jopsa.org

I’d love to see implementers and tool-providers shift away from data collection and towards creating local action. Here’s my reasoning:

1. Especially in the case of mobile technology, data can be a byproduct of action – e.g. clinicians and community health workers can use mobile phones to coordinate patient care, use identifiers, and produce a longitudinal record of care that can be aggregated with other records, analyzed, visualized, etc.

2. These systems involve people. Imagine someone tells you, “Please report, because we need data.” Now imagine they say, instead, “We need information.” Or, “We need to know.” Finally, “We need to act.” Which framing builds the greatest incentive for participation?

3. It takes action to achieve impact. Yes, large data sets that inform policy are important — reiterating my first point, I’d argue we’ll have that data anyway due to the nature of the technology. At the end of the day, for a situation to change, someone needs to do something differently. Our programs should reflect that reality.

Local actors want change through action. Policymakers want data to inform decisions. Luckily, technology can deliver both.

Me + Henry of Ushahidi

So much has happened in the last few weeks, I’ll have to retro-blog a bit. I’ll be sharing stories from my time at clinics in Bushenyi, Uganda, soon – but I wanted to post quick thoughts on a particular discussion thread at the recent W3C workshop, “Africa Perspective on the Role of Mobile Technologies in Fostering Social and Economic Development.”

More often than I expected, a very simple question arose, “What should we be talking about?” The easiest response is that it depends – on who “we” are and what we’re doing. I heard three broad options, each of which is tied to different factions I’ve encountered throughout my brief but focused mobile tech sprint, of late:

1. Talk about what works, right NOW. This is what interests NGOs and clinics working on the ground. They want to know what they can do, today, to impact operations and better serve beneficiaries (e.g. treat patients and support community health workers). For the most part, this group doesn’t know exactly what tech’s available, and they doubt they can afford it. I’ve heard from numerous clinics, “We knew mobile technology initiatives existed, but we never expected them to come – or work – here.”
2. Talk about where the technology is going, and what it’s going to take to get there. I turn to others for this information. Tech junkies rule this domain, and rightly so. They should be the ones leading the charge to construct and control tech advancements (e.g. WiMAX). This conversation excites and is clearly best suited for techies. End users are less interested.
3. Talk about theory and long-term convergence of best practices and competing technologies. Tech experts and NGOs with case studies can offer bits and pieces, but this conversation lends itself to policy experts, economists, and fortune tellers. If we can find technology and implementation strategies that scale horizontally, outputs of these conversations may be less prescriptive and more descriptive – with organizations choosing and mixing available technologies at their liking. Erik Hersman of White African and Ushahidi wrote about this idea a little while ago.

Obviously, there’s some crossover. But it may be useful to keep in mind how we fit into these discussions.

Those who know me will understand why I’m most passionate about the first discussion. We don’t have great means by which to connect “on-the-grounders” with tools they can use. The clinics and service providers I talk to certainly feel disconnected from what they perceive to be immense, unattainable opportunity to utilize mobile tools. Their perception of value may be accurate and real, but the barriers to entry need not be.

A lot has happened in the last 6 months. Everyone involved with Mobiles in Malawi is excited by the SMS program’s impact on healthcare delivery at St. Gabriel’s Hospital. The medical staff and participating CHWs have taken ownership of the initiative, and what’s left of a public service grant will keep the program running for 10-15 years – at just $500 per year.

St. Gabriel’s is certainly not alone in the challenges it faces as a rural healthcare provider. After witnessing the effects of simple ideas and equally uncomplicated technology on medical care, one thing was clear – each day that a clinic goes without tools they want and need is a day with undue hardship.

To date, I am talking with healthcare organizations working in eleven countries (Burundi, Malawi, Uganda, Zambia, Mozambique, India, Kenya, Ghana, South Africa, Peru, and Haiti) about partnerships to expand the tools and strategies used at St. Gabriel’s to their respective sites. Details regarding these organizations and my role in supporting them will be expounded upon at jopsa.org in the coming weeks. In addition to healthcare providers, I’m honored to be collaborating with kiwanja.net, The kiwanja Foundation, the FrontlineSMS team, and MobilizeMRS.

