O-01: Circadian medicine across pediatric populations

Sleep-Wake Circadian Rhythm in Children and Adolescents with Epilepsy

Presenting Author

Antonella Iadarola (Italy)

Authors

Antonella Iadarola (Italy), Marco Veneruso (Italy), Ramona Cordani (Italy), Margherita Mancardi (Italy)

Introduction

Epilepsy is a neurological disorder characterized by recurrent seizures that can significantly affect cognitive, emotional, and behavioral functioning. Sleep disturbances are common in individuals with epilepsy, and the relationship is bidirectional: seizures and interictal epileptic discharges can disrupt sleep architecture, while poor sleep may lower the seizure threshold. Although previous studies have explored sleep quality and architecture in pediatric epilepsy, the impact on rest-activity and sleep-wake rhythm regulation remains less understood. This study aimed to investigate sleep patterns and circadian rhythm characteristics in children and adolescents with epilepsy compared to healthy controls.

Materials and Methods

Eighty-six children and adolescents diagnosed with epilepsy according to the 2017 ILAE classification and 51 age-matched healthy controls were included in this observational cross-sectional study. Sleep and circadian parameters were assessed using actigraphy over 7–10 days and standardized questionnaires. Group differences were analyzed using ANCOVA, adjusting for confounders. A Functional Linear Model

(FLM) was applied to examine differences in activity patterns across the 24-hour cycle.

Results

Children with epilepsy showed significantly later wake times compared to controls (p<0.05). They also exhibited lower interdaily stability and higher intradaily variability, indicating weaker and more fragmented circadian rhythms (p<0.01). FLM analysis confirmed altered 24-hour activity patterns in the epilepsy group, showing reduced activity levels during the morning and afternoon and increased activity between 2 am and 4 am.

Conclusions

These findings highlight that children and adolescents with epilepsy not only experience altered sleep timing but also show less stable and more fragmented circadian rest-activity patterns compared to healthy peers. The combination of objective actigraphic measures and functional time-series analysis provides new insights into how epilepsy may disrupt biological rhythms beyond sleep architecture alone. Targeting circadian misalignment could represent an additional strategy to improve seizure control, daytime functioning, and overall quality of life in this vulnerable population.

Overview of circadian rhythm disorders in children with cystic fibrosis in the era of CFTR modulators

Presenting Author

Coutier Laurianne (France)

Authors

Chen Catherine (France), Guyon Aurore (France), Gronfier Claude (France), Taytard Jessica (France), Portefaix Aurelie (France), Ioan Iulia (France), Reix Philippe (France), Franco Patricia (France), Coutier Laurianne (France)

Introduction

The development of sleep disorders in adults with cystic fibrosis (CF) on ELEXACAFTOR-TEZACAFTOR-IVACAFTOR (ETI) has been reported. No chronotype studies have been performed with ETI in either children or adults.

The aim of the study was to evaluate the nature of the chronotype of children treated with ETI and their sleep quality.

Materials and Methods

A prospective three-center study in France (Lyon, Paris-Trousseau, Nancy) was conducted in 2024 in patients aged 2-18 years with CF under ETI. Questionnaires (chronotype, sleep disorders, sleepiness, hyperactivity, anxiety-depression), actimetry (7 days) and urinary melatonin (24h) were carried out.

Results

94 patients included (46 boys, median age [min-max]: 9.1 years (2.8-17.7) of whom 40 (43%) reported sleep disorders (mostly at 2-5 years). Among them, 32 (80%) had no sleep disorders before ETI or pre-existing sleep disorders amplified under ETI. 15 patients had a therapeutic modification with partial improvement. Sleep onset insomnia was the most frequent (29%). 13% were of an evening type according to questionnaires and 30% according to urinary melatonin. Actimetry showed a late bedtime (60%), a short sleep duration (30%) and a fragmented sleep (50%). 37% of patients had a high anxiety score.

Conclusions

There is a high incidence of sleep disorders in children and adolescents treated with ETI, occurring in the first 6 months after initiation. Final analyzes will help to better understand their pathophysiology (anxiety-depression and/or chronotype abnormalities).

Stability of Circadian Phase and Its Relationship with Sleep Timing Across Early Childhood

Presenting Author

Lauren Hartstein (United States)

Authors

Lauren Hartstein (United States), Monique LeBourgeois (United States)

Introduction

Early childhood represents a period of profound developmental changes in brain maturation and sleep consolidation. However, little is known about the developmental trajectory of the circadian system and its relationship with sleep timing across this period in which children begin school and transition to monophasic sleep. Understanding this trajectory is essential for identifying factors that contribute to behavioral sleep problems. Here we provide fundamental data on the development of circadian phase and its relationship with sleep timing across early childhood.

Materials and Methods

Healthy children participated in a longitudinal study assessing circadian phase and sleep timing at four ages: 2-3 years (T1, N = 58), 3-4 years (T2, N = 49), 4-5 years (T3, N = 32), and 5-6 years (T4, N = 35). Children followed their habitual sleep schedule for 5 days, confirmed via actigraphy and daily sleep diaries. They then completed an in-home assessment of their dim light melatonin onset (DLMO), with saliva samples collected every 30 min throughout the evening until 1 h past habitual bedtime. DLMO was defined as the linear interpolated clock time at which melatonin levels crossed (and remained above) 4 pg/ml. Linear mixed-effects models tested whether DLMO, sleep timing (bedtime, sleep onset, midsleep time, sleep offset), or phase angles of entrainment varied across timepoints.

Results

Average DLMO was 19:30 ± 0:47 at T1, 19:30 ± 0:59 at T2, 19:16 ± 1:06 at T3, and 19:21 ± 1:10 at T4. DLMO preceded bedtime by an average of 36 ± 47 min at T1, 43 ± 60 min at T2, 52 ± 63 min at T3, and 36 ± 48 min at T4. There was no significant effect of timepoint on DLMO, sleep timing, or phase angles of entrainment (all p > 0.19). Later DLMOs were associated with significantly later sleep onsets and midsleep times at all timepoints. Later DLMOs were also associated with significantly later bedtimes at T1 (r = 0.43), T2 (r = 0.30), and T3 (r = 0.45), but not at T4 (r = 0.27). Finally, later DLMOs were associated with later sleep offsets at T1 (r = 0.55) and T4 (r = 0.47), but not at T2 (r = 0.12) or at T3 (r = 0.33).

Conclusions

These findings indicate that both circadian phase and its relationship with sleep timing remain remarkably stable across early childhood. These foundational data on early circadian physiology can help guide the development of targeted interventions for children with behavioral sleep problems, supporting better alignment between circadian rhythms and sleep timing.

Acknowledgements

Data collection was supported by the National Institute of Mental Health (K01-MH07643, R01-MH08566) awarded to the late Dr. Monique LeBourgeois.

Differences in sleep spindle characteristics between children with narcolepsy and those with other causes of subjective sleepiness.

Presenting Author

Rosemary Horne (Australia)

Authors

Marisha Shetty (Australia), Nushrika Kadam Kishore  (Australia), Georgina Plunkett (Australia), Lauren Nisbet (Australia), Margot Davey (Australia), Gillian Nixon (Australia), Lisa Walter (Australia), Rosemary Horne (Australia)

Introduction

Central disorders of hypersomnolence are characterised by excessive daytime sleepiness. There has been growing interest in analysing sleep micro-architecture, to further understand how sleep is affected by these conditions. We investigated if overnight sleep spindle activity differed between children diagnosed with narcolepsy, idiopathic hypersomnolence (IH), and subjectively sleepy children (defined as having a non-diagnostic multiple sleep latency test (MSLT)), and matched controls.

Materials and Methods

All children having an overnight polysomnographic (PSG) study followed by a MSLT for investigation of excessive daytime sleepiness (EDS) between May 2010 to December 2023 were eligible. Children were excluded from analysis if they had an obstructive apnoea-hypopnoea index (OAHI) >5 events/h, or were using any medications with the potential to affect REM sleep on the day of the MSLT. Children with narcolepsy had a mean sleep latency ≤8 min (an average of the time taken to fall asleep over 4 or 5 naps on their MSLT) AND ≥2 sleep onset REM periods (SOREMPs) within the first 15 minutes of falling asleep (on MSLT or overnight PSG). Children with IH had a mean sleep latency ≤8 min AND <2 SOREMPs. A third group of children, who did not meet the MSLT criteria for either narcolepsy or IH was also included, the subjectively sleepy group. Control children had been referred for PSG for reasons other than assessment of EDS and were individually matched for OAHI, age and sex. Spindle activity was characterised as spindle duration (in seconds), spindle density (number of spindles/h) and intensity (density x average duration) on central (C4) and frontal (F4) EEG electrodes across the sleep period and also when separated into sequential NREM sleep periods (NREMP) across the night.

Results

28 children were diagnosed with narcolepsy, 11 with IH and 26 with subjective sleepiness (aged 6-19 years). Spindle duration was longer in children with narcolepsy compared to their controls (p<0.05 for all) and compared to those with subjective sleepiness (p<0.05 for all). In the children with narcolepsy and subjective sleepiness, C4 spindle intensity was greater (p<0.05 for all) compared to controls. C4 spindle density and intensity were lower at the beginning compared to the end of the night in the children with narcolepsy and subjective sleepiness (p<0.05 for all).

Conclusions

Children with narcolepsy have longer spindles and more intense spindle activity than controls. The subjectively sleepy group also had higher spindle intensity than controls, despite the fact that they had a non-diagnostic MSLT, suggesting that this group do have differences in their sleep micro-architecture despite not meeting the strict MSLT criteria. This group likely includes young people who have evolving narcolepsy or IH but who do not meet the current adult criteria for diagnosis. Understanding the microstructure of sleep in combination with clinical history may aid in distinguishing between groups of children referred for assessment of EDS, particularly those who do not meet strict diagnostic criteria for a primary disorder of hypersomnolence but did have abnormal MSLT results, such as our subjectively sleepy group.

Acknowledgements

We wish to thank all the parents and children who took part in this study and acknowledge the staff of the Melbourne Children's Sleep Centre where the studies were carried out. Prof Rosemary Horne is supported by a National Health and Medical Research Council of Australia Investigator Grant (1195453).

Circadian Misalignment and Cognitive Performance in Adolescents: Evidence from a School-Age Cohort

Presenting Author

Tabata Luna Garavazzo Tavares (Brazil)

Authors

Tabata Luna Garavazzo Tavares (Brazil), Larissa Nicolini de Santa (Brazil), Marina Foresti dos Santos (Brazil), Paula Azevedo Kageyama Longo (Brazil), Fabio Carmona (Brazil), Alan Luiz Eckeli (Brazil)

Introduction

Evening chronotype has been associated with poorer academic performance and increased risk of mental and physical health problems in adolescents, likely mediated by misalignment between biological rhythms and early school schedules (social jetlag). Although many studies link circadian misalignment to cognitive and academic impairment, the relationship between chronotype and intelligence remains controversial. Evidence also suggests that alignment between sleep timing and circadian preference may be more relevant to cognition than sleep quantity alone. In this study, our objective was to evaluate the association between chronotype, circadian misalignment, and cognitive and academic outcomes in adolescents from a birth cohort followed into school age.

Materials and Methods

This analytical retrospective cohort study used data from the BRISA project – school-age phase (Ribeirão Preto, Brazil). Cognitive performance was assessed using the WISC-IV, and chronotype was measured by the Children’s Chronotype Questionnaire (CCTQ). Additional data included school shift, grade repetition, and sociodemographic characteristics. Statistical analyses included log-binomial regression and beta-binomial regression models.

Results

The sample included 3,258 adolescents, 51.7% male, mean age 11.8 years. Chronotype distribution was 42.6% morning, 10.6% intermediate, and 46.7% evening types. Circadian alignment showed 65.2% aligned and 34.8% misaligned with school schedules. Grade repetition occurred in 9.5% of students, with no significant difference between aligned and misaligned groups. Mean Full-Scale IQ was 96.75 (SD 16.30). Misaligned adolescents showed lower IQ (95.9) compared to aligned peers (97.2, p = 0.03), but the mean scores of both groups fall within the same WISC-IV classification.

Conclusions

Although WISC-IV scores from both groups fall within the same classification, circadian misalignment between chronotype and school schedules is associated with lower cognitive performance in adolescents, although not with grade repetition. These findings highlight the importance of educational policies that consider adolescent sleep biology to reduce social jetlag and support improved cognitive and academic functioning.

Acknowledgements

The authors thank the University of São Paulo, the BRISA research team, the Brazilian Ministry of Health, and CAPES for their support.

The long-term clinical outcome of pediatric Narcolepsy type 1

Presenting Author

Antonella Barbieri (Italy)

Authors

Antonella Barbieri (Italy), Martina Gnazzo (Italy), Fabio Pizza (Italy), Francesco Biscarini (Italy), Stefano Vandi (Italy), Lino Nobili (Italy), Giuseppe Plazzi (Italy)

Introduction

Narcolepsy type 1 is a rare sleep disorder primarily characterized by excessive daytime sleepiness (EDS) and cataplexy, with higher incidence in childhood and adolescence. Despite that, little is known about the long-term outcomes of pediatric narcolepsy and the predictors of progression: our goal was describing symptoms’ long-term evolution and identifying prognostic factors that may influence long-term evolution of sleepiness and cataplexy.

