BACKGROUND: Maternal shared reading practices predict emergent literacy, but fathers' contributions are less certain. We examined whether fathers' shared home reading activities at 2 years (1) predict language and emergent literacy at age 4 years, when controlling for maternal contributions; and (2) differentially benefit these outcomes in disadvantaged children.

METHODS: Design: Two-parent families recruited from 5 relatively disadvantaged communities for the universal Let's Read literacy promotion population-based trial (ISRCTN 04602902) in Melbourne, Australia. Exposure, 2 years: Home reading practices via self-reported maternal and paternal StimQ-Toddler questionnaires, dichotomised at study median (high vs. low). Outcomes, 4 years: Receptive and expressive language (Clinical Evaluation of Language Fundamentals-4), emergent literacy (Sunderland Phonological Awareness Test-Revised).

ANALYSES: Aim 1: Linear regression, adjusted for mothers' home reading, 2-year-old vocabulary and communication skills and family disadvantage. Aim 2: Interaction of disadvantage [yes vs. no] with high home reading by (a) fathers and (b) at least one parent.

RESULTS: Data were available for 405 (62.3%) families. High father reading at 2 years (reference: low) predicted better expressive (mean difference 8.0, 95%CI 4.5 to 11.5) and receptive (mean difference 7.3, 95%CI 4.1 to 10.5) language at 4 years (both p<0.001) but not emergent literacy skills. Similar patterns were observed in families with at least one parent with high home reading. Father reading did not differentially benefit outcomes in disadvantaged children.

CONCLUSION: Fathers' involvement in reading at 2 years predicted better language but not emergent literacy at 4 years, and did not protect against adverse effects of socioeconomic disadvantage.