Preliteracy Intervention: Lessons to be Learned From Seemingly Discrepant Results

At face value, promoting early literacy skills is a bit of a “no-brainer.” We all want to maximize the life opportunities for children, and early intervention is immensely appealing for parents, practitioners, and policy-makers alike. Early intervention is increasingly supported by neurobiological evidence for the negative consequences of restricted environments1 and the economic evidence for the value of early intervention.2 Yet, knowing that early intervention has been shown to work for one aspect of development at a given age and dosage is not the same as saying that such results will be universally applicable. Indeed, as the 2 studies discussed in this issue of Pediatrics demonstrate,3,4 the results of effectiveness studies may seem contradictory; in this case, results of 1 study indicate that a preliteracy intervention does not work, and results of the other study indicate that it does. The research community has the responsibility of teasing these issues apart: how much does it work and for whom?The differences between the programs in question are instructive in helping to take the science of early intervention forward. At face value, the interventions themselves, “Let's Read” and “Little By Little,” a development of “Reach Out and Read,” are comparable in that they used nonspecialists to target preliteracy skills, although aspects of the delivery (location, intensity, and duration) differed. The study designs differed too. The Australian study3 used a conventional prospective design with details of allocation and statistical power. The sample came from “relative disadvantage,” although 80% of the parents concerned had more than 12 years of education. The US study,4 by contrast, randomly assigned subjects from a large existing database of children specifically identified because they are socially disadvantaged. In particular, the parents of their Spanish-speaking participants had the lowest educational level but their children responded most dramatically to the intervention.At face value, the US study seems to deliver more bang for its buck, but care should be taken not to overinterpret the results. Even if we assume that there were no systematic biases to distort the findings, it would seem that the Australian study specifically excluded those who the US study's results suggest are most likely to respond positively to the intervention. In both studies the authors controlled for various factors in accounting for their results, and it is clearly important to examine whether the effects of an intervention are direct or work through a third factor such as the child's communication environment. Similarly, the models may need to be expanded to better capture both parental characteristics such as motivation and mental health5 and child characteristics such as tested, rather than parental report of, IQ and language development.Finally, the results of these studies highlight the interesting tension between targeted and universal interventions in relation to the demographic characteristics of the populations concerned. It has been suggested that the relationship of preschool achievement to social disadvantage may be closely associated with the level of income inequality in the country concerned; such inequalities are more pronounced in the United States and the United Kingdom than they are in Australia or Canada.6 It may be that this type of intervention works better in the most disadvantaged populations, which rather suggests that universal interventions may not be the way forward for 2 seemingly contradictory reasons. On the one hand, there is a tendency for them to exacerbate health inequalities because those least in need of the messages may respond to the intervention more readily.7 On the other hand, care needs to be taken to ensure that the parents are not already using the intervention strategies of their own volition. It is important to go beyond oversimplistic notions of whether an intervention does or does not work to explore why such putatively similar interventions achieve such divergent results.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.