Materials and Methods

We retrospectively selected a continuous cohort of 296 patients diagnosed with pediatric onset of Narcolepsy Type1 from 1990 to 2025 at the Narcolepsy Center at IRCCS Istituto delle Scienze Neurologiche in Bologna. Information was collected from diagnosis and follow-up at 1-, 5-, and 10-years post-diagnosis, including demographic and anthropometric data, diagnostic delay, clinical features, orexin levels, polysomnographic data, and Multiple Sleep Latency Test (MSLT) results. At each follow-up, we recorded anthropometric measures, ongoing treatments, the Epworth Sleepiness Scale (ESS), symptom changes, and global clinical impressions, along with any medical and psychiatric comorbidities present at onset or during follow-up.

Using the collected data, we performed longitudinal assessments (Cochran’s Q test) to evaluate the clinical course of all narcolepsy symptoms. We then conducted preliminary univariate analyses to identify factors associated with the control of daytime sleepiness. Based on these results, we selected the variables that showed at least a trend toward association (p < 0.1) with sleepiness control and included them in a mixed linear model to determine which factors were independently related to better or poorer symptom control.

Results

Longitudinal analyses demonstrated a statistically significant improvement in all symptoms. Furthermore, the study highlights that when considering the long-term evolution of the main symptom of NT1, namely excessive daytime sleepiness, specific clinical and therapeutic factors are associated with disease outcome. Over the ten-year follow-up period, we observed a general trend toward improved control of daytime sleepiness, although some negative prognostic indicators emerged, including greater baseline severity of symptoms such as sleep paralysis and cataplexy, as well as the development of comorbidities such as obesity. Conversely, treatment with sodium oxybate was consistently associated with better long-term symptom control, confirming its key role in the management of narcolepsy type 1.

Conclusions

The wide population of this study allowed a meaningful assessment of changes in evolution of pathology over time and in relation to different factors, showing an overall improving trend in all narcolepsy symptoms and it also allowed us to identify several factors associated with the clinical outcome of daytime sleepiness specifically. In the future, our goal is to expand these analyses to determine which factors are associated to other symptoms control and with the overall course of narcolepsy, ultimately aiming to improve the clinical and therapeutic management of these patients.

O-02: Pediatric sleep disordered breathing: Mechanisms, outcomes, and interventions

Unravelling the Inflammatory Signature of Paediatric Obstructive Sleep Apnoea: Connecting Upper Airway and Circulation

Presenting Author

Ming Yang (China)

Authors

Ming Yang (China), Joseph Gar Shun Tsun (China), Kathy Yuen Yee Chan (China), Ebe Tsz Yau Jan (China), Chun Ting Au (Canada), Nobel Tsz Kin Yuen (China), Ting Ting Chen (China), Albert Martin Li (China), Kate Ching-ching (China), Natalie Moon Wah Leung (China), Alice Kwai Yee Siu (China)

Introduction

Paediatric obstructive sleep apnoea (OSA) is a prevalent sleep-related breathing disorder characterised by recurrent upper airway obstruction, intermittent hypoxia, and sleep disruption. Emerging evidence suggests that inflammatory processes may contribute to both the airway pathology and systemic consequences observed in affected children. Nasal mucosal inflammation may have a bidirectional relationship with upper airway resistance and may be potentially driven by the vibration and mechanical stress caused by snoring. Systemic inflammatory changes could signal broader physiological effects with potential relevance to cardiovascular and metabolic health. This study aimed to investigate cytokine profiles in both the upper airway and circulation of children with OSA, and to examine the relationship between nasal inflammation and systemic inflammatory response.

Materials and Methods

Children aged 5 to 12 years underwent overnight polysomnography. OSA was diagnosed based on the presence of OSA symptoms and an obstructive apnoea-hypopnoea index (OAHI) ≥ 1 event/hour. Nasal epithelial lining fluid (NELF) and plasma samples were collected for cytokine analysis. Correlation analysis was conducted to examine the relationship between cytokine levels and the severity of OSA. Additionally, quantile regression was employed to assess this relationship across the distribution of cytokine levels, while adjusting for potential confounding factors.

Results

One hundred and thirty-one children (69.9% boys) with a mean age of 8.78 years (SD = 1.80) were recruited in this study. Fifty-nine children (55.7%) suffered from OSA, 99 (75.6%) participants had allergic rhinitis, and 19 (14.5%) had physician-diagnosed asthma. IL-10, IL-12p70, and IL-17A levels in NELF were moderately correlated with their corresponding plasma cytokines (r = 0.45, p < 0.001; r = 0.41, p < 0.001; r = 0.41, p < 0.001). IL-5 and IL-6 showed weaker but significant correlations between NELF and plasma (r = 0.39, p < 0.001; r = 0.29, p = 0.002). Nasal IL-1β was significantly associated with OAHI in children at the upper tail of the cytokine distribution, after adjusting for potential confounders (β = 0.230, 95% CI:

(0.06,0.4), p = 0.011; bootstrap 95% CI: (0.09,0.37), p = 0.001), independent of age, sex, BMI z-score, and allergic rhinitis. In contrast, plasma IL-17A was associated with OSA severity at the 25th percentile (β = 0.734, 95% CI: (0.44,1.02), p < 0.001; bootstrap 95% CI: (0.25,1.22), p = 0.004) and the 50th percentile (β = 0.556, 95% CI: (0.29,0.83), p < 0.001; bootstrap 95% CI: (0.06,1.05), p = 0.030), but no significant association was observed at the upper quantile.

Conclusions

Our findings demonstrate a moderate correlation between inflammatory patterns in NELF and plasma. Children with OSA exhibit evidence of localised inflammation in the upper airway, characterised by elevated IL-1β. Systemic inflammation presented a different cytokine profile characterised by IL-17A, aligning with some current evidence. The differential cytokine patterns observed between NELF and plasma highlight the importance of considering both local and systemic inflammatory responses in paediatric OSA, suggesting mechanisms beyond a simple spill-over effect.

Acknowledgements

Commissioned Paediatric Research at Hong Kong Children’s Hospital, Health and Medical Research Fund, Health Bureau, The Government of the Hong Kong SAR (Ref: PR-CUHK-4).

Surgery rates after intranasal sprays for sleep disordered breathing in children

Presenting Author

Gillian M Nixon (Australia)

Authors

Gillian M Nixon (Australia), Alice Baker (Australia), Andrew Davidson (Australia), Anneke Grobler (Australia), Karen Davies (Australia), Amanda Griffiths (Australia), Harriet Hiscock (Australia), Haytham Kubba (United Kingdom), Sarath Ranganathan (Australia), Joanne Rimmer (United Kingdom), Elizabeth Rose (United Kingdom), Katherine Rowe (United Kingdom), Chris Selman (United Kingdom), Catherine Simpson (United Kingdom), Kirsten P Perrett (United Kingdom)

Introduction

Intranasal mometasone furoate and saline sprays resolve symptoms in almost half of children with sleep disordered breathing (SDB). However, the impact of medical treatment on the subsequent need for surgical treatment is unclear.

This study assesses whether intranasal mometasone furoate or saline reduces the rate of surgery for SDB over three years, and whether relevant clinical or other factors are associated with requiring surgery.

Materials and Methods

This study is the long-term follow-up of the MIST Study, a multicentre, randomised, double-blind, placebo-controlled trial involving children aged 3-12 referred to a specialist (otolaryngology or respiratory/sleep medicine) for SDB symptoms. Children were randomised to receive either intra-nasal mometasone furoate or saline daily for 6 weeks. The long-term outcome was surgery for SDB at 36-months following randomisation. Pre-specified variables at baseline and 6-week follow-up were assessed for their association with surgery.

Results

276 children were enrolled and 138 randomised to each treatment arm. The proportion of children who had surgery for SDB over three years was similar in the mometasone and saline groups (33% and 31% respectively; risk difference 1.6%; 95% CI -9.7 to 12.9%; p=0.78) with 15 patients overall lost to follow-up. Symptom resolution at six weeks was strongly associated with a lower likelihood of surgery (risk difference (RD) -18.5%; [95%CI -30.6%-6.3%]; p=0.003). Clinical factors that were associated with increased likelihood of surgery were mouth breathing on examination at both baseline (RD 22.3% [95%CI 9.8-34.8%]; p<0.001) and after treatment (RD 25.2% [95%CI 12.6-37.9%]; p<0.001); surgeon assessed need for surgery at baseline (RD 14% [95%CI 2.2-25.8%]; p=0.02) and after treatment (RD 22.5% [95%CI 10.8-34.2%]; p<0.001); and parent assessed need for surgery at baseline (RD 21.6% [95%CI 10.3-32.9%];

p<0.001) and after treatment (RD 26.3% [95%CI 14.4-38.1%]; p<0.001). Severity of SDB symptoms using a Pediatric Sleep Questionnaire-SDB subscale score ≥0.33 at the post treatment visit was associated with a higher risk of surgery (RD 22.7% [95%CI 10.6 – 34.9%]; p<0.001), but symptoms at baseline was not (RD 12.5% [95%CI -3.8 – 28.8%]; p=0.133). However, severe symptoms based on an OSA-5 score ≥5 at both baseline (RD 18.7% [95%CI 7.3% - 30.2%]; p=0.001) and after treatment (RD 21.1% [95%CI 8.1 – 34.2%]; p=0.001) were associated with surgery.

Conclusions

Children with SDB whose symptoms respond to six weeks of nasal spray treatment, with either mometasone furoate or saline, are less likely to need surgery over the following three years, adding further evidence for the use of nasal spray as a first line treatment for SDB. Mouth breathing, surgeon and parent preference and symptom burden were associated with a higher proportion of children having surgery.

Acknowledgements

The study was funded by a research grant from the Garnett Passe and Rodney Williams Memorial Foundation.

Comparison between Chan-Parikh and IPSES scales in Predicting persistent pediatric Obstructive Sleep Apnea after adenotonsillectomy

Presenting Author

Neemias Santos Carneiro (Brazil)

Authors

Neemias Santos Carneiro (Brazil), Silke Ana Theresa (Brazil), Renato Batistel Santana (Brazil), Antonio Carlos Marão (Brazil), Camila de Castro Corrêa (Brazil), Danilo Weber Nunes (Brazil)

Introduction

Adenotonsillectomy (T&A) is the primary treatment for pediatric obstructive sleep apnea (OSA), yet persistent disease affects 25-30% of patients, highlighting the need for reliable preoperative predictors. While drug-induced sleep endoscopy (DISE) enables dynamic airway assessment, the optimal classification system for predicting surgical outcomes remains uncertain. This study directly compared the established Chan-Parikh scale against the newer, multidimensional IPSES scale to determine their relative accuracy in stratifying persistent OSA risk.

Materials and Methods

It's a prospective interventional study involving 67 children (mean age 6.57 ± 2.01 years) with Grade III-IV tonsillar hypertrophy. All participants underwent preoperative Level III polysomnography (PSG) followed by DISE immediately before T&A. The DISE protocol featured standardized anesthesia induction with 6% sevoflurane and maintenance with intravenous propofol (1 mg/kg). All procedures were video-recorded for blinded subsequent analysis.

Two independent reviewers simultaneously applied both classification systems to the DISE recordings. The Chan-Parikh scale evaluated five anatomical levels (adenoids, velopharynx, lateral oropharyngeal wall, tongue base, and supraglottis) using a 0-3 scoring system. The IPSES scale incorporated dynamic and circumferential collapse patterns through its multidimensional criteria. Postoperative PSG was performed 6-9 months after surgery, with patients categorized into Persistent OSA (AHI ≥5 events/hour) or Resolution (AHI

Results

The comparative analysis of the predictive performance of the Chan-Parikh (CP) and IPSES scales revealed statistically significant differences. The IPSES scale demonstrated superior overall discriminatory capacity, with an Area Under the Curve (AUC) of 0.812 (95% Confidence Interval - CI: 0.670–0.954). In comparison, the Chan-Parikh scale achieved an AUC of 0.749 (95% CI: 0.588–0.910). DeLong's test for correlated ROC curves confirmed that this difference of 0.063 in AUC was statistically significant (p = 0.032).

Regarding classification metrics, the IPSES scale showed a sensitivity of 72.2%, correctly identifying the majority of persistent OSA cases. The Chan-Parikh scale, in turn, recorded a sensitivity of 66.7%. However, the Chan-Parikh scale maintained a slight advantage in specificity, reaching 95.9% compared to 91.8% for IPSES. The Positive Predictive Value (PPV) was 85.7% for the Chan-Parikh and 81.2% for the IPSES. The overall accuracy was 88.0% for CP and 86.0% for IPSES.

