Preterm Children in Iceland: Neonatal Morbidities and Motor Development in the First Two Years, and Later Disability Diagnosis Authors: Hildur Bjork Adolfsdottir, Edda Thorvaldsdottir, Helga Bogadottir, Bjorg Gudjonsdottir Published: 2026 ABSTRACT ABSTRACTAimIcelandic data on outcomes after preterm birth are limited. The aim was to describe neonatal morbidities, motor development up to 24 months, and examine whether morbidities predicted disability.MethodIt was a retrospective cohort study of 318 children born < 32 weeks of gestation and/or weighing < 1 000 g, who were followed at the University Hospital of Iceland from January 2012 to December 2020. Motor development was assessed using the Alberta Infant Motor Scale to 15 months, and the Bayley Scales of Infant and Toddler Development–III motor scales at 12 and 24 months. Associations between neonatal morbidities and outcomes were examined using correlation and multivariable regression.ResultsNeonatal morbidities were more common in children born < 28 weeks of gestation and/or weighed < 1 000 g, who also had higher rates of later disability diagnoses. Motor performance to 24 months was below test norms in this group, but morbidities were not associated with motor scale outcomes. In the cohort of children born before 32 weeks of gestation, cumulative neonatal morbidity burden independently predicted later disability, with adjusted odds increasing by approximately 60% per additional morbidity.ConclusionChildren born more prematurely had greater morbidity and disability risk. Cumulative neonatal morbidity predicted later disability, supporting long‐term follow‐up. Abbreviations Alberta Infant Motor Scale The Bayley Scales of Infant and Toddler Development –III confidence interval Cerebral palsy grams odds ratio variance inflation factor World Health Organisation Introduction Preterm birth affects more than one in ten infants worldwide and remains a major contributor to neonatal morbidity and long‐term neurodevelopmental impairment [1]. Advances in perinatal and neonatal care, such as antenatal steroids and exogenous surfactant administration, have improved their survival in the last two decades [2]. However, the risk of neonatal morbidity and subsequent developmental impairment remains strongly related to gestational age and birth weight [3]. The World Health Organisation defines preterm birth as delivery before 37 completed weeks of gestation and is classified as extremely preterm (< 28 weeks of gestation), very preterm (28 to < 32 weeks of gestation), and moderate to late preterm (32 to < 37 weeks of gestation) [1]. Low birth weight is defined as < 2 500 g, with very low birth weight < 1 500 g and extremely low birth weight < 1 000 g [3]. Low birth weight reflects prematurity or growth restriction, or both, and has been linked to adverse outcomes [3]. Both pathological conditions and external stimuli can interfere with the infant's neurological development. As a result, preterm infants are at increased risk of complications such as respiratory distress syndrome, bronchopulmonary dysplasia, intraventricular haemorrhage, sepsis, periventricular leukomalacia, and patent ductus arteriosus. The prevalence of these morbidities increases with decreasing gestational age and birth weight [4, 5, 6, 7, 8, 9, 10] and is associated with both major outcomes, such as cerebral palsy, and more subtle motor and sensory impairments [11, 12]. An increasing number of neonatal morbidities have been associated with a stepwise increase in the risk of adverse neurodevelopmental outcomes in very preterm and extremely preterm infants. Studies have suggested that cumulative morbidity burden provides clinically meaningful prognostic information beyond any single morbidity alone [13, 14, 15]. Motor Development Across studies, preterm infants score lower than full‐term peers on the Alberta Infant Motor Scale (AIMS) up to 18–19 months, despite age correction. They appear to follow a similar but slower trajectory compared with full‐term peers [16, 17, 18]. Poorer AIMS outcomes have been associated with low birth weight and morbidities such as respiratory distress syndrome, intraventricular haemorrhage, and retinopathy of prematurity, whereas bronchopulmonary dysplasia has shown no significant effect [16, 17]. Periventricular leukomalacia has been associated with additional reductions at 6, 12, and 18 months [18]. A systematic review confirmed that lower AIMS scores in preterm infants up to 18 months are associated with shorter gestation, lower birth