Moderate emotional reactions to stressful life events are associated with lowest risk of increased alcohol consumption: evidence from the Whitehall II study

Abstract Background Research investigating the associations between stress-related negative emotions and alcohol consumption often assumes a linear dose–response relationship. Based on the current theories of emotions, we questioned this assumption and hypothesized that both very low and very high levels of negative emotional response (NER) to stressful life events are related to increased alcohol consumption. Methods We used data from Phases 1 (1985–88) and 2 (1989–90) of the British Whitehall II study. At both phases, participants reported on their alcohol consumption, recent stressful events and the NER to the events. Two thousand and sixteen participants without recent stressful events at baseline were selected. Logistic regression was used to model the association between emotional response at baseline and increased frequency of alcohol consumption between the two phases. Results The likelihood of increased alcohol consumption increased with the number of recent stressful events. Among participants with at least one event, 17% increased alcohol consumption, compared with 14% of those who reported no events. Participants with average NER were at lowest risk (14%) of increasing alcohol consumption after major life events. Those with highest NER were significantly more likely to increase alcohol consumption (20%) than those with average NER, but the difference between those with highest NER and those with lowest NER (17%) was not statistically significant. Conclusions Increases in alcohol consumption following stressful events are least likely if people experience moderate levels of NER to events. Negative emotions in moderate doses should not be regarded as a risk factor for unhealthy behaviours, but rather a potentially protective factor.