Our group of HEWs and midwives have now been using the ODK-based patient management tools for just over a year – that is using the tools with actual patients, they were using the tools for several months before this just for testing and for refining the patient visit forms.
Despite only working with a small group of HEWs and midwives (although recently approx 20 more joined the project), we now have over 1000 patients (pregnant mothers) registered who have attended ante-natal care or post-natal care visits. We also have nearly 300 deliveries recorded, although given we have 1000 pregnant mothers, we would expect this to increase substantially in the coming months.
Looking back over the patient visits forms submitted each day, we can see that the HEWs and midwives are making use of the tools more regularly, although we still see spikes in the number of forms submitted on some days – likely to coincide with ‘immunization days’ when more people attend the health centres/posts.
Now that we’re building up a lot of experience, data and knowledge in how the tools are being used, As I’ve already posted up (here and here) Araya has already had a couple of research papers published in this area, and we’re now working on completing the feasibility study and technical papers.
Araya has just had his second paper published: “Knowledge and performance of the Ethiopian health extension workers on antenatal and delivery care: a cross-sectional study”. The full paper is available from Human Resources for Health, provisional abstract:
In recognition of the critical shortage of human resources within health services, community health workers have been trained and deployed to provide primary health care in developing countries. However, very few studies have investigated whether these health workers can provide good quality of care. This study investigated the knowledge and performance of health extension workers (HEWs) on antenatal and delivery care. The study also explored the barriers and facilitators for HEWs in the provision of maternal health care.
In conducting this research, a cross-sectional study was performed. A total of 50 HEWs working in 39 health posts, covering a population of approximately 195,000 people, were interviewed. Descriptive statistics was used and a composite score of knowledge of HEWs was made and interpreted based on the Ethiopian education scoring system.
Almost half of the respondents had at least 5 years of work experience as a HEW. More than half (27 (54%)) of the HEWs had poor knowledge on contents of antenatal care counseling, and the majority (44 (88%)) had poor knowledge on danger symptoms, danger signs, and complications in pregnancy. Health posts, which are the operational units for HEWs, did not have basic infrastructures like water supply, electricity, and waiting rooms for women in labor. On average within 6 months, a HEW assisted in 5.8 births. Only a few births (10%) were assisted at the health posts, the majority (82%) were assisted at home and only 20% of HEWs received professional assistance from midwives.
Considering the poor knowledge of HEWs, poorly equipped health posts, and poor referral systems, it is difficult for HEWs to play a key role in improving health facility deliveries, skilled birth attendance, and on-time referral through early identification of danger signs. Hence, there is an urgent need to design appropriate strategies to improve the performance of HEWs by enhancing their knowledge and competencies, while creating appropriate working conditions.
We’ve just set up a demonstration site of our analytics and mobile site for the patient management tools. Previously if someone wanted to test out the tool for themselves, we could really only give them the mobile application and the protocols to look at, but we didn’t have a demo area for the server side. The only demo was on my laptop, and we can’t give access to the live site as it has real patient cases. I took this opportunity to look at using Amazon Web Services (EC2) for setting up the demo server – it all worked out really well and very easy to use.
You can log into the analytics/scorecard site at:
http://odk-demo.digital-campus.org/scorecard/ (username/password is demo/demo)
and the mobile version is at:
http://odk-demo.digital-campus.org/scorecard/mobile (same username/password)
The demo user has supervisor privileges, so is able to see all the data entered, usually health workers logging in would only get to see the data directly related to their patients.
If you would like to see the whole process, from entering the protocols on the smartphone, all the way through to seeing the cases on the analytics scorecard and mobile site, I also set up a demo ODK Aggregate server for submitting protocols. To set this up:
- Download and install on your phone our version of ODK
- start the app and enter the following settings (go to menu > change settings):
- Server: http://odk-demo.digital-campus.org/ODKAggregate (note that this is case sensitive)
- Username: demo
- Password: demo
- Go to ‘get blank form’ – this should connect to the server and show all the available protocol forms – select and download the ones you would like to try out
- Enter and submit a few protocols from your phone
- You will then be able to see the forms you have entered on the analytics scorecard, and the mobile version – note that the forms don’t appear instantly on the scorecard or mobile site, it may take a couple of hours, as we have some caching running, to make the site run more quickly
Please let us know how you get on – especially if I need to add some more info to the instructions above.
