DMK Internship Story

| August 20, 2014 | 0 Comments

By Nathan Georgette

Credit: Nathan Georgette
Nathan & Maja, along with their fellow DMK intern Sarah MacVicar ’13, at the Koinonia Conference & Retreat Centre, at which they stayed in South Africa.

            Khwezi clenched her small hands as her mother coughed violently. The mother had tuberculosis (TB), and the girl might soon not have a mother. Outside the South African family’s rondavel, I tried to entertain Khwezi as the community health worker counseled the mother. The mom had fallen ill after she stopped taking her HIV medicine, which is known as defaulting. There are 5.6 million people infected with HIV in South Africa alone1. The government provides life-long treatment for free, which requires quarterly visits to the physician and pharmacy. Treatment side effects, relocation, and lack of social support can all contribute to a patient interrupting their treatment, or defaulting2. About 5% of African patients default each year on their HIV treatment, making them vulnerable to life-threatening diseases3. Behind these statistics are thousands of individual tragedies. During a Harvard Global Health Institute internship in South Africa during the summer of 2012, I witnessed far too many.  The internship was based at the Don McKenzie TB Hospital near Durban, but we were also rotated out to live and work in the village of KwaXimba for a week at a time with the Umndeni Care Program.  It was there that I met Khwezi and resolved to launch my best effort against the defaulting problem. Majahonkhe Shabangu ‘14, a fellow intern, and I realized that early intervention was imperative. But it is difficult to know when a patient begins defaulting. That is, few patients call in to the clinic to announce that they have stopped treatment despite the doctor’s orders. Thus, we developed a computer program to automatically identify defaulting patients based on the electronic pharmacy database. This gave us a list of the defaulters at hand – but what could we do with this information?  After perusing the medical literature, we learned that text messaging has been proved effective at improving a patient’s adherence to their HIV medication4. Thankfully, mobile phones are nearly ubiquitous in South Africa, and text messaging is low cost. We coded another program to personalize and deploy text messages to each defaulting patient, encouraging them to reinitiate treatment, and funded the distribution with some of our internship stipend.

Credit: Nathan Georgette Sunset over the Valley of 1000 Hills in outer-west Durban. The Don McKenzie Hospital and KwaXimba are both located in this valley.

Majahonkhe spoke Swati (a cousin tongue of Zulu), so he followed up with individual calls to the patients. One afternoon, an outgoing call failed – insufficient account balance. Reloading the airtime requires buying vouchers, but the only nearby vendor had run out. We sprung into action, asking the nurses and staff where we could find them. Our search led us to Ward 3, where one entrepreneurial young patient had been selling the vouchers from her hospital bed. Here, supply met demand and the calls to defaulters could continue with only the briefest of interruptions. Majahonkhe was able to get through the entire list of defaulters before we had to depart. Although we were not able to follow up with Khwezi and her ill mom before we left South Africa, the memories of them and the many other patients stuck with us. We teamed up with our more technically savvy classmates, Dario Sava ’13 and Roy Zhang ’13, to augment and scale the program. And so, Sawubona was born.             Sawubona, which means “I see you” in Zulu, builds upon the prototype to reduce default through a personalized and triaged approach. For example, patients with good adherence will be contacted weekly with supportive messages. Those who have previously defaulted will receive more tailored treatment reminders. In the future, currently defaulting patients will be engaged with a menu-based dialogue to ascertain the reason for default and compose specific suggestions and encouragement. For example, if a patient had to travel for work, the program will ask for the patient’s new postal code and then reply with contact information of the nearest HIV clinic. We will further augment the program with automated calls to the least responsive defaulting patients. The automation limits costs by minimizing staff time required, and the

Credit: Roy Zhang The members of Team Sawubona: (from left) Roy, Maja, Nathan, and Dario.

personalized approach further improves cost-effectiveness by allocating the most intensive (and expensive) interventions to those in greatest need. We entered the i3 startup competition and won $5000 in seed funding from the Technology and Entrepreneurship Center at Harvard. Thanks to this, and generous contributions from private philanthropists, we were able to install the program at two partner sites in South Africa in June. Our long-term plan is to scale throughout the Durban area and eventually offer the service nationwide. When Majahonkhe and I arrived in South Africa for our internship, we lacked even an inkling that our experience would lead to founding a nonprofit. We learned that openness to unforeseen possibilities is invaluable on any program abroad. We are excited to see where this venture takes us, and certainly hope that Sawubona can make a positive impact in the lives of people living with HIV around the world. If you’d like to learn more about our program, please visit us at www.SawubonaHealth.org.

  1. UNAIDS. “South Africa.” http://www.unaids.org/en/regionscountries/countries/southafrica/ 9 Jul 2013.
  2. Alamo, Stella T., Robert Colebunders, Joseph Ouma, Pamela Sunday, Glenn Wagner, Fred Wabwire-Mangen, and Marie Laga. “Return to normal life after AIDS as a reason for lost to follow-up in a community-based antiretroviral treatment program.” JAIDS Journal of Acquired Immune Deficiency Syndromes60, no. 2 (2012): e36-e45.
  3. Dalal, Rishikesh P., Catherine MacPhail, Mmabatho Mqhayi, Jeff Wing, Charles Feldman, Matthew F. Chersich, and Willem DF Venter. “Characteristics and outcomes of adult patients lost to follow-up at an antiretroviral treatment clinic in Johannesburg, South Africa.” JAIDS Journal of Acquired Immune Deficiency Syndromes 47, no. 1 (2008): 101-107.
  4. Horvath, Tara, Hana Azman, Gail E. Kennedy, and George W. Rutherford. “Mobile phone text messaging for promoting adherence to antiretroviral therapy in patients with HIV infection.” Cochrane Database Syst Rev 3 (2012).

Category: Student Submissions

About the Author ()

Leave a Reply