weight, intraventricular haemorrhage, white matter injury, and chronic lung disease [17]. Serenius et al. [19], found that infants born before 27 weeks of gestation scored significantly lower than term peers on the Bayley Scales of Infant and Toddler Development–III (BSID‐III) at 2.5 years, thereby providing evidence that delays persist beyond the first year. Mean BSID‐III increased with longer gestation for each additional week, and motor scores rose by 2.5 points. In the preterm group, 84.8% had normal or mildly impaired motor development, compared with 98.6% among term‐born children. Perinatal and neonatal care in Iceland aligns with international standards. Follow‐up of preterm infants was previously led by paediatricians, but since 2012, the University Hospital of Iceland has provided multidisciplinary follow‐up tracking of the health and developmental outcomes of preterm children to age six. Infants born < 28 weeks of gestation and/or those with birth weight < 1 000 g receive standardised motor assessments with the AIMS at four, eight, and 12 months and BSID‐III at 12 and 24 months. Very preterm infants (28–32 weeks of gestation) receive routine paediatric follow‐up care and are referred to physiotherapy if motor delay is detected. Children with significant motor or cognitive delays are referred to the Counselling and Diagnostic Centre. This also applies to children with confirmed or suspected neurodevelopmental diagnoses. The Centre is a national institution providing multidisciplinary assessment, early intervention, and resources for children with developmental disabilities and their families. Upon referral, hospital follow‐up ceases. This follow‐up programme enables local evaluation and action based on the findings. The purposes of this study were fivefold. First, we assessed the prevalence of neonatal morbidities in children born before 32 weeks of gestation. Second, we determined the proportion of these children who received a neurodevelopmental diagnosis or were referred to interventions. Third, we examined whether neonatal morbidities predicted later disability. Fourth, we tracked gross motor outcomes over two years in extremely preterm children and/or those with extremely low birth weight. Finally, we examined how neonatal morbidities related to motor development at 12 and 24 months' corrected age. Materials and Methods Design This was a retrospective longitudinal cohort study that was based on data from medical records at the University Hospital of Iceland and from the Counselling and Diagnostic Centre in Iceland. Participants and Procedures Information was collected on 318 children who were born before 32 weeks of gestation at the University Hospital during the period 2012–2020. The children were divided into two groups by gestational age and birth weight. Group one included extremely preterm children and/or those with extremely low birth weight (< 1000 g), and group two consisted of very preterm children. Data Collection We worked with the Children's Hospital's preterm team to extract data for extremely preterm infants, very preterm infants and/or those with extremely low birth weight. Variables included gestational age, sex, multiple birth, and maternal age. Birth characteristics included weight and length. Clinical variables included ventilator days and neonatal morbidities. These comprised sepsis, respiratory distress syndrome, bronchopulmonary dysplasia, intraventricular haemorrhage grade, cystic periventricular leukomalacia, and patent ductus arteriosus. Cystic periventricular leukomalacia is a severe form of periventricular leukomalacia. Physiotherapy was also recorded. For group one, standardised motor assessment results were also retrieved. A list of children born before 32 weeks of gestation who were referred to the Counselling and Diagnostic Centre was compiled to obtain information on disability diagnoses and services received. Outcome Measures The AIMS is a standardised observational measure of movements in infants from birth to independent walking. It has 58 items across four positions: Supine, prone, sitting, and standing. Each item scores one point (total 0–58). Norms come from 2202 infants, with one‐month age bands to 19 months and corrected age for preterm. Scores are interpreted against age norms: 1–2 standard deviations below the mean suggests delay, and > 2 standard deviations indicate a strong suspicion of delay [20]. The BSID‐III is a standardised tool designed to assess development from 1 to 42 months across five domains [21]. In this