Even though until now Akvo FLOW has mainly been used in the water sector, we at Akvo have always believed that this is a very powerful tool that should be made available for the development aid sector as a whole. Just last week, Amitangshu elaborated about the first roll-out in Indonesia for a baseline study for food security. In this post I’m sharing the experiences of using Akvo FLOW in the health sector in Jinja, Uganda.
Above: Enumerator practising Akvo FLOW with a sister from St. Benedict health clinic near Jinja, Uganda. Photo by Luuk Diphoorn.
In all the health programmes that are taking place in Uganda (as in many other nations), ‘Patient Satisfactory Surveys’ need to be done to assess the quality of the health services provided by a certain health facility. These surveys have always been done by hand. This research was set up to look at what the impact would be of using other types of data collection methods. Together with the Diocese of Jinja, a main partner of Cordaid within the Connect4Change Alliance in Uganda, three health centres were selected near Jinja to conduct this research. The collection methods tested were Akvo FLOW, Text to Change’s Call Centre Approach, and the traditional approach by hand. To conduct this research we got help from an independent consultant, in the person of Anneke Maarse, to coordinate and carry out this research.
I won’t elaborate into what the results of the research have been, as the report still needs to be finalised. In this post I am however sharing how Akvo FLOW was used and some of the challenges we faced.
As mentioned earlier, three health centres were selected, namely one hospital (Buluba Hospital) and two community clinics (St. Benedict and Nawanyago). About three days after being treated, patients would then be approached by members of Village Health Teams (VHTs) at their homes, and the surveys would be done on the spot using Akvo FLOW. A number of exit interviews at the health centres themselves were also conducted using Akvo FLOW.
Training workshop in Jinja:
We had decided on using eight smartphones for the data collection, four for Buluba Hospital and two for each of the clinics. We had organised a training workshop in Jinja for all the data collectors in the first week of October 2012. The participants include the members of VHTs, who were going to collect all the data, along with some key people at the health centres. Below you can see all the participants that joined.
At the very start of the training workshop, I asked if any of the participants had ever used a smartphone. The response was a staggering zero! Being used to a Kenyan context, where most people either have a basic smartphone or know someone who does, I was a bit anxious on how to approach this training. Adding to the complexity, after the lunch break we also had no power. Luckily with help from my colleague Francis Warui and Text to Change’s Eunice Gnay, we managed to teach the participants the basic functionalities of a smartphone and do a run-through of conducting the surveys using Akvo FLOW.
The second day we travelled to the premises of the Buluba Hospital and St. Benedict clinic. Here we did a recap of the things we had learnt from the previous day of training and then did more one-on-one training exercises with the VHTs. We managed to also conduct initial test surveys in the area around the health centres and show the results directly on a Google map. On day three we did the same for the Nawanyago clinic. Below you can see two video interviews I did after the training workshop with participants Cleoehas Nanzushi and Patricks Omogi.
The data collection started officially on the 15th of October and ended on the 29th of November 2012. In total 347 surveys were collected, with a distribution of 144 for Buluba Hospital, 99 for Nawayango clinic, and 104 for the St. Benedict clinic. Below you can click on the map and see for yourself where and what kind of data was collected. We managed to separate the surveys done as exit interviews (yellow dots) and follow-up (green dots). From the map we generated you can see what the power is behind collecting the exact GPS coordinates. Some of the green and yellow dots show that data was not collected at the places where they should have been collected. Due to privacy issues I have not included the names of the patients and their pictures. We have made a selection of some of the most important data – just click on the map below.
Even though for all the enumerators it was the first time using a smartphone, they managed to collect all the data as planned. We did encounter one major obstacle in this data collection exercise, and that was the availability of a good working mobile internet connection. This meant that many of the pictures did not manage to get through with the rest of the data. During our training exercise at the Nawanyago clinic we had already noticed the poor connection, and subsequently none of the pictures came through from there at all.
We therefore needed to collect all the phones and manually upload the pictures that did not make it. All in all this has been a very important lesson for us, and all future data collection and training workshops using Akvo FLOW will be adapted to try and minimise this issue.
This was a first for using Akvo FLOW in the Connect4Change Alliance and I’m sure that in 2013 many more will take place. With a new and improved Akvo FLOW 1.5 release becoming available to us in January 2013, many of the issues we have faced will no longer be relevant.
I also want to personally thank Paula Mommers for her continued support, Anneke Maarse for all the time she spent coordinating this research, and the people at the Diocese of Jinja for all the trouble shooting they needed to do locally.
Luuk Diphoorn leads Akvo’s East Africa Hub, and is based in Nairobi.