ORIGINAL ARTICLE

 

Evidence of training influence on infant manual behavior: a systematic review

 

 

Andressa Lagoa Nascimento; Natália Matos Tedesco; Daniele de Almeida Soares-Marangoni

Programa de Pós-Graduação em Saúde e Desenvolvimento na Região Centro-Oeste, Universidade Federal de Mato Grosso do Sul - Brasil

Correspondence

 

 

---

ABSTRACT

INTRODUCTION: Researchers have widely investigated how interventions by means of training can improve manual behaviors in infants. However, no systematic review has been found on this topic.
OBJECTIVE: To analyze the quality of scientific evidence considering the methodological quality and level of evidence by type of study in research on training of object-directed manual behaviors in infants in the first 18 months of life.
METHODS: National Library of Medicine (PubMed/MEDLINE), Latin American and Caribbean Health Sciences (LILACS), Virtual Health Library (BIREME/BVS), Science Direct, SciELO, and Physiotherapy Evidence Database (PEDro) databases were used. Only clinical trials that assessed the benefits of manual object-directed training in infants and were published up to February 2018, in English, were included. The Cochrane Collaboration Model was adapted to extract bibliographical data from the articles and their methodological quality was assessed using the PEDro scale and the Oxford Centre for Evidence-Based Medicine's Levels of Evidence.
RESULTS: Twenty one clinical trials were included. Studies investigated typically developing full-term infants, preterm infants, and infants at risk for autism spectrum disorders. Trainings were administered to infants by means of "sticky mittens" paradigm, task-specific practice, or contingency reinforcement. Most of the studies presented fair or poor methodological quality. Only studies that used task-specific active practice presented high methodological quality.
CONCLUSIONS: The results indicate there is high quality evidence that task-specific training improves object-directed manual behaviors in typically developing infants and preterm infants in the first 2-4 months of life. Studies addressing infants with established diagnoses of developmental dysfunction are lacking.

Keywords: child development, infant, early intervention, motor skills.

---

RESUMO

INTRODUÇÃO: Pesquisadores tem amplamente investigado como o treino pode melhorar comportamentos manuais em lactentes. No entanto, nenhuma revisão sistemática foi encontrada sobre este tópico.
OBJETIVO: Analisar a qualidade da evidência científica considerando a qualidade metodológica e o nível de evidência por tipo de estudo em pesquisas sobre treino de comportamentos manuais direcionados a objetos em lactentes nos primeiros 18 meses de vida
MÉTODO: Foram utilizadas as bases de dados da Biblioteca Nacional de Medicina (PubMed / MEDLINE), Literatura Latino-Americana e do Caribe (LILACS), Biblioteca Virtual em Saúde (BIREME / BVS), Science Direct, SciELO e Physiotherapy Evidence Database (PEDro). Apenas ensaios clínicos que avaliaram os benefícios do treino manual orientado a objetos em lactentes e que foram publicados até fevereiro de 2018, em inglês, foram incluídos. O Modelo de Colaboração Cochrane foi adaptado para extrair dados bibliográficos dos artigos e sua qualidade metodológica foi avaliada pela escala PEDro e pelo Nível de Evidência Científica por Tipo de Estudo de Oxford.
RESULTADOS: Vinte e um ensaios clínicos foram incluídos. Os estudos investigaram lactentes nascidos a termo, prematuros e lactentes de risco para distúrbios do espectro autista. Os treinos foram administrados aos lactentes por meio das abordagens do paradigma de "luvas aderentes", prática específica da tarefa ou reforço de contingência. A maioria dos estudos apresentou qualidade metodológica razoável ou fraca. Apenas os estudos que utilizaram a prática ativa específica da tarefa apresentaram qualidade metodológica alta.
CONCLUSÃO: Há evidência de alta qualidade de que o treino específico da tarefa aprimora comportamentos manuais orientados a objetos em lactentes com desenvolvimento típico e lactentes prematuros nos primeiros 2-4 meses de vida. Estudos abordando lactentes com diagnósticos estabelecidos de disfunção do desenvolvimento não foram encontrados.

Palavras-chave: desenvolvimento infantil, lactente, intervenção precoce, destreza motora.

---

 

 

Authors summary

Why was this study done?

This is the first systematic review that addresses the methodological quality of clinical trials that investigated the effects of different trainings of early manual behaviors. This review can be used as a guide for decision-making on stimulation or sensorimotor intervention strategies focused on manual behaviors in the first months of life.

What did the researchers do and find?

Main methodological characteristics and findings of clinical trials that examined the effects of different types of manual behavior trainings in typical and at-risk infants up to 18 months of age were detailed. The quality of evidence from these studies was classified and discussed, and recommendations for clinical practice were finally provided. Studies that used active task-specific training presented high quality of evidence, while training using sticky mittens or contingent reinforcement approaches presented moderate or poor quality.

What do these findings mean?

Since outcomes from task-specific training are likely to present high internal validity and excellent level of evidence, this approach is best recommended for clinical practice if one aims to stimulate and improve reaching in typical and premature infants at the emergence of this skill.

 

INTRODUCTION

Infants are able to interact with objects since the first days of life¹. Gradually, by means of spontaneous practice over days and months, infants become skilled in reaching, grasping and exploring objects. These early object-directed manual behaviors are essential motor milestones as they allow infants to gather new information about the environment and learn further motor strategies from that^2-4.

