Some years ago we published an editorial setting out a checklist for authors writing articles in the field of addiction . We believe that now is a good time to re-assess our priorities when considering submitted articles and to ensure that our editors, reviewers and potential contributors are aware of these. That way we hope to maximise consistency of decision-making and help researchers send us papers that have a higher probability of acceptance. We have arrived at the classification shown in Table 1. These priorities reflect Addiction's emphasis as an international journal in an applied area of study on communicating to the scientific, clinical and policy community important findings relating to the:
|Usually high priority|
|1. Reports of primary results from adequately powered controlled trials of interventions to prevent or treat an addictive disorder or reduce harmful consequences of a potentially addictive behaviour|
|2. Large epidemiological studies establishing in a population of interest the current extent of an addictive behaviour, a problem relating to an addictive behaviour, and/or attempts at reducing an addictive behaviour or problems associated with the behaviour|
|3. Studies of any kind (e.g. laboratory-, survey- or clinic-based, secondary analyses) that provide substantial new information on the development of, maintenance of, or recovery from an addictive disorder|
|4. Studies that substantially advance methods in the field of addiction, including the development and assessment of measures, sampling strategies or methods of data analysis|
|5. Qualitative studies and case studies that provide genuine insights into phenomena relating to addiction or point to important phenomena the existence of which was previously unsuspected |
|6. Major ‘position pieces’ that provide an overarching review of an important area and systematic reviews of research literatures that include meta-analyses where appropriate|
|7. Articles that develop important new theories in the field or provide definitive data supporting or calling into question currently influential theories|
|8. Large-scale policy-relevant studies that use randomized as well as quasi-experimental designs (e.g. interrupted time series, regression discontinuity, before-after panel surveys) to evaluate the impact of interventions on population level morbidity, mortality and substance use or addictive behaviour|
|9. Adequately powered studies of vulnerability to substance use disorders using behavioural and genetic epidemiology designs (e.g. twin studies)|
|10. Treatment systems research on the organization and effectiveness of clinical interventions and mutual help organizations at the population level|
|11. High quality policy research that provides new insights into how policies are developed or can be influenced, including research on the role of tobacco, alcohol and gaming industries|
|Usually low priority|
|1. Uncontrolled descriptions of outcomes of interventions unless the results show something very novel and interesting (e.g. very high success rates for a new intervention or very low success rates for an established intervention) or there is an unusually strong quasi-experimental design that enables major threats to validity to be well addressed|
|2. Small scale epidemiological studies or large studies that confirm what is already established or are in populations of limited international interest (e.g. South London)|
|3. Laboratory studies with marginal findings that only emerge from post-hoc analyses (e.g. found only in left-handed men) and/or outcome measures that are not clearly linked to ‘real-world’ variables or are subject to ‘demand characteristics’ (e.g. hypothetical choices)|
|4. Studies on putative new measures or methods that do not demonstrate an advance on existing measures and methods or fill an important gap; in general simply demonstrating the reliability or a specific factor structure of a questionnaire will not be sufficient|
|5. Qualitative studies and case studies that illustrate the existence of a phenomenon that is already known about or that attempt to make any kind of quantification of a phenomenon or where the phenomenon is only of local or regional interest|
|6. Reviews that do not adopt a systematic replicable approach to study identification and selection|
|7. Articles that develop new theories without clearly indicating how they relate to existing theories and why they represent an improvement|
- 1extent and nature of the problem worldwide and attempts made to mitigate it
- 2effectiveness of interventions that aim to mitigate the problem
- 3evaluation and improvement of research methods that support aims 1 and 2.
We have no priorities regarding the type of addictive behaviour studied (alcohol, illicit drug use, tobacco use, gambling etc.) nor the specialism from which authors come (genetics, public health, psychology, psychiatry etc). The exception is that we do not publish primary research that involves non-human species: that is properly within the province of our sister journal, Addiction Biology.
We recognise that there are trade-offs between originality, the importance of a topic and the rigour with which it can be investigated. While we prioritise originality and importance the conclusions of all papers must appropriately reflect the level of confidence surrounding the conclusions.
We also value the iconoclasm that has been a feature of this journal for decades. The field of addiction is not unique in beliefs being firmly held for which there is insufficient evidence and this needs to be pointed out on a regular basis. This may involve reporting negative findings where positive findings would be expected or pointing out that the literature is inconsistent in an area where it is assumed to be consistent.