The multivariate logistic regression analysis, adjusted for age and BMI, reinforced that both scales are significant independent predictors. However, IPSES obtained a higher Odds Ratio (OR) of 4.82 (95% CI: 2.17–10.71; p=0.008), indicating a stronger association with the persistence outcome, compared to the OR of 3.95 (95% CI: 1.89–8.24; p=0.015) for severe lateral wall collapse on the Chan-Parikh scale.

Conclusions

The IPSES scale outperforms Chan-Parikh in predicting persistent pediatric OSA post-T&A, demonstrating superior sensitivity and discriminatory power. While Chan-Parikh retains value for its high specificity, IPSES is better suited for preoperative risk stratification and personalized surgical management, supporting its adoption in clinical practice to enhance patient selection and outcome prediction.

Attentional function and nocturnal electroencephalography theta/beta ratio in children with rapid-eye movement sleep-related obstructive sleep apnea

Presenting Author

Dandi Ma (China)

Authors

Dandi Ma (China), Yunxiao Wu (China), Yao Song (China), Xiaolin Ning (China), Zhifei Xu (China)

Introduction

To investigate daytime attentional function in children with rapid eye movement (REM) sleep-related obstructive sleep apnea (OSA; REM-OSA) and its relationship with sleep electroencephalography theta/beta ratio (TBR).

Materials and Methods

One-hundred-and-three children (aged 6–11 years) with snoring/mouth breathing were recruited from

Beijing Children’s Hospital. Participants completed the Attentional Networks Test for Interactions and Vigilance - executive and arousal components (ANTI-Vea) and underwent polysomnography. Groups were classified according to the obstructive apnea/hypopnea index (OAHI): 20 non-OSA, 53 non-REM-OSA, and 30 REM-OSA. The TBR of the frontal, central, and occipital regions was analyzed.

Results

Children with REM-OSA exhibited higher reaction time variability for executive vigilance and more errors during invalid condition ANTI-Vea trials than the non-REM-OSA group. The REM-OSA group exhibited higher TBRs in the frontal and central regions during all sleep stages than the non-OSA and non-REM-OSA groups and a higher TBR in the frontal and central region during REM sleep than the non-OSA group. Occipital TBR did not differ among the three groups. Reaction time variability for executive vigilance correlated positively with OAHI during REM sleep (OAHIREM). TBR in the frontal and central regions during all sleep stages and TBR in the frontal region during REM sleep correlated positively with the number of errors during invalid condition ANTI-Vea trials.

Conclusions

Children with REM-OSA exhibited impaired attentional function, characterized by increased reaction time variability for executive vigilance and poorer performance in the invalid condition attention task trials. The OAHIREM and TBR reflect distinct dimensions of attentional impairment in pediatric OSA.

The Association between Insomnia, Chronotype, and Positive Airway Pressure Adherence in Children and Adolescents

Presenting Author

Lena Xiao (Canada)

Authors

Lena Xiao (Canada),  Rianna Sarbajna (Canada), Sarah Kuyntjes (Canada), Nisha Cithiravel (Canada), Reshma Amin (Canada), Jackie Chiang (Canada), Adele Baker (Canada), Indra Narang (Canada)

Introduction

Positive airway pressure (PAP) therapy is highly efficacious for the treatment of sleep-disordered breathing in children but is limited by poor adherence. We sought to evaluate the relationship between insomnia and chronotype on PAP adherence in children.

Materials and Methods

This is a cross-sectional study conducted at the Hospital for Sick Children (Toronto, Canada) of children aged 4-18 years old prescribed PAP for a minimum of six months. Self-reported and/or caregiver-reported questionnaires including the Pediatric Insomnia Severity Index and Children's Chronotype Questionnaire were completed. PAP therapy usage was measured using objective download data.

Results

There were 159 participants included (median age=14.2 years, females=38%). Median PAP usage was 436.0 minutes/night (IQR 147.0, 516.0) for the morning group, 417.0 minutes/night (IQR 189.2, 538.8) for the intermediate group, and 161.5 minutes/night (IQR 6.0, 405.2) for the evening group. Children with an evening chronotype used PAP therapy for a median of 264.4 minutes less per night than children with an intermediate chronotype (95% CI 65.8, 397.1; p=0.0018). Children used PAP therapy for 37.0 minutes less per night for each 1-point increase in the sleep maintenance problems score (95% CI 10.0, 47.7; p<0.001) and 33.7 minutes less per night for each 1-point increase in the sleep onset problems score (95% CI 20.0, 39.5; p<0.001). Sleep onset and sleep maintenance problems significantly influenced the relationship between chronotype and PAP usage.

Conclusions

We found that insomnia and evening chronotype were associated with reduced PAP adherence in children. These may be modifiable factors to promote PAP adherence.

Acknowledgements

The authors acknowledge the International Pediatric Sleep Association and the Sleep Research Society Foundation for funding this study. We thank the patients and families participating in this study. We thank Clodagh M Ryan for reviewing the study design. We also thank the Hospital for Sick Children Sleep Clinic and Long-term Ventilation Clinic for assistance with patient recruitment.

Tongue function assessment: reliability and normative values for further use in pediatric obstructive sleep-disordered breathing rehabilitation

Presenting Author

Pierre Cnockaert (Belgium)

Authors

Pierre Cnockaert (Belgium), William Poncin (Belgium)

Introduction

The tongue plays a key role in obstructive sleep-disordered breathing (SDB). The lack of pediatric normative data and reliability information limits investigations of tongue function and its use in rehabilitation. This study proposes a set of tongue function outcomes, including strength, endurance, and frenulum assessment, along with normative values and reliability data to guide pediatric SDB assessment and rehabilitation.

Materials and Methods

Tongue function assessment was performed using the Iowa Oral Performance Instrument (IOPI). The IOPI measured tongue pressure and endurance during protrusion (pProt, eProt) and elevation (pElev, eElev), and the tongue swallowing pressure (pSwal). The Quick Tongue-Tie Assessment (QTTA) evaluated the extent of tongue mobility restriction due to the frenulum. Data will be collected in 300 healthy children, with a subset (n=56) reassessed twice, once with the same rater and then with a different rater, to compute the intra- and inter-rater reliability using intraclass correlation coefficients (ICCs). Normative data were stratified into four age groups: 4 to 6 (A), 7 to 9 (B), 10 to 13 (C), and 14 to 17 years (D).

Results

A total of 248 children (83% of the target sample) were included: 87, 60, 56, and 45 children for groups A, B, C, and D, respectively. All measures related to tongue muscles were influenced by age (p<.001). Post-hoc comparisons showed that strength measures (pProt, pElev, and pSwal) increased between age groups A and B (p<.001), while endurance measures (eProt, eElev) increased between groups C and D (p<.05). QTTA values remained stable across age groups, with a global mean (SD) mobility ratio of 62.4 (13.0)%. Reliability analyses were performed on the planned subset (n=56). Intra-rater reliability ranged from good to excellent, with ICCs between 0.83 for eElev and 0.93 for the QTTA. Inter-rater reliability ranged from moderate to good, with ICCs between 0.71 for pProt and 0.86 for eElev.

Conclusions

Tongue strength and endurance showed age-related increases, while frenulum measures remained stable. The reliability findings support the use of this set of outcomes in both pathophysiological and interventional research, as well as in clinical practice for pediatric SDB.

Acknowledgements

P.C. is a Research Fellow (ASP-REN mandate) of the F.R.S. – FNRS (Fonds de la Recherche Scientifique, Belgium).

O-03: Sleep and neurodevelopment in pediatric neurology

Supervised EEG-Based Vigilance State Classification in Preterm Infants

Presenting Author

Gabriele Arnulfo (Italy)

Authors

Gaia Burlando (Italy), Valentina Marazzotta (Italy), Sara Lanino (Italy), Sara Uccella (Italy), Annalisa Barla (Italy), Luca Antonio Ramenghi (Italy), Lino Nobili (Italy), Gabriele Arnulfo (Italy)

Introduction

Sleep plays a crucial role in early brain development, offering a window into the maturation of neural circuits during a highly dynamic period. In preterm infants, whose brains are still undergoing rapid developmental changes, the organization of sleep and its EEG signatures provides key information about neurological health. However, sleep staging in this population remains challenging: existing scoring standards and most automated tools are designed for term-born infants and do not capture the distinctive EEG patterns seen before term age. As a result, preterm sleep is difficult to classify reliably, and manual annotation is time-consuming and dependent on specialized expertise.

Materials and Methods

EEG recordings from twenty-two very preterm infants were segmented into 30-second epochs and independently annotated by two neonatal sleep experts. Only artifact-free epochs with full inter-rater agreement were retained, yielding 1500 labeled segments (658 AS, 515 QS, 327 QW). We trained two elastic-net logistic-regression classifiers: a baseline model using conventional time-domain EEG descriptors, and an extended model integrating advanced physiological metrics, including phase synchronization (wPLI), bistability, narrowband spectral power, aperiodic components, phase–amplitude coupling, and heart rate. Model hyperparameters were optimized through nested cross-validation, and generalization was evaluated on a subject-held-out blind test set.

Results

The baseline model, relying solely on time-domain metrics, achieved an accuracy of 85% (F1: 84%, MCC: 0.75). Feature-importance analysis highlighted amplitude variability and measures of signal unpredictability as the most informative predictors. Applying a confidence threshold to exclude low-certainty epochs further improved accuracy to 92%.

The extended model integrating advanced EEG descriptors markedly enhanced performance, reaching 97% accuracy (F1: 97%, MCC: 0.96) on the blind test set, and up to 99% accuracy (MCC: 0.98) after excluding uncertain predictions. Advanced metrics also contributed to clearer separation among vigilance states through differences in bistability, spectral power, aperiodic activity, and signal mobility.

To assess real-world applicability, the classifier was evaluated on an independent three-hour continuous recording, achieving 65% accuracy overall and 87% (MCC: 0.79) on confidently labeled segments, demonstrating strong generalization beyond curated datasets.

Conclusions

Overall, this study introduces a simple, interpretable, and physiologically informed approach for automated sleep staging in preterm infants. By integrating advanced EEG metrics within an elastic-net framework and explicitly accounting for prediction uncertainty, our method provides robust and clinically meaningful annotations of vigilance states. These results highlight its potential for bedside monitoring, large-scale sleep analysis, and early screening of atypical brain development in neonatal intensive care settings.

Acknowledgements

Work supported by #NEXTGENERATIONEU (NGEU) and funded by the Ministry of University and Research (MUR), National Recovery and Resilience Plan (NRRP), project MNESYS (PE0000006) – A multiscale integrated approach to the study of nervous system in health and disease (DN. 1553 11.10.2022).

Early Sleep-EEG Dynamics Reflect Cortical Maturation and Predict Neurodevelopmental Trajectories in Preterm Infants

Presenting Author

Gaia Burlando (Italy)

Authors

Gaia Burlando (Italy), Sara Uccella (Italy), Valentina Marazzotta (Italy), Chiara Scotto (Italy), Sheng H Wang (Finland), Luca Antonio Ramenghi (Italy), Lino Nobili (Italy), Gabriele Arnulfo (Italy)

Introduction

Sleep provides a powerful window into early brain maturation, as the emergence of distinct vigilance states reflects the progressive integration of cortical and thalamocortical networks. In preterm infants-those born before 37 completed weeks of gestation-the electrophysiological dynamics underlying sleep-wake activity encodes critical information on neuronal synchronization and large-scale connectivity, serving as early signatures of neurodevelopmental integrity. Deviations in these oscillatory and connectivity patterns may hold prognostic value, anticipating individual trajectories of neurodevelopmental outcome.

Materials and Methods

We first analyzed video-polysomnographic EEG recordings from 22 very low birth weight (VLBW, <1500g) preterm infants, obtained at 34.0 ± 1.4 weeks postmenstrual age (PMA). For each vigilance state—active sleep (AS), quiet sleep (QS), and quiet wakefulness (QW)—we quantified neuronal oscillatory dynamics—neuronal bistability and local cross-frequency phase–amplitude coupling (PAC). These metrics were assessed globally, separately for frontal and posterior regions, and correlated with PMA to characterize maturational trends.

To explore the prognostic potential of these early EEG markers, we then trained a disease progression model on longitudinal Griffiths developmental scores from an independent cohort of 196 preterm infants to estimate an average developmental trajectory. The model was then personalized on a subset of 10 preterm infants from our original cohort with baseline EEG recordings, yielding individualized parameters for each infant: a time shift (τ), reflecting early or delayed developmental onset, and an acceleration factor (ξ), indicating the rate of developmental progression. Finally, we correlated these individualized parameters with frontal and posterior BiS values measured from neonatal EEG across vigilance states.

Results

Both BiS and PAC reflected well-established features of early cortical development. Spatially, δ-band bistability was consistently lower in posterior regions across all vigilance states, reflecting the posterior-to-frontal maturation gradient, while δ-PAC exhibited state-dependent regional differences: it was lower posteriorly during sleep but reversed during wakefulness. Importantly, both bistability and PAC decreased with increasing postmenstrual age.

