Neurodevelopmental outcomes and predictors among late preterm infants: a 6-month prospective cohort study Authors: Shrisruthi Suresh, Vadivelan Kanniappan, Sushmitha Kannairam Published: 2025 ABSTRACT AbstractBackgroundLate preterm infants (33–36 weeks gestation) are at increased risk for neurodevelopmental delays, but early signs often go unrecognized due to subtle presentations and inconsistent follow-up. While screening tools are available, few studies have examined how well parents identify developmental concerns compared to trained professionals.AimTo evaluate neurodevelopmental outcomes in late preterm infants at 3 and 6 months of corrected age by comparing parental and therapist-reported developmental scores, and to identify predictors influencing developmental scores.MethodsSixty late preterm infants were screened using the Trivandrum Developmental Screening Tool (TDST) at discharge and followed up at 3 and 6 months of corrected age. Both therapists and parents completed the TDST independently during follow-up. Developmental scores were compared using Wilcoxon signed-rank tests. Regression analyses were conducted to identify predictors that influence parent and therapist scoring. A p-value of < 0.05 was considered statistically significant.ResultsAt 3 months there is no significant difference between the parent and therapist TDST overall scores, while at 6 months, therapist and parent TDST overall scores showed significant difference (p < 0.001). Regression analysis identified rolling and object transfer as strong predictors of overall scores for both parent and therapist scores at 6 months.ConclusionsDespite similar overall scores at 3 months, significant discrepancies between parent and therapist scoring emerged by 6 months. These findings emphasise the importance of caregiver education in early neurodevelopmental surveillance.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12887-025-06391-0.Keywords: Late preterm, Neurodevelopment, Developmental delay, Parent report, Early intervention, Trivandrum developmental screening tool (TDST) Introduction According to the World Health Organization (WHO), preterm birth is defined as the birth of a baby before 37 completed weeks of gestation [1]. Preterm infants are further classified based on gestational age into three subcategories: extremely preterm (< 28 weeks), very preterm (28 to < 32 weeks), and moderate to late preterm (32 to < 37 weeks) [2]. Although they are often perceived as being at lower risk, late preterm infants still face significantly higher rates of morbidity and mortality compared to their full-term counterparts (Fig. 1). Their vulnerability is largely due to incomplete brain maturation during the final weeks of gestation, which predisposes them to long-term neurodevelopmental impairments [3]. Conceptual framework for neurodevelopmental problems for preterm infants In 2020, an estimated 13.4 million preterm births occurred globally, with India accounting for approximately 3 million (22%) of these cases. In the Indian state of Tamil Nadu, the preterm birth rate is reported to be as high as 14% [4]. While improvements in neonatal intensive care have increased survival rates, the risk of developmental impairments, particularly in neurological domains, remains high. Predicting neurodevelopmental outcomes in these children is challenging due to the complex interplay of biological, environmental, and psychosocial factors, among which parental involvement plays a pivotal role [5]. Parental inclusion is essential in optimizing neurodevelopmental outcomes for preterm infants. Actively involving parents in early screening, therapy, and follow-up enhances their ability to detect early warning signs and participate in decision-making regarding their child’s care [6]. Such engagement is particularly important in low-resource settings where parental awareness and participation can bridge service gaps. Early screening is critical for the timely identification of developmental abnormalities and the initiation of targeted interventions [6]. The first year of life is a crucial period for brain development, necessitating structured monitoring during this window. Neurodevelopmental assessments conducted at 3, 6, 9, and 12 months of age help identify delays in motor, cognitive, language, and social domains [7]. These periodic evaluations offer valuable opportunities to address emerging concerns and implement early interventions. To support early identification, the Trivandrum Developmental Screening Tool (TDST) has been developed as a culturally appropriate, valid, and easy-to-use tool for the Indian population [8]. It consists of a single pictorial page with developmental milestones across key domains and requires minimal training, making it suitable for administration by both healthcare professionals and parents. TDST’s simplicity, affordability, and suitability for mass screening make it a preferred tool in Indian clinical and