We finally finished working on the back end and manual for DataPall earlier this week! Since the staff trainings are next week, we’ve taken the week to explore the hospital more. On Monday, I got to poke around the hospital lab a bit. I’m pretty sure they thought I was crazy for taking so many pictures, but the technicians were eager to talk to me about what they like and don’t like about some of their equipment.
Before I came here, previous interns told me that St. Gabriel’s has a well-equipped lab for a private hospital of its size. Even still, early on in our time here I was a little surprised to hear in morning report that the hospital was monitoring blood sugar and hemoglobin levels for some of its patients. Two of the most exciting projects that I’ve worked on as part of BTB (yes, I’m biased) are the HemoSpec and its bilirubin analog, and both are based on the need for low-cost point of care blood tests in developing-world settings; I had hoped that St. Gabriel’s would be a good place to see the kind of impact those technologies could have.
I wasn’t disappointed for too long. When I visited the hospital lab this week, I realized that while the lab is indeed fairly well-equipped for a private hospital in terms of what it can test, it has to scramble to make up for what it lacks in how many samples it can test. The hospital has a lot of donated lab equipment, but usually it only has one of a type. Particularly with older machines, the equipment they get often only runs a few samples at a time. Medical wisdom in the States, as far as I know, is that the surer you can be of your diagnosis, the better—therefore, the more tests and the more timepoints, the better. Here, though, real-time patient monitoring is incredibly difficult because of the backlog in the lab. You might only get one shot at getting a patient’s blood sugar level, which doesn’t really tell you much about if your treatment is working.
Before I left, I thought what defined the HemoSpec was its cheapness: it’s designed to replace the expensive, currently used plastic cuvettes with cheap paper ones. When I got here, though, I realized that “expensive” and “currently used” are oxymoronic. They only use the quicker hemoglobin-only plastic cuvette-based spec when the power is out in the lab or if they have a blood donor. Otherwise, they just run the whole panel every time. That means that a simple quantitative check for anemia could easily take long enough for the patient to decide to give up and go to a traditional healer. Most of the providers here have to rely instead on checking the color of the conjunctiva—red, healthy; pale, anemic. I’ve yet to see them find red conjuctiva. The Hemocue is small enough to fit in a doctor’s bag or a generous white coat pocket; what it stands to offer St. Gabriel’s is not a new metric, but a much more dynamic and therefore useful one.
The problem with reagents is they have to be restocked.
The bilirubin analog is much earlier in the design process, but I think the single-question approach can offer the same benefits on the wards here. One of the design questions our team faced was whether centrifuges were available at hospitals in the developing world, but I was pleased to note that St. Gabriel’s actually has 4, two in the main lab and two in the blood donation room.
On the far right: the Dremofuge
I was even more pleased when I saw the fourth one, stationed in the blood donation room. The technician seemed to especially want to show me how this one worked: a wooden base, plastic shield, and battery pack, looked like a BTB device. I asked him the same question I’d asked about every device in the lab: Do you use it? He emphatically agreed. The one next to it, he said, had a broken brake, and this one worked just as well even when the power was out.
Yes, that’s a floppy disk drive on the bottom.
The device the techs wanted to spend the most time talking to me about, though, was the CD4+ counter. CD4+ counting is very important for HIV/AIDS initiatives, and it’s a known problem in the developing world. Even with all its donation connections, though the hospital still only has one running CD4+ counter, and can only run one sample at a time. In busy times, it’s impossible to keep up with the demand. The outpatient HIV clinic has to rely on counting how many pills the patients have left, which is unreliable and imprecise. They are very envious, they said, of hospitals with machines that can run more than one sample at a time.
In a “well-equipped” lab like this one, BTB can’t provide new diagnostic capabilities. What we can do, though, is make what metrics they have more robust and clinically relevant. It’s not about taking clearer or fancier pictures, it’s about putting the camera in a clinician’s pocket so that it’ll actually get used.