The purpose of this study was to assess the biological and behavioural characteristics associated with the variation in the awake-sleep blood pressure (BP) difference in a large series of patients who had ambulatory BP monitoring. The 1186 subjects in the study (689 men and 497 women) each wore an ambulatory BP monitor over one 24 h period. All patients had been off antihypertensive medication for at least 3 weeks prior to the BP monitoring. The extent of the difference in awake-sleep BP was defined by the percentage change to sleep pressure from the average awake value. The results showed that when the average levels of office, awake and sleep pressures are compared across groups defined by this percentage change, there is no statistically significant association between the level of office pressure and the extent of the awake-sleep pressure difference. In accordance with this finding, the relation between ambulatory awake BP and the awake-sleep difference was relatively weak, although significant (P < 0.01). In striking contrast, the ambulatory sleep pressure was a very strong predictor of the awake-sleep difference (P < 0.0001), such that subjects with the lowest sleep pressures had the biggest diurnal change. In examining the factors associated with the awake-sleep change in pressure, the most important predictors were a low sleep pressure and alcohol intake, both of which were associated with bigger diurnal changes of both SBP and DBP in both men and women. A large pulse pressure while awake (that is, isolated systolic hypertension) was also associated with bigger diurnal changes in both sexes. The only other predictive factor which applied in both sexes was a low awake BP variability which was associated with a greater diurnal variation of DBP. Among men but not women, older age, smoking cigarettes, and being overweight were associated with bigger diurnal changes of both SBP and DBP, and a family history of hypertension was also associated with a bigger diurnal change of systolic pressure. Finally, among women but not men, being measured in winter months was associated with a greater awake-sleep difference in diastolic pressure. These results show that there is a general pattern of factors which predict awake-sleep changes in men and women, but there are also contributing behavioural factors which are sex specific. As awake-sleep changes of BP may be of prognostic significance, behavioural factors need to be considered when evaluating this change.