community settings. The success of early screening and intervention efforts also depends on the skills and training of healthcare professionals, including paediatricians, therapists, and primary care providers [9]. Proficiency in administering screening tools, interpreting results, and delivering family-centred care is essential. Continued professional development and multidisciplinary collaboration ensure that preterm infants receive comprehensive and individualized care [10]. While some developmental issues are evident in early infancy, others, particularly cognitive and executive function deficits, may not emerge until later in childhood or adolescence—a phenomenon known as “growing into deficit” [11]. Therefore, long-term monitoring and cognitive assessments are necessary to identify subtle but significant developmental challenges. Despite growing awareness, research on moderately and late preterm infants remains limited, with most long-term studies focusing on extremely preterm populations [12]. The findings from existing studies are often inconclusive; some report minimal deficits among late preterm infants, while others suggest a higher risk of behavioural and learning difficulties. This highlights the urgent need for structured cohort studies that track neurodevelopmental trajectories and identify risk and protective factors unique to this subgroup [13]. Although there is substantial literature on child development monitoring, many studies lack an integrated approach that includes the perspectives and active roles of both therapists and parents [14]. A triangular model that combines inputs from the child, therapist, and parent is essential for developing and implementing customized care plans. Within this framework, our study aims to contribute evidence that can inform such individualized, holistic strategies for developmental support in late preterm infants [15]. The conceptual framework (Fig. 1) was designed based on the perinatal and postnatal influences on various factors such as gestational age, APGAR, birth weight, Intraventricular Haemorrhage (IVH). Although periventricular leukomalacia (PVL) and cerebral ischemia are also critical contributors to neurodevelopmental sequelae, were not included in this model as their incidence was low in our setting. This study focuses on the parental factors represented in the conceptual framework. The current study adopts a prospective cohort design to follow late preterm infants over the first six months of life, a key developmental period. This approach allows for systematic assessment of neurodevelopmental outcomes and exploration of predictors influencing early development [16, 17]. The novelty of the study lies in its exclusive focus on late preterm infants in the Indian context, use of a locally validated screening tool - TDST, and inclusion of parental and therapist variables. These insights are vital for designing contextually appropriate early intervention strategies in low- and middle-income countries. Aim of the study The study aims to evaluate neurodevelopmental outcomes in late preterm infants at 3 and 6 months of corrected age by comparing parental and therapist-reported developmental scores, and to identify predictors influencing developmental scores under a blinded protocol. Materials and methods Study design This prospective, non-experimental cohort study focused on identifying the neurodevelopmental outcomes and their predictors among late preterm infants (33 ± 0 days to 36 ± 0 days) over the first six months of the child’s life. Each child’s timeline includes three main points: Initial enrolment (at the time of discharge), follow-up at the 3rd and 6th months of corrected age (Fig. 2). Developmental Surveillance Timeline for Late Preterm Infants Study setting and participants The study was conducted from the Department of Pediatrics, SRM Medical College Hospital and Research Centre, Kattankulathur, Tamil Nadu, India. 60 late preterm infants and their primary caregivers have been recruited after a detailed explanation of the study within the first week of delivery. At 3rd and 6th month, 41 infants were followed. Inclusion criteria Infants eligible for inclusion in the study were those born between 33 and 36 completed weeks of gestation and were clinically stable at the time of discharge from the neonatal unit. Additionally, the presence of a primary caregiver, typically a parent, who was willing to participate in the scheduled follow-up assessments was required [18]. Exclusion criteria Children with congenital anomalies, genetic abnormalities, and a history of recurrent NICU admissions. Neurodevelopmental assessment Developmental assessments were conducted using the Trivandrum Developmental Screening Tool (TDST), a culturally appropriate and pictorial tool designed for Indian children. The TDST is a culturally adapted screening tool validated for general developmental delay with a sensitivity of 85% and specificity