Due to developmental importance of these manual behaviors, researchers have widely investigated how interventions by means of induced practice (training) could improve them in infants. Several training approaches have been addressed in literature, including the "sticky mittens" paradigm, task-specific training, or contingent reinforcement. In the "sticky mittens" paradigm, 3-month-old infants were seated on a parent's lap in front of a table, on which Velcro covered toys are placed. As infants wore Velcro covered mittens and were encouraged to reach for the toys, the toys adhered to the mittens as if the infants had grasped them^5-7. During task-specific trainings, pre-reaching infants were induced to repeat toy-oriented arm movements organized in steps of activities. Activities included haptic contact with the toy and assisted and active toy-oriented hand movements^8-10. In the contingent reinforcement approach, 3-month-old infants were exposed to moving/sounding toys or a mobile which could be activated contingently upon infants' upper limb actions¹¹.

These movement experiences have been used as induced trainings to advance or enhance the amount and duration of object-directed manual behaviors in typically developing infants and at-risk infants over 1-2 brief-term sessions or daily sessions. In general, authors have suggested these trainings provide infants with opportunities to improve upper limb control^8,9,11, select more efficient movement patterns^6,8,9,11, become motivated and engaged^10,12, and learn to couple object perception and movement action^6,10. These trainings, therefore, can be considered interesting tools to potentiate early manual development in typical infants and to prevent or minimize manual impairment in infants at risk for developmental disorders during early intervention practice.

Despite the relevance of this topic, only one article has characterized and discussed studies related to training for upper extremity behaviors in infancy¹³. No systematic review has been found on this topic. Therefore, this article systematically reviewed studies that used training of object-directed manual behaviors to advance their acquisition or improve their development in infants over the first year and half of life. Its purpose was to identify the studies methodological characteristics and analyze their quality of evidence considering methodological quality and level of evidence by type of study.

 

METHODS

Definition of the research questions

The following questions guided this review:

1) Which methodological aspects the studies have been using to examine the influence of trainings of object-directed manual behaviors in infants?

2) What have these studies found?

3) What is the quality of evidence on the effects of those trainings?

Inclusion criteria

Clinical trials that assessed the efficacy of manual object-directed training were included. The World Health Organization¹⁴ defines clinical trial as a research study that assess the effects of health-related interventions using human participants. For this review, non-randomized, randomized or quasi-randomized clinical trials with at least one control group were included. At least one object-directed manual behavior (reaching, grasping, or object exploration) should be an outcome variable. The target population was infants from 0 to 18 months of age with typical development, at risk or with established diagnosis of developmental disorders. The studies should have been published up to February 2018, in English. All of the criteria needed to be met for inclusion in this review.

Exclusion criteria

Studies were excluded if/in case of: 1) at least one manual behavior was not an outcome variable; 2) no manual behaviors were trained; 3) animal studies; 4) project reports, review articles or case studies.

Search strategy

The articles were selected from the following electronic databases: National Library of Medicine (PubMed/MEDLINE), Latin American and Caribbean Health Sciences (LILACS), Virtual Health Library (BIREME/BVS), Science Direct, SciELO, and Physiotherapy Evidence Database (PEDro). The search employed keywords and Pubmed medical subject headings (MeSHs) terms, combined by the Boolean operators AND and OR: reaching, grasping, exploration, training, intervention, practice, learning, motor skills, infant, movement, hands, movement training. In addition, the reference list of the selected studies were used to retrieve potential articles. Each reviewer conducted the search several times to avoid losing articles and the final search was on March 2018.

Selection and data extraction

Two independent reviewers worked on the searches and initial selection of the articles. The reviewers selected articles based on their titles, excluding duplicate articles and those that clearly were not related to the review subject. After reading the abstracts and the entire articles independently, the reviewers discussed together, reached a consensus and determined a final selection. A third reviewer was consulted when necessary. The Cochrane Collaboration Model¹⁵ was adapted to extract bibliographical data from

Infants are able to interact with objects since the first days of life¹. Gradually, by means of spontaneous practice over days and months, infants become skilled in reaching, grasping and exploring objects. These early object-directed manual behaviors are essential motor milestones as they allow infants to gather new information about the environment and learn further motor strategies from that^2-4.

Due to developmental importance of these manual behaviors, researchers have widely investigated how interventions by means of induced practice (training) could improve them in infants. Several training approaches have been addressed in literature, including the "sticky mittens" paradigm, task-specific training, or contingent reinforcement. In the "sticky mittens" paradigm, 3-month-old infants were seated on a parent's lap in front of a table, on which Velcro covered toys are placed. As infants wore Velcro covered mittens and were encouraged to reach for the toys, the toys adhered to the mittens as if the infants had grasped them^5-7. During task-specific trainings, pre-reaching infants were induced to repeat toy-oriented arm movements organized in steps of activities. Activities included haptic contact with the toy and assisted and active toy-oriented hand movements^8-10. In the contingent reinforcement approach, 3-month-old infants were exposed to moving/sounding toys or a mobile which could be activated contingently upon infants' upper limb actions¹¹.

These movement experiences have been used as induced trainings to advance or enhance the amount and duration of object-directed manual behaviors in typically developing infants and at-risk infants over 1-2 brief-term sessions or daily sessions. In general, authors have suggested these trainings provide infants with opportunities to improve upper limb control^8,9,11, select more efficient movement patterns^6,8,9,11, become motivated and engaged^10,12, and learn to couple object perception and movement action^6,10. These trainings, therefore, can be considered interesting tools to potentiate early manual development in typical infants and to prevent or minimize manual impairment in infants at risk for developmental disorders during early intervention practice.

