How to learn from an epic fail

From the mHealthNews archive
By Eric Wicklund
09:40 am

Editor's Note: It's often been said that we learn more from our mistakes than we do from our successes, and in the ever-growing field of global mHealth, that may be more true than ever. Cultural challenges, language barriers and climate differences often play a huge part in an mHealth project's failure – sometimes with hilarious results. At last year's Global mHealth Forum, part of the 2014 mHealth Summit, one entire session was devoted to projects gone bad, and it proved to be the hit of the conference. Here's a look back at that session. This article orginally appeared on Dec. 11, 2014 – Eric Wicklund, Editor, mHealth News. 

In 2005, health officials launched an SMS program in Pakistan targeting mothers and children in the wake of a devastating earthquake. The effort was heavily advertised, using an attractive woman doctor to get the word out.

The program fielded 1 million phone calls and messages within days, said Syed Ali Hussain, a Michigan State University researcher who was involved with the program. But only 5 percent of those calls were relevant. The rest were from men looking to meet the model.

Not every mHealth project achieves success. And some turn into what might be called "epic fails."

That was the gist of a session Wednesday evening at the mHealth Summit's Global mHealth Forum, in which several project leaders gamely explained why their noble efforts fell flat – sometimes with humorous results – and, more importantly, what they learned from the experience.

"We know we're going to fail – that's not the question," said Peter Benjamin of the South Africa-based mHELP program, the session's moderator. "The question is whether we're going to learn from it."

Benjamin's own example? A 2002 program in which cellphones were handed out to female health workers heading out to rural communities. The program withered, he said, because the phones were too big. The health workers, who hadn't been consulted when the project was being planned, said they would have preferred phones small enough to fit in their bras.

In Uganda, an automated text messaging program education health workers on infection prevention didn't differentiate between A.M. and P.M., said Pamela Riley of Abt Associates. The result? Health workers were getting messages in the middle of the night, and one's spouse even called the program after wondering if his wife was having an affair.

Particularly in global projects, mHealth messages can be lost if the translation isn't right, or if accent marks are discarded (as one project did to save money), or if the organizers forget about a national holiday, or if the message is too long and broken into three parts – and then sent out in non-chronological order. In Kenya, a project that was based on a 4-digit text code fell apart when the country switched to 5-digit codes right before the program was to begin.

The lessons learned? Spend as much time as possible in planning, Benjamin advised. Make sure you know your target audience, and especially your target audience's native languages and customs. And sometimes, forget about the technology.

'Paper's not going to fail, whereas a phone will fail," advised Gillian Javetski of Dimagi, who found out that a public health program in Benin worked much better if it was printed on sheets of paper that health workers could carry with them from village to village.

Susan Thompson, of Health Alliance International, ran into several "whoops moments" while launching the first-ever mHealth project – called Mobile Moms – in the tiny island country of Timor-Leste. She thought they'd need to buy phones for the project and spent time educating people, but found out that almost 70 percent of her target audience already had phones and even more knew how to text. She also didn't take into account the fact that not everyone on the island had ready access to electricity, or that her service, which promised ambulances for pregnant women going into labor, was dependent on a fleet of ambulances that was in pretty bad shape.

That story had a happy ending. The World Bank helped to buy fuel for the ambulances, while the Australian government sent over a mechanic to fix them, and to set things up so that the fleet would be continually maintained.

'We can learn from them as much as they can learn from us," Thompson said, while advising others to "stay nimble and flexible, because you're going to need to change things up."

Whether you're battling spambots in Nigeria or developing an HIV program for clinics that don't understand the meaning of confidentiality ("HIV-positive patients go to this line, while non-HIV-positive patients go to this line"), the message is simple: be prepared to fail, and to know what to do when that happens. Learn from those mistakes. And have a sense of humor about it, too.

After all, you never know when your health message might translate to something like: "Do you like homemade bread with pork?"