Building on these observations, we next examined whether early BiS values could predict individualized neurodevelopmental trajectories. Frontal BiS during active sleep showed a strong negative correlation with ξ in the δ band (ρ = –0.84) and a positive correlation with τ in the θ and α bands (ρ=0.81 and 0.93, respectively). Moreover, posterior θ-band BiS during quiet wakefulness negatively correlated with ξ (ρ = –0.82). Since higher ξ values reflect accelerated clinical worsening and lower τ indicates earlier onset, these associations suggest that reduced bistability in neonatal EEG is linked to more adverse developmental outcomes.

Conclusions

These findings highlight the value of early EEG markers as sensitive indicators of cortical maturation and potential predictors of neurodevelopmental trajectories in preterm infants. This integrative framework may enable earlier identification of at-risk infants, supporting timely and personalized monitoring and intervention strategies.

Acknowledgements

Work supported by #NEXTGENERATIONEU (NGEU) and funded by the Ministry of University and Research (MUR), National Recovery and Resilience Plan (NRRP), project MNESYS (PE0000006) – A multiscale integrated approach to the study of nervous system in health and disease (DN. 1553 11.10.2022).

Clinical and Polysomnographic Characterization of Sleep Disorders in Patients With Hypothalamic Damage Secondary to Craniopharyngioma

Presenting Author

Lorenzo Chiarella (Italy)

Authors

Lorenzo Chiarella (Italy), Antonella Barbieri (Italy), Anita Orrera (Italy), Marco Veneruso (Italy), Antonio Verrico (Italy), Antonella Iadarola (Italy), Valentina Marazzotta (Italy), Ramona Cordani (Italy), Lino Nobili (Italy)

Introduction

Craniopharyngiomas (CP) are rare brain tumors characterized by a severe clinical course due to hypothalamic–pituitary involvement. In patients with CP, excessive daytime sleepiness caused by sleep-disordered breathing and central hypersomnia is frequent yet often underdiagnosed.

Materials and Methods

Patients underwent medical history taking, neurological examination, and completion of questionnaires assessing daytime sleepiness (Epworth Sleepiness Scale - ESS), circadian rhythm (Morningness-Eveningness Questionnaire - MEQ), and sleep quality (Pittsburgh Sleep Quality Index - PSQI). Home-based actigraphy was performed. During inpatient evaluation, patients underwent polysomnography (PSG) and Multiple Sleep Latency Test (MSLT). Correlations among collected variables were analyzed. Actigraphic data were compared with those of an age- and sex-matched healthy control group.

Results

The pilot sample consisted of 10 patients (7 M, 3 F) aged 7.7 to 29.4 years. Subjective excessive daytime sleepiness was reported in 30% of patients, whereas MSLT demonstrated pathological sleepiness in 70%. PSG identified sleep-disordered breathing in 20% of cases. Narcolepsy due to another medical condition was diagnosed in 40% of patients, and Hypersomnia due to another medical condition in 20% of patients. Actigraphic analysis revealed statistically significant differences compared with controls in daytime sleep, Interdaily Stability, Intradaily Variability, Relative Amplitude, and mean afternoon motor activity patterns.

Conclusions

This study highlights a marked underestimation of sleep disorders when relying on subjective reports in patients with CP, even in cases where a sleep-disordered breathing condition has been excluded or treated. A multidimensional objective assessment enables the identification of otherwise unrecognized sleep disturbances, particularly within the spectrum of central hypersomnias.

Acknowledgements

We sincerely thank the Neuro-Oncology Unit at the IRCCS Istituto Giannina Gaslini for their invaluable support and the patients’ families for their cooperation and trust.

 

Sleep disorders in children with epilepsy in Latvia

Presenting Author

Una Purvina (Latvia)

Authors

Marija Kairane (Latvia), Marta Celmina (Latvia), Jurgis Strautmanis (Latvia)

Introduction

Epilepsy and sleep medicine center, Children's Clinical University Hospital, Riga, Latvia

Sleep is essential for mental and physical health. Sleep and epilepsy form a strong connection: bad-quality sleep can affect the frequency of seizures, seizures can affect sleep patterns, or both. One of the objectives of the study was to compare whether children with epilepsy had more frequent sleep disorders than the overall pediatric population.

Materials and Methods

A cross-sectional study was performed. Online questionnaires about various sleep problems were sent to all educational institutions of Latvia. Parents of children aged 2-18 years were asked to complete a questionnaire. The children were then divided in three groups according to their age: Group1 (2-5 years), Group2 (6-12 years) and Group3 (13-18 years). Data was analyzed using IBM SPSS Statistics software.

Results

Overall, 7618 parents participated in the study; 1.3% (n=102) kids had epilepsy. In Group2 and Group3, most of the sleep complaints were statistically significantly more frequent in children with epilepsy: difficulties in falling asleep (28.1% vs 18.1% and 60.0% vs 31,9%), excessively tired during the day (19.3% vs 10.3% and 84.0% and 51.1%), restless sleep (38.6% vs 17.7% and 60.0% vs 16.9%), nightmares (28.1% vs 16.5% and 24.0% and 10.8%), sleep terrors (12.3% vs 4.8% and 12.0% vs 1.7%), snoring (29.8% vs 17.3% and 40.0% vs 18.8%), sleep well (78.9% vs 93% and 44.0% vs 84.7%), waking up during the night (21.1% vs 9.0% and 40.0% vs 10.6%). In Group1, only three sleep complaints were statistically more frequent in children with epilepsy: snoring (45.0% vs 19.0%), strange behaviour during sleep (50.0% vs 19.9%), and sleeping in parents’ bed (85.0% vs 58.1%).

Conclusions

According to parents, children with epilepsy have more frequent sleep problems, especially after the age of six. Knowing this, specific questions about sleep problems should be asked for such children. Potentially, parents observe their children more closely if they have epilepsy.

Acknowledgements

I am deeply thankful to my colleague, M.Kairane, M.Celmina and J.Strautmanis for their work in this study.

Shared local brain dynamics in pediatric and adult NREM parasomnias

Presenting Author

Marco Veneruso (Switzerland)

Author

Marco Veneruso (Switzerland), Julian Amacker (Switzerland), Simone Ulzega (Switzerland), Samuel Wherli (Switzerland), Sven Hirsch (Switzerland), Lino Nobili (Italy), Silvia Miano (Switzerland), Mauro Manconi (Switzerland), Anna Castelnovo (Switzerland)

Introduction

Disorders of arousal arise from incomplete awakenings during slow-wave sleep. Their cortical dynamics, differences across age and from physiological arousals remain unclear.

Materials and Methods

We enrolled 19 children (mean age 10.9 ± 3.0 years) and 22 adults (mean age 30.3 ± 5.2 years) with disorders of arousal. High-density electroencephalography (256 channels) and video-polysomnography yielded 84 parasomnia episodes and 146 physiological motor arousals. Source-space spectral analysis quantified delta (0.5–4 Hz) and beta (18–30 Hz) activity 5 seconds before and 15 seconds after movement onset. Linear mixed-effects modeling with permutation testing was applied with multiple-comparison correction.

Results

Compared with stable slow-wave sleep, adults showed a pre-onset delta increase over frontal–midline regions with widespread beta enhancement. After onset, frontal delta activity persisted, accompanied by centro-parietal delta suppression and diffuse beta increases. Children exhibited a similar pattern, with a broader pre-onset delta build-up and less extensive beta enhancement, followed by persistent frontal delta and robust post-onset beta increases. Relative to physiological motor arousals, children displayed significantly higher pre-onset delta power in posterior cortices, whereas in adults this effect was nonsignificant; beta activity tended to be lower in disorders of arousal. After onset, both groups exhibited greater delta and beta activity during parasomnia episodes, with more widespread changes in children and a predominantly anterior distribution in adults.

Conclusions

Disorders of arousal share a cortical signature across age, marked by local dissociation between motor-control and awareness networks. These findings may aid differential diagnosis from other nocturnal motor phenomena.

Correlation of sleep architecture with cognition, behaviour and executive function in 5-18-year-olds with pharmaco-responsive non-lesional epilepsies: a prospective observational study

Presenting Author

Deepthi Krishna (India)

Authors 

Deepthi Krishna (India), Biswaroop Chakrabarty (India), Sheffali Gulati (India), Prashant Jauhari (India), Pratap Sharan (India), Atin Kumar (India), Maroof A Khan (India)

Introduction

Epilepsy is a chronic neurological condition, with a prevalence of 1-2%. Pharmaco-responsive non-lesional epilepsy (normal MRI Brain epilepsy protocol) constitutes approximately 50% of childhood and adolescent epilepsies. The representative generalised epilepsies are juvenile myoclonic epilepsy (JME), childhood and juvenile absence epilepsy (CAE/JAE), and generalised tonic-clonic seizures alone (GTCA). The focal epilepsy subtypes are self-limited epilepsy with centrotemporal spikes (SELECTS) and focal epilepsies with electrical occipital, frontal, or temporal focus. Children and adolescents with pharmaco-responsive non-lesional epilepsies often present with academic underachievement despite borderline to normal full-scale intelligence quotient (FSIQ), consequent to domain-specific cognitive deficits. The impact of pharmaco-responsive non-lesional epilepsy on cognition, executive function, and behaviour is underestimated. Sleep is a modifier of cognition, and studies have shown its impact on cognition in epilepsy, more so in drug-refractory epilepsies and epileptic encephalopathies. Hence, we conducted a study to investigate the association between sleep architecture and cognition, executive functions, and behavior in children and adolescents with pharmaco-responsive non-lesional epilepsy.

Materials and Methods

An observational study was conducted in a tertiary care teaching hospital in Northern India.

The primary objective was to evaluate the correlation between sleep parameters on PSG with cognition, behaviour, and executive function in 5-18-year-olds with pharmaco-responsive non-lesional epilepsy. Additionally, sleep, cognition, behaviour, and executive function were compared pre- and post-standard of care in subjects with sleep disorders.

Children aged 5-18 years with pharmaco-responsive non-lesional epilepsy, seizure-free for over 12 months, and an IQ > 70 were enrolled. Baseline cognition, behavior, executive function, and sleep were assessed using Wechsler’s intelligence scale for children-IV (WISC-IV), Stroop test, Child Behaviour Checklist (CBCL), Child and Adolescent Sleep Evaluation Questionnaire (CASEQ), and overnight polysomnography (PSG). Subjects with baseline sleep disorders were reassessed after six months of standard care.

Results

Forty-nine children (mean age 11.5 ± 2.5 years; 53.06% male; 53.06% generalized, 46.94% focal epilepsy) were enrolled. Sleep disorders were identified in 59.18% (29/49), most commonly sleep-related movement (48.98%) and breathing disorders (30.61%).

Significant negative correlations were noted of

- Periodic Limb Movement Index with Verbal Comprehension Index (VCI) (r=-0.335, p=0.019),

- Apnea-Hypopnea Index (AHI) with processing speed index (PSI) (r=-0.406, p=0.004), working memory index (WMI) (r=-0.362, p=0.01), Full Scale IQ (FSIQ) (r=-0.318, p=0.03), and Stroop colour T-score (r=-0.347, p=0.01).

Statistically significant positive s were noted between Periodic Limb Movement Index and CBCL externalising (r=0.477, p=0.001), and attention scores (r=0.440, p=0.002), and Apnea- Hypopnea Index with CBCL externalising score (r=0.338, p=0.018).

After six months of standard of care in those with sleep disorders (n=29), PSG showed improved sleep efficiency (p=0.031) and REM sleep (p=0.004); WISC-IV showed improvement in VCI (p=0.013), perceptual reasoning index (PRI) (p

Conclusions

Sleep architecture significantly correlates with cognitive and behavioural outcomes in pharmaco-responsive non-lesional epilepsies. Treating sleep disorders improves sleep, along with cognition and behaviour in children and adolescents 5-18 years of age with pharmaco-responsive non-lesional epilepsy.

Acknowledgements

Krishna D, Chakrabarty B, Gulati S, Jauhari P, Sharan P, Kumar A, Khan M.

O-04: Parenting and pediatric sleep: Influence, expectations, and outcomes

Parent-Reported Infant Sleep Characteristics and Behaviors at 3 and 12 Months and Predictors of Perceived Sleep Problems in a Canadian Cohort

Presenting Author

Elizabeth Keys (Canada)

Authors

Elizabeth Keys (Canada), Mahtab Matin (Canada), Tai Lin Michon (Canada), Kyle Dewsnap (Canada), Avaline Konkin (Canada), Catherine Lebel (Canada), Gerald Giesbrecht (Canada), Lianne Tomfohr-Madsen (Canada)

Introduction

Infant sleep impacts, and is impacted by, a variety of socio-ecological factors. Limited contemporary data exist on Canadian infant sleep characteristics and few studies have explored what socio-ecological factors predict parental perceptions of problematic sleep. This study aimed to (1) describe sleep characteristics and behaviors of Canadian infants at 3 and 12 months and (2) identify predictors of parental perceptions of problematic sleep across the first postpartum year.

