Sara co-leads MED International, an interdisciplinary team developing IT and engineering solutions for resource-limited hospitals. She is a sophomore concentrating in Engineering at Brown. 

We navigated our way through the sprawling city of Accra, Ghana via tro-tro, a privately owned public transport vehicle with no signage or fixed schedules, which leaves only when the van has exceeded maximum capacity. During my summer with MED International, we visited 40+ hospitals across Tanzania and Ghana. Four other Brown University students and I toured and inventoried ‘medical equipment graveyards’ in the hospitals we visited—rooms filled with medical equipment that was no longer functional. By meeting with biomedical equipment technicians, facilities managers, hospital directors, and technical directors, we gained a better understanding of each hospital’s challenges and the overall state of the health care system in Ghana.

Broken medical equipment hurts everyone. Patients lose access to needed medical care, and hospital earnings are diminished as they spend more and more on maintenance and repairs. Hospitals in low-resource settings have an abundance of broken medical equipment, much of which only requires a spare part or simple fix. However, these hospitals lack access to trained technicians and management tools. MED International looks to maximize the life-saving potential of a hospital’s existing inventory using IT and engineering solutions. The software we have developed tracks medical equipment through its lifecycle in order to minimize the downtime and financial output of operating the machines.

At nearly every hospital, the staff and doctors were excited to use the new software within their facilities and pointed to specific ways these technical solutions would benefit their hospitals. The positive response was very encouraging. However, as we began our umpteenth tro-tro trip stopping at hospitals throughout Accra, things moved more slowly than we had hoped. Two major obstacles we encountered:

1. The hospitals’ inability to pay – some hospitals simply do not have the means to afford the software. In forming the business plan for our organization, we had originally agreed that if a hospital was not willing or able to put money towards purchasing the software, they probably wouldn’t be likely to use it. Our hope was that once they made the initial investment, revenue from the software would enable the organization’s sustainability and adequate customer support for hospitals. In reality, we found that because many hospitals could not afford the initial software, they also would not be able to benefit from the long-term revenue we had anticipated, requiring us to rethink our funding model.
    
2. Inadequate technical infrastructure – we observed a huge variation in technical literacy and capabilities across hospitals. This ranged from hospitals with just a single computer for the hospital administrator to hospitals with full in-house IT departments. These technical constraints limited which hospitals were feasible customers for the software we were offering.

What we learned: Local fieldwork, customer research, observation, and understanding—not just product usability and quality testing—are critical to success.

While we tested the product among users at Brown—a predominantly computer literate crowd—and on Brown’s Wi-Fi, we were not able to test the product with our target users and in the actual, developing world climate. If we could have defined a comparable population and environment beforehand, we would have been able to shape the product to better meet the needs of hospitals in Sub-Saharan Africa. We hired a biomedical engineer, Kevin, to start using an early version of the software while piloting a technician training program in Zanzibar. It was not enough to translate the software to a local dialect and make it simple enough for a beginning computer user. The software also needed to function in resource limited settings, where power and Internet were no guarantee.

While at Brown, it was difficult to communicate with hospitals across the world. Phone calls and emails rarely elicited responses. While we spoke to several advisors familiar with the region, we did not truly get a sense of the local climate for hospital management software.

Because of these new insights, we have decided to open-source the software developed by computer science students at Brown, which means the source code is freely and publicly available for developers to modify and hospitals to use. Hospitals who would like to customize the medical inventory software and integrate it with their other management software can do so. In our experience, the private hospitals were more likely to be able to afford inventory management software, yet it was the large, public hospitals that would most benefit, and lead to greater impact. As a result, we have decided to transform the software we developed in Rhode Island into an open-source web application with the hopes that we can generate the most impact in the communities for which it was intended.

Summer on-site research was essential in uncovering the roots of several technical challenges hospitals face in Sub-Saharan Africa. For the software to be effective, we need to respond to the realities of the technology available to hospitals in resource-limited settings.