We appreciate Bailey et al's interest in our study ^{Reference Boschloo, Vogelzangs, Van den Brink, Smit, Veltman and Beekman1} and respond to their comments regarding the found differential effects of alcohol dependence v. alcohol abuse and alcohol consumption.

It is evident that heavy alcohol consumption has a major impact on public health because of its negative consequences on the onset and course of various diseases. ² However, this does not necessarily imply that the level of alcohol consumption is also causally related to the onset and persistence of psychopathology such as depression and anxiety. For example, Haynes et al ^{Reference Haynes, Farrell, Singleton, Meltzer, Araya, Lewis and Wiles3} showed that heavy alcohol consumption was not a risk factor for the onset of depression and anxiety in a national sample of adults in Great Britain. Furthermore, additional analyses in our sample showed that neither heavy alcohol consumption at baseline (odds ratio (OR) = 0.98, 95% CI 0.71–1.36) nor at 2-year follow-up (OR = 0.98, 95% CI 0.69–1.41) predicted the persistence of depression and anxiety.

In contrast, DSM-IV alcohol dependence has strong links with depression and anxiety. For example, previous epidemiological studies have demonstrated robust cross-sectional associations ^{Reference Boschloo, Vogelzangs, Smit, Van den Brink, Veltman and Beekman4,Reference Hasin, Stinson, Ogburn and Grant5} and, in addition, our study showed that alcohol dependence at baseline was an important risk factor for the persistence of depression and anxiety during 2-year follow-up (OR = 1.69, 95% CI 1.04–2.75, adjusted for basic covariates). ^{Reference Boschloo, Vogelzangs, Van den Brink, Smit, Veltman and Beekman1} Additional analyses further showed that this association remained significant after controlling for the possible effects of alcohol consumption (OR = 2.05, 95% CI 1.19–3.53, additionally adjusted for alcohol consumption). This indicates that impairments related to alcohol dependence, rather than consumption per se, have an effect on the course of depression and anxiety. Support for this hypothesis was also provided by our finding that persistence rates of depression and anxiety increased with the severity of alcohol dependence (i.e. number of alcohol dependence criteria). ^{Reference Boschloo, Vogelzangs, Van den Brink, Smit, Veltman and Beekman1}

Previous studies have raised the question whether DSM-IV alcohol abuse should be considered a genuine psychiatric disorder as it was characterised by low severity in the general population and showed limited reliability and validity (see Boschloo et al ^{Reference Boschloo, Vogelzangs, Van den Brink, Smit, Veltman and Beekman1} for a discussion on this topic). This notion received further support by our finding that 95.7% of patients with alcohol abuse met only one of four abuse criteria. ^{Reference Boschloo, Vogelzangs, Van den Brink, Smit, Veltman and Beekman1} Additional analyses showed that alcohol abuse was not related to impairments in daily functioning (assessed with the World Health Organization Disability Assessment Schedule II) and had a favourable course (remission rate after 2 years: 93.5%). Apparently, alcohol abuse is characterised by only minimal alcohol-related impairments, which may clarify our finding that it was not related to the persistence of depression and anxiety. Note that alcohol abuse in our sample of out-patients with depression or anxiety or in the general population might differ substantially from alcohol abuse in clinical samples of severe abusers with regard to associated levels of impairment.

Taken together, these findings emphasise the importance of severity indicators, rather than dichotomous diagnoses of alcohol dependence or abuse, in the assessment of alcohol problems. This is in line with the proposal of the DSM-5 Work Group to distinguish different levels of severity within the diagnosis of an alcohol use disorder (www.dsm5.org).