Araya, one of the PhD students from Ethiopia we have been working with for the last few years, has just this week had his first journal article published: “The role of health extension workers in improving utilization of maternal health services in rural areas in Ethiopia: a cross sectional study”. More details on BioMed Central or PubMed. Abstract:
Community health workers are widely used to provide care for a broad range of health issues. Since 2003 the government of Ethiopia has been deploying specially trained new cadres of community based health workers named health extension workers (HEWs). This initiative has been called the health extension program. Very few studies have investigated the role of these community health workers in improving utilization of maternal health services.
A cross sectional survey of 725 randomly selected women with under-five children from three districts in Northern Ethiopia. We investigated women’s utilization of family planning, antenatal care, birth assistance, postnatal care, HIV testing and use of iodized salt and compared our results to findings of a previous national survey from 2005. In addition, we investigated the association between several variables and utilization of maternal health services using logistic regression analysis.
HEWs have contributed substantially to the improvement in women’s utilization of family planning, antenatal care and HIV testing. However, their contribution to the improvement in health facility delivery, postnatal check up and use of iodized salt seem insignificant. Women who were literate (OR, 1.85), listened to the radio (OR, 1.45), had income generating activities (OR, 1.43) and had been working towards graduation or graduated as model family (OR, 2.13) were more likely to demonstrate good utilization of maternal health services. A model family is by definition a family which has fulfilled all the packages of the HEP.
The HEWs seem to have substantial contribution in several aspects of utilization of maternal health services but their insignificant contribution in improving health facility delivery and skilled birth attendance remains an important problem. More effort is needed to improve the effectiveness of HEWs in these regards. For example, strengthening HEWs’ support for pregnant women for birth planning and preparedness and referral from HEWs to midwives at health centers should be strengthened. In addition, women’s participation in income generating activities, access to radio and education could be targets for future interventions.
I’ve recently been rewriting the mobile HEAT application and I now have a first version ready for people to test out. You can get the app from Google Play for installing on you Android phone. As it’s just a first version, I’d really appreciate any feedback (either email me directly or post a comment below), which is also why I’ve just posted it up here, rather than making it available on the Google Play store – which I’ll do once it’s been tested a little more.
Few notes comments on how to use the app and what it does:
- As with the previous version, you’ll need a MQuiz login account (though you can register for this directly in the app). So you do need an active internet connection on your phone for this step.
- Unlike the previous version, this app comes with no course content. This was one of the main reasons for rewriting this app: to decouple the content from the app. So after you’ve first logged in, click on the ‘manage modules’ button and you can get a list of the available modules you can install, so you can select which content you’d like to download. You’ll also need an active internet connection for doing this, though I have built in a way in which you can just put the course package directly on the phone SD card and it will auto install.
- Once you’re logged in and have downloaded some content, an active internet connection is no longer needed.
- For testing out the video content (in the “video demo” course), the videos are not included in the course download package (as it makes the download packages too large). If you’d like to test the video content, please download the .m4v files and place them all in the /mtrain/media/ directory on the phone sd card.
There are 3 main areas I’d really appreciate feedback on:
- Tracking: as you navigate through the content (play videos and take assessment exercises) the app records your activity to submit back to the server (for example, so your course tutor/supervisor can see how you’re doing). The app tries to do this whenever you complete an activity (so connects to the internet at this point), but it may be that you’re offline at the time. Note that a text page is only considered as being completed if you have spent at least 3 seconds on the page before moving to the next one (to at least give some pretence that you may actually have read the content ;-)). To cover the possibility that you may be offline when using the app, but you still want your activity logged, the app also installs a service to try to connect once an hour to submit your activity, even if you’re not using the app. I’m a little unsure that this is the best approach, since I’m not too keen on applications connecting to the internet in the background, but I haven’t yet thought of a better way to handle this. I may just add a preference to allow you to decide if this service is allowed to connect or not. But any comments/thoughts on this appreciated.
- Navigation: once inside a course module, I’d like some feedback on the navigation between the activities for each section in the course. Currently you can’t see a full list of all the activities for a given section, without clicking on the previous/next arrow buttons, as I wanted to avoid the user needing to go through another list selection page to get to the activity, but I’m not sure that what I’ve done so far is quite right.
- App name: I’m not too keen on the app name ‘mTrain’, so any suggestions for alternative names welcome!
In the next days, for those of you who don’t have an Android phone to test this on, I’ll post up a video of the app in action.