Materials and Methods

This study used data from the Canadian Pregnancy During COVID-19 Pandemic longitudinal cohort, established in April 2020. Parent-reported infant sleep was assessed at 3 and 12 months postpartum using the Brief Infant Sleep Questionnaire–Revised Short Form (BISQ-RSF). Sleep characteristics, sleep-related parenting behaviors, and parental perceptions of sleep problems were analyzed. Decision tree analysis was conducted to explore which variables (e.g., gestational age, prior sleep difficulties, parental mental health, social support, couple satisfaction, socioeconomic status) were most likely to predict parental perception of infant sleep as a moderate or severe problem at 12-months.

Results

At 3 and 12 months, 3689 and 4069 birthing parents provided infant sleep data, respectively. Observed changes in infant sleep characteristics were aligned with expected developmental changes (e.g., nighttime sleep duration increased, daytime sleep decreased, fewer awakenings). Women were most often involved with the child at bedtime at both 3- and 12-months (96.8% and 92.5%, respectively). Approximately 75% of infants slept in their parents’ room at 3-months, compared to only 21.8% at 12 months. A large decrease in the proportion of infants breastfeeding to sleep was observed from 3- to 12-months (75% to 42.4%); however, the proportion of infants who bottle fed or used a sippy cup to fall back asleep was relatively stable (21.5% compared to 18.5%). The proportion of infants who were put back down to sleep while still awake doubled from 3- to 12 months (14.8% and 29.5%, respectively), while the proportion of participants indicating they did not respond to night awakenings tripled (from 6.9% to 24.6%).

The proportion of participants rating their infant’s sleep as a problem was relatively stable from 3 to 12 months. Specifically, about half of participants reported their infant’s sleep was not a problem at 3- or 12-months (47.6%, and 53.3%, respectively), while about 10% rated their infant’s sleep as a moderate or severe problem at 3- or 12-months (10.7% and 9.0%, respectively). Decision tree analyses identified couple satisfaction, tangible social support, and high infant surgency as key predictors of parental perception of problematic sleep at 12 months.

Conclusions

This study provides novel Canadian data on infant sleep. Findings are consistent with prior research on infant sleep patterns and describe important trends in parenting practices related to sleep. Our results also suggest that factors such as couple satisfaction, tangible social support, and perceptions of infant temperament may help identify families at greater risk for infant sleep problems at 12 months and could be explored in future intervention development.

Acknowledgements

The authors gratefully acknowledge the Pregnancy during the Pandemic study team and the study participants and their families. Funding for initial PdP data collection was provided by the Owerko Centre and the Alberta Children’s Hospital Research Institute. The Canadian Institutes of Health Research supported ongoing data collection. Analyses were supported by a UBC Hampton New Faculty Grant and EK is supported by a Michael Smith Health Research BC Scholar Award.

Parental Expectations of Sleeping Through the Night

Presenting Author

Marie-Helene Pennestri (Canada)

Authors

Marie-Helene Pennestri (Canada), Malka Hershon (Canada), Michelle Ly (Canada), Evelyne Touchette (Canada), Marjolaine Chicoine (Canada)

Introduction

The age at which their infant will sleep through the night is one of the most common concerns of parents in the perinatal period, but there is not a lot of data available regarding parental expectations related to this developmental process. The objectives of the present study were to 1) assess parental expectations about the age at which an infant should sleep through the night in a large representative sample of parents in Quebec (Canada) and 2) to describe sociodemographic and family factors associated with these expectations.

Materials and Methods

A total of 3866 families participated in the Quebec Longitudinal Study of Child Development. Measures included in the present study were collected when the infant was 17 months with parental reports (mostly mothers). Parents were asked at what age they think an infant should sleep through the night. They were able to select a specific age or to answer that it varies for each infant. Other sociodemographic and family variables used in the present study were infant sex, immigration status of the mother, family type (nuclear, single family or blended family), birth order, socioeconomic status, level of education, maternal depression, general infant health, and feeding method. Parents were also asked if they thought their own infant was sleeping through the night currently. First, descriptive statistics were conducted to describe parental expectations. Then, for parents who selected a specific age (as opposed to answering that it varies for each infant), a multinomial logistic regression was used to assess the association between sociodemographic/family factors and the expected age (less than 6 months, more than 6 months but less than 12 months, more than 12 months).

Results

Among the 3866 parents, 66.1% answered that the age varies for each infant, 9.2% thought that an infant should sleep through the night by 6 months, 15.2% between 6 and 12 months, and 9.4% after 12 months. An expectation to sleep through the night earlier in development (before 6 months compared to after 12 months) was associated with mothers born in Quebec (as opposed to mothers who immigrated to Quebec; OR = 1.67; p < .01), receiving formula (OR = 2.56; p < .001), and having an infant currently perceived as sleeping through the night (OR = 4.60; p < .001). Although some other factors were individually associated with parental expectations, they were not statistically significant when entered in the multinomial logistic regression all together.

Conclusions

Parental expectations about the age at which an infant should sleep through the night are quite variable in a large representative population of parents, with two thirds of parents thinking that this developmental process varies across different infants. Expectations varied as a function of immigration status, suggesting a cultural component to sleep-related parental expectations. Feeding method, and even more importantly, current infant sleep patterns seem to influence parental expectations. Whether these expectations were similar or different during pregnancy and at different stages of development, remains to be clarified.

Acknowledgements

We acknowledge the support of the Canada Research Chair in Pediatric Sleep and of the Quebec Longitudinal Study of Child Development (Institut de la statistique du Quebec).

The Interplay of Socioeconomic Risk, Sleep, and Parenting Quality in Predicting Academic Success During Kindergarten

Presenting Author

Ekjyot Saini (United States)

Introduction

Several family factors, such as socioeconomic risk (SER) and parenting quality are known to be determinants of children’s academic adjustment. Recent research highlights sleep as an additional contributor to cognitive and academic functioning, particularly during the transition to kindergarten. However, few studies have examined these predictors together or explored how they interact to confer risk or resilience in adverse contexts. This study addresses these gaps by investigating the independent and interactive effects of SER, actigraphic sleep, and observed parenting quality at bedtime prior to the start of kindergarten on academic outcomes across the kindergarten year.

Materials and Methods

Participants were 225 families from the northeastern United States that had a child entering kindergarten (mean age = 5 years; 50% girls). Families were predominantly White (77%) and socioeconomically diverse. Children and families participated in assessments at four time points: pre-kindergarten (August), start of kindergarten (September), mid-year (early-mid November), and near the end of kindergarten (April). Observations of parenting at pre-K were scored from video recordings of children’s bedtimes, using reliable coders trained on the Emotional Availability Scales (EA) (Biringen et al., 1998). Mothers and fathers EA scores were related (r =.59, p <.001) and averaged to create a composite EA score. Socioeconomic risk (SER) was a composite of family income-to-needs ratio, parental education, and subjective reports on monetary resources. Children wore AW-64 actiwatches (Philips Respironics, Inc) for 7 consecutive nights at pre-K, which were used to derive average sleep minutes (minutes scored asleep from sleep onset to morning wake time). Teachers reported on academic outcomes across the year using validated and reliable measures via the Academic Competence Evaluation Scale (DiPerna & Elliot, 2000), Academic Performance Rating Scale (DuPaul et al, 1991) and School Readiness Questionnaire (Bierman et al., 2008).

Results

A series of multilevel models were fit to examine how SER, parent EA, and child sleep individually and interactively predicted outcomes at the end of kindergarten (intercept) and trajectories of academic outcomes. Across the kindergarten year, children demonstrated significant improvements in academic competence, performance, and school readiness trajectories. Higher SER predicted lower scores on all outcomes at the end of kindergarten, where greater sleep minutes before kindergarten was associated with higher academic competence and performance. Parenting quality (EA) predicted higher school readiness and academic performance. Moderation analyses revealed that EA buffered the negative effects of SER and poor sleep on academic trajectories. Specifically, children from high SER families with emotionally available parents showed greater improvement in academic performance and school readiness compared to those with lower EA. Similarly, EA mitigated the adverse impact of shorter sleep duration on academic performance over time.

Conclusions

Findings underscore the importance of family-level factors prior to kindergarten in shaping academic trajectories. SER, sleep, and parenting quality each uniquely predicted outcomes, while parenting quality emerged as a key protective factor in contexts of socioeconomic adversity and sleep insufficiency. These results highlight modifiable targets for intervention which may include promoting consistent sleep routines, sleep hygiene education, and enhancing parenting quality at bedtime to bolster school readiness and academic success.

The Association between Sleep Architecture and Glycemic Variability in Children and Adolescents with Type 1 diabetes

Presenting Author

Cecilie Paulsrud (Denmark) 

Authors

Cecilie Paulsrud (Denmark), Steffen Ullitz Thorsen (Denmark), Nanette Mol (Denmark), Jannet Svensson (Denmark) 

Introduction

Objective evidence linking nocturnal glucose variability (GV) and sleep architecture in children and adolescents with type 1 diabetes (T1D) is limited. We investigate associations be-tween GV and sleep quality and architecture during the first four hours of the night, using home-based polysomnography (HST REM+) and continuous glucose monitoring (CGM).

Materials and Methods

In this cross-sectional study, 62 children and adolescents with T1D (233 nights) wore their own CGM and performed up to five consecutive nights of home sleep testing. Mixed-effects models examined associations between nightly GV (standard deviation [SD] and coefficient of variation [CV] of glucose), time in hypo- and hyperglycemia, and sleep parameters and stages. Models were adjusted for age, sex, pump type, weekday/weekend and mean glucose, except for CV where we did not adjust for mean glucose.

Results

53.2 % were female, mean age was 12.8 (2.9) and diabetes duration 5.9 (3.5) years. Clinically, doubling SD of glucose corresponded to a 3–4 % lower sleep efficiency (p=0.016) and 30–40 second longer sleep onset latency (p=0.002). Hypoglycemia (

Conclusions

Nocturnal GV was modestly, but consistently associated with poorer sleep quality and subtle redistribution of sleep stages, particularly among adolescents. Home-based sleep assessment proved feasible and informative, supporting future research into sleep–glucose interactions in pediatric T1D.

Acknowledgements

Danish Diabetes and Endocrine Academy (DDEA)

Determinants of toddler sleep and parent level targets to guide adaptation of a behavioral sleep intervention in Bangladesh

Presenting Author

Ayesha Sania (United States)

Authors

Ayesha Sania (United States), Nicolo Pini (United States), United States (United States), Tasnim Kabir (Bangladesh), Reun Tanzin (Bangladesh), Snigdho Sikder (Bangladesh), Chanel Malette (United States), Shafiqul Ameen (Bangladesh), Fahmida Tofail (Bangladesh), Monica Ordway (United States)

Introduction

Prior studies reported late bedtimes and short nighttime sleep in Asian countries. Our data shows that toddlers in Bangladesh had late and variable bedtime (average midnight), and one fifth did not meet the WHO recommendation of 11 to 14 hours. Behavioral sleep interventions can shift bedtime earlier and increase sleep duration. However, there are no culturally relevant programs available for the context. We are adapting Sleep Well, Bee Well, a behavioral sleep promotion program originally developed for low-income families in the United States. To inform the adaptation, we examined parental and household determinants of toddler sleep and parent-level behavioral targets.

Materials and Methods

We measured sleep using wrist actigraphy and the Brief Infant Sleep Questionnaire (BISQ). Data on parental education, income and occupation were collected by questionnaire. Mothers’ perceptions of toddler sleep, their knowledge of recommended sleep duration, and their confidence in moving bedtime earlier were assessed using the Knowledge, Attitudes, Self-Efficacy, and Beliefs (KASB) questionnaire. Analysis included survey data from 300 mothers and actigraphy data from 177 toddlers (median 6 days, interquartile range 5–7). We used multiple linear regression to model predictors of bedtime and bedtime variability, night sleep duration, and nap duration adjusted for toddler age and sex.

Results

Parental and household routines were linked to toddlers’ sleep timing and distribution. Compared with low-income families, toddlers in higher-income households went to bed 60 minutes later (95% confidence interval, CI 14, 117, p=0.01) and had more regular bedtimes, with the standard deviation of bedtime 53 minutes lower (95% CI −90, −15, p=0.006). Toddlers of working mothers slept 24 minutes less at night

(95% CI −44, −4.4, p=0.017) and napped 17 minutes longer during the day (95% CI 5.2, 29, p=0.005) than toddlers of stay-at-home mothers. Evening activities occurred late, with mean child dinnertime at 10 pm (SD 2.2 hours) and mean family dinnertime at 10:30 pm (SD 2.2 hours), and later dinner was associated with later bedtime (p

Conclusions

Toddler sleep in Bangladesh is shaped by family income, maternal work status, and late meal timing. Adaptation of a sleep promotion intervention must consider parental knowledge, expectations about bedtime, timing of family meals, and constraints of family environments. We are currently conducting in-depth interviews and focus groups with mothers, fathers, and grandparents to understand how cultural practices and beliefs shape toddler sleep. We will share these qualitative insights during the conference.

Acknowledgements

We thank the participating families and the field teams who identified families through door-to-door visits and support the ongoing retention efforts of this cohort.

