We studied a large sample of healthy children in four countries with different cultural and linguistic characteristics to examine the development of children in the first 3 years of life. Our study provides information on developmental milestones that might be used across populations to assess development and also on those that require further investigation or elimination from international instruments.
The aim of most research comparing early childhood development across populations has been to describe cultural and ethnic variations and their association with contextual differences.20x20Fernald, LCH, Kariger, P, Engle, P, and Raikes, A. Examining early childhood development in low-income countries: a toolkit for the assessment of children in the first five years of life. World Bank,
Washington DC; 2009http://siteresources.worldbank.org/INTCY/Resources/395766-1187899515414/Examining_ECD_Toolkit_FULL.pdf. ()
Crossref | Google ScholarSee all References,24x24Henrich, J, Heine, SJ, and Norenzayan, A. The weirdest people in the world?. Behav Brain Sci. 2010;
Crossref | PubMed | Scopus (1981) | Google ScholarSee all References Most studies have included children from high-income countries, ethnic minorities, and small samples from LMICs. By contrast, our objective was to describe the variability in the ages of attainment of milestones and to establish whether enough similarities exist to guide the development of universal instruments, to avoid the costly restandardisation and revalidation of instruments. We therefore used definitions of equivalence to interpret our data rather than statistical significance alone.
In a cross-sectional study9x9Lansdown, RG, Goldstein, H, Shah, PM et al. Culturally appropriate measures for monitoring child development at family and community level: a WHO collaborative study. Bull World Health Organ. 1996;
PubMed | Google ScholarSee all References with a similar goal led by WHO in the 1990s, approximately 28 000 children aged 0–6 years were tested in China, India, and Thailand. The prevalence of health risks in the samples was not described and different developmental instruments were applied across sites. Both factors might have accounted for differences in the median age of attainment of milestones across countries and within country urban and rural sites. Nevertheless, when comparing the study led by WHO9x9Lansdown, RG, Goldstein, H, Shah, PM et al. Culturally appropriate measures for monitoring child development at family and community level: a WHO collaborative study. Bull World Health Organ. 1996;
PubMed | Google ScholarSee all References and our study, the median ages of attainment for the milestone of saying one meaningful word in our sample and the samples from urban China and India in the WHO study are similar (9·3 months in our study vs 9·7 months in China and 9·3 months in India) and for the milestone of saying two meaningful words (21·5 months in our study vs 20·1 months in China and 18·7 months India). The WHO Motor Development Study13x13WHO Multicentre Growth Reference Study Group. Assessment of sex differences and heterogeneity in motor milestone attainment among populations in the WHO Multicentre Growth Reference Study. Acta Paediatr Suppl. 2006;
PubMed | Google ScholarSee all References assessed the ages of attainment of six gross motor milestones in healthy children in five countries. This study used both caregiver report and direct observations to establish when children attained milestones. Again, the median age of attainment for the milestones common to both studies (ie, sits without support, stands alone, and walks alone) in our study compared with the Motor Development Study are similar (6·5 months vs 5·9 months for the sitting alone milestone, 10·0 months vs 10·8 months for standing alone, and 12·9 months vs 12·0 months for walking alone). Furthermore, to compare our data with data obtained in high-income countries, we examined the Denver II developmental screening test,25x25Frankenburg, WK, Dodds, J, Archer, P, Shapiro, H, and Bresnick, B. The Denver II: a major revision and restandardization of the Denver Developmental Screening Test. Pediatrics. 1992;
PubMed | Google ScholarSee all References an instrument developed in the USA. The median ages of attainment of our total sample were almost identical for milestones such as “uses six meaningful words”, “walks alone”, “kicks ball”, “reaches to objects”, and “holds pencil and scribbles”. These striking similarities provide further support for the universality of development across countries for some milestones, and also for the validity of the open-ended question technique used in our study.
Our study advances the understanding of early childhood development by showing that many milestones in numerous domains are similarly attained across sexes and countries. We found that the attainment of almost all milestones is similar in the first year when environmental and cultural influences might have the smallest effect. The similarity of play across our country samples parallels earlier studies.26x26Cote, LR and Bornstein, MH. Child and mother play in three US cultural groups: comparisons and associations. J Fam Psychol. 2009;
Crossref | PubMed | Scopus (14) | Google ScholarSee all References The difference in ages of attainment for pretend play between girls and boys emerging in the third year of life might reflect cultural influences with regard to how boys and girls are expected to play. The ages of attainment of play milestones in healthy children across countries is of utmost importance to integrated interventions that include play and are being highly promoted in LMICs.3x3Richter, LM, Daelmans, B, Lombardi, J et al. Investing in the foundation of sustainable development: pathways to scale up for early childhood development. Lancet. 2017;
Summary | Full Text | Full Text PDF | PubMed | Scopus (27) | Google ScholarSee all References,27x27Vaivada, T, Gaffey, MF, and Bhutta, ZA. Promoting early child development with interventions in health and nutrition: a systematic review. Pediatrics. 2017;
Crossref | PubMed | Scopus (1) | Google ScholarSee all References
A large proportion of the differences in ages of attainment of milestones was associated with timing of children’s exposure to experiences. For example, South African children could drink from a cup at a median age of 8 months compared with Argentinian children who reached this milestone at a median age of 16 months. In South Africa, where early independence is encouraged,28x28Roman, NV. Maternal parenting in single and two-parent families in South Africa from a child’s perspective. Soc Behav and Personal. 2011;
Crossref | Scopus (6) | Google ScholarSee all References children attained most self-help milestones at an earlier age than children in the other three countries, whereas in Argentina a more protective parenting style is generally adopted,29x29de Minzi, MCR, Lemos, V, and Rubilar, JV. Argentine culture and parenting styles. in: H Selin
Parenting across cultures. Childrearing, motherhood and fatherhood in non-Western cultures. Springer,
Netherlands; 2014: 277–292
Crossref | Google ScholarSee all References which might explain later attainment of these milestones. Culture is not the only factor that determines experiences. South African and Argentinian children attained the milestone of climbing up and down stairs at an older age than Indian and Turkish children, which is probably because most children included in these samples were more likely to live in single-storey houses, whereas Indian and Turkish children were more likely to live in apartment buildings with stairs.
