The Tangible Common Denominator of Substance Use Disorders: A Reply to Commentaries to Rehm et al. (2013a)

Abstract

In response to our suggestion to define substance use disorders via 'heavy use over time', theoretical and conceptual issues, measurement problems and implications for stigma and clinical practice were raised. With respect to theoretical and conceptual issues, no other criterion has been shown, which would improve the definition. Moreover, heavy use over time is shown to be highly correlated with number of criteria in current DSM-5. Measurement of heavy use over time is simple and while there will be some underestimation or misrepresentation of actual levels in clinical practice, this is not different from the status quo and measurement of current criteria. As regards to stigma, research has shown that a truly dimensional concept can help reduce stigma. In conclusion, 'heavy use over time' as a tangible common denominator should be seriously considered as definition for substance use disorder.

Prevalence of and Potential Influencing Factors for Alcohol Dependence in Europe

“…Accordingly, it seems that AUD prevalence is very culture-specific, and thus correlations with other health indicators are biased downwards because of the high culture-specific variability. Heavy drinking rates may be the better indicator for public health and healthcare planning [11,79,80]. Other factors contributing to the low prevalence of AUDs in the South of Europe may be lack of training in physicians, leading to perceptions that some less severe AUDs may be ‘normal drinking' [81].…”

The images theory of addiction

“…Heavy use over time is a better predictor of substance-related health and social problems, and does not require diagnoses of dependency or addiction . Epidemiology has no need for "an explanation in terms of a mysterious force that cannot be fully modelled in animal or neurobiological research" (Rehm et al, 2013b). A related argument is that attempts to control the behavior of people labelled as addicts are often rooted in biases against specific groups, and have less to do with the harm caused by the behavior than with disparities between those people's observed lifestyles and what is expected of them by those in power.…”

Commentary on Cherpitel et al. (2015): Improving global estimates of alcohol‐attributable injury

“…[6]). The risk curves become quite unreliable for higher drinking levels, and while high or very high drinking levels are neither untypical nor infrequent for people with severe alcohol use disorders in North America or Europe [7,8], they are not typical in emergency room study respondents.The research also does not take into account the heterogeneity of drinkers and their impact on risk at different levels. As Krüger and colleagues have shown for the specific injury category of drunk driving, there is a substantial difference if people drive with a blood alcohol level of 0.09% by volume if, on average, they are very moderate drinkers, heavy drinkers or alcohol-dependent [9], and these differences have not been incorporated so far in country-specific or global estimates of alcohol-attributable risk.…”

“…[6]). The risk curves become quite unreliable for higher drinking levels, and while high or very high drinking levels are neither untypical nor infrequent for people with severe alcohol use disorders in North America or Europe [7,8], they are not typical in emergency room study respondents.…”

“…All data in the analysis from Cherpitel and colleagues [1] came from emergency room studies, and were analysed by pairmatched case-crossover design [5], a statistical technique based on self-report with inherent biases (see, for instance [6]). The risk curves become quite unreliable for higher drinking levels, and while high or very high drinking levels are neither untypical nor infrequent for people with severe alcohol use disorders in North America or Europe [7,8], they are not typical in emergency room study respondents.…”