Q: How should I answer regarding selection bias comment?
In my paper, I said that patients were divided into one of the three choledochorraphy techniques based on surgeon preference in charge of the procedure. The reviewer's comment was "The study design is prospective, although non randomized, but there is an important selection bias in the study analysis". How do I respond to this?
Thank you for your question. We will try to answer your question by providing different scenarios explaining how selection bias can be a potential limitation in studies and by mentioning a solution on how to overcome the same:
Scenario 1: During patient recruitment, some patients are less likely to enter a study whereas others are more likely. Such a sample is therefore not representative of the population in which this research is conducted. Subjects less likely to enter the study will be under-represented and those more likely to enter the study will be over-represented relative to others in the general population, to which the conclusions of your study are to generally applicable to. This is what is referred to as “selection bias.”
Scenario 2: Selection bias can occur when people volunteer to participate in a study. Those who choose to join may share a characteristic that makes them different from non-participants. Consider a study to assess a program for improving the eating habits of shift workers. You put up posters where many people work night shifts and invite them to participate. However, those who sign up may be very different from those who don’t. They may be more health conscious, which is why they were interested in a program to improve eating habits. If this was the case, one cannot conclude that the program was effective because the health of those who took part in the program was better than that of those who did not.
Solution: To minimize selection bias in your study, patients can be randomly assigned to the study groups. If this is not possible, inform the reviewer why the study design included selection bias and mention to the reviewer why the distribution could not be random.
We hope this helps you respond to your reviewer’s comments.