Mobile technology could help solve Africans’ breastfeeding dilemma

For infants in Sub-Saharan Africa who are born pre-term, with low birth weight or with HIV, access to human breast milk can mean the difference between life and death. Human milk banks have been established to solve this problem, but they tend to be expensive, requiring electricity, computer access and clean water. These are often scarce commodities in this part of the world.

Faculty and students at the UW are rapidly innovating to solve problems like this. The prevailing attitude among these motivated faculty and students: a good idea is a good idea regardless of the source and collaboration—especially novel collaboration—produces better solutions than a scientist working in isolation.

A collaboration between UW Computer Science and Engineering (CSE) and PATH, a Seattle-area non-governmental organization, has led to a simple, ingenious solution to the breastfeeding dilemma. While medical care and safe water are not always available, most Africans today have smartphones.

Rohit Chaudhri, ’11, ’14, developed a tool called FoneAstra that can be used to connect a smartphone with readily available temperature probes. Along with a heat source and some glass jars, the system pasteurizes the stored breast milk, killing harmful pathogens but leaving the beneficial immunological and nutritional properties intact. A phone app guides the user through the experience. The heat sensor attached to the phone beeps when the milk is the right temperature. Then it is cooled and ready to use. The solution is a practical and economical alternative to commercial processing that requires power to work.

“The goal here was to use it in a rural clinic with no power. It’s all run off a cell phone, which is portable,” says Gaetano Borriello, professor of CSE. Borriello guides students from many disciplines—public affairs, public health, business, communication, information and computer science—to use technology to solve problems.

Borriello developed the Open Data Kit, which is a free set of cloud-based and smartphone tools. The tools help organizations and research scientists use smartphones to collect data, conduct health surveys and help clinicians decide on a course of action by providing algorithms programmed into the phone. The breast-milk pasteurization solution is just one example of how UW faculty, students and partners like PATH and VillageReach are reaching out and working with one another and with global communities to solve problems.

Jennifer Unger, assistant professor of obstetrics and gynecology, is studying the use of mobile-phone technology to improve maternal and neonatal health. Unger’s study is being conducted in a large slum in Nairobi, Kenya. To understand Unger’s research project, it’s helpful to consider these facts: In Kenya, maternal mortality is high, with 530 deaths for every 100,000 births. Pregnant women typically attend one or two out of every four recommended clinic visits and 40 percent of births occur without the aid of a skilled birth attendant. On the upside, female literacy is about 80 percent and cell-phone coverage is reasonably high. About 90 percent of women in the slum have smartphone access.

Unger wanted to know if sending women messages about appointments and encouraging them via text message would increase prenatal and antenatal visits, use of skilled delivery services, family planning and infant health services.

“We did a lot of formative work asking women what they wanted in the messages and talking to providers about what messages they wanted to send,” she says. Brian DeRenzi, ’08, ’11, who now works for IBM in Nairobi, and Trevor Perrier, a third-year graduate student, managed the technical aspect of the project.

The trick is to find the sweet spot that provides just the right engagement to affect behavior.

One group of women has two-way communication. They receive messages and can text back to a clinic nurse. The nurse can respond to them and Unger monitors the texting and communicates with the nurse from Seattle. The two-way messages also serve as a kind of medical record. Women both respond to the texts sent by Unger’s group but also spontaneously send messages announcing the birth of their baby, asking questions about family planning methods, test results, clinic logistics and individual problems. The other group of women receives automated messages once each week but there is no two-way communication capability. The messages are timed to the patient’s week of pregnancy. Thus far, the two-way messaging seems to be quite engaging with about 60 percent of women regularly communicating with the nurse.

“Our real interest is in learning how much interaction is required to see a difference in having the women deliver in a facility, participate in family planning, exclusively breast feed and attend four antenatal care visits,” she says. The trick is to find the sweet spot that provides just the right engagement to affect behavior.

Carey Farquhar, professor in medicine, epidemiology and global health, is the investigator of the Home-based, Partner, Education and Testing (HOPE) study in cooperation with the Kenya Ministry of Health. Health workers collect information on smartphones that also have a GPS feature to locate the homes of the women for later counseling of the woman and her partner. The workers upload the data to a server at the Kenya Ministry of Health. The study assigns one group of couples to standard pregnancy services and routine prenatal care. The other group of couples has been randomly assigned to receive follow-up and HIV testing at home. Farquhar suspects that the home-based partner education and testing will result in finding and preventing HIV as well as linking those with HIV to available treatment. Using smartphones to gather and enter the data is a huge timesaver.

“Formerly, we had data on paper and then entered it manually and triple checked it. It would take weeks. Now we get weekly summaries of the data results and we can improve and tweak the study as needed,” says Farquhar.

The cross-departmental collaboration at the UW has involved undergraduates as well as graduate students and faculty. Saloni Parikh, who is majoring in public health and computer science, developed the app for the HIV data-collection study. Parikh traveled to western Kenya and worked on the IT part of the project and she trained data and technology managers in Kisumu. “I made the app work without a network connection and I showed their manager how to pull up the data and import and filter it,” she says.

Borriello, who has taught at the UW for 25 years, says that students are quite different now. Like Parikh, DeRenzi and Chaudhri, they want to use technology to work on things that really matter. “During the dotcom boom, people were in it for the money. They wanted the degree to get into that world and cash in,” says Borriello. He says there undoubtedly will be more projects emerging from the UW that help research efforts and provide answers that contribute to improved global health.