Objective: To modify and standardise the Denver Developmental Screening Test (DDST) and Denver II for developmental screening of children in Singapore.
Method: The study used a quota sample of 2,194 Singapore children aged 4 weeks to 6 years. Logistic regression analysis established the 25th, 50th, 75th, and 90th percentile passing age for achieving the test tasks. Subgroup differences in Sex, Ethnicity, Social Class and Mother's Education were analysed by stepwise logistic regression; the composite norms for items with statistically significant subgroup differences (p < or = 0.10), were then adjusted by weighting based on the composition of Singapore children. The study protocol was based on the DDST (1975) and Denver II (1990), the latest version of the DDST. Modifications were introduced to improve on the sensitivity of the test and to make the test more suited to Singapore culture.
Main findings: Out of the 215 items studied, 115 items were selected to form the new test, DDST, Singapore, DDST, Singapore shares 63% of the items with DDST (1975) and 67% of the items with Denver II. Among the comparable items, differences between the norms of Singapore and Denver children greater than 10% were demonstrated in more than 30 items, and differences of greater than 20% in 10 items. Within the study sample of Singapore children, there were relatively smaller differences among the subgroups studied. Only 10 items had clinically significant subgroup differences of more than 10%. None had more than 20% difference.
Conclusion: DDST, Singapore is substantially different from the DDST (1975) and Denver II (1990). The use of the local standardised version for developmental screening of Singapore children is justified.