Science starts with counting.

Control central for Bhutan's national cancer registry

 

 

I wouldn't normally be confused with an IT expert, and in fact in most cases I'm the one asking for help.  But earlier this week I had a chance to test my skills on the brand new National Cancer Registry.  Just started 6 months ago, there are over 450 patients logged.  The data at this point is little more than tumor site, date of diagnosis and some demographics, but if science starts with counting then this little room in the hospital basement is where counting starts.  The one and only resgistrar painstakingly copies material from scrawled paper charts onto a data sheet, and then enters the codes into the computer.  I sat with her for over an hour, trying to figure out how extract and filter the data--so far no one had actually tried to produce a report.  I alternated between the help file and trial and error, back and forth for over an hour.  Picture Han Solo steering the Millenium Falcon with smoke coming out of the dash...  Finally, I hit the right combination and Viola!!  A list of all 40 breast cancer patients!  Fist pump!  I made a cheat sheet for the registrar, and even though she speaks almost no English she had been watching me closely and with just a few hints she was able to duplicate the process. 

Medical Records

 

Piles of data to enter.  Dr. Tashi Dendup and the regstrar contemplate the challenge

 

 

Later in the week I returned to the catacombs with the HOD (Head of Department), Bhutan's only Surgical Oncologist, Dr. Tashi Dendup.  With him driving at the console I led him through the steps, and then we quickly expanded our tricks to multiple filters, excel spreadsheets, etc.  With yellow-jacketed paper charts stacked four feet high behind us we were able to glean information without turning a page.  The reports at this point only give a hint of information, but also serve as a way to do quality control.  For example, less than 10% of the records include tumor staging, critical to determining progress or planning interventions.  Also, there is no follow up data of any kind.  We brainstormed possible ways to use the national census data or death records to at least document year by year whether patients are dead or alive.  The registrar will need training to find staging information, but this can be done.   Hint:  the nice neat charts on cancer survival in Bhutan that you might see on the WHO website are just extrapolations from data in India.  Soon, we can do better.

Recycled bottles line floor to ceiling shelves in pathology.