Splash image

Millennium Villages Project

Gordon Grand Fellow

After graduating from college in 2010, I spent a year in East Africa working with the Millennium Villages Project, an international non-governmental organization focused on achieving the Millennium Development Goals in a series of sites across sub-Saharan Africa. My goal for the year was to help build and deploy an electronic patient tracking system, called ChildCount, which aimed to support the work of community health workers. I spent the majority of my time in the town of Mbarara in southwestern Uganda, which borders the Millennium Villages' Ruhiira site.

Ruhiira's community health workers and health managers. Back Row (from left): Rose Kyomuhangi, Robinah Uwizeye, Elisaf Muriisa, Harriet Twongeire, Clare Ayebare, Chrisestome Muhereza, Edward Kadogo Bananuka, Emmanuel Toko, John Asiimwe Museveni, Dr. Emmanuel Atuhairwe, Joshua Arinaitwe, Esther Musenero, Geraldine Tusiime. Kneeling (from left): Winnie Kemigisha Ka-Office, Shallon Tusiime, Henry Corrigan-Gibbs, Annet Katungi. Not pictured: Pamela Besigye, Anita Katusiime, Michael Mugisha, Patricia Namakula.

The purpose of the ChildCount system was to keep basic health information on every child under five years of age and on every woman in the Project's catchment area. For children, the system tracked whether the child had received all of his or her immunizations on schedule, whether the child had been screened for malnutrition recently, and whether the child needed a follow-up visit from a community health worker (after the child was treated for malaria, for example). For women, the system tracked the woman's ante-natal care visits and whether the woman had been offered family planning services by the Project's health staff. By tracking women and children, the Project hoped to improve the quality of care provided to these high-risk populations to avoid preventable deaths.

When I arrived in Uganda in August 2010, the ChildCount patient tracking software was partially functional but lacked many of the features that would make it useful to clinicians and community health workers. For example, there was an interface that allowed health workers and data clerks to input health information into the system's database but there was no means for health workers to information from the system. My first task, then, was to work with the Project's doctors, community health workers, and data clerks to design a system for extracting printed weekly, monthly, and annual health indicator reports from the system.

Ruhiira Health Center. Photo by Dr. Maya Arii.

After getting the reporting system up and running, I worked with the Project's health systems manager to organize monthly meetings with the community health workers at each of the Project's six health centers. During the first rounds of meetings, we distributed the ChildCount reports to the health workers and explained how the system calculated each of the key health indicators. Once the community health workers became familiar with the format of the reports, we used the meetings as an opportunity to identify and discuss areas of healthcare delivery that needed improvement. These discussions helped the community health workers to identify gaps in their own knowledge (e.g., a few health workers did not know how often women were supposed to see a midwife for ante-natal care) and the discussions also helped me understand how to shape the ChildCount reports to better serve the health workers' needs.

In parallel with this process, I negotiated with Uganda's leading mobile phone operators to secure low-cost SMS services for our ChildCount deployment. Although the Millennium Villages Project had imagined that community health workers would interact with the ChildCount system via structured text messages, when I arrived in Uganda, the Project had not been able to find a mobile phone company that was willing to provide SMS services for the ChildCount pilot. As a result, community health workers filled out their patients' health information on paper forms, Project cars carried these forms to a field office, and data-entry clerks would key the forms into a Web interface. Getting the SMS service up and running would allow the ChildCount server to send automated reminders to health workers about patients who needed care and would allow the health workers to interact with the ChildCount database in real time.

Over the course of my time in Uganda, I led negotiations with four mobile operators and two local SMS service providers. Half-way through my year in Uganda we signed a contract with a local phone company for a SMS services, and I then trained health workers on how to interact with ChildCount via SMS.

After the ChildCount deployment in Uganda was running, I visited Millennium Village sites in Kenya, Rwanda, Ghana, and Ethiopia to help debug deployments at those sites.

Slides from my final presentation to the Millennium Villages Project Ruhiira site staff, which I gave on October 7, 2011, are online here. My colleague Emmauel Toko and I also gave a presentation at a Maternal Health Workshop in Kampala, Uganda on November 9, 2010. Those slides are online here.

Controversy over the Millennium Villages Project
The Millennium Villages Project has generated an enormous amount of press and controversy. Here are a few articles (in reverse chronological order) that give a flavor of the debate:
Funding

The Gordon Grand Fellowship covered my travel and living expenses during my 12+ months in Uganda and elsewhere in East Africa. I am so thankful to the to the fellowship committee and to Yale's Office of Fellowship Programs for making my trip possible.

Relevant Links