Week 2

Hi blog readers!
The week started off great because I was able to see yet another patient, still learning new things along the way. I also got this very official looking badge made for me! Sorry that the picture isn't clearer since I took it from my phone. The badge doesn't change much other than the path I take to get to my work area. However, I'm still at the point where I get super excited to scan my badge to get passed all the locked doors.

My very own badge!

Another important development: I was able to start data collection this week!  I now have access to a computer, and will be collecting data for at least the next few weeks. I can't show you what it looks like to navigate through patient medical records to find the data I'm looking for since it has some classified information but I am allowed to show you what my data looked like. Here is a quick snapshot!

Click on this picture to enlarge it!

I think the spreadsheet itself is pretty self explanatory. On it, is the height of the patient, and weights at various stages around the operation. I used Excel to add one hundred days to the date of the transplant so it was easier for me when I was looking through the records. It might look pretty and fun because it's all colorful, but every time there is a color, there is something wrong. I'll briefly discuss some of the majors problems I've faced so far.

Problem #1: Missing Data 

There are many times where the height/weight of a patient was simply not measured or recorded in the patient record upon admission. There have also been times when a patient has recovered rapidly after the transplant and the cancer is in remission. Although a visit back to the hospital one year after the transplant is recommended, it is not mandatory if that patient's health is in good condition. As a result, one year weight is sometimes missing.  This is all missing data I would need to calculate weight loss in terms of BMI. Missing data is signified by an "X" in the spreadsheet.

Problem #2: Early Patient Death

What happens if a patient died prior to their one hundred day or one year checkup? How would we evaluate weight loss then? This is signified by a "D" in the spreadsheet. 

Problem #3: Data Outside of Time Parameter

Patients don't always come exactly one hundred days after transplant or exactly one year after transplant. Therefore, prior to data collection we set some time parameters. Pre-transplant weight could be measured on the day of the transplant or two weeks prior. One hundred day weight could be two weeks before or after the hundredth day (so 86-114 days after the transplant). One year weight could be 11-13 months after the transplant. Obviously, we would always take the closest date to the time we wanted, but there were many times where the closest date did not fall within the time parameter we set. 

Problem #4: Patient Gets Multiple Transplants

Obviously, this would be a major factor when it comes to the reliability of our study population. Patients who get multiple transplants would be in a far worse condition because it means either the first transplant failed or the cancer has relapsed. We should expect that their weight loss would be much more severe and this would skew the overall survival rates, cancer relapse rates, etc.

Conclusion

It's really hard to deal with the issues without collecting all the data first. The goal would be to make the data as clean as possible, but taking out too many patients would make the results unreliable due to a small sample size. I can't resolve these issues until my data collection is complete.

Another thing I wanted to do was spice up the blog a bit by sharing a bit more of the information I've learned about my topic. As a result, I'd like to add a new feature to my blog: a drug of the week! I don't know how I'll format this in yet but I'll have it up sometime by the end of week 3. As a preview, I will reveal the drug of the week for weeks 1 and 2.

Week 1: Prograf

Week 2: Sprycel

If you guys have any suggestions for things you would want added to the blog please post it in the comments!