study, only the motor scale was used. Each item is scored as 0 or 1, where 0 indicates that the child cannot perform the item and 1 indicates that the child can perform it. Testing stops after three consecutive scores of 0, and total scores are converted to age‐specific scaled scores using conversion tables. Normative data from 1700 American children produce scaled scores ranging from 1 to 19 with a mean of 10 ± 3. For preterm infants, age is corrected up to two years [21]. The psychometric properties of the AIMS and the BSID‐III have been studied and have demonstrated good validity and high reliability [20, 21]. Data Analysis Data processing and statistical analysis were conducted using Microsoft Excel (Microsoft Corp., USA), Jamovi 2.3.28 (The Jamovi project, Australia) and the R (version 4.3.3; R Foundation for Statistical Computing, Austria) with the mvtorm package (version 1.2–4). Descriptive statistics summarised information about the sample. The prevalence of later disability diagnosis and intervention referrals of children born before 32 weeks of gestation was calculated. Categorical variables were presented as numbers and percentages, and the chi‐square test of independence or Fisher's exact test was used when appropriate. Fisher's exact test was used when any expected cell count fell below 5. A two‐tailed significance level of 0.05 was used for the tests. To assess whether overall neonatal morbidity burden was associated with later disability, we constructed a morbidity count (range 0–5) by summing five binary neonatal morbidities: Sepsis, patent ductus arteriosus, respiratory distress syndrome, bronchopulmonary dysplasia, and severe brain injury. Intraventricular haemorrhage (grade III) and cystic periventricular leukomalacia were combined into a composite severe brain injury variable because both represent major neonatal brain injury associated with adverse neurodevelopmental outcome, and the low frequency of each diagnosis separately limited statistical power. We fitted the binomial logistic regression model with disability (yes/no) as the dependent variable and morbidity count, gestational age (expressed in decimal weeks), and sex as predictors. Results are presented as odds ratios (OR) with 95% confidence intervals (CI). All tests were two‐sided, with statistical significance set at p < 0.05. To address whether children in group one had delayed gross motor development compared to age‐adjusted peers, motor assessments were grouped according to the AIMS age categories. Each month was defined as 30 days. Dunnett's T3 test was used to compare the mean AIMS scores of the preterm infants to reference values for each age group. Each preterm age group was compared to its corresponding AIMS reference group to determine whether the mean difference differed from zero. The significance level was set at ≤ 5%. Dunnett's T3 is similar to a two‐sample t‐test but accounts for dependent age groups, unequal variances, and sample sizes. Due to non‐normal distribution, the Wilcoxon signed‐rank tests were performed to compare the BSID‐III total scores, gross motor scores, and fine motor scores of preterm infants of around 12 and 24 months of corrected age to the mean scores for full‐term peers. The children were not all exactly 12 or 24 months of age when the test was administered. Measurements for the 12 months were taken at 10, 12, and 15 months of corrected age. For the 24 months, the ages ranged from 24 to 36 months. The significance level was p ≤ 0.05 in the analyses. A correlation test examined the relationship between BSID‐III outcomes at 12 and 24 months of age and gestational age and birth weight, using Spearman's correlation due to non‐normal distribution. A multiple linear regression model assessed the impact of birth weight and neonatal morbidities on BSID‐III total scores. BSID‐III total scores were the explanatory variables. The models included birth weight, sepsis, respiratory distress syndrome, bronchopulmonary dysplasia, patent ductus arteriosus, and grade III intraventricular haemorrhage as independent variables. Of the 87 infants in group one who were in the follow‐up during the first 18 months of life, 66 required ventilations, averaging 14.67 ± 14.15 days. However, due to incomplete records on ventilation days, these data were not used in the analysis. The variance inflation factor (VIF) was used to check for multicollinearity. An F‐test was used to evaluate if predictor coefficients were zero, testing to verify there was no relationship between the