After speaking with global health organizations and the clinics they’re linked to, it is clear that 2009 must be a year of action. Everything that happens will be posted here. I hope you’ll join me on this journey, and share your thoughts along the way.

Why was FrontlineSMS the right tool for St. Gabriel’s Hospital?

I wrote a bit about the virtues of the software in July, amidst the implementation period. The program has been running for six months, and my latest trip allowed me to document the project’s impact on primary care, treatment coordination, and hospital efficiency. Further, the SMS program has saved thousands of hours of travel time for CHWs and hospital staff, and bolstered the CHWs’ status within their respective communities – which are now connected to the hospital’s resources.

As Ken Banks recently stated, whether or not a mobile tool is suitable is “all about the context of the user.” In succinct terms, here are a few reasons that Ken’s tool was a the right fit for St. Gabriel’s:

1. It’s free.
2. It works with simple, readily-available hardware. I used recycled phones and didn’t need to worry about the donated laptop’s specifications.
3. It doesn’t require an internet connection. The hospital’s web access is shaky, at best.
4. It is incredibly intuitive and easy to use. The nurse running the program had never used a computer in his life. After one hour of training, he was off and FrontlineSMS-ing.
5. The hospital found the tool – not the other way around. After I spent time at the hospital two summers ago, the need for connectivity was clear. To meet that need, a tool was employed. It wasn’t forced on me or the hospital.
6. Straightforward features allowed the hospital to take ownership and get creative, which encouraged user-driven functionality. One example: Auto-replies set to provide CHWs with immediate drug dosage and usage information.
7. A ten-second demonstration can illustrate the program’s purpose. This hooked clinical staff working within various hospital programs (e.g. HIV treatment coordinator, TB officer, PMTCT director).
8. Past text messages, and vital patient informaiton, are just two clicks away.

In examining the success of the SMS initiative over the last half-year, the appropriateness of the technology cannot be ignored.

A few, very committed individuals – my mother and sister – set out to answer that question. My mom, Casey Nesbit (DPT), receives every message that is sent to the hospital, via email (thanks to a simple forwarding command in FrontlineSMS). Those messages are in Chichewa. For four months, she translated every message to English.

My sister, Elizabeth Nesbit, decided to code and organize every SMS sent by the CHWs. She’s a sophomore at Rice University, making her way to medical school. She categorized messages by keywords and/or phrases (e.g. symptoms, supplies, patient updates/referrals, deaths, requests for help, requests for visits, meeting coordination).

Under this introduction is a list of all the symptoms found in messages communicated to the hospital. Elizabeth sorted these symptoms out into categories (body pains, digestive and urinary tract, respiratory tract, swelling, skin and sores, malaria and fever, weakness, heart problems, cancer, and other). She broke apart every incoming message this way.

Below the symptom list, you’ll find the fruit of their combined efforts – charts explaining the subject matter of texts to the hospital. Click on any of the charts to view a larger version. These messages fell between mid-August and early December. Shoot me an email if you want to see more of Elizabeth’s analysis.

Enjoy!

___________________________________________________

This morning, we distributed the first batch of solar panels from G24 Innovations. I was also able to reconnect with CHWs I hadn’t seen in a while. Everyone was extremely happy – I’ll let the pictures tell the story.

Of course, Alex ran the session. The instructions were quick and easy, and everyone was rather celebratory:

To close the meeting, the CHWs sang a song for me that they had prepared – I’m not sure what the lyrics were, but “phones” and “messages” were included! Afterward, I traveled with Dickson Mtanga and Mary Kamakoko to their villages – it took a good two hours on a bicycle. We spent another 3 hours biking around and seeing patients, before I started back to the hospital. Dickson and Mary, using their new panels:

On Christmas morning, Silia stopped by the guest house to talk about the SMS program. He’s responsible for testing, drug provision, and follow-ups for TB patients. He described how he’s using FrontlineSMS and the network of cellphone-wielding CHWs. Almost all of what follows developed in my absence.

