From the Guidelines
Bias Type
The type of bias that occurs when a researcher only publishes a trial showing a link between Medication X and reduced cholesterol levels, while discarding prior trials with negative results, is publication bias.
- This bias arises from the non-publication of full studies and selective publication of results in relation to their findings, as noted in 1.
- Publication bias can lead to an overestimation of the effectiveness of a treatment, as only positive results are published, while negative results are discarded, as seen in a review of trials of antidepressants where the effect estimates were 32% larger on average when only published trials were analyzed compared to when all trials were included 1.
- It is essential to assess the risk of bias across studies, including publication bias, to ensure the validity of a systematic review and meta-analysis, as emphasized in 1.
- Methods to investigate possible bias across studies, such as funnel plots and statistical tests, should be reported in detail to address the issue of publication bias, as suggested in 1 and 1.
From the Research
Type of Bias
The type of bias that occurs when a researcher only publishes a trial showing a link between Medication X and reduced cholesterol levels, while discarding prior trials with negative results, is known as publication bias.
- This bias occurs when studies with statistically significant results have an increased likelihood of being published, as stated in the study by 2.
- Publication bias is commonly associated with inflated treatment effect, which lowers the certainty of decision makers about the evidence, according to 2.
Characteristics of Publication Bias
- Publication bias is more likely to occur when the results of a trial are statistically significant or show a positive direction of effect, as found in the study by 3.
- Trials with positive findings are more likely to be published than trials with negative or null findings, with an odds ratio of 3.90, as reported in 3.
- The study by 4 also found that nonpublication of trials was primarily due to "negative" results and lack of interest.
Consequences of Publication Bias
- Publication bias can lead to distorted scientific literature and mislead researchers, doctors, and policymakers in their decision-making, as discussed in 5.
- The under-reporting of negative results can waste other researchers' time, money, and manpower, and introduce bias in meta-analysis, according to 5.
- The study by 6 found that selective outcome reporting is prevalent in trials, and proposed methods to diminish the occurrence of this bias in future research.