To examine whether items comprising a preschool well-child check for use by family doctors in Australia with 4–5-year old children predicts health and academic outcomes at 6–7 years. The well-child check includes mandatory (anthropometry, eye/vision, ear/hearing, dental, toileting, allergy problems) and non-mandatory (processed food consumption, low physical activity, motor, behaviour/mood problems) items. The predictive validity of mandatory and non-mandatory items measured at 4–5 years was examined using data from the Longitudinal Study of Australian Children. Outcomes at 6–7 years included overweight/obesity, asthma, health care/medication needs, general health, mental health problems, quality of life, teacher-reported mathematics and literacy ability (n = 2,280–2,787). Weight or height >90th centile at 4–5 years predicted overweight/obesity at 6–7 years with 60 % sensitivity, 79 % specificity and 40 % positive predictive value (PPV). Mood/behaviour problems at 4–5 predicted mental health problems at 6–7 years with 86 % sensitivity, 40 % specificity and 8 % PPV. Non-mandatory items improved the discrimination between children with and without mental health problems at 6–7 years (area under the receiver operating characteristic curve 0.75 compared with 0.69 for mandatory items only), but was weak for most outcomes. Items used in a well-child health check were moderate predictors of overweight/obesity and mental health problems at 6–7 years, but poor predictors of other health and academic outcomes.