The differences between countries in language milestones must be interpreted with caution. Receptive language is known to be difficult to assess because it is dependent on what caregivers expect and think children understand.30x30Tomasello, M and Mervis, CB. The instrument is great, but measuring comprehension is still a problem. Monographs of the Society for Research in Child Development. 1994;
Crossref | Scopus (72) | Google ScholarSee all References Consistent with these recognised difficulties in assessing the attainment of receptive language is the finding that most language milestones attained at different ages were associated with caregivers’ understanding of children’s speech and their interpretation of what children understand. More objectively, interpretable expressive language milestones such as the use of pronouns, the use of past tense, or the ability to recount a story or event were attained at nearly identical ages across countries, suggesting that overall language acquisition was similar. Milestones on acquisition of sentences might reflect differences in syntax. Furthermore, considerable differences were found across countries in maternal and paternal education. Whether differences in language milestones reflect true differences in children, cultural and ethnic differences in caregivers’ interpretations of what children convey or understand, caregivers’ use of language with young children, or the effect of psychosocial factors (eg, caregivers’ education) requires further study.
Our study has important strengths. First, the cross-sectional design avoids potential biases of repeated questioning and retention of compliant families. Second, the countries included are from diverse geographical areas of the world with ethnic, cultural, and language differences. Third, the sample of almost 5000 children is one of the largest to date, providing information on multiple domains of development of healthy children younger than 3 years. Fourth, our criteria for a healthy sample were more stringent than criteria used in previous research.31x31Wijnhoven, TM, de Onis, M, Onyango, AW et al. Assessment of gross motor development in the WHO Multicentre Growth Reference Study. Food Nutr Bull. 2004;
Crossref | PubMed | Google ScholarSee all References,32x32Gladstone, M, Lancaster, GA, Umar, E et al. The Malawi Developmental Assessment Tool (MDAT): the creation, validation, and reliability of a tool to assess child development in rural African settings. PLoS Med. 2010;
Crossref | PubMed | Scopus (52) | Google ScholarSee all References,33x33Lejarraga, H, Pascucci, MC, Krupitzky, S et al. Psychomotor development in Argentinean children aged 0–5 years. Paediatr Perinat Epidemiol. 2002;
Crossref | PubMed | Scopus (37) | Google ScholarSee all References We excluded children with health conditions associated with potential adverse developmental outcomes.34x34Walker, SP, Wachs, TD, Gardner, JM…, and International Child Development Steering Group. Child development: risk factors for adverse outcomes in developing countries. Lancet. 2007;
Summary | Full Text | Full Text PDF | Scopus (760) | Google ScholarSee all References The fact that half of the recruited sample was excluded supports the high prevalence of health problems in LMICs that has been reported previously,2x2Black, MM, Walker, SP, Fernald, LC et al. Early childhood development coming of age: science through the life course. Lancet. 2017;
Summary | Full Text | Full Text PDF | PubMed | Scopus (47) | Google ScholarSee all References which has been shown to adversely affect children’s development. More girls than boys were excluded from our study because of health problems, which might support evidence for sex-associated health disparities. Further research using such milestones that are attained similarly in healthy children will enable the development of common methods to examine the effect of health-associated risk factors on child development, and comparisons of child development between populations with differences in the prevalence of such risk factors.
Our study has important limitations. We did not include a large number of LMICs, particularly those with lower incomes. We chose four countries that were culturally distinct and had collaborating teams with the capacity to do rigorous research and to provide services for children identified with risk factors. Another limitation is that the sample did not include rural sites. Thus, the applicability of our results to rural populations needs to be established. The small sample size—particularly the small number of older children (aged 25–42 months) enrolled in South Africa—is a limitation that is reflected in the larger confidence intervals in South Africa for some of the milestones attained at an older age, and might require repetition in larger samples. The number of children who were excluded because of health problems was more than we expected in all countries, but particularly in South Africa, where we could not change our recruitment strategy as we did in India, because the sociodemographic characteristics of children attending private paediatric clinics would have been substantially different. We recruited children from health clinics and not from homes to enable application of health criteria. This approach might decrease generalisability because our sample might have included more children with health problems using the clinics than children with health problems in the general population, or an increased number of healthy children that access primary care. Bias in either direction should not affect the results of the healthy sample. Direct measurements of undernutrition and anaemia, detailed questioning of caregivers about birthweight, perinatal and chronic illness, and a health checklist provided by clinicians were the most rigorous health criteria we could apply. Nevertheless, we might have erroneously included some children with unknown health conditions. We did not exclude children with psychosocial risk factors such as poverty, a low level of caregiver education, or depression.34x34Walker, SP, Wachs, TD, Gardner, JM…, and International Child Development Steering Group. Child development: risk factors for adverse outcomes in developing countries. Lancet. 2007;
Summary | Full Text | Full Text PDF | Scopus (760) | Google ScholarSee all References Further research is required to define the effects of psychosocial risk factors on the ages of attainment of developmental milestones.
Our study has identified the median age at which healthy children of both sexes and from four countries attain milestones in multiple developmental domains. These findings might contribute to the construction of internationally applicable tools to assess children’s development to guide policy, service delivery, and intervention research that might help narrow the gap between high-income countries and LMICs in addressing early childhood development.