Background and Objective: Childcare provides the first opportunity to stimulate children’s cognitive, social, and emotional development outside the home. We know little about how childcare is associated with mental health in childhood and adolescence. The Millennium Cohort Study, a large UK longitudinal study, was used to examine the relationship between age of starting childcare and weekly hours in formal childcare between birth and 5 years and internalizing and externalizing behavior trajectories from ages 5 to 14 years. Methods: The sample comprised 6194 children. Parents reported externalizing and internalizing behaviors with the strengths and difficulties questionnaire at 5, 7, 11 and 14 years. Associations were analyzed using a multilevel general linear regression model, with adjustment for socio-economic position, maternal mental health, demographics, and child temperament. Results: Later entry into childcare was associated with more internalizing but not externalizing behaviors, with the effect strengthening from 5 years to 14 years. In the fully adjusted model, starting childcare between 1-2 and 4-5 years compared to starting between 0-1 year was associated with 0.39 (-0.01, -0.79) and 0.98 (0.24, -1.71) higher internalizing behavior scores at 14 years. Children who spent more than 40 hours per week in childcare between birth and 3 years had 1.23 (0.58, 1.87) higher externalizing behavior scores at 5 years than children who spent no time in childcare. The association disappeared by 14 years. Weekly hours were not associated with internalizing behavior. Conclusions: Results show that childcare is associated with subsequent mental health and that the timing matters. Childcare could play an important role in a population’s mental health promotion, however so far this is only the case for internalizing behaviors, and high intensities might even be associated to more problem behaviors.