Data visualization and correct analysis for BUBBLE treatment

Hi everyone,

I am working at a major research institution in the US and couldn’t find a satisfying answer to my question about my research methodology. I want to compare certain type of treatment (hereinafter referred to as “BUBBLE”) with the risk of developing a rash. This treatment, BUBBLE, is used to treat depression in kiddos and is applied to the head. The individuals need to visit the clinic twice a week. However, based on their response, we can increase or decrease the frequency of their visits. In addition, their visits were disrupted due to the COVID pandemic in early 2020.

I have a list of 2 groups of patients: Patients who received BUBBLE and developed rash and patients who received BUBBLE but did not develop a rash.

I have the following information available:


  • Biological sex

  • Their index BUBBLE treatment date

  • List of dates for each of their visits for BUBBLE.

  • The dates when they first noticed rash

  • Concomitant medications right before they developed rash

-Depression scores here and there during their treatment course

-Comorbid medical conditions

-Comorbid psychiatric conditions

  • Photos of rash

We have several questions:

1- Does BUBBLE increase the risk of developing a rash?

2- Does the frequency or number of BUBBLE treatments increase the risk of rash?

3- Is there a subgroup of patients who are more likely to develop a rash with BUBBLE?

What would be a good methodology in this particular case?

This is a binary outcome as a response or dependent variable with treatment or dependent variables including categories and scores, which might be continuous or categorical—can’t tell. The first method that should be applied is logistic regression. I have a brief introduction using a coronary heart disease example. The site at the time of posting the site has an issue with its certificate, so you might need to visit the unsecured version

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.