Am back in Ethiopia for the next couple of weeks – whilst writing this, I’m listening to a presentation about blogging in Ethiopia. Currently there are very few blogs running in Ethiopia (for a list of the main ones visit http://ethiopian-blog.com)
The BarCamp has been very good so far, several hundred staff, students and others from universities and other organisations across Ethiopia, although I arrived slightly late this morning – the traffic was pretty bad – but fortunately I didn’t miss too much. The sessions this year seem to be even more varied than last year, though still quite technology focussed. Quite a few staff and students have made the trip down from Mekelle Uni, so very good to see them here – especially the elearning team and lab attendants.
Have just been to presentation about localisation by Google. Tomorrow I’ll give my presentation about using mobile technologies (smartphones) to improve maternal healthcare. Hoping to get plenty of people to come along – though Google are giving another presentation at the same time, so hope they don’t lure too many people away from attending the other sessions!
As I’ve mentioned before many of the Health Extension Workers (HEWs) we’re working with have trouble using the Gregorian calendar. Much of the work we’re doing with the HEWs depends on them having a good idea of the expected date of delivery for pregnant women. To help the HEWs calculate the expected delivery date, in Ethiopian date format, for the pregnant women they are working with, we have developed a small application which gives them the expected delivery date based on the date of the last menstrual period.
With the application the HEW can enter the last menstrual period in Ethiopian date format and it will give them and the pregnant woman the expected delivery date in the date format they are used to using in their day-to-day lives.
We’ve made the application available for anyone to download and use on their Android phone – just download the app and install on your Android smartphone.
I should give a big thanks to the developers of Joda Time (a java library for working with alternative calendaring systems) which made the development of the application so much easier!
Any feedback or comments welcome. If anyone is interested in developing the application further then we’re very happy to give access to the source code – just contact me
I’m travelling over to got to the conference and going to give a presentation about the work we have been doing recently on using smartphones for data collection and reporting to improve maternal healthcare in the rural areas of Ethiopia, specifically working with the Health Extension Workers. Will post the presentation up soon!
Some of the feedback we had from the initial HEW training last week was that some of the HEWs had difficulty in using the Gregorian calendar. Ethiopia has it’s own calendar which is the normal calendar used for the vast majority of Ethiopians (Ethiopian calendar entry on Wikipedia). Given that we’re asking the HEWs to collect date information, we need to make this as easy and understandable for them as possible – for example, to enter appointment dates a few weeks or months in advance. So over the past couple of days I’ve been looking at creating a date picker widget for integration into ODK Collect that will allow dates to be entered using the Ethiopian calendar format, but will store the date in the database as Gregorian. This transformation is hidden from the user and storing Gregorian dates in the database means we can manipulate and compare dates for reporting purposes, which we’d be unable to do storing Ethiopian dates as strings in the database (we can transform the dates back into Ethiopian calendar for final display).
It ended up being much more straightforward than I though it would be, especially with much help from the ODK Community and with the Joda Time java library already built into ODK Collect. I’ve got a first version ready for testing, so if anyone else is interested in having a look you can download the .apk file for installation on your Android phone.
To see the Ethiopian date picker working you’ll need to load up a form which specifies data picker. You can connect to our ODK Aggregate server at: http://hew-datacollect.appspot.com and download the EthioDateTestV1.5 form. Or you can download the original form XML to put on your own server.
This is only a first version, so any feedback is very welcome – or if you’d like the source code then please feel free to contact me (will put it up somewhere once I’ve tested it a bit more). [Update 19-Aug-11: I’ve now put all the code up as a clone on Google Code at https://code.google.com/r/alextlittle-dc-odk/, so you can see the full changes I’ve made.]
Finally, here are a couple of screenshots of the Ethiopian date picker running in my Android emulator:
I just posted this up on our Digital Campus blog:
Last weekend Araya and Florida ran the first training course for Health Extensions Workers (HEWs) who will participate in our feasibility study. We are starting with a very small group of HEWs – just 5 in this first training session – and the 2 day training consisted of:
- the basic functions of the HTC hero phones, contacts, making calls, messaging and switching between Amharic and English keyboard layouts
- charging the phones using the solar lamps and chargers we provided – as most of the HEWs do not have electricity in their Health Posts
- using the EpiSurveyor client application for entering and sending data, plus using the GPS
- practice completing and sending the Ante Natal Care protocol forms set up in EpiSurveyor
This first training is simply to get the HEWs used to using the phones, find out what problems and issues they may have, especially with battery life, recharging, completing forms etc – at this stage we’re not looking to collect real data – though we hope they can practise using the forms with real patients.
The feedback we have received is that the training went very well, the HEWs seem very and eager to learn how to use the phones and soon became quite comfortable using them. Over the coming months we’ll introduce the phones to another couple of small groups of HEWs and we’ll find out what challenges may exist and the feasibility for using smartphones for protocol and data collection in this environment.