of 91% in children aged 0–2 years across gross motor, fine motor, language, and personal-social domains. In this study, the TDST was administered at the 3rd and 6th months of corrected age. The scoring is binary (Achieved/Not achieved). When a child failed to achieve one or more age-appropriate milestones, this was considered indicative of a developmental delay, warranting referral to a healthcare professional [19]. Blinding procedure This study employed a dual-blinded methodology to reduce both observer and responder bias. At discharge, the primary investigator educated parents on how to use a simplified TDST-based checklist to observe and score their child’s milestones at home. Parents completed this checklist at the 3rd month of corrected age and submitted it during the scheduled follow-up visit. The primary investigator collected the parent-reported scores and referred the infant to an evaluating therapist for a separate TDST assessment. The scoring was performed by a single trained therapist. The therapist was blinded to the parents’ scores, and likewise, the parents did not have access to the therapist’s evaluation. The same procedure was followed at the 6th-month follow-up. This allowed the primary investigator, who was the only person with access to both datasets, to compare parental and therapist scoring to assess any discrepancies. Ethical approval Ethical clearance was obtained from the Institutional Ethics Committee (IEC) of SRM Medical College Hospital and Research Centre (Approval No: SRMIEC-ST0224-1174), and the study was conducted following Declaration of Helsinki [20]. A written informed consent was obtained from all parents or legal guardians before enrollment. Data collection The baseline child data, namely gestational age, birth weight, head circumference, APGAR scores at 1 and 5 minutes, and NICU stay duration, were collected at discharge. Parental demographic details such as age, educational status, socioeconomic status, and consanguinity marriage were also recorded. Factors influencing maternal mental health or parenting stress were not included in this study. Statistical analysis The data was processed with SPSS version 21. The demographic variables were descriptively analyzed. Wilcoxson's sign rank test was conducted to investigate the significant difference between the parent and the therapist score at the 3rd and 6th months. Linear regression analysis was used to determine potential predictors of parent and therapist scores. The demographic variables influence was checked by Mann-Whitney U test and Kruskal-Wallis’s test. A p-value of less than 0.05 was considered statistically significant. Results This prospective, non-experimental cohort study followed 41 late preterm infants (born between 34 and 36 weeks of gestation) over six months, using corrected age for developmental assessment. Demographic characteristics As detailed in Table 1, the mean paternal age was 30.5 ± 4.2 years, while the mean maternal age was 28.3 ± 3.9 years. Among the parents, 36(60%) of fathers and 39(65%) of mothers held graduate-level qualifications. Consanguineous marriages were reported in 7(12%) of cases. In terms of socioeconomic classification, 30(50%) of families were from middle-income backgrounds, 18(30%) from low-income backgrounds, and 12(20%) from high-income groups. Table 1Demographic and perinatal characteristics of study participantsVariableSub- variablesMean (SD)/n (%)Parental Age (years)-30.5 ± 4.2Maternal Age (years)-28.3 ± 3.9Mean Birth Weight (kg)-2.4 ± 0.3Mean Gestational Age (weeks)-35.2 ± 1.2NICU Stay Duration (days)-4.2 ± 2.1Paternal Literacy LevelHigh School24(40%)Graduate36(60%)Maternal Literacy LevelHigh School21(35%)Graduate39(65%)Consanguineous MarriageYes7(12%)No53(88%)Socioeconomic StatusLow18(30%)Middle30(50%)High12(20%)Participants Follow upAt discharge100% (60 samples)At 3rd month68% (41 samples)At 6th month68% (41 samples)Infant Biological sexBoy33 (55%)Girl27 (45%)APGAR Score1 min7.5 ± 0.95 min8.8 ± 0.6 Demographic and perinatal characteristics of study participants All infants (N = 60) were enrolled at the time of discharge, with a same cohort of infants (n=41) retention of 68% was assessed at both the 3rd and 6th months, ensuring consistency in follow-up. The initial drop-out occurred between discharge and 3 months. After 3 months, those who returned continued through 6 months, aided by proactive reminders, parental rapport, and scheduled follow-up calls. The cohort consisted of 55% male and 45% female infants. The mean birth weight was 2.4 ± 0.3 kg, and the average gestational age was 35.2 ± 1.2 weeks. The mean APGAR scores were 7.5 ± 0.9 at 1 minute and 8.8 ± 0.6 at 5 minutes, and the average NICU stay was 4.2 ± 2.1 days. A subgroup analysis revealed that among graduate fathers (n = 25), only 36% accurately identified motor milestones when cross-checked against therapist assessments, suggesting that formal education alone may not be sufficient