Despite the relevance of this topic, only one article has characterized and discussed studies related to training for upper extremity behaviors in infancy¹³. No systematic review has been found on this topic. Therefore, this article systematically reviewed studies that used training of object-directed manual behaviors to advance their acquisition or improve their development in infants over the first year and half of life. Its purpose was to identify the studies methodological characteristics and analyze their quality of evidence considering methodological quality and level of evidence by type of study.

 

METHODS

Definition of the research questions

The following questions guided this review:

1) Which methodological aspects the studies have been using to examine the influence of trainings of object-directed manual behaviors in infants?

2) What have these studies found?

3) What is the quality of evidence on the effects of those trainings?

Inclusion criteria

Clinical trials that assessed the efficacy of manual object-directed training were included. The World Health Organization¹⁴ defines clinical trial as a research study that assess the effects of health-related interventions using human participants. For this review, non-randomized, randomized or quasi-randomized clinical trials with at least one control group were included. At least one object-directed manual behavior (reaching, grasping, or object exploration) should be an outcome variable. The target population was infants from 0 to 18 months of age with typical development, at risk or with established diagnosis of developmental disorders. The studies should have been published up to February 2018, in English. All of the criteria needed to be met for inclusion in this review.

Exclusion criteria

Studies were excluded if/in case of: 1) at least one manual behavior was not an outcome variable; 2) no manual behaviors were trained; 3) animal studies; 4) project reports, review articles or case studies.

Search strategy

The articles were selected from the following electronic databases: National Library of Medicine (PubMed/MEDLINE), Latin American and Caribbean Health Sciences (LILACS), Virtual Health Library (BIREME/BVS), Science Direct, SciELO, and Physiotherapy Evidence Database (PEDro). The search employed keywords and Pubmed medical subject headings (MeSHs) terms, combined by the Boolean operators AND and OR: reaching, grasping, exploration, training, intervention, practice, learning, motor skills, infant, movement, hands, movement training. In addition, the reference list of the selected studies were used to retrieve potential articles. Each reviewer conducted the search several times to avoid losing articles and the final search was on March 2018.

Selection and data extraction

Two independent reviewers worked on the searches and initial selection of the articles. The reviewers selected articles based on their titles, excluding duplicate articles and those that clearly were not related to the review subject. After reading the abstracts and the entire articles independently, the reviewers discussed together, reached a consensus and determined a final selection. A third reviewer was consulted when necessary. The Cochrane Collaboration Model¹⁵ was adapted to extract bibliographical data from the included articles. Data consisted of methodological characteristics of the studies and the results of training.

Assessment of the quality of evidence

The methodological quality of the studies was assessed using the PEDro scale^16-18. The scale consists of a checklist of scored dichotomous (yes or no) questions/criteria that predict bias. It guides users to trials that are more likely to be valid (internal validity) and to contain sufficient information to base clinical practice. The following criterion comprise the scale: 1) Subjects were randomly allocated to groups (quasi-randomized allocation does not satisfy this criterion), 2) Allocation was concealed; 3) Groups were similar at baseline regarding the most important prognostic indicators; 4) There was blinding of all subjects; 5) There was blinding of all persons who administered the intervention; 6) There was blinding of all assessor who measured at least one key outcome; 7) Measures of at least one key outcome were obtained from more than 85% of the subjects initially allocated to groups; 8) All subjects for whom outcome measures were available received the treatment or control condition as allocated or, where this was not the case, data for at least one key outcome was analyzed by "intention to treat", 9) The results of between-group statistical comparisons are reported for at least one key outcome; 10) The study provides both point measures and measures of variability for at least one key outcome. Each criterion is awarded with 1 (yes) or 0 (no). Thus, the PEDro score is out of 10. Points are only awarded when a criterion is clearly satisfied. The methodological quality of each study was classified as "high" (6-10 points), "fair" (4-5 points), or "poor" (below 4 points). Three examiners (authors) applied the scale together and reached 100% agreement regarding the studies scoring in each criterion.

In addition, the studies were analyzed using the Oxford Centre for Evidence-Based Medicine's (CEBM) Levels of Evidence¹⁹. The tool was designed so that it can be used as short-cut for clinicians, researchers and patients with limited time. It consists of a hierarchy of the likely best evidence that considers mainly the type of the study methodological design (eg.: cohort study, systematic review, randomized trial, etc.). The studies can be classified within a hierarchy of levels of evidence (1a, 1b, 1c, 2a, 2b, 2c, 3a, 3b, 4, 5) and grades of recommendation (A, B, C, D). Level 1 studies, the best level of evidence, present grade of recommendation A; level 2 or 3 studies present grade of recommendation B; level 4 studies present grade of recommendation C; and level 5 studies present grade of recommendation D. Therefore, studies with evidence level 1 and grade of recommendation A present the best evidence. For this review, only the two possible levels of evidence for clinical trials (1b and 2b) were considered as only clinical trials were included. Individual randomized controlled trials with narrow Confidence Interval were classified as level 1b; low quality clinical trials (e.g., <80% follow-up, non-randomized) were classified as level 2b. In these cases, the grades of recommendation could be A (consistent level 1 studies) or B (consistent level 2 studies). Three examiners (authors) reached 100% agreement regarding the classifications.

