This prospective longitudinal birth cohort follow-up analysis from urban Vellore evaluated developmental trends in the domains of cognition, language, motor and social skills; and predictors for individual domain change over time. All developmental domain quotients showed a decreasing trend over time with cognition and language domains dropping more than 15 points between 6 and 36 months of age. Children in the highest tertile of SES had better cognition and language scores over time. Stunted children had poorer cognition and motor scores over time. Home environmental factors of parental responsivity affected changes in cognition, language and motor scores, while punishment avoidance and toy availability influenced the motor score change. Higher maternal cognition affected changes in both language and motor domains, while maternal depression adversely affected language scores over time. Body iron levels was associated with changes in all cognition and motor skills. There was good level of follow-up in the current study with 94% at 6 months, 91% at 24 months and 87% at 36 months.
Child development evolves over time and developmental measures such as BSID-III have shown weak predictability of school age cognition [35], motor skills [36], and behaviour [37]. Despite this, BSID is the most common developmental tool used across the globe, as it is a sensitive measure of child development [23]. Though child developmental process and sequence are uniform across the world, there can be differences in the rate of achievement between populations. Trends of all developmental domains in an urban low-income setting as shown in this study can be useful for clinical practice in such settings, academic understanding, and analysing the risks contributing to any setback. Though the current analysis has not done a complete trajectory exploration, the birth cohort showed a decreasing trend in all developmental domains’ quotient. This is in discordance with another large rural birth cohort study reported from China, where more than 90% showed an increasing trend in cognitive development between 6 and 24 months of age [9]. The Vellore birth cohort is predominantly a low-income urban slum setting with additional environmental and nutritional challenges, which have been explored in the current analysis. Similar results were reported from another study conducted in rural India, where recruited infants showed a decline in scores in fine motor, receptive and expressive language skills and visual reception over a 6-month follow-up period [38].
The SES including household wealth affects linear growth [39, 40], development [2] and cognition [15, 41] in children. Childhood SES is shown to influence neural development particularly of the language and executive function areas in the brain [42]. Better SES can result in improved nutrition, better sanitation, less infection and enhanced interactive experiences, all of which can improve child development especially language as shown in the current study.
Stunting had independent associations with cognition and motor scores despite correction with SES, HOME status and micronutrient levels in the current study. Stunting has been shown to affect development and cognition; thus integrated interventions have been suggested to improve both linear growth and child development [2, 5]. A rural study conducted in Telangana, South India found similar results where height z-scores at enrollment had a positive association with all child development domains, the effect of which was attenuated by a nurturing home environment [38].
In the current study, positive home environmental factors influenced developmental gains over time, despite correction with SES. Home environment can mediate the relationship between SES and early childhood development [15, 16] as well as between linear growth and development [38]. The two components of home environment – physical and relational modulate each other with the sensitive and interactive relational factor overriding the negative effects of a sub-optimal home milieu. The effect of physical overcrowding on childhood development and cognition has been mediated by maternal responsiveness [43].
Maternal factors can influence child development, cognition and behaviour directly and indirectly through better home environmental factors. The effect of maternal depression can even start in the ante-natal period not only through altered placental functions, but also through foetal epigenetic changes and stress reactivity [14]. Maternal cognition also affects childhood development and cognition through genetic factors as well as enriching interacting experiences [8, 15, 43].
Iron deficiency in early childhood, when there is increased iron demand to optimize neuronal maturation, neurotransmitter synthesis, mitochondrial function and other iron dependent enzymes [44], can be detrimental as shown in the current analysis. This early childhood iron deficiency can have persisting effects on cognition in later life as evidenced by another analysis on the same cohort showing cumulative iron deficiency negatively impacting verbal, performance and processing speed domains of cognition at five years of age [34]. That assessment had shown that more than 40% of children had iron deficiency at 15 and 24 months of age. Later childhood iron supplementation improves body iron stores, but not the persisting effect of early onset iron deficiency on cognition [44, 45].
There are limitations for the current study including a comparatively small sample size. Though BSID-III is a sensitive and adapted measure, there can be variations (within normal limits) in developmental achievements. However the low-income urban slum setting of the cohort, its high early childhood iron deficiency and a significant decline in developmental quotients over time warrant an assessment as in the current study. Strengths of the study include good follow-ups of a longitudinal birth cohort, strong data granularity during early childhood period, standardized developmental, home environment and SES assessments, blood results from a national reference laboratory and rigorous quality control measures internally and externally.