The purpose of this study was to discover whether greater sleep problems are associated with reduced heart rate variability during working hours and at night, and to determine whether this association is in part mediated by experienced affective states. This study involved 199 working women with a mean age of 33.8years. Sleep problems were assessed with the Jenkins Sleep Problems Scale, and the Day Reconstruction Method was used to measure positive affect and stress on the evening before and during the working day. Heart rate variability was indexed by the mean square root of the successive standard difference in heart period. Disturbed sleep was inversely related to heart rate variability during the working day (P=0.022), independently of demographic and behavioural confounders. Additional adjustment for positive affect and stress did not lead to further reductions in the association between sleep problems and reduced heart rate variability over the work day. Sleep problems were not predictive of reduced night-time heart rate variability. This report extends the findings from experimental studies and clinical samples, and suggests that disturbed sleep might impair heart rate variability in real life settings, in particular during working hours. Reduced heart rate variability might be a potential pathway linking sleep problems with cardiovascular disease. Based on the current data there was little evidence that the inverse associations between sleep problems and heart rate variability were mediated by experienced affective states.