 

RESULTS

The electronic search found 367 potential articles, all published in English. Initially, 82 articles were selected based on their titles and 25 studies were added from their references lists, thus totaling 107 articles. The abstracts were read and 85 articles were excluded for the following reasons: manual behaviors were not outcome variables (n=20), manual behaviors were not trained (n=29), infants' age over 18 months (n=4), animal studies (n=3), project reports, review articles or case studies (n=14); two or more of these criteria (n=15). Hence, 22 articles were read in full. Of these, 1 was excluded because the outcome measure was the infant's brain response during the training. After the final selection, 21 articles were included. The most recent study was published in February 2017. Seventeen studies were carried out in the United States and 6 in Brazil (Table 1).

Ten of the included studies could be characterized as randomized controlled trials. Sixteen studies investigated only typically developing full-term infants. Their sample sizes ranged from 16²⁰ to 90²¹. Other 5 studies included infants at risk for developmental disorders: three studies with preterm infants with less than 34 weeks of gestation and low birth weight, with sample size from 16 to 39^9,20,22; one study with 32 late preterm infants (34 a 36 6/7 weeks of gestation) with adequate birth weight¹⁰; and one study with 17 infants with familiar history (high risk) for autism spectrum disorders (ASD) mostly born full-term⁵. This latter study included 72 additional infants from previous research. Infants' age ranged from around 2 to 15 months (Table 1).

Table 2 shows data on studied manual behaviors and main methodological procedures. The studies addressed 3 types of early object-directed manual behaviors: reaching, grasping (holding), and object manual exploration. Although some studies did not label their dependent measures as "reaching", we considered them as reaches (or pre-reaches) because they were characterized as the movement of one or both hands towards a toy in an attempt to obtain it²³. Reaching was assessed in 17 studies as a dependent measure. Infants were mostly assessed and trained seated. The studies predominantly used similar toys for assessments and training. However, most of the studies did not provide complete information on the physical characteristics (color, texture, size, rigidity and type of material) of the toys used as stimulus (Table 2).

All 21 studies presented at least one assessment after training (post-training). However, only 5 studies^8,10,24-26 presented at least one retention measure (after the post-training) or weekly follow-up assessments during or after the training weeks. Most of the studies analyzed the infants' behaviors using video recording (Table 3).

In 12 studies, training was administered by the researcher; in 9 studies training was administered by parents/caregivers. As to training duration, the protocols were constituted of daily practice of 10-20 minutes over 2-8 weeks; single 3-10 minutes session of practice; or 3 sessions of 4-minute practice divided in two consecutive days. The protocols varied from daily practice throughout weeks to a single session. Velcro covered mittens (sticky mittens paradigm) were used to train infants in most of the studies (n = 11). The other studies trained infants by means of task-specific training in itself without mittens (n = 8) or contingent reinforcement (n = 2) (Table 3).

Table 3 shows the outcomes of the studies. In summary, increased reaching, grasping and exploratory activity were found after daily training over 2 weeks and single-session training of 3-10 minutes using the sticky mittens paradigm. Task-specific training resulted in improved reaching performance after daily sessions over 2-8 weeks and single session training of 4-5 minutes. Contingent reinforcement leaded to improvements in reaching after daily training over 2 weeks and single session training of 9 minutes.

Table 4 presents the methodological quality of the studies classified using the PEDro scale and the Oxford (CEBM) Levels of Evidence. Three studies^10,22,26 presented high methodological quality and level of evidence 1b (grade of recommendation A) (Table 4).

 

DISCUSSION

This systematic review examined articles that investigated the effects of training on object-directed manual behaviors in typically developing infants and at-risk infants in the first 18 months of life. All studies reported enhancement of object-directed manual behaviors as a result of active experience exploring Velcro covered objects with Velcro covered mittens, task-specific experience in itself, or contingent reinforcement. All studies with high methodological quality and best level of evidence used task-specific training.

Which methodological aspects the studies have been using to examine the influence of trainings of object-directed manual behaviors in infants?

Few studies included infants at risk for developmental disorders^5,9,10,20,22. They addressed infants born preterm or at risk for ASD. Studies with established disorders which potentially hamper early manual function, such as Down syndrome or cerebral palsy, are lacking. This is not surprising as learning mechanisms need to first be tested in typically developing infants before atypical samples. However, we highlight the essential importance of promoting further research on early interventions for manual behaviors in those population with established disorders.

Among the studied manual behaviors, reaching was the most assessed and trained one. Most of the studies started the training before or within a few days after reaching onset, thus minimizing the influence of spontaneous practice as a confounding factor that could bias the outcomes. On the other hand, some studies did not report on the infants' skill level^21-23,27, which makes it more difficult to conclude on how much the training improved reaching regardless of the influence of spontaneous practice. Future studies must be attentive to this issue.

Another important issue is that most of the studies did not report the size of the toys used as stimulus. In addition, two studies did not provide specific information or at least illustration on the main physical characteristics (color, texture, rigidity, type of material) of the toys^24,28. Such characteristics should be always clear as they may influence the outcomes of the studies. For example, malleable toys can produce more reaches than rigid ones, whilst big rigid toys can produce more bimanual than unimanual reaches²⁹. Considering studies that have found influence of the physical properties of objects on manual skills in infants^29,30, reporting a maximum of physical characteristics of the used toys in the studies are of essential relevance to allow methodological replications in other research and clinical practice.

