PHLB09H3 Chap.9 Research Ethics and Clinical Trials
Research Ethics and Clinical Trials
Research ethics in week ten
Week ten covers research ethics alongside the second consultation simulation. The readings include Freedman on equipoise and the ethics of clinical research, Savulescu on how ethics committees should weigh harm after a fatal gene therapy trial, a podcast on male birth control and a news report on controversial genetic research.
Savulescu's paper anchors this chapter because it supplies a method you can apply to cases.
The Gelsinger case
Jesse Gelsinger, eighteen and living with a mild, controllable form of a metabolic disorder, died in 1999 after receiving an experimental gene therapy in an early-phase trial that offered participants no expected benefit.
The trial had enrolled consenting adults rather than newborns with the lethal form of the disease, who could not consent.
Expected harm before consent
Savulescu argues that ethics committees have two duties: to ensure that the expected harm of participating is reasonable and to obtain valid consent. Expected harm multiplies how likely a harm is by how large it would be.
Because a dying newborn would lose far less than a healthy young adult, he concludes that, if the trial was justified at all, it should have enrolled the newborns, and that committees erred by giving consent more weight than harm.
What this chapter covers
- 01
Equipoise
- 02
The Gelsinger Trial
- 03
Expected Harm
- 04
Quality-Adjusted Life Years
- 05
Role of Ethics Committees
- 06
Consent and Disclosure in Research
Worked example · free
Compute and compare expected harm
- 2Compute the size of the loss for each group.
- 2Multiply by the probability to get expected harm.
- 2Apply the principle and note what it leaves open.
Key terms
- Equipoise
- Equipoise is genuine uncertainty about which of the treatments being compared in a trial is better.
- Quality-Adjusted Life Year
- A quality-adjusted life year is a measure of health benefit that scores a year of perfect health as one and death as zero.
- Phase 1 Trial
- A phase 1 trial is an early study testing safety and dosing, in which, as in the Gelsinger case, no benefit to participants may be expected.
Research Ethics and Clinical Trials FAQ
Why does Savulescu think the Gelsinger trial enrolled the wrong people?
The committee preferred consenting adults with the mild form over newborns with the lethal form, who could not consent. Savulescu argues that the newborns, already dying, would lose almost nothing if the trial killed them, while a healthy young adult would lose decades of life. Their expected harm was therefore far lower, and choosing more harm over less because of consent is irrational.
How is expected harm calculated?
To find expected harm, multiply the chance of the harm by its size. Savulescu measures size in quality-adjusted life years. For Gelsinger he uses a quality of 0.8 over about 50 remaining years, giving a possible loss of 40, and a risk of death of one in ten thousand, so the expected harm is 0.004 of a quality-adjusted life year.
Does Savulescu think consent does not matter?
No. He says ethics review has two goals: ensuring the expected harm of participation is reasonable and obtaining valid consent. His complaint is that committees increasingly treat consent as the main protection and neglect harm. He also argues that fuller disclosure would probably not have saved Gelsinger, because volunteers would have enrolled anyway.
What went wrong in the Gelsinger trial besides the choice of participants?
According to the reading, investigations and a lawsuit raised several concerns: questions about whether he met the eligibility criteria, financial conflicts of interest held by the lead researcher, the omission of animal deaths from the final consent form, and the claim that he was led to expect a benefit from a trial in which none was envisaged.
Exam move
Learn the formula for expected harm and practise it on three invented trials with different populations, writing every step in quality-adjusted life years. Summarise the Gelsinger case in five facts you could state in a quiz answer.
Then write a consultation note on a trial design choice: pose the ethics question as a choice between populations, compare expected harm, address consent and disclosure, and answer an objection from the consent side.
