Well-developed theory implies that the human secondary sex ratio moves inversely over time with the level of anxiety and depression in the population. Few tests of this hypothesis, however, appear in the voluminous literature concerned with the sex ratio. These tests, moreover, employ designs that allow only weak inference. We contribute to the literature by applying time-series methods to Swedish data for the 276 months beginning January 1974 to detect a relationship between the sex ratio and defined daily doses of antidepressants and anxiolytics dispensed to women. Consistent with theory, we find the drug variable inversely related to the sex ratio. We argue that the discovered association cannot be attributed to shared trends, cycles, or other forms of autocorrelation in the data, or to the problem of endogeneity that necessarily plagues studies based on samples of individual women and births. Implications include that surveillance systems might monitor dispensing of anxiolytics and antidepressants as a marker for population stress thought to be a risk factor for adverse pregnancy outcomes such as preterm delivery.