We suggest that an over-arching "fitness factor" (an index of general genetic quality that predicts survival and reproductive success) partially explains the observed associations between health outcomes and intelligence. As a proof of concept, we tested this idea in a sample of 3654 US Vietnam veterans aged 31-49 who completed five cognitive tests (from which we extracted a g factor), a detailed medical examination, and self-reports concerning lifestyle health risks (such as smoking and drinking). As indices of physical health, we aggregated "abnormality counts" of physician-assessed neurological, morphological, and physiological abnormalities in eight categories: cranial nerves, motor nerves, peripheral sensory nerves, reflexes, head, body, skin condition, and urine tests. Since each abnormality was rare, the abnormality counts showed highly skewed, Poisson-like distributions. The correlation matrix amongst these eight abnormality counts formed only a weak positive manifold and thus yielded only a weak common factor. However, Poisson regressions showed that intelligence was a significant positive predictor of six of the eight abnormality counts, even controlling for diverse lifestyle covariates (age, obesity, combat and toxin exposure owing to service in Vietnam, and use of tobacco, alcohol, marijuana, and hard drugs). These results give preliminary support for the notion of a superordinate fitness factor above intelligence and physical health, which could be further investigated with direct genetic assessments of mutation load across individuals. (Contains 3 tables and 1 figure.)