Challenges and limitations in early intervention


Dr Mijna Hadders-Algra, University Medical Center Groningen, Developmental Neurology, Hanzeplein 1, 9713 GZ Groningen, the Netherlands. E-mail:


Research over the past three decades has shown that early intervention in infants biologically at risk of developmental disorders, irrespective of the presence of a brain lesion, is associated with improved cognitive development in early childhood without affecting motor development. However, at present it is unknown whether early intervention is also able to improve developmental outcome in infants with a serious lesion of the brain. This paper discusses factors that might play a role in the effect of early intervention. The following picture emerged from the limited evidence available: (1) coaching of parents seems an effective means of intervention; (2) our understanding of the plasticity of the developing human brain is currently too limited to allow a direct practical implementation in early intervention; (3) intervention before term age should primarily focus on stress reduction, intervention after term age on stimulation of infant development; and (4) our knowledge of the best ways to stimulate infant development is scant. Nevertheless, preliminary data suggest that offering the infant ample opportunities to explore by self-produced motor activities the borders of their own abilities might be a good strategy for promoting developmental outcome, including functional mobility.

Infants might be at risk of developmental disorders for a variety of reasons. The risk might consist of social or biological disadvantage or a combination of both.1 Well-known examples of biological disadvantage are prematurity or low birthweight, whether accompanied by a serious lesion of the brain or not.2 The usual care of infants at risk of developmental disorders includes early intervention. This comprises ‘the multidisciplinary services provided to children from birth to 5 years of age to promote child health and well-being, enhance emerging competencies, minimize developmental delays, remediate existing or emerging disabilities, prevent functional deterioration and promote adaptive parenting and overall family functioning’.1 Today a wide variety of early intervention programmes exists.3–6 The programmes have been shown to be effective in promoting developmental outcome, in particular cognitive outcome, in infants socially at risk of disadvantaged development.7 Gradually, evidence is also emerging of the effect of early intervention in infants biologically at risk of developmental disorders, i.e. those with prenatal, perinatal, and neonatal complications. The effect of early intervention in these groups has been studied most often in preterm infants. Recent systematic reviews, including meta-analyses, on the effect of early intervention in infants born preterm indicate that early intervention is associated with improved cognitive development during infancy and preschool age.5,6 Evidence that the effect persists into school age or later is lacking. In addition, the reviews indicate that early intervention in general has little effect on motor development. Interestingly, the beneficial effect of early intervention on cognitive development appears to be relatively independent of the specifics of the intervention programme.5 However, it should be realized that few studies investigated relationships between the contents of the intervention programmes and developmental outcome.

The aim of the present paper is to discuss factors that might play a role in the effect of early intervention in infants biologically at risk of developmental disorders. The following issues will be addressed: (1) the role of parent–infant interaction; (2) the type and timing of an early lesion of the brain; (3) the age at which the intervention is applied; and (4) the contents of intervention.

Parent–Infant Interaction

Parents are the key persons in a child’s life. Awareness of the impact of parents, caregivers, and the family on child development has resulted in the recognition of the importance of family-centred intervention services in health care for infants with or at risk of neurodisability.8,9 However, the role of the family, parents, and caregivers in the various forms of family-centred service varies substantially. The interventions differ for instance in the extent to which family autonomy is respected, in the degree of parental instruction or parental coaching, and in the attention paid to parent–infant interaction.10 The effect of most components of family-centred service on the child’s developmental outcome is unknown. An exception to this rule is the effect of parent–infant interaction on developmental outcome. Various studies have indicated that greater dyadic tuning between parent and infant, and more responsive, positive, warm, and sensitive parenting, is associated with better developmental outcome at preschool and school age.11,12 However, it is important to realize that association does not automatically imply causation. The association between parental behaviour during interaction with the infant and developmental outcome might in part be mediated by the characteristics of the infant. For instance, infants with an early lesion of the brain might show more stereotyped and less adaptive behaviour than typically developing infants.13 This, in turn, might induce less positive parental behaviour and impaired dyadic interaction between parent and infant. Nevertheless, the recent randomized controlled study of Koldewijn et al.14 suggests that parental behaviour also has an independent effect on child outcome. The study indicated that the preterm infants who had received intervention that paid specific attention to the infants’ self-regulation and sensitive parent–infant interactions had a significantly better motor outcome at the age of 2 years than the comparison group of infants. In contrast to most other studies, the intervention was not associated with improved cognitive outcome.