explanatory variables and BSID‐III scores. Results Neonatal Morbidities The cohort comprised 318 children, split by gestational age. Group one comprised 130 infants born before 28 weeks of gestation and/or weighing < 1 000 g. Of these, 25 died, and one had no available data, leaving 104 survivors for analysis. Group two comprised 188 infants born at 28–32 weeks of gestation, of whom seven died. Table 1 summarises the demographics and morbidities of the groups. Group differences in neonatal morbidity diagnoses were significant since group one had higher rates of neonatal morbidities. TABLE 1Information about participants.Extremely preterm infants and/or those with extremely low birth weight (< 28 weeks of gestation/< 1 000 g)Very preterm infants (28–32 weeks of gestation) N = 104 N = 181Number%Number% p Male5048.110558—Female5451.97642—One child8177.913876.2—Twins2120.24323.8—Triplets21.900—Sepsis a 3735.6116.1< 0.001* Respiratory distress syndrome a 8177.96636.5< 0.001* Bronchopulmonary dysplasia a 5351168.8< 0.001* Intraventricular haemorrhage (grade I–III) a 26251910.50.001* Intraventricular haemorrhage (grade I) a 1110.6105.50.116Intraventricular haemorrhage (grade II) b 76.742.20.105Intraventricular haemorrhage (grade III) b 87.752.80.075Periventricular leukomalacia43,810.60.062Patent ductus arteriosis a 4442.3116.1< 0.001* a A chi‐square test of independence. b Fisher's exact test.*A significant group difference. The significance level was p ≤ 0.05. Information about participants. A chi‐square test of independence. Fisher's exact test. A significant group difference. The significance level was p ≤ 0.05. Diagnosis of Disabilities and Referrals Group differences in diagnoses were significant. Children in group one had higher referral rates to the Counselling and Diagnostic Centre. They also had higher rates of cerebral palsy and other neurodevelopmental disorders. In addition, they received more intervention referrals, except for speech therapy. The category of neurodevelopmental disability was heterogeneous and included diagnoses such as cerebral palsy, intellectual disability, autism spectrum disorder, attention deficit hyperactivity disorder, and various syndromes. Table 2 shows the referral rates and the three main diagnosis categories. Some children had more than one diagnosis, resulting in overlap between categories. TABLE 2Service referrals and diagnosis.Extremely preterm infants and/or those with extremely low birth weight (< 28 weeks of gestation/< 1 000 g)Very preterm infants (28–32 weeks of gestation) N = 104 N = 181Number%Number% p‐valuesThe Counselling and Diagnostic Centre a 2524179.4< 0.001* CP a 1312.5105.50.037* Neurodevelopmental disorders a 1110.631.7< 0.001* Motor impairments other than CP b 76.742.20.105Referrals to physical therapy a 51494022.1< 0.001* Referrals to occupational therapy a 1817.3116.10.003* Referrals to speech therapy b 312.5950.545Personal assistance in kindergarten/preschool a 2322.1147.7< 0.001* a A chi‐square test of independence. b Fisher's exact test.*A significant group difference. The significance level was p ≤ 0.05. Service referrals and diagnosis. A chi‐square test of independence. Fisher's exact test. A significant group difference. The significance level was p ≤ 0.05. In the binomial logistic regression model, adjusted for gestational age and sex, neonatal morbidity burden (0–5) was associated with later disability. Each additional neonatal morbidity increased the odds of disability by approximately 60% (adjusted OR = 1.60, 95% CI 1.119–2.15, p = 0.001). This is shown in Table 3. TABLE 3Binomial logistic regression for disability (morbidity burden model).PredictorsAdjusted OR95% confidence interval p LowerUpperMorbidity count1.601.202.150.001Gestational age1.060.901.250.478Sex – female (vs. male)1.060.552.080.847 Note: Outcome definition (disability, yes/no); Morbidity count (5 morbidities) (per 1 additional morbidity); Gestational age in decimal weeks. Binomial logistic regression for disability (morbidity burden model). Note: Outcome definition (disability, yes/no); Morbidity count (5 morbidities) (per 1 additional morbidity); Gestational age in decimal weeks. Motor Development In group one, 87 children had at least one recorded motor development assessment during the first 18 months of life. The number of infants in group one assessed with AIMS over the first year of life varied at each time point. Information about this group is presented in Table 4. TABLE 4Extremely preterm infants and/or those with extremely low birth