I have not had any experience yet on clinical studies but I would like to develop a little more on blinded clinical studies. In this type of study, subjects are blind to which treatment is given. There are three types of blind clinical studies,
1) Single Blind- The subject, but not the observer, does not know which of the possible treatments he is receiving.
2) Double Blind- Trial in which neither the subject nor the observer know which treatment is being administered.
3) Triple blind clinical studies - the participating subject, the observer-researcher and the researcher who analyzes the data do not know which treatment is being received.
These types of studies help control biased results. In a single study, the advantage is that by keeping the information secret, you can avoid any potential biased reactions or responses from the subject. Nevertheless, the observer runs a risk of consciously or subconsciously affecting the subject’s responses. This is why double blind is more beneficial. Double blind studies are the most convincing research design, in which randomly assigning the intervention can eliminate the influence of unknown or confounding variables that may lead to biased and incorrect estimate of treatment effect. A triple blind study are probably the most thorough about eliminations biased results. If the subject, observer and scientist do not know what the study is testing then the true and raw results can be determined thus avoiding omitting results because of biased experiments. Nevertheless, triple blind studies can be potentially dangerous when testing medical devices or pharmaceuticals.

I've been ok an investor double blind study ( data was collected in real time and we didn't know who had placebo and who didn't). On the scientist side of things it's exciting to have samples run it and see natural true results, it's a good feel to see that something is working. I remember my coworkers and I wouldn't constantly try to guess which sample belonged in which group. It's scary when we get negative result because it can be good or bad thing depending on the group and drug. It was fun anxiety if that is a thing. When everything was done, we were unblinded and everything went well.

Recently, I was involved in a single blind user study at work that involved our production parts and an R&D part that will be available on the market within the next year or two. As Dr. Simon mentioned in his lecture, the study involved a concealed identity of a product—in this case it was 3 prototypes of the new product—and a known identity of a product—the company’s production part that is already on the market. We are looking to improve the production part, and thus, for this user study, nurses and pharmacy technicians were asked to come in to participate and give their feedback on all of the devices that were given to them. The person collecting the data knew the differences between the prototypes, while the users did not and this resulted in an unbiased study and honest feedback. The data that was collected by the observer helped the team make final decisions on which prototype would be better than the current production part and would also meet the customer needs. Unfortunately, I have not been part of a single study that involved any sort of drug, but being a part of a R&D user study in a medical device company was extremely exciting and an unforgettable experience.

I’ve been involved in double blind and randomized studies. In the randomized study, we actually ran into an issue where the trial was cancelled. For this study, active ingredient material was administered to some patients and placebo to others. The problem was that the placebo had a very minor difference to the active material and when patients compared medication, they were able to identify who was in the control group and who was in the experimental group, creating a bias. The study ended up being repeated, with different placebo material, but not before a great deal of money was lost and paperwork was completed to amend the changes.

Hi mjf34,

That is a really interesting example. I have always wondered what the negative outcome would be of attempting to complete a double blind study. However your example of the group becoming bias based on the experimental group and placebo group, ruining the double blind as well. I have always thought this would be the best type of study; however seeing the negatives shows how careful consideration must be done before conducting a trial.

-Andrew Nashed

At the lab where I used to work, we had both prospective clinical study which is collecting data in real time or retrospective which is using old data to be analyzed. The way the studies were organized were so that the previous study can build the groundwork for the next study either as pilot data for a new grant or just a side dataset contributing to the next project. Doing it this way is more efficient in terms of time. In many cases, more data is always better than less data because we can (most of the time) analyze it in the future which could be helpful in a grant writing or writing a journal paper.

The type of clinical trial my group is involved with is a prospective clinical study which is collecting data in real time. We are 20 locations that enroll at least 10 qualified patients into the trail with a max of 3 roll-ins at each sight. We collect data the day after the operation, 30, 60, 90, 120, and 360 days after the operation of the patients. This gives us real live data to see how our device is doing.

Example: Placebos Study
Placebos are necessary for studies and clinical trials. When doing statistical analysis, you’ll have a controlled-group that’ll receive the drug or device in question for actual treatment and perhaps another group that’ll receive a type of “sugar pill” or something else that won’t treat the illness or condition, but the patient or subject is led to believe that it will do something. It is interesting to note in an experiment where patients are given a placebo, whether they are told it is fake or not, a given percentage will still have positive physiological responses. Placebos have their appropriate times to be used while other times it should not be used at all.
https://www.ncbi.nlm.nih.gov/pmc/articles