An influential literature on the Barker's hypothesis (or the Developmental Origins of Health and Disease, DOHaD) has documented that poor conditions in utero lead to higher risk of hypertension, diabetes, stroke and heart disease in middle age in middle- and high-income contexts. One of the main explanations is that periods of high calorie intake after birth are inconsistent with the adaptations that the fetus makes to prepare for a poor resources environment (thrifty phenotype hypothesis). Using data from a persistently low-income country, Malawi, we find that individuals exposed in utero to a substantial famine in 1949, have lower levels of blood pressure and blood sugar and less symptoms associated with stroke over half a century later. These findings may be explained by a prolonged period of malnutrition following the famine in contrast to most of the contexts studied in the previous literature.