BLOG: Dissecting DREAM, part 3: Wanted: an effective placebo

ByDarrell E. White, MD

Placebo: a substance or treatment of no intended therapeutic value. It can be made to resemble an active treatment or substance so that it functions as a control in scientific studies.

Much has been made about the choice of olive oil as the placebo or control in DREAM, and rightly so. Those who designed the study went to great lengths to disguise the placebo so that neither the study subjects nor any of the participating eye doctors would be able to tell which pills were active study agents and which were the controls. The pills were identical in size and shape, and were given a lemon flavor additive to further mask any differences. As a final precaution, the pills were shipped directly to the homes of the individuals taking part in the study and were never seen by the investigators.

If only as much thought had been given to the choice of what was actually going into those pills.

When you are designing a study, your first priority is to be clear on what you are evaluating. DREAM designers felt they were putting together a study that would evaluate “real-world” dry eye care. Studies such as this need impeccable statistical chops. There are several stats wonks looking at this in DREAM, and I will report their findings when available. If you, the authors, wish to publish a study that you hope will change how we, the readers, address a certain disease, it is my strong opinion that anticipated objections to your findings, positive or negative, should be considered during the development phase. What you choose as your placebo control is one of these factors.

Sheppard’s HydroEye study was already published when DREAM was taking shape. No one batted an eyelash at their choice of placebo, sunflower oil. Two massive cardiac studies looking at the effect of EPA on survival in otherwise fully treated heart disease, at least one of which was funded by NIH, were also enrolling and launching during the DREAM design period. REDUCE-IT used mineral oil as a control, and VITAL used olive oil. Were the DREAM authors aware of either? If so, was this awareness a factor in choosing olive oil? It is interesting to note that REDUCE-IT met its primary endpoint (overall reduction in mortality) as well as several secondary endpoints. VITAL, on the other hand, did not meet a similar primary endpoint (although several subgroups did).

Does olive oil affect the symptoms of moderate to severe dry eye disease? Heck if I know. Heck if the DREAM (or VITAL) authors do, either. There is certainly a sense in the wider eye and general medical community that olive oil has multiple benefits. What I do know is that by choosing olive oil as the placebo, the DREAM authors opened their study to criticism that was avoidable. When the next study is done — and you know there will be a next study — let’s hope that this particular shot to the shoe is avoided by choosing sunflower (Sheppard), mineral (REDUCE-IT) or linoleic (Epitropoulos) oil so that we can move on.

There might actually be a silver lining here. I’ve long held that it is nearly impossible to treat vegans who have meibomian gland dysfunction and resulting evaporative dry eye because it’s so difficult to get any effective oil into the mix. Starting today, they all get a couple of tablespoons of olive oil as part of their basic treatment.

Disclosure: White reports he is a consultant to Allergan, Shire, Sun, Kala, Ocular Science, Rendia, TearLab, Eyevance and Omeros; is a speaker for Shire, Allergan, Omeros and Sun; and has an ownership interest in Ocular Science and Eyevance.