As to assessments, only few studies included at least one retention test or follow-up assessments after the first post-training assessment^8,10,24,26. This hampers the conclusions that could be drawn on the duration of observed effects of training, since it is not possible to evidence whether initial changes in motor behavior were only immediate or more long lasting.

Training was predominantly administered by parents. Parents were trained, monitored and could consult with the researchers about questions regarding the execution of the training protocol. When the training is administered by the researcher, the chance of experimental error is minimized; however, the infant may not respond naturally to the intervention administered by an unfamiliar person. Training administration by parents may be more natural for infants and more convenient for researchers; however, in this case there is more risk of error during the procedure as parents could offer excessive stimuli to the infant, for example. Despite this point of view, it is valuable that studies investigate trainings that can be easily administered by parents and caregivers so that they can apply the intervention at home without difficulties. This is especially important because some at-risk infants do not qualify for professional assistance in early intervention programs due to their clinical similarities to health infants at birth^31,32.

What have these studies found?

All training protocols were based on the repetition of a set of hand movements towards toys offered on a table or at the infants' trunk midline. The protocols varied from daily practice throughout weeks (long-term)^{5-7,9,11,24, 26,28,33} to single session practice (short-term)^{8,10, 22-25,27,34,35}.

Velcro covered mittens (sticky mittens paradigm) were used during training in most of the studies^{5,7,10,21,24,25,36,37}. Infants wore Velcro covered mittens and were offered Velcro covered toys on a table. As infants moved their arm while wearing the mittens, the toy could easily adhere to the mittens, thus providing opportunities for infants to "grasp" the toys before acquiring this skill. In these studies, training was based on the active or passive (observation) repetition of toy-mitten sticking. Ten minutes of daily active training over 2 weeks benefited the number of reaches and grasps and favored the interest to explore objects and start contacting them in pre-reaching typically developing infants^6,24,33,36. Two to 10 minutes of the training in a single session^21,27,35 or 3 sessions²⁵ improved the infants' engagement contacting toys manually during the active training with the mittens. The studies also suggested that the observed effects in typically developing infants can last 2 to 12 months after training ceases^11,24. One sole study used the sticky mittens paradigm in at-risk infants. Daily 10 minutes of the active training over 2 weeks benefited grasping activity in infants at familial history (high risk) for ASD⁵. One study found no clear advantage of daily 10-minute training with open-fingers sticky mittens over 2 weeks in pre-reaching typically developing infants and concluded that providing opportunities for infants to repeat active hand-toy interaction is possibly enough to enhance reaching regardless of the toy-mitten sticking provision⁷.

In most of the other studies, training was based on the repetition of active hand-toy interaction in itself (task-specific experience), mostly in the supine position. In general, these trainings were organized in sequential steps of activities essentially constituted by presentation of the toy and the infant's hand at his/her visual field for a few seconds, assistance of the infant's hand movement to his/her midline to touch the toy, and offering the infant opportunities to reach actively for the toy over some minutes. Two weeks of this daily training for 20 minutes followed by active manual and oral exploration of toys with varied texture, sizes and rigidity advanced the emergence of reaching and increased the number of hand-toy contacts in 2 month-old typically developing infant³⁵. In addition, 3 weeks of 10-15 passive toy-oriented reaches followed by 5 minutes of active toy-oriented reaching enhanced reaching with more functional behaviors (with open hand and hand closer to the toy) in 2 month-old typically developing infants²⁶. In pre-reaching preterm infants born with less than 33 weeks of gestation and low birth weight (<2500g), daily training during 15-20 minute over 4 weeks leaded to more reaches than untrained controls after 4 weeks⁹.

Infants also received task-specific training in steps of activities during a single session. The activities were constituted by assisted and active hand-toy contact at the infant's midline, tactile stimulation of the infant's upper limb with the toy, followed by offering the infant opportunities to perform toy-oriented reaching actively for some minutes, in the reclined seated position. In typically developing infants, 4-5 minutes of the training in a single session resulted in more functional behaviors to reach^8,34,37. These effects were likely to be specific to body position when infants were trained in the reclined position; on the other hand, when infants were trained in supine, the effects were found in both supine and reclined positions³⁴. In newly reaching preterm infants born from 30 to 36 6/7 weeks of gestation, a similar training applied in a serial sequence of activates resulted in immediate increase in the number of reaches, mainly bimanual reaches¹⁰, and in the amount of motor strategies to gain success in reaching and pre-grasping attempts^10,20,22. The effects of these brief-term trainings, however, seem to be only temporary as they were not maintained in the following day after the session, at least in the preterm infants¹⁰. Hence, whilst task-specific adaptations in manual behavior start and can be observed in the initial minutes of motor experience¹⁰, long-term training is required for long lasting adaptations¹⁰.

The other approach used to train infants' manual behavior was contingent reinforcement, in which infants were induced to notice the consequences their upper limbs' actions have on nearby objects by means of a mobile¹² or moving/sounding toys¹¹. Active training moving a mobile with the arm during a single session of 9 minutes benefited the number of hand-toy contacts in pre-reaching typically developing infants¹². Three-month-old infants' hand-toy contacts were also benefited from a daily 10-minute training actively boosting a sounding and moving toy with the hand over 2 weeks¹¹.

Overall, the studies have suggested offering opportunities for infants to self-produce toy-oriented movements potentiates their ability to learn from the sensory properties of the objects and the biomechanical strategies than can be adopted by their upper limbs for manual action. As a result, infants seem to become more engaged and improve perception on objects and motor control to move the hands toward the toys to attempt to touch and explore them.

