Dementia is one of the most common, still incurable diseases of old age, with high costs for the patient, the family and the society. Scholars of dementia have identified important risk factors in mid-life but little attention has been devoted to the earliest period in life. This article extends the research about risk factors of dementia to the time around birth. We start by discussing the influential theories of fetal programming and of the thrifty phenotype, and briefly contrast these theories with life-course approaches and the theory of the accumulation of advantage or disadvantage over the life course. Our own empirical study is based on an exogenous indicator of the early life environment, namely the month of birth. We present the theoretical background of using this indicator and discuss others such as famines, epidemics, and macro-economic fluctuations. Using data from the largest public health insurer in Germany we explore the relationship between the month of birth and dementia and find that the winter-born have the lowest risk of developing dementia. We present the pros and cons of using claims data in the study of dementia and discuss research findings that may shed light on the biological mechanisms underlying our finding.