to ensure accurate developmental monitoring. Component-wise comparison between therapist and parent scores at 3 months At the 3-month follow-up, Wilcoxon signed-rank tests were conducted to compare therapist and parent scores (Table 2) for individual developmental components. There was a statistically significant difference between therapist and parent ratings for eye-follow (Z = −2.121, p < 0.034) and head holding (Z = −2.887, p< 0.004), suggesting discrepancies in parental recognition of these milestones. However, no significant difference was observed in social smile (Z = 0.000, p >1.000) or in the overall TDST score (Z =−1.155, p >0.248). Table 2Comparison of parent and therapist developmental scores at 3 monthsComponentZ-valueAsymp.Sig(2-tailed)InterpretationSocial Smile0.0001.000No significant differenceEye Follow−2.1210.034*Significant differenceHead Holding−2.8870.004**Highly significant differenceOverall TDST score (Therapist vs. Parent)−1.1550.248No significant difference*p < 0.05**p < 0.01 Comparison of parent and therapist developmental scores at 3 months *p < 0.05 **p < 0.01 Component-wise comparison between therapist and parent scores at 6 months At the 6-month follow-up, the Wilcoxon signed-rank test revealed significant differences between parent-reported and therapist-assessed developmental milestone scores (Table 3). Notably, discrepancies were found in the domains ofrolling from back to stomach (Z = −3.207, p<0.001), transferring objects from one hand to another (Z = −2.496, p<0.013), and the overall TDST score (Z = −3.477, p<0.001). These findings suggest parents may under- or overestimate certain milestones compared to therapists. However, no statistically significant differences were found in social smile, eye following, head holding, or turning head to sound, indicating consistency in these observations. Table 3Comparison of parent and therapist developmental scores at 6 monthsComponentZ-valueAsymp.Sig(2-tailed)InterpretationSocial Smile0.0001.000No significant differenceEye-follow0.0001.000No significant differenceHead holding0.0001.000No significant differenceRolls from back to stomach−3.2070.001**Highly significant differenceTurns head to sound−1.1340.257No significant differenceTransfers object from one hand to another−2.4960.013*Significant differenceOverall TDST score−3.4770.001*Highly significant difference*p < 0.05**p < 0.01 Comparison of parent and therapist developmental scores at 6 months *p < 0.05 **p < 0.01 The detailed parent therapist score (Achieved/Not Achieved), percentage of agreement and table of atypical scores are provided in supplementary file 1. Regression analysis Linear regression analysis was conducted only for the 6-month follow-up period (Supplementary file 2), as regression at the 3rd month was not feasible. At 3 months, the component scores, lacked sufficient variability and interpretive value for predictive analysis. The regression models conducted for the 6-month follow-up revealed that two developmental milestones—rolling from back to stomach and transferring objects hand to hand—were statistically significant predictors of overall developmental scores, as reported by both parents and therapists. For parental scores, both rolling from back to stomach (B = 0.78; 95% CI, 0.35–1.21; p = 0.001) and transfer of objects from hand to hand (B = 0.78; 95% CI, 0.52–1.03; p < 0.001) were significant predictors. The model was statistically significant, F (2, 38) = 22.707, p < 0.001, explaining 54.4% of the variance (R² = 0.544). Predictive equation:parent score (6 months) = 4.222 + 0.778(rolling) + 0.778(transfer) For therapist scores, the same two components were also significant predictors: rolling from back to stomach (B = 0.95; p < 0.001; 95% CI, 0.75–1.15) and transfer of objects from hand to hand (B=1.01; p < 0.001; 95% CI, 0.81–1.20). This model was highly significant, F (2, 38) = 107.626, p < 0.001, accounting for 85% of the variance (R² = 0.850). Predictive equation: therapist score (6 months) = 3.930 + 0.951(rolling) + 1.008(transfer) The parental score model explained 54.4% of the variance, indicating a moderate level of predictability. In contrast, the therapist score model explained a substantial 85% of the variance, suggesting a stronger and more consistent relationship between these milestones and overall therapist-assessed development. These results highlight the importance of gross and fine motor skills, particularly the ability to roll over and transfer objects between hands, as reliable markers for developmental surveillance at 6 months of corrected age. The stronger predictive power in therapist assessments may reflect greater scoring consistency and observational expertise compared to parent-reported evaluations. Influence of demographic variables Statistical tests were conducted to examine whether demographic and perinatal factors influenced developmental scores. Using the Mann–Whitney U test, no significant differences