Some sputum-positive patients don’t turn up to receive their medication. It’s Silia’s job to track these patients and get them back on their drug regimens. Before the SMS program, he was visiting an average of 17 patients per week – this took him three trips on his motorbike. Each trip would take ~9 hours. That’s 27 hours per week spent tracking patients in various villages.

The SMS network has allowed Silia to share his workload with the CHWs. He now tracks an average of 20 patients per week via SMS. He simply texts CHWs nearby patients that haven’t turned up. As Silia says, the CHWs provide “immediate feedback.”

The program has been running for roughly 26 weeks. With the shift to SMS-based patient tracking, Silia had an additional 700 hours to utilize. Not surprisingly, he’s been using FrontlineSMS to supplement other areas of his work.

He now visits an average of 4 patients per week, for different reasons. Some messages from the CHWs tell of patients who are too ill to travel to the hospital. Silia will respond by bringing a new supply of drugs. Other messages relay symptoms of community members – e.g. “A man has a chronic cough, and we suspect TB.” Silia will visit the patients, and collect a sputum sample. He’ll return to the hospital to do testing and send the results, by SMS, back to the CHW.

Finally, when patients at the hospital test positive for TB, they’re told which CHWs near their home have cell phones.

Coming up: Impact on Home-Based Care, PMTCT, Public Health, and HIV/AIDS programs

Here’s the truth – this project involves people, rich in character and experience. It’s not only about the technology. If I’m interested in the tech fulfilling its potential, I’ve got to pay attention to the people.

Case in point:

I spent yesterday mulling over text messages sent through FrontlineSMS over the last four months, noting which CHWs had communicated least. I put together a list of a few CHWs I suspected might be having signal issues. Looking at the map, three of six CHWs on this list were clumped together – clearly, they must not have good reception.

I told Alex about my findings, this morning. He took a look at the names and said, “Well, Bernadeta took her phone with her to Zambia, we’ve discovered that Chrissy is not able to write her own name, and Jereman’s phone battery was stolen while it was charging at the local barber shop.” My time away from the hospital almost made me forget the multitude of stories swirling around these phones and the hospital they’re linked to. With 100 phones in the field, three random problems are to be expected.

Whether or not everyone agrees, I think personal stories convey a project’s successes, as well as their failures. Silia, a hospital attendant who runs the hospital’s TB program, said yesterday, “The SMS project is very, very good – I can get much more work done, instead of driving the motorbike everywhere. It’s very simple – we can expect feedback about patients immediately.” I met the new hospital administrator today, and his second sentence was, “You know, it’s not only beneficial for communication. The volunteers are now committed to their work, and more will follow.”

I’m letting stories from patients, CHWs, and the medical staff at St. Gabriel’s drive my exploration into this project’s value. I turned to people for the direction of the initiative, and I’m turning back to them to measure part of its impact.

The first batch of solar panels arrives tomorrow.

That’s the headline the CNN Technology site is using to pull visitors to a story covering the project. Here’s the link:

Text service provides more than a Band-Aid for rural health service by Steve Mollman

It’s amazing to see Alex and Grace on the site, and I think Steve wrote a wonderful article. I’m headed back to the hospital for Christmas break, and I’ll be updating the blog regularly – starting now. Here’s a (very) quick 30-second video, as a re-introduction to the project and a few of the people involved.

Here’s some information on conferences and events I’m set to attend in the next month:

October 9th – Stanford Undergraduate Research and Public Service Symposium

October 11th – BarCamp Africa @ Google

October 12-14th – Alex Ngalande from St. Gabriel’s Hospital will be speaking at MobileActive08

October 14-15th – Institute For The Future’s ‘Reinventing Health Care in a Mobile World’

October 22nd – Stanford Community Health and Public Service Fall Forum

October 23rd, 9AM – Health 2.0 Conference