Could the use of biochemical validation of smoking abstinence introduce a confound into the interpretation of randomized controlled trials of tobacco cessation?
Bibliographic record
Abstract
An important component of randomized controlled trials (RCTs) of tobacco cessation interventions is to confirm the validity of the primary outcome variable—smoking abstinence. The use of biochemical validation is central to determining the accuracy of smoking abstinence claims and has become almost a prerequisite for the publication of RCTs of tobacco cessation interventions in high-impact journals. However, does the use of biochemical validation come without cost? Depending on how the biochemical validation is implemented, it is possible that such validation of smoking abstinence could introduce an alternative explanation (i.e. a confound) into the interpretation of results of tobacco cessation trials. Specifically, the researcher might decide, as a cost-saving measure, to collect samples (e.g. saliva or blood) for biochemical validation only from participants who state that they have stopped smoking. In an RCT where one experimental condition is receiving an intervention and the other is not, it would be expected that more participants will claim abstinence in the intervention condition than in the control condition. This could become a problem if there are multiple follow-up points. In RCTs of tobacco control interventions there is often an immediate post-intervention follow-up and a more extended follow-up (e.g. 8 weeks and 6 months). If the researcher collects samples for biochemical validation at the immediate post-intervention follow-up, then it would be expected that more participants in the intervention condition would have taken part in this research procedure (the collection of samples) than those in the control condition (that is, if the intervention has any impact). Such a situation would introduce a post-randomization difference between experimental conditions in addition to the intervention under study. This post-randomization difference could then serve as an alternative explanation for observed differences in smoking abstinence at the extended follow-up (which is often the primary outcome for tobacco cessation RCTs). Why should such a post-randomization difference be a concern? There is a fairly extensive research tradition demonstrating the impact of the procedural aspects of a research trial on the behaviour of its participants. Often termed the study of Hawthorn effects 1-3, this research demonstrates how participants are not passive recipients of interventions but, instead, are engaging actively in the study procedure. From this perspective, the collection of a biochemical sample has the potential to be a meaningful event for the participant. This event may then highlight to the participant that the researcher places special importance on their report of abstinence, and could then act as an extra motivator to maintain their abstinence for those who are asked for such samples. One of the difficulties with interpreting the results of RCTs for tobacco cessation is that, while it is stated most often that smoking abstinence was validated biochemically, there are rarely any details of the specific procedure involved. Thus, the reader cannot determine whether only those participants who claimed smoking abstinence were asked to provide a sample for biochemical validation. As a first step in rectifying this limitation, the researcher should publish their procedure for sample collection so that the reader can be aware of this potential confound in the interpretation of long-term outcomes. In addition, this confound could be alleviated in one of two ways. First, all participants could be asked to provide samples whether or not they claim abstinence. Second, if the researcher can only afford to collect samples for biochemical validation from those claiming smoking abstinence, then these samples could be collected only at the final follow-up point. The collection of biochemical validation at just the final follow-up point is acceptable, and is the recommended procedure in the Russell Standard for outcome criteria for smoking cessation trials 4. In this way, there is no chance that the collection of samples could impact research outcomes differentially between experimental conditions through the introduction of a post-randomization difference to the research trial which is not part of the planned intervention under study. None. This research is funded by the Canadian Institutes of Health Research (CIHR). Support to CAMH for salary of scientists and infrastructure has been provided by the Ontario Ministry of Health and Long Term Care. The views expressed in this article do not necessarily reflect those of the Ministry of Health and Long Term Care. John Cunningham is supported as the Canada Research Chair in Brief Interventions for Addictive Behaviours.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.760 | 0.913 |
| Meta-epidemiology (narrow) | 0.004 | 0.004 |
| Meta-epidemiology (broad) | 0.012 | 0.012 |
| Bibliometrics | 0.007 | 0.010 |
| Science and technology studies | 0.003 | 0.029 |
| Scholarly communication | 0.011 | 0.020 |
| Open science | 0.009 | 0.008 |
| Research integrity | 0.019 | 0.015 |
| Insufficient payload (model declined to judge) | 0.009 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".