What is the quality of evidence on the effects of those trainings?

Considering the PEDro scale, the majority of the studies presented fair methodological quality^{7,8,11,12,21,24,25, 27, 34,37}. They especially failed on reporting or performing two or three key criteria required for avoiding biased outcomes: randomization, allocation concealment, and blinding of outcome assessors. Proper randomization in clinical trials ensures that subjects in groups are systematically equal, which avoids biased results due to the influence of imbalance of covariates³⁸. In addition, allocation concealment and blinding of outcome assessors ensure that researchers and participants do not know to which group subjects were assigned and therefore avoid selection and measurement biases that tend to overestimate treatment effects^38,39. As a general consequence, treatment groups may have been systematically different in those studies and their results may be biased. On the other hand, among those studies with fair methodological quality, seven studies^{8,9,11,24,27,28,35} presented level of evidence 1b, grade of recommendation A, which represents the clinical trials with best quality according to the Oxford CEBM Levels of Evidence. These differences between the tools classifications occur because the PEDro scale evaluates the reporting and performance of several methodological criteria that allow appraising the study internal validity, as mentioned earlier, whilst the Oxford CEBM Levels of Evidence was designed for rapid appraisals and considers the study design type as the main criteria to classify the likely best evidence. Therefore, considering both classifications is important to a more definitive judgment of the quality of evidence in clinical decision-making.

In this sense, we highlight the three randomized controlled trials that used task-specific active training and presented not only high methodological quality (PEDro scale) but also the best level of evidence (Oxford CEBM Levels of Evidence)^10,22,26. Their results are likely to be valid for clinical practice. In these studies, which presented background from Physical Therapy, training was administered to preterm infants in a single 4-5 minute session^10,22 or to typically developing infants in a daily 10-minute basis over 2 weeks²⁶. This latter protocol could be recommended by clinicians when the intention is to use stimulating play in daily life to improve reaching and object exploration in typically developing infants as health promotion strategy. When the intention is to benefit reaching behavior in preterm infants, clinicians could use the brief-term training protocol as part of a session of early intervention practice.

The studies with fair or poor methodological quality^{5-9,11,20,24,27,25-33,35} and/or level of evidence "2b", degree of recommendation "B",^5-7,20-33, addressed the three approaches discussed earlier - sticky mittens paradigm, task-specific training, and contingent reinforcement -, but most of them used the sticky mittens paradigm. It calls attention that none of the studies using the sticky mittens paradigm scored highly in the methodological quality. Considering their background from Psychology, it is possible that these studies were not directly intended to guide intervention for clinical practice and therefore researchers may not have engaged in a full clinical trial standard. They were predominantly non randomized controlled trials. Despite of this, it is valid to observe the high degree of agreement between these studies outcomes, with almost all showing some benefit of the training for manual behaviors. This may suggest some consistence in the effectiveness of their manipulation.

Overall, this systematic review shows there is strong evidence for the effectiveness of training for object-direct manual behaviors, specially reaching. It is important to caution readers that the PEDro scale and the Oxford CEBM Levels of Evidence should not be used as a measure of the validity of the study conclusions as they do not judge, for instance, whether the study results are clinically important. However, this review suggests that future studies report more detailed and clear information of their methods and try to follow methodological standard criterion known for reducing bias.

Recommendations for clinical practice and future research

The outcomes of the studies with high methodological quality are likely to present high internal validity and excellent level of evidence to guide clinical practice. Therefore, considering their main methodological settings (infant, therapist and toy position; type of activities), intervention to stimulate and improve early manual behaviors should be constituted by these basic points:

(1) infants placed supine, reclined on the therapist's thighs (both face-to-face), or seated in a baby seat slightly reclined; (2) toy offered in the infant's midline at an arm-length distance; (3) repetition of assisted and active hand and arm movements towards the toy; (4) tactile stimulation of infants' upper limbs; (5) active upper limb movements towards the toy.

The expected effects may be: a) more functional hand shape and position to act upon small toys, and b) increased number and time of hand-toy exploration. Such effects can be temporary (intra-session) or more lasting and seem to result from improvement of motor control, selection of more efficient movement patterns, intrinsic motivation and better perceptual and motor coupling.

We stress the lack of studies addressing infants with established diagnoses of developmental dysfunction, such as Down syndrome and cerebral palsy. Considering these conditions affect early manual behavior and have impaired a growing population of children worldwide^40,41 addressing them in future research on this topic will be of valuable importance for clinical practice. In addition, as the majority of the studies did not use or report basic methodological criteria known to reduce biased results (eg.: randomization, allocation concealment), we recommend researchers engage in high quality standards for clinical trials in future studies. To allow proper replication, it is also important to provide detailed information on the objects used as stimuli during training. Moreover, as few studies performed follow-up tests, future studies could include at least one additional assessment after post-training.

 

CONCLUSION

There is high quality evidence for the benefits of task-specific training for improving manual behaviors in typically developing infants and preterm infants in the first 2-4 months of life. Additional randomized controlled trials that use high methodological standards to minimize bias may help to endorse or refute the studies findings.

 

Acknowledgment

Study supported by the Foundation to Support the Development of Education, Science and Technology of State of Mato Grosso do Sul (FUNDECT) in partnership with the National Counsel of Technological and Scientific Development (CNPq), Brazil (protocols: 59/300.051/2015; 59/300.347/2015).