weight (< 28 weeks of gestation and/or < 1 000 g).NumberMean (SD)SexBoys42Girls45MultiplesTwins17Triplets2Gestational age in weeks8726.25 (1.98)Classification of gestational age< 241224–< 251225–< 261626–< 271527–< 281328–< 2910* 29–< 306* ≥ 303* Birth weight in grams87779.62 (184.18)Classification of birth weight< 1 000791000–15008Birth length in centimetres63** 33.26 (3.40)Number of motor assessments for each child873.64 (1.89)Classification of motor assessment1 measurement42 measurements33 measurements74 measurements175 measurements56Referrals to physiotherapyPhysiotherapy46No physiotherapy41Mothers Ages (years)82*** 31.22 (6.64)*Infant with extremely low birth weight born before 28 weeks of gestation.**Measures of birth length were missing for 24 infants.***Five mothers had two children in the group. Extremely preterm infants and/or those with extremely low birth weight (< 28 weeks of gestation and/or < 1 000 g). Infant with extremely low birth weight born before 28 weeks of gestation. Measures of birth length were missing for 24 infants. Five mothers had two children in the group. The infants in group one had significantly lower AIMS scores at age ranges 4–7, 8–9, and 12–13‐months corrected age (Figure 1 and Table 5). Mean AIMS scores of the extremely preterm children and the norm‐referenced group. TABLE 5Results from the Dunnett T3 test. The mean difference was calculated by subtracting the mean of the preterm infants in each age group from the mean of the reference group in the AIMS test, and the 95% confidence interval is provided for the difference.Age in monthsNumberMeanStandard deviationDifference between means p 95% confidence intervalLower limitsUpper limits2− < 368.001.901.800.427−1.75.33− < 42410.922.861.680.193−0.393.764− < 53913.773.344.13< 0.001* 2.196.075− < 61117.733.585.47< 0.005* 1.489.466− < 71821.114.527.19< 0.001* 3.5510.837− < 81827.945.764.360.072−0.278.988− < 93029.807.0210.00< 0.001* 5.714.39− < 10938.446.657.060.121−1.5515.6610− < 11738.8612.6510.440.45410.4231.3111− < 12846.386.094.920.413−3.7713.6212− < 134147.736.286.87< 0.001* 3.6710.0713− < 141046.009.399.600.097−1.4720.6714− < 15947.4410.979.460.262−4.8523.76*A significant difference. The significance level was p ≤ 0.05. Results from the Dunnett T3 test. The mean difference was calculated by subtracting the mean of the preterm infants in each age group from the mean of the reference group in the AIMS test, and the 95% confidence interval is provided for the difference. A significant difference. The significance level was p ≤ 0.05. For BSID‐III at approximately 12 months, we assessed 83 children, and 64 at around 24 months. The attrition reflected prior diagnoses of disability with referral to the Counselling and Diagnostic Centre, missed visits, or loss to follow‐up care. Some families may have declined follow‐up if the child seemed to be typically developing. Table 6 shows that in the Wilcoxon signed‐rank results, BSID‐III total and gross motor scores were significantly lower than the norms at both 12 and 24 months' corrected age. Fine motor scores were significantly lower at 12 months of age, but not at 24 months. TABLE 6The results from BSID‐III at 12 and 24 months in extremely preterm children. and/or those with extremely low birth weight.ScoresMean ± SD95% confidence interval p 12 monthsBSID‐III total scores—normative score is M = 10088.8 ± 16.985.2–92.5< 0.001 * BSID‐III fine motor scores—normative score is M = 109.44 ± 2.748.85–100.036 * BSID‐III gross motor score—normative score is M = 107.08 ± 2.66.52–7.65< 0.001 * 24 monthsBSID‐III total score—normative score is M = 10093.3 ± 11.890.6–96.5< 0.001 * BSID‐III fine motor score—normative score is M = 1010 ± 2.729.37–10.70.929BSID‐III gross motor score—normative score is M = 107.78 ± 1.917.78–8.26< 0.001 * * A significant difference. The significance level was p ≤ 0.05. The results from BSID‐III at 12 and 24 months in extremely preterm children. and/or those with extremely low birth weight. A significant difference. The significance level was p ≤ 0.05. BSID‐III Total Scores and Gestational Age and Birth Weight Spearman's correlation between gestational age or birth weight and BSID‐III total scores was not significant at either age. At 12 months, the correlation between gestational age and BSID‐III was −0.079 (p = 0.482), while the correlation between birth weight and BSID‐III scores was 0.006 (p = 0.959). At 24 months, the correlation between gestational age and BSID‐III scores was 0.149 (p = 0.256), and between birth weight and