Background: Vaccinating children against influenza virus may reduce infections in immunised children and household contacts, thereby reducing the household-based cost associated with respiratory illnesses. Objective: To evaluate the impact of influenza virus vaccination of daycare children on costs of respiratory illnesses of the children and their household contacts from the household and societal perspective. Study design: Cost analysis of data from a randomised controlled trial covering the period November to April of 1996-7 and 1998-9. Children (127 in 1996-7 and 133 in 1998-9) from daycare centres in Californian (USA) naval bases received influenza virus vaccine (inactivated) or hepatitis A virus vaccination. Outcome measures: Direct and indirect costs (1997 and 1999 US dollars) of respiratory illnesses in households of vaccinated and not vaccinated daycare children, excluding the cost of vaccination. Results: There were no statistically significant differences in household costs of respiratory illness between households with or without influenza virus-vaccinated children ($US635 vs $US492: p = 0.98 [1996-7]; $US412.70 vs $US499.50: p = 0.42 [1998-9]). In 1996-7, adult and 5- to 17-year-old contacts of vaccinated children had lower household costs than contacts of unvaccinated children ($US58.50 vs $US83.20, p = 0.01 and $US32.80 vs $US59.50, p = 0.04, respectively), while vaccinated children 0-4 years old had higher household costs than unvaccinated children in the same age group ($US383 vs $US236, p = 0.05). In 1998-9, there were no differences within individual age groups. Results from societal perspective were similar. Conclusions: Overall, from both the household and societal perspectives, there were no economic benefits to households from vaccinating daycare children against influenza virus. However, we found some over-time inconsistency in results; this should be considered if changing recommendations about routine influenza virus vaccination of healthy children. Our study size may limit the generalisability of the results.