What doctors don't know about the drugs they prescribe | Ben Goldacre

New Section

The speaker introduces the topic of precognition and skepticism towards predicting the future, highlighting biases in academia and medicine.

Nostradamus and Precognition

• Introduction to Nostradamus and skepticism towards predicting the future.

• Discussion on precognition research by Daryl Bem with evidence of precognitive powers in undergraduate students.

• Bias in academic publishing where negative data is often not published, affecting the true representation of scientific studies.

Academic Bias and Research Replication

The discussion delves into biases in academia, particularly in psychology and cancer research, impacting replication studies and publication practices.

Academic Publishing Biases

• Challenges faced in replicating precognition study findings due to publication bias.

• Example of unreplicable basic science cancer studies leading to recommendations for easier publication of negative results.

• Proposal to change incentives for scientists to publish more negative results for a comprehensive scientific landscape.

Publication Bias Impact on Medicine

The impact of publication bias on medical research is explored through a case study involving a drug trial, emphasizing the consequences of unpublished negative data.

Medical Research Consequences

• Discussion on replication challenges in cancer research due to biased publishing practices.

• Case study on lorcainide drug trial revealing increased death rates but remained unpublished due to commercial reasons.

• Consequences of delayed detection of safety issues with anti-arrhythmic drugs leading to unnecessary deaths.

Evidence-Based Medicine Challenges

The importance of addressing publication bias in evidence-based medicine is emphasized through a personal experience with prescribing antidepressants based on limited published data.

Addressing Publication Bias

• Reflection on prescribing reboxetine based on limited positive studies without considering unpublished negative results.

Misleading Practices in Medical Research

The speaker discusses the issue of publication bias in medical research, highlighting how selective publishing can distort the perception of a drug's effectiveness.

Publication Bias in Medical Research

• Seven trials on reboxetine were conducted, but only the positive one was published, while six negative ones were left unpublished.

• Publication bias is a well-studied phenomenon where positive results are favored over negative ones.

• Studies on antidepressants approved by the FDA showed a 50-50 split between positive and negative trials.

• Despite an equal number of positive and negative trials, most negative trials remained unpublished compared to positive ones.

• Publication bias affects around half of all medical trials, with positive findings being twice as likely to be published.

Research Misconduct in Medicine

The speaker delves into the ethical implications of publication bias and its impact on evidence-based medicine.

Ethical Implications of Publication Bias

• Selective publishing distorts the true efficacy of treatments, akin to withholding coin toss results to fake outcomes.

• Withholding data from studies is considered research fraud; however, selectively publishing favorable results is often overlooked.

Challenges in Accessing Clinical Trial Data

The speaker highlights challenges faced when accessing complete clinical trial data for drugs like Tamiflu.

Accessing Complete Clinical Trial Data

• Governments spent billions stockpiling Tamiflu based on incomplete trial data regarding its efficacy in reducing influenza complications.

• Cochrane reviewers faced difficulties obtaining complete trial data for Tamiflu due to inconsistent and unavailable results.

Dissemination of Clinical Trial Data

In this segment, the speaker emphasizes the critical importance of having access to all information regarding clinical trials to understand the true effects of prescribed medicines. They highlight the need for transparency in publishing trial data conducted on humans.

Importance of Publishing All Clinical Trials

• "We cannot know the true effects of the medicines that we prescribe if we do not have access to all of the information."

• Emphasizes the necessity to publish all trials conducted in humans, including older trials.

• Criticizes the FDA Amendment Act for only requiring publication of trials conducted after 2008.