Bibliographic record
Abstract
The author responds: Dr. Wolfson1 highlights important nuances to my discussion2 of using randomized clinical trials (RCTs) to estimate causal effects in the clinical setting. We agree on the following points: A deception RCT (where patients are deceived into believing a treatment is effective) matches the clinical context when a patient is offered a choice between two “active” treatments, with only one actually being effective. The unblinded RCT matches the clinical context where patients are offered active treatment or no treatment. The unblinded RCT cannot distinguish between effects mediated through physiological processes versus psychological processes (which themselves, may be mediated through physiological or behavioral processes). A “covert” RCT measures a biological effect corresponding to the context where subjects believe they will receive inactive treatment. A covert RCT is one type of deception RCT. In a deception RCT, subjects believe they will receive a treatment (reference); the reference treatment may be active (eg, caffeinated coffee) or inactive (eg, decaffeinated coffee). Furthermore, subjects may not even know they are in a study. We differ in some minor areas. Dr. Wolfson’s definition of biological effect1 is restricted to causal effects when subjects do not know they are receiving active treatment. However, one could envision a medication that lowers sympathetic activity (heart rate) when sympathetic activity is elevated (eg, subjects believe they will receive caffeine) but not when sympathetic activity is “normal” (eg, subject believe they will not receive caffeine). The “biological” effect would be different for each level of the variable “sympathetic activity,” with each effect measured by a different deception RCT. Considering one variable level as more important than another variable level is a value judgment. Although deception RCTs may “generally be regarded as unethical”1 when evaluating treatments with unknown side-effect profiles, the method is applicable where the risk of adverse effects is nil or the minimal effects considered minor. Although clinicians may sometimes choose between offering versus not offering treatment, I believe they usually choose between several possible treatments that are potentially active. For example, there are many exercise programs promoted as effective treatment for ankle sprains, and effective treatment options for an anterior cruciate ligament tear include exercise, as well as several surgical procedures. In these contexts, the deception RCT where subjects are told a treatment is effective mimics the context of interest. One should evaluate on a case-by-case basis whether traditional RCTs or observational studies provide the less-biased estimate for the causal effect of interest. Ian Shrier Centre for Clinical Epidemiology, Jewish General Hospital, Montreal, QC, Canada, [email protected]
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.019 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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; both teacher heads 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".