were found in developmental scores based on infant biological sex at either follow-up point for both parent and therapist assessments. Similarly, the Kruskal–Walli’s test indicated no significant influence of mode of delivery on developmental scores at both 3rd and 6th months. The educational level of fathers also did not show a statistically significant impact. Furthermore, APGAR scores at both 1 and 5 minutes were not significantly associated with developmental scores, nor was socioeconomic status at any time point. Discussion This prospective cohort study aimed to assess neurodevelopmental outcomes among late preterm infants at the 3rd and 6th months of corrected age, focusing on comparisons between parental and therapist-reported developmental scores and identifying key predictors of scoring. The findings revealed both areas of alignment and significant discrepancies in milestone recognition between parents and therapists, highlighting the critical need for enhanced caregiver education and early intervention strategies. At 3 months, parents and therapists demonstrated moderate agreement in overall developmental assessments, suggesting a baseline awareness of key early milestones among caregivers. However, domain-specific discrepancies were evident, especially in visual and motor behaviours such as eye-following and head-holding. These mismatches imply that while parents observe their infants closely, they may not fully interpret the developmental significance of these behaviours, particularly those involving postural control and visual-motor integration. Similar findings have been reported in Chinese samples, where parents and teachers showed low sensitivity in identifying motor difficulties, even among children performing below the 15th percentile (Suspected Developmental coordination disorder) on the Movement ABC-2 performance test [21]. By 6 months, the divergence between parent and therapist assessments widened further, with statistically significant discrepancies noted in complex motor tasks such as rolling from back to stomach and transferring objects from hand to hand. These skills, which involve midline crossing and bilateral coordination, are known indicators of neurological maturation. Under-recognition by parents may stem from the increasing complexity of motor milestones, consistent with global evidence showing that as task demands increase, parental accuracy declines. For example, in an Australian cohort, parent proxy reports explained only ~11% of the variance in actual motor performance assessed using the TGMD-3 [22]. Similarly, Zysset et al. (2018) reported a weak correlation (r ≈ 0.23) between parental ratings and standardized scores using the Zurich Neuromotor Assessment (ZNA3–5), reinforcing that caregiver reports often lack precision in motor domains [23]. Regression analysis in our study revealed that rolling and object transfer were the most significant predictors of developmental outcomes at 6 months for both therapist- and parent-reported models. However, the variance explained was markedly higher in the therapist model (R² = 0.850) compared to the parent model (R² = 0.544), reflecting the advantages of structured training and standardized assessments in capturing developmental delays more reliably. Therapist and parent scores at 6 months showed persistent mismatches especially in motor domains underscore the limitations of relying solely on self-usage of assessment tool rather than observational learning. This is further enhanced by previous research indicating that while parents may become more aware over time through repeated interactions with healthcare professionals, exposure alone is insufficient to develop the interpretive skills required for accurate milestone tracking. According to Akkineni et al. (2020) reported that 36% of parents possessed knowledge of child development, despite 46.2% of mothers and 52.3% of fathers having attained secondary education, highlighting that formal schooling does not guarantee developmental literacy [24]. Several validated tools attempt to address these challenges. The Early Motor Questionnaire (EMQ) has shown promise in eliciting useful parental insights into motor development when milestones are clearly defined and contextualized (Libertus & Landa, 2013) [25]. Similarly, Dite and Amick (2021) found that daily checklists improved recall accuracy for gross motor skills, though variability persisted without professional guidance. However, tools like the CDC Milestone Tracker (version 2022) [26]and ASQ-3, unless accompanied by training and cultural adaptation, may fall short in low- and middle-income settings where developmental literacy is variable and milestone interpretation may be culturally influenced [27]. This study results highlighting the difficulty in identifying more complex or subtle motor skills at 6 months namely transferring objects from one to other, rolling from back to stomach, is consistent with