We do not have a specific category of ‘brief reports’ but generally take the view that less is more. Long introductions and discussions are not required in Addiction. We do not require speculation in discussion sections that an intelligent reader could well engage in for him or herself.
We do not insist on high response rates in surveys but authors need to persuade us that the low response rate is unlikely to undermine the specific claims being made.
We may consider papers from researchers who are directly employed or are consultants to producers of tobacco or alcoholic beverages, or organisations involved in the representation of these industries and the gambling sector. However, we will bear in mind the increased risk of bias either in the choice of research question, the interpretation of the results, or the decision concerning which data or analyses to publish. We recognise that researchers with ties to other industries and receiving government funding may also be subject to similar biases [3, 4]. Individuals employed by the tobacco, alcohol or gaming sector will not normally be allowed to contribute to editorials and commentaries.
We welcome papers from non Anglophone and low and middle income countries. We will prioritise studies that make a positive virtue of the geographical location where this shows that findings that we take for granted in one domain do not necessarily generalise worldwide or which point to novel phenomena that are little known outside a given country or culture.
Authors are reminded of a few specific matters to which we attach high importance:
- 1Non-experimental studies must always express findings in terms of associations and not cause and effect – authors must argue the case for a causal interpretation.
- 2Evaluations of behavioural interventions of any kind must lodge full manuals or complete descriptions in our supplementary materials archive or provide a reference to a permanent record that will be readily available to readers.
- 3All statistical comparisons must make it clear whether they were part of a planned analysis or post-hoc.
- 4All comparative statements must be supported by statistical tests or confidence intervals (as set out in our guidelines for authors).
- 5Avoid preferentially citing work from one's own country. It is imperative that researchers from all countries capture the international evidence base on the topic and appropriately credit the work of others in the field. We are prepared to publish long reference lists where these are appropriate. The exception is editorials and commentaries which for logistical reasons should normally not exceed 19 citations.
- 6The conclusion sections of abstracts should read as stand-alone statements of the most defensible generalisable statement of what was found: what we have come to call ‘citable statements’. These are sentences of the kind that would have the paper in question as the citation. Prefacing conclusions with redundant phrases such as ‘This study found . . .’ is not acceptable. Neither are conclusions that do not say anything substantive. Table 2 shows examples of unacceptable conclusions and their acceptable variants.
|Unacceptable conclusion: We conclude that the public should be informed that the addition of caffeine to alcohol does not enhance driving or sustained attention/reaction time performance relative to alcohol alone.|
|Acceptable conclusion: The addition of caffeine to alcohol does not appear to enhance driving or sustained attention/ reaction time performance compared to alcohol alone.|
|Unacceptable conclusion: At present, it is not possible to interpret the evidence with any degree of certainty. Future research should consider that the distribution of alcohol consumption data is likely to be skewed and that appropriate measures of central tendency are reported.|
|Acceptable conclusion: Computer-based interventions may reduce alcohol consumption compared with assessment only but the conclusion remains tentative because of methodological weaknesses in the studies. Future research should consider that the distribution of alcohol consumption data is likely to be skewed and that appropriate measures of central tendency are reported.|
|Unacceptable conclusion: The observed pattern of results indicates that it is the intake of nicotine from nicotine gum that helps mitigate weight gain during cessation. The findings strongly suggest that active gum provides a pharmacological benefit of limiting weight gain. Using the recommended number of pieces of nicotine gum per day can help to reduce weight gained during quitting.|
|Acceptable conclusion: There is a dose-response association, not observed with placebo gum, between nicotine gum use and mitigation of weight gain in the first 30 days of an attempt to stop smoking.|
|Unacceptable conclusion: Results indicated that stress coping was associated with improved affect but it did not protect against later smoking. Coping may help smokers trying to quit feel better but may not help them stay smoke-free.|
|Acceptable conclusion: Coping responses early in a quit attempt may help smokers trying to quit feel better but may not help them stay smoke-free.|
Readers who are interested in learning more about writing articles in the field of addiction are recommended to read Publishing Addiction Science.
In sum, we are interested in publishing addiction science that is methodologically sound and original and is likely to have a substantial impact for the better on the health and wellbeing of individuals and populations. We want our science to be relevant globally, and supportive of global initiatives (such as the FCTC and the Global strategy to reduce the harmful use of alcohol) that aim to translate research into effective policy at the local, national and international levels.