 

REFERENCES

1.Rönnqvist L, von Hofsten C. Neonatal finger and arm movements as determined by a social and an object context. Infant Child Dev. 1994;3(2):81-94. DOI: https://doi.org/10.1002/edp.2430030205        [ Links ]

2.Connolly K, Dalgleish M. The emergence of a tool-using skill in infancy. Dev Psychol. 1989;25(6):894-912. DOI: http://dx.doi.org/10.1037/0012-1649.25.6.894        [ Links ]

3.Corbetta D, Thelen E, Johnson K. Motor constraints on the development of perception-action matching in infant reaching. Infant Behav Dev. 2000;23(3):351-74. DOI: https://doi.org/10.1016/S0163-6383(01)00049-2        [ Links ]

4.Gibson JJ. The ecological approach to visual perception. London: Lawrence Erlbaum Associates, 1986.         [ Links ]

5.Libertus K, Landa RJ. Scaffolded reaching experiences encourage grasping activity in infants at high risk for autism. Front Psychol. 2014;5:1071. DOI: https://doi.org/10.3389/fpsyg.2014.01071        [ Links ]

6.Needham A, Barrett T, Peterman K. A pickme-up for infants' exploratory skills: early simulated experiences reaching for objects using "sticky mittens" enhances young infants' object exploration skills. Infant Behav Dev. 2002;25(3):279-95. DOI: https://doi.org/10.1016/S0163-6383(02)00097-8        [ Links ]

7.Williams JL, Corbetta D, Guan Y. Learning to reach with "sticky" or "non-sticky" mittens: A tale of developmental trajectories. Infant Behav Dev. 2015;38:82-96. DOI: https://doi.org/10.1016/j.infbeh.2015.01.001        [ Links ]

8.Cunha AB, Lobo MA, Kokkoni E, Galloway JC, Tudella E. Effect of short-term training on reaching behavior in infants: a randomized controlled clinical trial. J Mot Behav. 2015;48(2):132-42. DOI: https://doi.org/10.1080/00222895.2015.1050549        [ Links ]

9.Heathcock JC, Lobo M, Galloway JC. Movement training advances the emergence of reaching in infants born at less than 33 weeks of gestational age. Phys Ther. 2008;88(3):310-22. DOI: https://doi.org/10.2522/ptj.20070145        [ Links ]

10.Soares DA, van der Kamp J, Savelsbergh GJ, Tudella E. The effect of a short bout of practice on reaching behavior in late preterm infants at the onset of reaching: a randomized controlled trial. Res Dev Disabil. 2013;34(12):4546-58. DOI: https://doi.org/10.1016/j.ridd.2013.09.028        [ Links ]

11.Willians JL, Corbetta D. Assessing the impact of movement consequences on the development of early reaching in infancy. Front Psychol. 2016;7:587. DOI: https://doi.org/10.3389/fpsyg.2016.00587        [ Links ]

12.Needham A, Joh AS, Wiesen SE, Willians N. Effects of Contingent Reinforcement of Actions on Infants' Object-Directed Reaching. Infancy. 2014;19(5):496-517. DOI: https://doi.org/10.1111/infa.12058        [ Links ]

13.Lobo MA, Galloway JC, Heathcock JC. Characterization and intervention for upper extremity exploration & reaching behaviors in infancy. J Hand Ther. 2015;28(2):114-25. DOI: https://doi.org/10.1016/j.jht.2014.12.003        [ Links ]

14.World Health Organization (WHO). International Clinical Trials Registry Platform [cited 2018 oct 28] Available from: https://www.who.int/ictrp/en/        [ Links ]

15.Higgins JPT, Green S. Cochrane handbook for systematic reviews of interventions. Cochrane Collaboration. Version 5(0). 2011.         [ Links ]

16.Maher CG, Sherrington C, Herbert RD, Moseley AM, Elkins M. Reliability of the PEDro scale for rating quality of randomized controlled trials. Phys Ther. 2003;83(8):713-21. DOI: https://doi.org/10.1093/ptj/83.8.713        [ Links ]

17.Morton NA. The PEDro scale is a valid measure of the methodological quality of clinical trials: a demographic study. Aust J Physiother. 2009;55(2):129-33. DOI: https://doi.org/10.1016/S0004-9514(09)70043-1        [ Links ]

18.Sherrington C, Herbert RD, Maher CG, Moseley AM. PEDro. A database of randomized trials and systematic reviews in physiotherapy. Man Ther. 2000;5(4):223-6. DOI: https://doi.org/10.1054/math.2000.0372        [ Links ]

19.Oxford Centre for Evidence-Based Medicine. Levels of evidence [cited 2019 jan 23] Available from: http://www.cebm.net/oxfordcentre-evidence-based-medicine-levels-evidencemarch-2009/        [ Links ]

20.Guimarães EL, Cunha AB, Mira DM, Tudella E. Influence of short-term training on the distal adjustments of reaching in preterm infants. J Hum Growth Dev. 2015;25(3):263-70. DOI: http://dx.doi.org/10.7322/jhgd.89621        [ Links ]

21.21. Gerson SA, Woodward AL. Learning from their own actions: The unique effect of producing actions on infants' action understanding. Child Dev. 2014;85(1):264-77. DOI: https://doi.org/10.1111/cdev.12115        [ Links ]