In the Groningen Vroege Interventie Project (VIP project), a study on intervention from 3 to 6 months corrected age in infants at high biological risk of developmental disorders, we recently evaluated the contribution of specific elements of intervention to developmental outcome. The results suggested in particular that coaching of parents was associated with improved developmental outcome at the age of 18 months.15,16 Coaching is defined as professional guidance aiming to empower caregivers so that they can make their own decisions during daily care activities. This implies that coaching differs largely from instruction.17

Type and Timing of the Lesion of the Developing Brain

Over the years, animal data have demonstrated that the effect of a lesion of the developing brain depends on the point in time at which the lesion occurs. Originally, it was thought that ‘the younger the age at insult, the better the outcome’ (the so-called Kennard-principle).18 However, gradually it became clear that the consequences of a lesion of the developing brain depend on developmental stage at insult, the site and the size of the lesion, animal species, exposure to chemical substances before and after the insult, and environmentally induced experience. It transpired that each stage, each neural system, and each species has specific vulnerabilities and resources of resilience to cope with the effects of an early lesion.19

Our knowledge on how this information translates to the human situation is limited. Recent advances in brain imaging and neurophysiological techniques have furnished some insights. It became clear that insults occurring during (the period equivalent to) the early third trimester of pregnancy usually affect the periventricular areas, including the periventricular white matter; those occurring near term more often affect the cortical grey matter.20 Staudt21 demonstrated that the organizational processes in response to an early lesion not only vary with the timing of the lesion but also with the neural system. For instance, in case of a unilateral lesion of the brain, the chance of motor recovery is higher for early third-trimester lesions than for lesions occurring near term. This is because, at early age, recovery might be mediated by persisting ipsilateral corticospinal projections that compensate for the lost ones on the side of the lesion. For the sensory systems the effect of an early third-trimester lesion is different. At that age the ascending thalamo-cortical somatosensory projections have not yet reached the cortex, allowing the system to use local deviations that bypass the lesion in order to reach the cortex.

Little is known, however, about the way in which we might facilitate the processes that mediate functional recovery in the developing human brain. Our knowledge is most advanced in unilateral spastic cerebral palsy (CP). Animal data pointed to the importance of balanced activity in both hemispheres and to the consequences of lesion-induced unbalanced activity during early development.22 The animal data, recent findings on the pathophysiology,23 and the effect of intervention in unilateral spastic CP (constraint-induced movement therapy or stimulation of bimanual activity)24 suggest that bilateral motor activity at early age is an important factor in functional recovery after an unilateral lesion of the brain.

Age at Which the Intervention is Applied

The human nervous system changes and develops throughout life.25 But the way in which it changes varies with age. During prenatal life, focus is on neuron and glial cell proliferation, cell migration, neural apoptosis, and axon and dendrite formation. During infancy, dendrite formation, synapse production, and myelination are most prominent. In later childhood, the major developmental processes consist of synaptic reorganization and myelination. This implies that the nervous system has age-specific forms of plasticity. For specific lesions and functions this plasticity is associated with critical windows during which intervention might have an effect that no longer might be obtained once the window is past. Critical periods have been demonstrated in the treatment of amblyopia and for the effect of cochlear implants on cortical processing of auditory information and speech development.26,27 Interestingly, the critical periods of the two sensory functions share the following characteristics: (1) the earlier the intervention is started, the larger is the functional effect; (2) the critical period of substantial functional effect ends around the age of 7 years; and (3) also beyond the critical period, intervention might still induce functional changes.26,27 It has been hypothesized that in the development of unilateral spastic CP a similar critical period might exist,23 but further research is required to corroborate this suggestion. Extrapolation of some animal data that implied recovery of function after an early lesion of the brain is best during the period of dendritic outgrowth and formation, suggests that the period ranging from the third trimester of pregnancy to the postnatal age of about 1.5 years offers the best opportunities for effective intervention.25,28

Before term age the effect of intervention might also be affected by stress. The stress is related to the difficulties of the preterm infant to adapt to the extrauterine situation owing to immaturity of vital physiological functions.29 The stress might enhance the capacity to cope with extrauterine life and thus promote survival. But stress before term age might also induce ‘developmental programming’.30 Animal data indicate that stress during early life gives rise to changes in serotonergic and noradrenergic activity in the cerebral cortex and alterations in dopaminergic activity in the striatum and prefrontal cortex.31–33 These changes have been associated with impaired development of the maps of body representation in the primary somatosensory cortex, inappropriately developed ocular dominance columns in the visual cortex, and mild motor problems.13 Gradually, evidence is accumulating that stress during early human ontogeny is also associated with long-term modifications of neurobehavioural development.32,34 In addition, animal research has indicated that the hormonal changes induced by stress might modify the effect of a lesion of the developing brain and the effect of early intervention.19 The sensitivity of the young nervous system to stress-related factors implies that reduction of stress is an important goal of early intervention before term age.