How Parental Warmth and Anger Shape Adolescent Sleep: Trajectories and Mechanisms of Influence

Presenting Author

Serena Bauducco (Sweden)

Authors

Serena Bauducco (Sweden), Terese Glatz (Sweden), Katja Boersma (Sweden)

Introduction

Adolescence is a period marked by profound physical, socio-emotional, and behavioral changes, all of which significantly impact sleep patterns (Crowley et al., 2018). While biological changes are the primary drivers of sleep alterations during this stage, the social context in which sleep occurs—particularly the family environment—also plays a critical role (Becker et al., 2015). This study investigated whether adolescents’ perceptions of mothers’ and fathers’ warmth and anger predict the development of sleep problems and whether sleep hygiene mediates these associations.

Materials and Methods

The sample for this study was drawn from a longitudinal study ('The three cities' study'), including 18 public schools in three middle-sized cities in Sweden. Participants answered questionnaires during school hours and included 7th grade students (age 13) in 2014 (T1), followed up one year later in 8th grade in 2015 (T2), and 9th grade in 2016 (T3). Among the participating 7th grade students (N = 1,457; 47.3% girls), only those with data on insomnia and sleep duration at T1 were included (n = 1,294, Mage = 13.2, SD = .42; 46.8% girls). Measures include self-reported sleep duration, insomnia symptoms, parent/caregiver dysregulated expressions of anger, parental warmth, and sleep hygiene. We modelled trajectories of insomnia symptoms and sleep duration separately using latent class analyses. Parental warmth and dysregulated expressions of anger were examined as predictors of trajectory membership, with gender and SES included as covariates. To investigate potential mechanisms, we conducted mediation analyses testing sleep hygiene as a mediator, estimated separately for mothers and fathers.

Results

Results showed that perceived parental warmth predicted membership in healthier sleep trajectories, whereas parental anger predicted chronic or worsening sleep problems. Mediation models revealed no significant indirect effects through behavioral or cognitive-emotional sleep hygiene.

Conclusions

Findings underscore parenting as a key contextual influence on adolescent sleep, while highlighting the need to investigate additional mechanisms underlying links between parental behaviors and youth sleep outcomes.

Acknowledgements

This study was made possible by access to data from the Three cities' study, a longitudinal research program at the department of Psychology at Örebro University, Sweden. The research program was supported by a grant from Formas.

O-05: Sleep and neurodevelopment in children and adolescents

LEPTIN AS A BIOMARKER OF SLEEP DYSREGULATION IN CHILDREN WITH ASD: A DRUG-FREE COHORT STUDY

Presenting Author

Marco Carotenuto (Italy)

Authors

Martina Gnazzo (Italy), Marco Carotenuto (Italy), Giuditta Bargiacchi (Italy), Valentina Baldini (Italy), Giuseppe Plazzi (Italy), Karen Spruyt (Italy)

Introduction

Autism spectrum disorder (ASD), typically diagnosed before the age of three, is characterized by deficits in social interaction, communication, and repetitive behaviors. Sleep disturbances are highly prevalent in ASD and have been associated with altered neuroendocrine regulation. Leptin, a hormone involved in metabolic balance and circadian rhythms, has been proposed as a potential biomarker of sleep and neurodevelopmental dysfunction. This study aimed to explore the association between plasma leptin levels, sleep habits, and autism symptomatology in a cohort of children with ASD.

Materials and Methods

A total of 76 medication-naïve children with ASD (mean age: 6.86 ± 1.88; range 4–11 years) were compared to 105 age-matched typically developing controls. ASD diagnosis was based on DSM-5 criteria and confirmed using ADOS-2, ADI-R, and SRS-2. Sleep patterns were assessed using the Sleep Disturbance Scale for Children (SDSC), and fasting morning blood samples were collected to measure plasma leptin levels.

Results

Children with ASD showed significantly higher SDSC total scores (p < 0.001) and leptin levels (p < 0.001) compared to controls. Strong positive correlations were found between leptin levels and both sleep disturbances (ρ = 0.59, p < 0.001) and autism severity, as measured by ADI-R, ADOS-2, and SRS-2 (all r > 0.80, p < 0.0001).

Conclusions

These findings highlight a robust association between elevated plasma leptin levels and both sleep disturbances and autism symptom severity in medication-naïve children with ASD. While causality cannot be inferred, the data support the potential role of leptin as a peripheral correlate of sleep and behavioral dysregulation in ASD. Further longitudinal studies are warranted to examine leptin’s utility as a monitoring biomarker within neurodevelopmental trajectories.

Developing Australian consensus guidelines for managing sleep problems in children with neurodevelopmental disorders: A modified Delphi study

Presenting Author

Jasneek Chawla (Australia)

Authors

Jasneek Chawla (Australia), Fiona Hudson (Australia), Grace Langdon (Australia), Kasey Fullwood (Australia), Karen Waters (Australia), Moya Vandeleur (Australia)

Introduction

Sleep disturbance affects up to 80% of children with neurodevelopmental disorders (ND) and are associated with significant behavioural, cognitive, medical, and family burden. Despite this, there are no Australian clinical guidelines to support health professionals in the assessment and management of sleep problems in this vulnerable population and international guidelines are limited. This study aims to evaluate current variations in clinical practice in the Australian setting and develop consensus guidelines for managing sleep problems in children with neurodevelopmental disorders.

Materials and Methods

A three-round modified Delphi study was undertaken with Australian paediatric sleep specialists. . Eligible clinicians held or were training toward, specialist accreditation in paediatric sleep medicine. Round 1 involved a REDCap survey capturing demographics of participants, current approaches to assessment and management of sleep-disordered breathing (SDB) and behavioural insomnia in ND, use of investigations and screening tools, and perceived challenges. Quantitative data were summarized descriptively and qualitative responses analyzed thematically. Round 1 findings informed the development of refined statements for Round 2. Participants rated agreement on a 5-point Likert scale, with ≥80% agreement defining consensus. Round 3 virtual roundtable with clinicians, was held to further refine remaining statements and prioritize actionable recommendations.

Results

Twenty-seven clinicians participated in Round 1. The sample comprised of consultant paediatric respiratory/sleep physicians working in metropolitan tertiary centres (96%). ND accounted for ≥25% of their caseload. While 70% reported confidence managing ND sleep problems, assessment practices varied widely, with infrequent use of questionnaire screening tools (59% rarely/never) and differing behavioural assessment approaches. Key qualitative themes included clinical complexity, limited evidence, caregiver burden, and inadequate behavioural sleep intervention support.

Twenty-seven clinicians completed Round 2. Screening and referral: Strong consensus that general paediatricians (100%) and subspecialists (92.6%) should routinely screen for sleep problems. There was consensus that indications for referral to a sleep physician include suspected SDB, the need for polysomnography and diagnostic clarification (agreement thresholds 85.2%, 96.2%, 96.3% respectively). Assessment: 92% of experts agreed that ND complexity of ND condition, child and family impact, comorbidities, caregiver capacity, medications, and goals of care were important considerations during assessment. SDB management: There was 100% agreement that referral to ENT for management of OSA was still 1st line management. Where CPAP/NIV was indicated, participants supported the use of an individualised approach compared to that used in typically developing children. Behavioural sleep problems: Sleep hygiene education (100%) and modified behavioural interventions (88.9%) were strongly endorsed as management options; clinicians highlighted insufficient evidence (66.7%) and need for more research (81.5%). Medication: Melatonin was commonly used (63%), with concerns about polypharmacy (67%), interactions (81.5%), and lack of prescribing guidance (88.9%). Service needs: Consensus supported the development of national guidelines (88.9%), a role for multidisciplinary teams with a strong preference for psychology (96.3%), longer consultations for management of children with ND (96.3%), accessible family resources (88.9%), and expanded training for specialists and non-specialists (89%)

Conclusions

This modified Delphi study provided clear consensus on priority areas for improving sleep care for children with ND despite substantial variability in current practice. Outcomes from a third, roundtable session will directly inform the development of an Australian position statement for the assessment and management of sleep problems in this population.

Acknowledgements

We thank the participating clinicians, the Australasian Sleep Association, TSANZ paediatric sleep network, and consumer representatives for their contribution to this project.

Sleep-Related Movement Disorders in Children and Adolescents with Iron Deficiency: A Pilot Study (1–18 Years)

Presenting Author

Maria Ausilia Musumeci (Italy)

Authors

Maria Ausilia Mususmeci  (Italy), Michela Ribersani (Italy), Valentina Baglioni (Italy), Oliviero Bruni (Italy)

Introduction

Several pediatric sleep-related movement disorders, including Restless Legs Syndrome (RLS) and hypermotor insomnia, share iron deficiency (ID) as a key pathogenetic factor. Caregivers of children with ID or iron-deficiency anemia (IDA) frequently report sleep-onset difficulties, leg discomfort relieved by movement, nocturnal hyperkinesia, frequent awakenings and daytime inattention. Central nervous system iron is crucial for dopamine synthesis, and ferritin levels below 35–50 µg/L are considered markers of poor CNS iron stores. This pilot study aimed to (1) assess the occurrence of RLS in a pediatric hematologic population with ID/IDA, and (2) explore whether iron supplementation improves hematological parameters and sleep quality.

Materials and Methods

We conducted a single-center, prospective longitudinal study at the Pediatric Hematology Clinic, Policlinico Umberto I, Rome (from October 2024; ongoing). Children and adolescents (1–18 years) with ID or IDA underwent baseline (t0) and 3-month follow-up (t1) assessments after oral iron therapy. Sleep was evaluated at t0 with the SDSC-R (Sleep Disturbance Scale for Children plus 7 RLS-specific items), and at t0 and t1 with sleep diary, CBCL, BISQ (for preschoolers), and actigraphy. Iron status (hemoglobin, ferritin, transferrin saturation [TSAT]) was measured at both time points. Nonparametric tests (Mann–Whitney, α = 0.05) were used. Healthy controls (N = 151; 15 with actigraphy) were also studied.

Results

Thirty-one patients (mean age 7.4 years; 16 females; 16 preschoolers) were enrolled; 30% screened positive for RLS symptoms. Caregivers reported nocturnal restlessness “often” in 45.2% of cases, and SDSC total scores were significantly higher than in controls (p < .001). The RLS-positive subgroup had significantly higher Sleep–Wake Transition Disorder scores on the SDSC compared with non-RLS children (p = 0.015). At t1 (N = 8), TSAT improved significantly (p = 0.016) and caregivers reported reduced nocturnal restlessness (p = 0.015), but ferritin levels did not change significantly.

Actigraphic recordings at t0 (N = 16) showed poorer sleep quality than controls, with worse total sleep time (SMIN), sleep efficiency (SEFF), wake after sleep onset (WASO), and wake episodes (WEP) (all p ≤ .006). In preschoolers (N = 8), mean activity (AMEAN, p = 0.021) and WASO (p = 0.028) were higher than in controls. No significant differences were observed in actigraphic recordings between T0 and T1, although the T1 group remained significantly different from healthy controls.

Conclusions

Children with ID/IDA showed more sleep initiation and sleep–wake transition disturbances and frequent nocturnal hyperkinesia than healthy peers; actigraphy confirmed poorer baseline sleep quality. After three months of oral iron therapy, subjective reports indicated reduced nocturnal and daytime hyperkinesia, while actigraphic improvements did not reach statistical significance, likely due to small follow-up numbers and incomplete iron repletion (ferritin rarely > 50 µg/L). These preliminary findings support systematic screening for sleep-related movement disorders and timely iron assessment in pediatric hematologic populations.

National Insights into ASD Diagnosis and Irregular Sleep: Evidence from the 2022 and 2024 National Health Interview Survey

Presenting Author

Chia-Shuan Chang (United States)

Introduction

Irregular sleep patterns among children are a growing health concern, particularly among those diagnosed with autism spectrum disorder (ASD). While prior research has established that poor sleep is prevalent among children with ASD, limited studies have examined whether an ASD diagnosis itself predicts irregular sleep timing on a national scale. Furthermore, the role of sex differences in the ASD-sleep relationship remains unclear, with mixed findings reported. This study aimed to investigate (1) whether an ASD diagnosis is associated with irregular sleep timing among U.S. children aged 0–17, and (2) whether this association differs by sex.

Materials and Methods

This study analyzed parent-reported data on 13,828 children from the 2022 and 2024 National Health Interview Survey (NHIS) using complete-case analysis (0.41% missing data). Descriptive statistics summarized continuous and categorical variables. Multinomial logistic regression assessed the association between ASD diagnosis and irregular sleep timing (bedtime and wake-up time), adjusting for demographic factors (e.g., sex, parental education) and psychological factors (e.g., anxiety, depression). An interaction term (sex × ASD diagnosis) was included to test potential sex differences.

Results

The final sample consisted of 51.05% boys, with a mean age of 9.74 years. After adjusting for covariates, children with ASD, compared to those without ASD, were less likely to have a regular bedtime on some days (AOR: 0.57, 95% Confidence Interval [CI]: 0.36–0.90) or on most days (AOR: 0.71, 95% CI: 0.57–0.97) compared to having a regular bedtime every day. In contrast, children with ASD were more likely than those without ASD to never have regular wake-up times compared to having consistent wake-up times every day (AOR: 2.24, 95% CI: 1.003–4.99). However, no significant moderating effect of sex was observed.

