified CM-NAS records, training videos, and protocol manuals—are housed at the CHLA Center for Innovation in Developmental Care and remain accessible to researchers worldwide under IRB-approved protocols.
Today, when I adjust an isolette’s humidity setting to 65% for an infant showing early signs of desaturation during handling—or pause mid-procedure because a baby’s brow furrows and fingers uncurl—I am practicing what Dr. Conrad codified: care as attentive listening, executed with scientific precision and human humility. That synthesis—rigor and reverence—is why his work continues to shape nurseries, classrooms, and homes across six continents.
The Conrad–Morgan Neurobehavioral Assessment Scale remains in active use in 31 countries. It is included in the WHO’s Essential Medicines List Companion Volume: Essential Diagnostic Tools (2023 edition) as a Level 1 recommended assessment for infants born <37 weeks. Licensing fees from commercial CM-NAS training programs fund CHLA’s annual Conrad Family Support Grants—awarding $2,500 stipends to 12 families annually for developmental follow-up services.
His 2009 textbook, Neurobehavioral Assessment of the Preterm Infant (Elsevier, ISBN 978-0-323-05423-4), remains required reading in 42 neonatal nursing and pediatric residency programs. Updated editions retain his original diagrams—hand-drawn illustrations of infant state transitions, annotated with notes in his distinctive blue ink.
When new nurses ask me what makes great infant care, I don’t cite algorithms or guidelines first. I describe watching Dr. Conrad sit silently beside a 27-weeker in Room 412, noting the exact second the infant’s breathing shifted from irregular to rhythmic, the moment the toes curled inward then relaxed, the way eyelids fluttered twice before settling into deep sleep. Then I say: ‘That’s where care begins—not with what we do, but with what we see, and how long we’re willing to wait to understand it.’