Contents of Early Intervention

Recent reviews3–6 have brought the limitations of our understanding on the effectiveness of specific components of early intervention clearly to light. Presumably the multifactorial composition of early intervention in children with or at risk of neurodisability is responsible for the lack of understanding. It is acknowledged that the effect of intervention might not only depend on the nature of the lesion of the brain and on family conditions, but also on the interaction of the intervention with the various personal and environmental factors.35

Still, the body of knowledge on early intervention steadily grows. It suggests that the nature of the most effective intervention before term differs from that applied after term age.3,4 Before term age, stress reduction appears to be essential. The Newborn Individualized Developmental Care and Assessment Program8 is a well-known example of a standardized intervention programme in which stress reduction is a major goal. This programme has been associated with a beneficial effect on developmental outcome in infancy, but the effects on outcome beyond infancy are controversial.3,4,36,37 The study of Guzzetta et al.38,39 suggests that, in addition to stress reduction, the application of infant massage during the preterm period might have a beneficial effect on development. The positive effect was found in low-risk infants. Further studies are required to investigate whether massage will have a similar positive effect on development in infants with a lesion of the brain. Animal data suggest that this might be the case.19

Little is known about the effectiveness of specific components of intervention applied after term age. A large diversity of general developmental programmes has been associated with better developmental outcome, in particular with improved cognition. The interventions presumably consist of a mix of effective, neutral, and counterproductive elements. Future research should aim at unravelling these components. The Groningen VIP project was a first step in this direction. In the VIP project the contents of intervention were systematically quantified and correlated to developmental outcome immediately after the end of intervention (at the corrected age of 6mo) and 1 year later. The data revealed (1) coaching of parents and (2) challenging the infant with a wide variety of self-produced motor activities were associated with better functional outcome at 18 months, in particular with improved functional mobility, (3) the proportion of time of the intervention sessions that was spent with handling techniques (hands-on) showed a negative association with functional outcome at 18 months, and (4) application of sensory and passive experiences was associated with a positive effect on cognition immediately after the end of the intervention, but these components of intervention showed a negative association with outcome at 18 months.15,16 The finding that self-produced activity in a variety of conditions is associated with better outcome is in line with the developmental principles of the neuronal group selection theory: the afferent information associated with self-produced trial-and-error activity guides the process of selection of the best strategy for each situation.13 The data suggest that coaching of parents to solve their own problems and to integrate motor activities in which the infant is allowed to explore the borders of its own abilities into daily routines might be a simple and effective means of promoting infant development.


Research over the past three decades has shown that early intervention in infants biologically at risk of developmental disorders is associated with improved cognitive development in early childhood. The effects have been shown for groups of infants with varying degrees of biological risk of whom only a minority had a serious lesion of the brain. This means that it is currently unknown whether early intervention is able to improve outcome in infants with a serious lesion of the brain. The data of the VIP project suggest, however, that this might be possible: the previously mentioned associations between components of intervention, such as parental coaching and promotion of the exploration of a variety of self-produced motor activities, and functional outcome were found in particular in the small subgroup of children who developed CP.15,16 Additional studies on the effect of early intervention in infants with a lesion of the brain are urgently needed. These studies also should pay attention to the contents of intervention. The LEARN 2 MOVE 0 to 2 years project is a step in the desired direction.40


I gratefully acknowledge Tineke Dirks, Ilse Ebbers-Dekkers, Elisa Hamer, and Tjitske Hielkema, for comments on a previous draft of the manuscript. The early intervention work of Mijna Hadders-Algra is financially supported by the Johanna KinderFonds, Stichting Fonds de Gavere, the Cornelia Stichting, ZonMW, Stichting Rotterdams Kinderrevalidatie Fonds Adriaanstichting, Phelpsstichting, Revalidatiefonds, Revalidatie Nederland, the Nederlandse Vereniging van Revalidatieartsen, and the Graduate School for Behavioural and Cognitive Neurosciences.