In Rwanda, Craig Spencer's Data Collection Project Fuels Innovation at the Point of Care

Published on May 13, 2014
Rwandan fellowship project gives insight into the value for reliable data and the need for experienced local partners.

Craig Spencer, Survey training in Burundi

Around the world, International Emergency Medicine (IEM) Fellowship programs provide unique opportunities for research and evaluation of emergency care systems. Last year, EP Craig Spencer, an IEM fellow from Columbia University in New York City, completed multiple projects in East Africa including a teaching curriculum and a patient monitoring program. The work gave Spencer an inside look at the importance of gathering reliable data, and of working with knowledgeable local partners.

A large part of Spencer’s work in Burundi was the routine care of sick patients, which was as eye-opening as it was frustrating. He was presented with a patient population where malnutrition caused daily complications and seemingly routine health concerns turned into emergencies because of a lack of supplies and follow-up. Spencer worked alongside NGO ‘Village Health Works’ (VHW) founded by another of Columbia’s 2013 IEM fellows, Dziwe Ntaba. The VHW clinic was one of the few places in South Burundi with access to oxygen. As a result it was often inundated with patents, forcing physicians to prioritize patient care.

“Many clinics would send their patients with pneumonia or anemia to the VHW clinic for oxygen. We only had a few canisters, so in the US five patients would have been given oxygen, a facemask or even intubated or on life support. In Burundi you had to choose which two of the same five patients needed it most. In one case we gave oxygen to a child with pneumonia over a woman who was very anemic from malaria.”

Spencer learned that in such difficult cases it was the experience of the Burundian and Congolese doctors that made all the difference.

“The physicians in Burundi are some of the best I’ve ever come across, I couldn’t have triaged and treated patients as well,” he says. “I don’t think I would have had the same success in deciding who needed oxygen to survive and who didn’t. But they had experience. They’d seen it before.”

While Spencer was able to introduce cutting edge empirical medical knowledge to many of his East African counterparts – in particular emergencies arising from long-term illnesses such as diabetes or stroke – he found that without adequate medical supplies, his impact was limited. “You can make a diagnosis but you can’t do much with it,” he says. Sometimes there was nothing to be done for stroke patients. “Sometimes even aspirin just weren’t available.”

Intore war dance

Surveying on the Ground

In addition to Spencer’s clinical work, he worked alongside renowned epidemiologist Les Roberts and VHW’s Ntaba to conduct an epidemiological fieldwork survey in South Burundi. He made a startling find which led him to an unorthodox conclusion.

“It was a household survey,” he explains. “We went into 30 different geographical locations chosen proportional to size, but otherwise randomly selected.”

After conducting epidemiology work in the eastern DRC, Roberts found that more children were dying from uvulectomy complications than from war-related conditions, so he proposed including a similar assessment in South Burundi.

“In addition to asking questions concerning mortality rate, we asked about uvulectomy. The practice is still active there but what we found was staggering – over 85% of children surveyed had had a voluntary, ‘therapeutic’ uvulectomy.”

Furthermore, patients who had undergone the procedure themselves were denying subjecting their children to the practice when questioned among procedural weight, vaccination and HIV discussions at VHW’s clinic discussions.

“It was strange,” says Spencer. “People were willing to accept vaccinations ... but they were unwilling to give up uvulectomy – a potentially dangerous and fatal procedure. They knew that they shouldn’t be doing it but they did it anyway.”

Though he admits the idea isn’t an appealing one, Spencer tentatively questioned whether western medicine may be better being incrementally introduced in the region.

“Would it make more sense for us to do uvulectomies in safe, sterile environments where people don’t have to pay? Would we be better off doing ‘A’ to get to ‘B’ rather than just heading to ‘B’ directly?”

“Long term change is going to be dependent on outside institutions being pragmatic for on-the-ground solutions. Future IEM success will depend on community education, covering topics like domestic violence, uvulectomies and HIV. For example, a lot of people thought seizures were contagious.”

Emphasizing the importance of surveys and local knowledge, Spencer also learned to question existing data.

“During a birth survey in the DRC we asked women where they gave birth. They usually said a clinic or hospital as that is the law. The numbers didn’t seem to add up so we went to various clinics and asked to see birth registers, finding they were over-reported. So, we established an ongoing surveillance system for continuous data collection in addition to the one-off survey, using a local monitor from the community. Then we triangulated the data with model predictions to get more accurate, 3D picture.”

Uvula assessment

NGO Benchmarking

Spencer believes it is not only patients who can give inaccurate information, expressing concern about the potential implications of health NGOs and initiatives in IEM after his experiences in East Africa.

“There needs to be a measure and proof that a program is working and serving the population, not only for financial donors . . . Without someone on-the-ground who knows the area and its people collecting information, the likelihood that you are going to do more harm than good is pretty high.”

e.g. Global, Research, or India
e.g. Features, Opinion, or Research
Browse by Category
    Most Popular
      Download Latest Issue
      Issue 19