BSID‐III scores it was 0.152 (p = 0.247). The Impact of Birth Weight and Neonatal Morbidities on the BSID‐III Total Scores The regression model results for the 12‐month‐old children are in the table in Appendix I. The regression analysis demonstrated that the intercept was statistically significant, with a baseline estimated outcome of 93.4 (95% CI 84.8–121.2, p < 0.001). None of the explanatory variables examined was significantly associated with the outcome. The explanatory variables did not yield statistically significant associations, as indicated by confidence intervals that encompassed zero. The result of the F‐test for the regression model was F = 0.806 (p = 0.568), which means that the explanatory variables did not account for the variation in the BSID‐III scores any better than a random distribution around the mean. The model's coefficient of determination, R2 was 0.061, which is low, meaning that the variables accounted for 6.1% of the motor scores. The regression model results for 24‐month‐old children are presented in a table in Appendix I. The F‐test for the regression model for this age group is F = 1.527 (p = 0.187), indicating that the explanatory variables explained little of the variation in BSID‐III total scores and no better than values randomly dispersed around the mean. The variables accounted for 14.5% of the BSID‐III total scores, with an R2 of 0.147. Overall, the findings suggest that within this cohort, neither birth weight nor major neonatal morbidities were predictive of the developmental outcome assessed. Discussion This study examined neonatal morbidities, early motor development, and later disability among children born preterm in Iceland. Children in group one were more frequently referred to the Counselling and Diagnostic Centre than those in group two. They also received more disability diagnoses, including cerebral palsy and other neurodevelopmental disorders. In addition, they were more often referred to physiotherapy, occupational therapy, and speech therapy. They also showed significant gross motor delay relative to AIMS norms at certain ages and scored below BSID‐III motor norms at 12 and 24 months. The explanatory variables examined did not significantly predict motor test outcomes. A significantly higher proportion of infants in group one had neonatal morbidities than those in group two, aligning with evidence that shorter gestation increased risk due to organ immaturity [4, 5, 6, 7, 8, 9, 10]. Notably, cystic periventricular leukomalacia and grade III haemorrhage were uncommon in our cohort, an encouraging finding given their strong association with later cerebral palsy. Compared to children in group two, children in group one had significantly higher referral rates to the Counselling and Diagnostic Centre and to physical, occupational, and speech therapy. This reflects the greater proportion of children in this group diagnosed with CP, developmental disabilities, and/or motor impairments, which in turn increases the need for rehabilitation services. In this study, an increasing number of neonatal morbidities was associated with higher odds of later disability, even after adjustment for gestational age and sex. This finding is in accordance with previous studies showing that the accumulation of major neonatal morbidities is associated with stepwise increases in the risk of adverse neurodevelopmental outcomes in extremely preterm infants [13, 14, 22, 23]. Although the specific morbidities included in composite counts vary somewhat across studies, the overall pattern appears consistent. A greater burden of neonatal morbidity was associated with progressively worse long‐term outcomes. The association between higher morbidity and later disability may have reflected the cumulative effect of multiple serious neonatal morbidities. Each morbidity may contribute to systemic instability, prolonged intensive care exposure, and heightened vulnerability of the developing brain. These findings suggest that morbidity count may be a useful marker of later risk, but it should be interpreted alongside other neonatal and postnatal factors when planning follow‐up in this population. Children in group one showed significantly delayed gross motor development at the ages of 4– < 5 months, 5– < 6 months, 6– < 7 months, 8– < 9 months, and 12– < 13 months compared to the AIMS test reference group. These age bands had the largest samples; other age intervals included fewer children and greater variability, reducing statistical power to detect real differences. AIMS