22.Guimarães EL, Tudella E. Immediate effect of training at the onset of reaching in preterm infants: randomized clinical trial. J Mot Behav. 2015;47(6):535-549. DOI: https://doi.org/10.1080/00222895.2015.1022247        [ Links ]

23.Carvalho RP, Tudella E, Savelsbergh GJP. Spatio-temporal parameters in infant's reaching movements are influenced by body orientation. Infant Behav Dev. 2007;30(1):26-35. DOI: https://doi.org/10.1016/j.infbeh.2006.07.006        [ Links ]

24.Wiesen SE, Watkins RM, Needham AW. Active Motor Training Has Long-term Effects on Infants' Object Exploration. Front Psychol. 2016;7:599. DOI: https://doi.org/10.3389/fpsyg.2016.00599        [ Links ]

25.Libertus K, Joh AS, Needham AW. Motor training at 3 months affects object exploration 12 months later. Dev Sci. 2016;19(6):1058-66. DOI: https://doi.org/10.1111/desc.12370        [ Links ]

26.Lobo MA, Galloway JC. Postural and object-oriented experiences advance early reaching, object exploration, and means-end behavior. Child Dev. 2008;79(6):1869-90. DOI: https://doi.org/10.1111/j.1467-8624.2008.01231.x        [ Links ]

27.Needham AW, Wiesen SE, Hejazi JN, Libertus K, Christopher C. Characteristics of brief sticky mittens training that lead to increases in object exploration. J Exp Child Psychol. 2017;164:209-24. DOI: https://doi.org/10.1016/j.jecp.2017.04.009        [ Links ]

28.Lobo MA, Galloway JC, Savelsbergh GJP. General and task-related experiences affect early object interaction. Child Dev. 2004;75(4):1268-81. DOI: https://doi.org/10.1111/j.1467-8624.2004.00738.x        [ Links ]

29.Campos AC, Costa CS, Savelsbergh GJ, Rocha NA. Infants with Down syndrome and their interactions with objects: development of exploratory actions after reaching onset. Res Dev Disabil. 2013;34(6):1906-16. DOI: https://doi.org/10.1016/j.ridd.2013.03.001        [ Links ]

30.Silva FPS, Rocha NACF, Tudella E. Can size and rigidity of objects influence infant's proximal and distal adjustments of reaching?. Rev Bras Fisioter. 2011;15(1):37-44. DOI: http://dx.doi.org/10.1590/S1413-35552011000100003        [ Links ]

31.Dusing SC, Lobo MA, Lee HM, Galloway JC. Intervention in the first weeks of life for infants born late preterm: A case series. Pediatr Phys Ther. 2013;25(2):194-203. DOI: http://dx.doi.org/10.1097/PEP.0b013e3182888b86        [ Links ]

32.Petrini JR, Dias T, McCormick MC, Massolo ML, Green NS, Escobar GJ. Increased risk of adverse neurological development for late preterm infants. J Pediatr. 2009;154(2):169-76. DOI: http://dx.doi.org/10.1016/j.jpeds.2008.08.020        [ Links ]

33.Libertus K, Needham A. Teach to reach: The effects of active vs. passive reaching experiences on action and perception. Vision Res. 2010;50(24):2750-7. DOI: http://dx.doi.org/10.1016/j.visres.2010.09.001        [ Links ]

34.Cunha AB, Woollacott M, Tudella E. Influence of specific training on spatiotemporal parameters at the onset of goal-directed reaching in infants: a controlled clinical trial. Braz J Phys Ther. 2013;17(4):409-17. DOI: http://dx.doi.org/10.1590/S1413-35552013005000099        [ Links ]

35.Sommerville JA, Woodward AL, Needham A. Action experience alters 3-month-old infants' perception of others' actions. Cognition. 2005;96(1):B1-11. DOI: https://doi.org/10.1016/j.cognition.2004.07.004        [ Links ]

36.Libertus K, Needham A. Encouragement is nothing without control: Factors influencing the development of reaching and face preference. J Mot Lear Dev. 2014;2(1):16-27. DOI: https://doi.org/10.1123/jmld.2013-0019        [ Links ]

37.Cunha AB, Soares DA, Ferro AM, Tudella E. Effect of training at different body positions on proximal and distal reaching adjustments at the onset of goal-directed reaching: a controlled clinical trial. Motor Control. 2013;17(2):123-44.         [ Links ]

38.Suresh K. An overview of randomization techniques: An unbiased assessment of outcome in clinical research. J Hum Reprod Sci. 2011;4(1):8-11. DOI: https://doi.org/10.4103/0974-1208.82352        [ Links ]

39.Grimes DA, Schulz KF. Bias and causal associations in observational research. Lancet. 2002;359:248-52. DOI: https://doi.org/10.1016/S0140-6736(02)07451-2        [ Links ]

40.Ozonoff S, Young GS, Carter A, Messinger D, Yirmiya N, Zwaigenbaum L, et al. Recurrence risk for autism spectrum disorders: a Baby Siblings Research Consortium study. Pediatrics. 2011;128(3):488-95. DOI: https://doi.org/10.1542/peds.2010-2825        [ Links ]

41.March of Dimes, PMNCH, Save the Children, WHO. Born too soon: the global action report on preterm birth. Geneva: World Health Organization, 2012.         [ Links ]

 

 

Correspondence:
daniele.soares@ufms.br

Manuscript received: September 2018
Manuscript accepted: August 2018
Version of record online: October 2019