prior literature (Miller et al, 2017) comparing parent reports and direct assessments in toddlers, where greater alignment is observed in language or social-emotional domains, but motor skill reports are often less reliable, reflecting variability in caregiver perception and reporting accuracy [28]. Our analysis of demographic and perinatal variables including child gender, mode of delivery, paternal education, APGAR scores, and socioeconomic status revealed no statistically significant influence on developmental scores. This finding is consistent with earlier studies suggesting that, within medically stable late preterm populations, these variables may have limited predictive value [29, 30]. Taken together, these findings underscore the urgent need for structured, milestone-specific, and culturally adapted parental education. The critical window for neuroplasticity in the first 3 months of life suggests that interventions delayed beyond this period may miss the optimal opportunity for effective early intervention[31]. Importantly, parental expertise developed after the third month may be too late to detect delays during the most critical window of neuroplasticity. Therefore, proactive caregiver education before the 3-month mark is vital to ensuring timely recognition and intervention for developmental delays. A key strength of this study lies in its prospective design, standardized corrected age assessment, and dual-informant comparison (therapist and parent) across multiple time points. The use of non-parametric and regression analyses provided robust insights into milestone recognition and predictor validity. The limitations of the study are as follows: First, 68% follow-up may limit the generalizability of findings, particularly if the dropouts were not random. Second, the single-centre design and relatively modest sample size restrict external validity. Third, the study did not assess maternal mental health or parenting stress, factors known to influence caregiver developmental perceptions. Future recommendations Future research should consider extending the follow-up period beyond infancy to determine whether discrepancies between parent and therapist assessments persist into toddlerhood. Community-adapted milestone tools using illustrations and lay terminology could improve usability among low-literacy populations. Additionally, the effect of structured parental training modules on improving milestone recognition accuracy can be considered. Policy makers should emphasize training parents in developmental surveillance protocols. Conclusion This study adds to growing evidence that while many parents of late preterm infants are attentive and engaged, they may struggle to recognize early signs of developmental delays especially as milestones become more complex. At 3 months, there was broad agreement between parents and therapists on general developmental progress. But by 6 months, clear gaps began to emerge, particularly in identifying motor skills like rolling over and transferring objects key indicators of neurological growth. The findings highlight a clear need: parents must be supported not only with information but with structured guidance tailored to their cultural and educational backgrounds. The importance of developmental monitoring along with signs of concern may enhance parent and therapist to work together for child betterment. Supplementary Information Supplementary Material 1 Supplementary Material 2 Acknowledgements We acknowledge and thank Dr. Raghul Gandhi Venkatesan (Assistant Professor), Center for Statistics, SRM Institute of Science and Technology, Kattankulathur, for providing consultations on statistical analysis. Abbreviations Trivandrum Developmental Screening Tool Institutional Ethical Committee Authors’ contributions All authors have contributed, read, and approved the final manuscript. Funding Open access funding provided by SRM Institute of Science and Technology for SRMIST – Medical & Health Sciences. The authors gratefully acknowledge the financial support by SRM College of Physiotherapy, Faculty of Medicine and Health Sciences, SRM Institute of Science and Technology (SRMIST), Kattankulathur, Chengalpattu 603203, India for bearing the defrayed costs of publishing this article. Data availability The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate Ethical approval was obtained from the Institutional Ethics Committee (IEC) of SRM Medical College Hospital and Research Centre (Approval No: SRMIEC-ST0224-1174), and conducted following the Declaration of Helsinki. A written informed consent was obtained from all parents or legal guardians before enrollment. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Conflict of interest The Author declares no conflicts of interest to disclose. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Supplementary Material 2 Data Availability Statement The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.