Conclusions

These findings reveal distinct patterns of sleep irregularity in children with ASD, showing that they were more likely to have irregular wake-up times but less likely to have irregular bedtimes. Difficulties with morning routines, such as inconsistent wake-up times, challenges getting out of bed, or variability in preparing for school, may therefore be a critical target for intervention. Future longitudinal studies should clarify underlying mechanisms and guide strategies to promote consistent sleep habits.

Acknowledgements

The authors gratefully acknowledge Amanda E. Ng, PhD, MPH, for her valuable guidance and support with data analysis throughout this study.

Associations Between Nocturnal Autonomic Activity in Infancy and Later Neurocognitive Development: The AuBE Cohort Study

Presenting Author

Lisa Brunel (France)

Authors

Lisa Brunel (France), Aurore Guyon (France), Laurianne Coutier (France), Hugues Patural (France), Vincent Pichot (France), Marion Comajuan (France), Marine Thieux (France), Sabine Plancoulaine (France), Patricia Franco (France), 

Introduction

The autonomic nervous system (ANS) reflects central nervous system (CNS) function and can be assessed non-invasively using heart rate variability (HRV). However, most studies assess HRV over 24 hours, mixing wake and sleep periods, limiting insights into sleep-specific ANS activities. Understanding how ANS activity during distinct sleep stages relates to neurocognitive outcomes is essential, given the critical role of sleep in early life for shaping brain development.

This study aimed to: (1) compare ANS activity between active sleep (AS) and quiet sleep (QS) at birth and at six months; (2) evaluate ANS maturation during sleep from birth to six months; and (3) investigate associations between ANS activity during sleep in early infancy and neurocognitive performance at three years of age.

Materials and Methods

Participants were part of the French AuBE birth cohort. Inclusion criteria for this analysis were to be born at term, to have overnight polysomnography (PSG; 8 pm–8 am) at birth (M0) and a cognitive assessment at three years using the Wechsler Preschool and Primary Scale of Intelligence-III (WPPSI-III). A subset of infants also underwent PSG at six months (M6).

Electrocardiogram (ECG) data were extracted to compute HRV indices in both time and frequency domains: mean heart rate (HR), SDNN, and low-frequency power (LF) as markers of sympathetic activity; pNN50, RMSSD, and high-frequency power (HF) as markers of parasympathetic activity; and the LF/HF ratio to evaluate sympathovagal balance. HRV parameters were compared between AS and QS at both M0 and M6, and changes from M0 to M6 were analyzed according to sleep stage. Associations between HRV indices and WPPSI-III scores (Verbal IQ, Performance IQ, Total IQ) were assessed using Spearman correlations.

Results

A total of 56 full-term infants were included at birth (48% girls), among which 34 had follow-up data at M6. HRV parameters did not differ by sex. Significant differences in ANS activity were observed between sleep stages at both M0 and M6. AS was characterized by higher sympathetic activity (elevated HR, SDNN, LFnu, and LF/HF ratio), while QS showed increased parasympathetic activity (higher RR intervals, pNN50, RMSSD, HFnu).

Between M0 and M6, ANS maturation was evident during QS, with an increase in RR intervals and HFnu, and a decrease in HR, LFnu, and LF/HF ratio, indicating an increase in parasympathetic regulation and a decrease in sympathetic dominance. No significant maturation was observed during AS. Importantly, total power during QS at M0 was positively correlated with Performance IQ at three years (r = 0.31, p

Conclusions

This study demonstrates that ANS activity in infants varies significantly according to sleep stage, with QS favoring parasympathetic predominance. ANS maturation primarily occurs during QS within the first six months of life. Furthermore, ANS function during QS at birth, as reflected by total power, is associated with later cognitive outcomes. These findings suggest that early sleep-specific ANS markers could serve as indicators of neurodevelopmental trajectories.

The Effect of Melatonin Usage on Children with Insomnia in Clinical Practice: A Retrospective Observational Study

Presenting Author

Phanthila Sitthikarnkha (Thailand)

Authors

Phanthila Sitthikarnkha (Thailand), Narong Simakajornboon (United States), Aisaku Nakamura (United States), Md Monir Hossain (United States)

Introduction

Melatonin is one of the most common treatments for insomnia in children. Due to recent concerns about the safety and efficacy of melatonin, the International Pediatric Sleep Association has published a consensus statement on the use of melatonin in children. However, there remains limited real-world data on outcomes and safety profiles in clinical practice. This study aimed to determine the clinical outcomes and adverse effects of melatonin in children with chronic insomnia in clinical pediatric sleep practice.

Materials and Methods

This retrospective observational study was performed in patients aged 2 months to 18 years at Pediatric Sleep Clinic, Cincinnati Children’s Hospital Medical Center, between January 1, 2000, and February 28, 2025. Patients diagnosed with chronic insomnia, defined by the International Classification of Sleep Disorders 3rd edition Text Review (ICSD-3-TR) and treated with melatonin, were included. Patients who did not follow up after their first visit would be excluded. Clinical outcomes and adverse effects were assessed at the follow-up visit.

Results

Of the 629 children diagnosed with chronic insomnia included in this study, the mean age was 9.5 ± 4.7 years, with 345 (55.1%) being male. The mean daily melatonin dose administered was 3.8 ± 2.5 mg. Most participants (506 children, 80.8%) used melatonin prior to bedtime. Clinical improvement in insomnia was observed in 477 children (76.2%) following melatonin treatment. Responders (R) to melatonin therapy had a longer median treatment duration (273, IQR 128-609 days) than non-responders (NR) (180, IQR 97-309 days), P < 0.001. Children with co-existing neurological conditions (37.6% [NR] vs 18.3% [R]; P=0.03) were more likely to show no improvement to melatonin, while children with co-existing sleep disordered breathing (30.2%[NR] vs 40.7%[R]; P=0.02) and sleep related movement disorders (16.1% [NR] vs 33.3% [R]; P<0.001) were more likely to show improvement to melatonin. Adverse effects were observed in 82 children (13.1%), with morning sleepiness being the most common (5%), followed by nightmares (4%).

Conclusions

Melatonin has been shown to be an effective pharmacological treatment for chronic insomnia in pediatric populations. Interestingly, children with certain medical conditions and co-existing sleep disorders may have different response rates to melatonin therapy. Clinicians should be aware of factors affecting melatonin responses and potential side effects in pediatric population.

Acknowledgements

This study was supported by the Cincinnati Children's Hospital Research Foundation

O-06: Technology and innovation in pediatric sleep

Associations Between Wearable Infant Sleep and Brain Function in the First Year of Life

Presenting Author

Nicolò Pini (United States)

Author

Nicolò Pini (United States), Ayesha Sania (United States), Lynn Chen (United States), Kirsten Donald (South Africa), Michal Zieff (South Africa), Seonjoo Lee (United States), William Fifer (United States)

Introduction

Sleep plays a foundational role in early brain development, yet the extent to which dimensions of sleep patterns shape emerging sensory cortical processing in infancy remains unclear. Visual evoked potentials (VEP) provide a sensitive index of cortical maturation, with shorter latencies and larger amplitudes reflecting more efficient neural signaling. Using multidimensional actigraphy-derived sleep and VEP measures, we examined whether daytime, nighttime, and total sleep duration were associated with VEP responses at 6 and 12 months. We hypothesized that lower daytime and greater nighttime and total sleep durations would be associated with more developmentally mature VEP profiles.

Materials and Methods

Sleep wearable and VEP data were collected from 102 infants at approximately 6 months (6M) and 12 months (12M) of age in a prospective cohort study conducted in Gugulethu, Cape Town, South Africa, as part of the 1kD program. Infants completed multi-day actigraphy recordings (range: 1–6 days), from which daytime, nighttime, and total sleep duration were extracted and averaged to produce a single estimate per participant. VEP responses were collected at the same timepoints using a high-impedance, high-density electroencephalogram system (128 channels), and N1, P1, and N2 amplitudes and latencies were derived. After merging sleep and electrophysiology datasets, multiple linear regression models were fit to examine associations between each VEP component and sleep measures, controlling for age at sleep/VEP acquisition, number of sleep days collected, and infant sex. Models examined both concurrent relationships (e.g., sleep and VEP at 6M and 12M) and lagged associations (e.g., sleep at 6M predicting VEP at 12M).

Results

Sleep and VEP at 6M: We detected marginal significant associations between (i) greater nighttime sleep duration and longer P1 latency (β = 0.061; 95% CI [–0.002, 0.124]; p = 0.084); (ii) greater daytime sleep and reduced N2 latency (β = –0.048; 95% CI [–0.097, 0.001]; p = 0.078); and (iii) greater daytime and total sleep duration and reduced N2 latency (β = –0.034; 95% CI [–0.069, 0.001]; p = 0.084).

Sleep and VEP at 12M: Greater total sleep duration at 12 months was significantly associated with shorter N1 latency at T3 (β = –0.019; 95% CI [–0.037, –0.001]; p = 0.032), indicating faster and more efficient cortical processing in toddlers who slept more at this time point.

Sleep at 6M and VEP at 12M: Greater total sleep duration at 6 months showed a marginal association with longer P1 latency at 12 months (β = 0.040; 95% CI [0.001, 0.079]; p = 0.072).

Conclusions

Together, these findings underscore the importance of investigating multidimensional sleep features and electrophysiological biomarkers for understanding early neurodevelopment. Marginal associations at 6 months indicate weak and inconsistent links between early sleep patterns and concurrent or subsequent VEP responses. In contrast, total sleep duration at 12 months was associated with faster N1 processing, pointing to a developmental window in which sleep consolidation may more directly support cortical efficiency. Further work integrating sleep, VEPs, and later cognitive and behavioral outcomes will help elucidate mechanistic pathways through which sleep contributes to brain development in infancy.

Performance of Pediatric-Trained Machine Learning Classifiers in Analyzing Actigraphy Data for Sleep-Wake Scoring

Presenting Author

Pin-Wei Chen (United States)

Author

Pin-Wei Chen (United States), Christopher Cielo (United States), Olivia Walch (United States), Peter X.K. Song (United States), Cathy Goldstein (United States), Jennette P. Moreno (United States), Erica C. Jansen (United States), Jonathan A. Mitchell (United States)

Introduction

Actigraphy methodologies are increasingly using raw acceleration data and machine learning scoring. However, these advancements have been predominantly developed in adults. We therefore trained pediatric-specific machine-learning classifiers. We evaluated their scoring performance by benchmarking them against standard GGIR algorithms and an existing adult-trained classifier. 

Materials and Methods

Sixty children (26 female, ages 5.3–17.7 years) completed in-lab overnight polysomnography at Children’s Hospital of Philadelphia and wore a GENEActiv device (3-axis accelerometer, 50 Hz) on their non-dominant wrist. Features were generated from the acceleration data in the time and frequency domains, resulting in 35 total features. All features were imputed with k-nearest neighbor and z-score normalized. Correlated features greater than 0.9 were removed. The Synthetic Minority Oversampling Technique was used to correct for the class imbalance. The features were in 30-second epochs and aligned with physician-scored sleep-wake data from electroencephalography. Six machine-learning models were trained and tested using a nested cross-validation approach with an inner and outer loop. The outer loop consisted of leave-one-subject-out cross-validation (LOSOCV), whereas the inner loop consisted of hyperparameter tuning with an additional 10-fold cross-validation. Hyperparameters were selected with grid searches based on maximization of the area under the curve score. For LOSOCV evaluation, epoch-by-epoch analyses generated performance metrics: sensitivity and specificity with balanced accuracy (BA) used to rank. Discrepancy analyses compared the overall sleep duration estimated.

Results

Overall, 560.1 hours of data were collected for the entire sample, with 74.4% of epochs scored as sleep. On average, sleep duration was 7.1 hours (SD = 1.9) per participant. Of the six pediatric-trained machine learning models, the top two were random forest (BA = 0.78; Sensitivity = 0.87; Specificity = 0.69) and neural network (BA = 0.77; Sensitivity = 0.83; Specificity = 0.73). These performance metrics exceeded that of an adult-trained neural net classifier applied to our data (BA = 0.71; Sensitivity = 0.93; Specificity = 0.49), but were comparable to the GGIR Cole-Kripke (GGIR-CK: BA = 0.79; Sensitivity = 0.75; Specificity = 0.85) and GGIR van Hees algorithms (GGIR-vH: BA = 0.78; Sensitivity = 0.84; Specificity = 0.73). Overall, sleep duration was underestimated by an average of 15 minutes using the random forest classifier and by an average of 37 minutes using the neural network classifier. For comparison, both GGIR-CK and GGIR-vH underestimated sleep duration by an average of 33 minutes. Wake after sleep onset (WASO) was overestimated by an average of 63.5 minutes using the random forest classifier and by an average of 82.6 minutes using the neural network classifier. For comparison, GGIR-CK and GGIR-vH overestimated WASO by an average of 99.6 and 54.0 minutes, respectively.

