Information is Beautiful

My initial work in Mozambique was a sort of side project (that has turned into a more primary project for the time being) to work with a non-profit here on a large household survey in the province of Zambezia.

For those less global-health-minded folks out there, household surveys are the bread and butter of international health programming. We take it for granted in the States that we have functioning vital registration and disease surveillance systems that keep policymakers and program managers abreast of mortality rates, causes of death, and burden of disease. Such systems provide data to make informed decisions, and also give us baseline data from which we can later evaluate whether or not interventions have really worked. In sub-Saharan Africa, vital registration systems are virtually non-existent, and where they do exist, they are largely unreliable. Until health information systems are more firmly in place, we use household surveys as a measure to obtain this necessary data.

Household surveys are long, unwieldy, complicated, and require enormous logistical effort. Our survey has over 500 questions, ranging from general demographic inquiry to questions about HIV/AIDS or women’s empowerment. It will be administered to a random sample of over 4,000 households in Zambezia province by the end of July. To top it all off, we’re administering it not on paper, but on Androids with ODK Collect.

Data collection, in a sense, is one of the most natural applications of mobile tech to public health, but the world of international public health feels a bit distanced from and intimidated by what it considers the world of technology. I’m working here to bridge the two. In this case, working with a data collection application on the mobiles is easier for the survey enumerators (can you imagine carrying around a stack of 33-page surveys in rural Mozambique?). Using the mobiles is safer in terms of data security, because the women simply submit the data to a server when they’re in an area that has cell coverage. The electronic survey automates skip logic, which is a mental jungle gym on the paper version of the survey (i.e. we do away with having to think about “If yes and the interviewee has children, skip to question 5, if no and the subject has children, skip to question 6a”, etc.) Additionally, we eliminate transcription errors that can occur with a paper survey, because we skip the step where paper survey results are entered into a database for analysis. And it uses less paper!

Training is obviously of a lot of importance here. We’re working with a group of Mozambican women to be the survey enumerators, and it’s their commitment and interest that makes the job fun.

Making sure that the survey enumerators know how to use the technology is important to my supervisors, but making sure that the technology is useful and useable to these women is more important to me. I’ve always thought that the job that’s hardest to do is the one where you feel as if you’re not contributing to a greater whole. And as someone who’s done data collection before, I know that it can be tedious and frustrating at times—and I was never sent out into the bush on foot to do it. To me, this system seems easier, better, and faster, but those qualifications don’t mean anything unless the enumerators think so too. In fact, I told everyone up front that if at the end of the initial week of training they didn’t like the phones, we should scrap it and go pen-and-paper. Who am I to make this job harder?

But things have gone well. Especially rewarding is watching a survey team leader (who was trained before the larger body of general enumerators) explain to their team that it might be confusing at first, but that the phone is actually much simpler and will make their job easier. Moreover, training has given us a way to elicit commentary from the enumerators so that we can modify and adapt the survey on the phone according to their needs and wants.

Most unexpected–and I think the best–part of all of this is that the enumerators are really connecting with the act of submitting their data. Instead of turning in paper forms to their team leader, who will turn them into a supervisor, who will in turn them in to the office ad infinitum until the point where they’ll eventually be analyzed, the women gather with their team at the end of the day to select the forms to send to the server. They push the button saying that the data is complete and ready to be used. It’s a small, empowering action that is not lost on them.

At the end of training, we had a small party for the enumerators in the district of Alto Molocue. We asked if anyone wanted to say a few words before we cut the cake. Perhaps the highlight of my time in Mozambique so far was when one enumerator stood to explain her gratitude for her time spent learning. Not just because of the job–which she admitted she wanted and needed–but because of the work she felt she was doing for her children, and their children in the future.

It doesn’t get much better than that.