DAY TO DAY

Before coming here we were all uncertain what our day-to-day life would be like. What we did know was that we had 500 charts of patient’s who had a chest tube (ICD) placed from GF Jooste hospital to review, those charts had to be cross checked with the tuberculosis (TB) database, all the data had to be inputted into the computer, and a statistical analysis of the data would be done.

Let me explain now how that has played out. The culture here is either very trusting or very lax, which ever you prefer! And all the records are paper copies, nothing is electronic. When a patient comes into the Casualty (ER) they go to the admission window and get a folder handed to them with preprinted labels. They bring that folder with them into the casualty room, which is an open pit. There is supposed to be a clerk by the door ‘checking’ patients in from admission but most of the time no one is there. So the patient comes up to anyone they see who looks like they work there and hand them their chart. We have had charts handed to us many times while we are doing chart reviews (naturally anyone reading a chart should in fact be a doctor). The patient finally finds a nurse who will do intake vitals on them and get them situated in the casualty. The nurse then registers the patient into the casualty registry book, a hand written log with patient name, patient folder number, reason for visit, and doctor assigned to the patient. Patients are then seen by the casualty doctor for treatment and referrals to an in-patient ward or other hospital.

Patients who need an ICD put in are sent to the surgical ward (10 feet away) and are registered into the surgical book. A chest tube is placed when there is risk or evidence of a collapsed lung from trauma. For our project we are only looking at penetrating trauma requiring an ICD, like that of a stabbing or gun shot wound.

Step one of our project was then going through the surgical registry books looking for all patients with an ICD placement. Our study protocol called for any patients with an ICD from 2004-present. From two registry books (the size of a very large yearbook) we had over 500 ICD’s. We made one complete list of all 500 of the patient folder numbers associated with the ICD patients. These happened to be all from 2006 and 2007.

Step two was going to the records department and pulling our charts. The first week or so we did it ourselves. I explained myself once the first day to one of the staff and they were okay with me coming in there. After we did two overnighters however we were told we must call the Director of Records and explain who we were and what we were doing. After that phone call we set up a wonderful system of the staff pulling 50 charts at night and 50 charts in the morning and just keeping them stacked up for us. (There is a sign saying “These charts are for a research project for doctors”.  That reminds me, in order to get access I am registered as a ‘visiting medical student’. I have a student ID card and everything. Who knew you could get a medical degree as a student at DU!) For the last 4 weeks that is what we have been doing. We had pulled 448 charts of 500 as of this Monday.

The first week or so we did our chart review at a tiny coffee table sized table in the records office. There were no chairs so we would either stand or end up sitting on the tile floor (not the cleanest!). When that got too much we moved out the casualty room and squeezed onto the side of the table in the center. There wasn’t’ ever a really conducive place to sprawl out with our folders. Through a series of events we eventually found the research resource room that has a conference room sized table. We have happily been using this room ever since.

Step three was cross-referencing our data with the electronic TB registry called ETR.net, which we have been trying to get access to since day one. That has been quite the fiasco and we have been given a bit of run around with it. In the mean time we have also been hearing varied stories of the reliability of TB data at Jooste. Apparently TB tracking has only really been happening within the last year - year and a half. Even then it also has low compliance because it relies on the doctors calling the research center and referring a patient for TB testing/counseling/treatment. Of all the hospitals in the Western Cape, Jooste has the poorest record for TB compliance.

On Tuesday we finally spoke with a women at the Department of Health who could get us access to ETR.net, kind of. We can’t access it ourselves as we thought, but we can give her a list of all the patients we need and she can pull a master report for us. That was time saving news to here. Also on Tuesday, through another source, we ultimately came to realize that a very, very small percentage of our ICD patients from 2006-2007 might in fact be in ETR.net due to the lack of records. Without the cross referencing capability there would be no way of confirming TB and/or HIV in our patients. Thus we could not test our study hypothesis that TB and/or HIV patients experience greater morbidity and complications associated with ICD placement. Not good news to hear after 4 weeks of work! But thankfully our team is pretty resilient and dedicated to this project.

We decided to recreate our list of ICD patients using only patients from 2008-2009, when ETR.net was ‘regularly’ used. We created a new master list of 500 patients, dropped it off at records Wednesday and started “Project #2”, as we are now calling it. We have a week left in our project/internship. It’s crazy to think about restarting the project that had taken us 4 weeks before, with only 1 week left. However there are some great learning lessons in this. We definitely feel that we worked out a lot of kinks in Project #1 including consistent data collection on our part, and learning more about the charts and thus adding a few items to our data collection forms for Project #2.  As our fearless leader back in the states said, “We have perfected this to a science”! Two eight-hour days with only tea breaks have already give us 200 charts.

Step four is data entry, the fun stuff. We just finished inputting the data from Project #1 last night. We are focusing on getting the charts reviewed now given our time crunch and then will do data entry for Project #2 next week. Then will turn around and give a report of patients to our Department of Health contact of all our patients and see which ones exist in ETR.net. All by next FridayJ

Step five will be the data analysis and I’ll do that back in the states. We are all every interested to see what results we get. Even looking at our current data regardless of TB status- see the ratio of men to women, mean age, whether there are more complications from ICD’s placed at the day hospitals than at Jooste, etc. And because we have Project #1 and Project #2 with two distinct time frames we can also look at differences over time. So all is not wasted here and we are all about slaphappy with exhaustion over it.