Conclusions

We developed pediatric sleep-wake classifiers demonstrating robust sleep detection and moderate-to-strong wake detection. Through epoch-by-epoch and discrepancy analyses, the random forest classifier emerged as the optimal model, surpassing GGIR-CK, GGIR-vH, and an adult-trained neural network. Future studies utilizing larger training and validation datasets may further minimize variability and enhance the precision of actigraphy-based pediatric classification.

Acknowledgements

The in-lab performance research was supported by the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of Health under Award Number R01HD100421. Dr. Jansen reports support from National Heart, Lung, Blood Institute grant K01HL151673. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Minimum Data Requirements for Automated Sleep Staging in the Pediatric Population

Presenting Author

Yuan Gao (Canada)

Author

Beth Payne (Canada), James Lee (Canada), Lena Xiao (Canada), Calvin Kuo (Canada), Mark Ansermino (Canada), David Wensley (Canada), Lyndia Wu (Canada)

Introduction

Sleep disorders are common in the pediatric population, and up to 28 % children may experience clinically significant sleep disorders. Polysomnography (PSG) remains the gold standard for diagnosis, and sleep staging is a key component of PSG analysis. However, sleep staging requires extensive multi-channel sensor setups and manual scoring, limiting its availability. Recent advances in machine learning have enabled automated staging. However, pediatric sleep staging is still a challenging task with existing models, especially for young infants. The optimal channel setup for accurate automated staging in children remains unclear. This study aims to identify the minimum electroencephalogram (EEG) / electrooculography (EOG) channel configuration required to maintain reliable staging performance.

Materials and Methods

A total of 2,153 pediatric PSG studies (6 months–18 years) from BC Children’s Hospital were analyzed

(Ethics protocol: H23-02977). The original data includes 7 EEG electrode locations (F3, C3, C4, O1, M1, M2, Cz), producing 21 channel options with re-referencing. The database was divided into four age groups: 0-2 (n=129), 2.1-6 (n=540), 6.1-12 (n=913), and 12.1-18-year-old groups (n=571). Signals were preprocessed with amplitude- and correlation-based noise rejection and filtered using a 0.3–30 Hz bandpass filter. The algorithm used is XGBoost with 26 features per channel, including statistical features, band power, entropy, and fractal dimension. The model was trained and evaluated with 5-fold cross-validation using clinical labels for 30-second epochs. To determine the minimal configuration, a forward-search approach was applied. We started from a single-channel model, and iteratively added channels until the accuracy reached ≥80% for 5-stage classification and ≥90% for sleep–wake detection.

Results

The full-age-group model achieved an average accuracy of 82.4% for 5-stage classification, comparable to inter-rater reliability among human scorers in adult staging. However, similar to previous work, the model performance had a notable decrease for infants and toddlers. For the age-grouped models, 80% accuracy was achieved with single-channel setup for older age groups (>6 years old). In single-channel evaluation, C3-M1 and C3-M2 outperformed other setups across all age groups. For the 2-6 age group, the model reached the target performance with a 2-channel setup. EEG-only configurations performed less robustly in the 0-2 age group, requiring 6 channels to reach the target performance. However, adding EOG channels improved accuracy by 4–5% for this group. For sleep-wake detection, single-channel models achieved 90% accuracy for all age groups above 2 years old, while the accuracy was 89.2% for the 0-2 age group. Overall, model performance improved with age, reflecting more consistent sleep features in older children.

Conclusions

This study presents a systematic evaluation of data requirements for automated pediatric sleep staging using a large clinical dataset. The work is performed across a wide range of sleep and developmental conditions, and our results demonstrate that simplified channel configurations can still yield acceptable accuracy across diverse pediatric patients. Our future work will further explore the relationship between data requirements and participant characteristics, including age, sex, and diagnostic indicators such as sleep apnea severity and other comorbidities. These results contribute to the development of simplified, home-deployable pediatric sleep monitoring systems.

Acknowledgements

The study was funded by the Canadian Institutes of Health Research (CIHR), Mitacs, and the BC Children’s Hospital Research Foundation. We would like to thank Ryan Lo for his assistance with the data transfer process.

Automatic video analysis and classification of disorders of arousal in children

Presenting Author

Anna Castelnovo (Switzerland)

Authors

Anna Castelnovo (Switzerland), Marco Veneruso (Switzerland), Mauro Manconi (Switzerland), Lino Nobili (Italy), Silvia Miano (Switzerland), Maria Camila Sebastiani (Switzerland), Alessandro Giusti (Switzerland), Ricardo Omar Chávez-García (Switzerland)

Introduction

Sleepwalking, sleep terrors, and confusional awakenings—collectively called disorders of arousal (DOA)—are complex and sometimes hazardous behaviors that arise from non-REM slow-wave sleep, most often in childhood. Although video-polysomnography (vPSG) is the diagnostic gold standard, it is expensive and time-intensive and usually misses DOA episodes, identifying them in only 30–60% of cases. For extended home-based monitoring, there is an increasing need for automated detection of potential clinical episodes. Automated systems can help both patients and clinicians by identifying cases that may benefit from further assessment in specialized sleep centers.

Materials and Methods

We introduce a custom multimodal Temporal Convolutional Network (TCN) architecture for detecting DOA events from whole-night video recordings. Temporal Convolutional Networks are specialized for processing sequential data, making them particularly suitable for identifying sleep disorders within video streams. Our TCN model analyzes entire night recordings alongside sleep staging information, detecting DOA events by evaluating temporal correlations between visual cues and sleep-stage labels across successive frames.

The model was trained and evaluated on 204 behavioral events from deep N3 sleep, identified in vPSG recordings from 18 children diagnosed with DOA. Three independent sleep experts, conducting blind evaluations, scored the selected events. Through consensus, 47 episodes were classified as DOA and 113 as physiological sleep-related movements; 44 events remained unclassified due to their brevity or ambiguity.

Results

We evaluated our trained model on 30 movement episodes, including 13 DOA and 17 physiological events. In classifying 30-second intervals of patient behavior as either DOA or physiological events, our model achieved an area under the receiver operating characteristic (ROC) curve (AUC) of 0.84. In comparison, a model relying solely on visual features attained an AUC of 0.78.

Conclusions

Our preliminary results suggest that both models effectively recognize visual patterns associated with DOA, and that incorporating sleep-stage data improves detection. The proposed pipeline is flexible, enabling integration of additional modalities, such as facial features and clinical indicators, including upper-torso movement and eye opening. Our model may help physicians identify and analyze DOA episodes, ultimately improving patient care.

A peer-led school-based approach to reduce night-time use of interactive devices and social media in early adolescence: CLOCK OFF intervention development and feasibility study

Presenting Author

Catriona Ewart (United Kingdom)

Authors

Catriona Ewart (United Kingdom), Anne Martin (United Kingdom), Xueqi Wu (Switzerland), Colin B Shore (United Kingdom), Benjamin Rigby (United Kingdom), Dawn Haughton (United Kingdom), Holly Scott (United Kingdom), Heather Cleland-Woods (United Kingdom), Sharon Simpson (United Kingdom), Sally Good (United Kingdom), Alyson O’Brien (United Kingdom), Rhiannon Evans (United Kingdom)

Introduction

Poor sleep among adolescents poses significant risks to their health and wellbeing. A key contributor to delayed bedtimes and disrupted sleep is night-time use of interactive electronic devices (IEDs), including social media use. The aim of this study was to develop and assess the feasibility and acceptability of CLOCK OFF, a peer-led school-based intervention combining teacher-led lessons and peer-nominated pupils trained as peer supporters to promote reducing night-time IED use and increasing sleep.

Materials and Methods

CLOCK OFF was developed by exploring the suitability and adaptability of combining three established programmes across 19 online workshops, with 24 pupils (aged 11-14 years) from four different high schools and eight adult stakeholders. Sixty peer supporters and 24 teachers were trained to implement CLOCK OFF with all 11–12-year-olds in four schools. Semi-structured interviews and focus groups were conducted with peer supporters (n=18), non-peer supporter pupils (n=9), teachers (n = 8), parents (n=8), and intervention providers (n=3). Thematic analysis was used to analyse data from both intervention development and feasibility testing.

Results

CLOCK OFF was developed by exploring the suitability and adaptability of combining three established programmes across 19 online workshops, with 24 pupils (aged 11-14 years) from four different high schools and eight adult stakeholders. Sixty peer supporters and 24 teachers were trained to implement CLOCK OFF with all 11–12-year-olds in four schools. Semi-structured interviews and focus groups were conducted with peer supporters (n=18), non-peer supporter pupils (n=9), teachers (n = 8), parents (n=8), and intervention providers (n=3). Thematic analysis was used to analyse data from both intervention development and feasibility testing.

Conclusions

CLOCK OFF is the first intervention developed to reduce night-time IED and social media use for improving sleep in early adolescence. CLOCK OFF is generally feasible and acceptable but requires optimisation to enhance intervention effectiveness and scalability.

A Pilot Feasibility Study of a Six-Week Digital Sleep Health Intervention for Caregivers of School-Aged Children

Presenting Author

Jessica Page (United States)

Author

Jessica Page (United States), Grace Wang (United States), Rebecca Robbins (United States), Judy Owens (United States)

Introduction

Insufficient sleep is a prevalent yet modifiable health issue among school-aged children and their caregivers. The Bedtime Stories-Caregiver program was developed as a six-week digital sleep health education intervention for caregivers of school-aged children from racial and ethnic minority backgrounds, focused on sleep health principles (RESTED domains: Routines, Expectations, Screen time, Timing, Environment, and Duration). This pilot study evaluated the program's feasibility, acceptability, and preliminary efficacy in improving caregiver and child sleep outcomes.

Materials and Methods

A single-arm, pre-post feasibility design was used. Eligible English-speaking caregivers of children aged 5-11 years were recruited from community health centers affiliated with a large Northeastern hospital (USA). Exclusion criteria included child neurodevelopmental or neurological disorders, diagnosed sleep disorders, or families experiencing homelessness. After completing baseline questionnaires (adult and pediatric PROMIS Sleep Disturbance and Sleep-Related Impairment, caregiver Knowledge Attitude Self-Efficacy Beliefs (KASB), Perceived Stress Scale (PSS), and Child and Adolescent Sleep Environment Scale

(CASES)) and a one-week digital sleep diary, participants received six weekly automated text messages linking to self-guided video modules, followed by weekly check-ins assessing satisfaction, clarity, and usefulness of the module. After completing the 6 weekly modules, participants completed follow-up assessments that mirrored baseline measures, except for the PSS and CASES. Feasibility outcomes included recruitment (target ≥ 60%), retention (≥ 75%), and adherence/engagement (≥ 4 of 6 modules completed). Acceptability was assessed through weekly satisfaction ratings. Exploratory pre–post changes (95% CI) were evaluated for caregiver sleep (PROMIS Sleep Disturbance and Sleep-Related Impairment), caregiver knowledge (KASB), and child sleep (PROMIS Sleep Disturbance and Sleep-Related Impairment). Exploratory models examined whether caregiver stress (PSS) and the child’s sleep environment (CASES) at baseline predicted caregiver or child sleep outcomes at follow-up.

Results

Twenty-two caregivers enrolled (mean age= 36.4 ± 6.1 years; 95% female). Feasibility targets were achieved: recruitment= 73%, retention= 86%, and 68% (15/22) completed ≥ 4 modules. Caregiver Knowledge (KASB) increased by 4.9 (95% CI =1.0 to 10.9) and caregiver Sleep-Related Impairment decreased by 5.7 (95% CI=10.1 to 1.3), indicating a statistically significant improvement in caregiver functioning. Both child Sleep Disturbance 2.5 (95% CI=5.7 to 0.7) and Sleep-Related Impairment 5.1 (95% CI= 8.4 to 1.8) decreased, indicating a statistically significant improvement in child sleep. Predictor analyses indicated that higher baseline caregiver stress and less optimal child sleep environments predicted poorer caregiver Sleep Disturbance scores at follow-up, while better baseline sleep environments predicted greater improvement in perceived sleep quality (digital sleep diary). Weekly satisfaction ratings averaged > 80%, supporting content clarity and acceptability. Digital delivery via automated texts achieved high engagement and minimal missing data.

Conclusions

The Bedtime Stories-Caregiver program demonstrated high feasibility, acceptability, and promising engagement among caregivers of school-aged children. Preliminary results suggest meaningful improvements in caregiver and child sleep-related impairment and small-to-moderate improvements in sleep disturbance and caregiver knowledge. The strong adherence and satisfaction metrics support the feasibility of digital sleep education in this population. Findings will inform refinement of intervention components, recruitment strategies, and outcome measurements for a future randomized controlled trial to establish efficacy and scalability.

Acknowledgements

This study was funded by a grant from the Allan and Gill Gray Philanthropies Foundation (90227-01). The Allan and Gill Gray Foundation had no role in the design or the conduct of the study. We would like to acknowledge the support and contributions from our community stakeholder advisory group- Diane Isaac, Kari Oakes, Dr Monica Ordway, Nancy Rothstein, and Pallas Snider Ziporyn.