is widely used in the preterm population, and comparisons with those studies support the findings of this study [16, 17, 24]. On the BSID‐III total motor scale, motor delays were evident at around 12 and 24 months. The sole exception was the 24‐month fine motor score, which did not differ significantly from normative values. This study reinforced an increased risk of delayed motor development after preterm birth. Consistent with previous studies, outcomes among extremely preterm infants and/or those with very low birth weight were heterogeneous, with some scoring high and others low at most ages, indicating substantial variability in motor developmental rates, as seen in full‐term children [16, 17]. Delays in motor skills can cascade to cognitive and language domains, and early motor performance in preterm infants predicts later motor, cognitive, and language outcomes [25]. Extremely preterm children may still lag behind full‐term peers at age three, even after age correction [26]. Therefore, systematic motor monitoring is warranted, and early intervention is supported by evidence of benefit in high‐risk infants, including those born preterm [27]. Birth weight and neonatal morbidities were not associated with BSID‐III motor outcomes at 12 or 24 months in extremely preterm and/or extremely low birth weight infants. Interpretation is limited by a small sample size and referral‐related loss to follow‐up, whereby children with disability diagnoses (and likely the lowest motor scores) were transferred to other services and no longer assessed with standardised motor measures in our dataset. Strengths and Limitations This was the first Icelandic study of motor development in the early life of extremely preterm infants and/or those with extremely low birth weight. The strength of the study was the use of reliable, valid motor assessments effective for detecting delay [20, 21]. The study also had a number of limitations. The retrospective design relied on routine clinic testing, which resulted in incomplete and uneven assessment due to illness and non‐attendance. Children were often tested at ages that differed from the intended assessment points. This may have biased age‐specific outcomes and risk factor analyses. Sample size was a limitation, particularly for comparisons with BSID‐III norms and risk factor effects, decreasing to 64 by the age of two years. Conclusions This study described neonatal morbidities, motor development up to 24 months, and whether morbidities predicted later disability. Neonatal morbidities were more common in group one than in group two. Group one also scored lower than norms on motor developmental tests and had higher rates of developmental disorders. Individual neonatal morbidities did not predict motor outcomes. However, cumulative morbidity burden among children born before 32 weeks of gestation independently predicted later disability. Longer‐term studies should evaluate whether neonatal morbidities influence motor outcomes beyond the age of two years. Author Contributions Hildur Bjork Adolfsdottir: investigation, writing – original draft, methodology, writing – review and editing, formal analysis, data curation, software, visualization. Helga Bogadottir: conceptualization, investigation, writing – review and editing, supervision, resources, writing – original draft. Bjorg Gudjonsdottir: conceptualization, investigation, funding acquisition, writing – original draft, writing – review and editing, formal analysis, data curation, supervision, project administration, methodology, validation, resources, software, visualization. Edda Thorvaldsdottir: investigation, writing – original draft, writing – review and editing, methodology, formal analysis, data curation, software, visualization. Funding This work was supported by the Research Fund for Women's and Children's Services at the University Hospital of Iceland. Ethics Statement The study was approved by the National Bioethics Committee (VSN‐23‐080) and the University Hospital of Iceland Health Research Ethics Committee. Approval included access to medical records for relevant data. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Appendix I Table Results from the multiple linear regression of the impact of birth weight and neonatal morbidities on BSID‐III total scores when the children were around 12 months. Appendix II Table Results from the multiple linear regression of the impact of birth weight and neonatal morbidities on BSID‐III total scores when the children were around 24 months.