Hi!
I work with Omar at IRD
@Neil - Most of your questions are program related so it'll take us a little while to find out some of the answers. Some things that I think are true, that Omar will have to confirm. Treatment supporters have a few patients, I think the max is 5-8 - being a treatment supporter for MDR-TB is time intensive. I'm not sure whether they're paid or not. If they're not paid they will be incentivized through food baskets similar to what our patients receive. They don't work on "getting new clients" no. They are not traditional community health workers and are not trained for case finding or anything like that. They are a nominated treatment supporter to help the patient through the 2 year regime. As Omar mentioned they would be looking out for symptoms within contacts.
We're focused on MDR-TB at the moment although the Indus Hospital (where the MDRTB program runs) also has a regular TB program. MDR is a massively unaddressed problem in Pakistan - we already see a few XDR patients - so there's an urgent need for scale-up and support on this v. dangerous and difficult to treat disease.
Omar will be able to comment on lessons learnt - he's been running the program from a technical stand point. As far as I can tell it's going quite well.
Paperless for this kind of activity seems like it would be a goal. The TB program itself has many paper-based reporting requirements for the national government etc. But, as you said, this is a management problem. If we're going to scale-up effectively, I believe this system is not much better than paper if we don't use intelligent flagging and reporting to trigger alerts to supervisors & co-ordinators. So I hope that these are the kind of things that will happen moving forward. 2011 is going to be a v. busy year for our TB program - so we'll keep you updated.
Regarding adherence, as you may know, we are part of the team developing XoutTB. The group in MIT is working on the strips and we are both developing the software - for reminders, verification & incentive delivery - as well as our program team doing a randomized control trial to study the effectiveness. This is the only other adherence work we're currently involved, but there's no particular reason for this. I think it's important to highlight the difference between what we're doing on mOpenMRS and XoutTB. The mobile application discussed here has come out of an immediate programmatic need; whereas the XoutTB system is still research at the moment. If the XoutTB system proves effective then it may well become part of programs, but it is still in testing. XoutTB and Smart Pill boxes aim to replace DOTS (and hence reduce costs and enhance scale-up opportunities); this application is designed to support DOTS in its current form. The two are not mutually exclusive as you'll never completely eliminate treatment supporters, so there's no reason why the two can't work together.
I think the important thing is that good technology doesn't make a bad program better, but it can make a good program more efficient. So for now, with mOpenMRS, we're just trying to help our TB program team get better access to data faster, while we continue to work on researching new alternatives.
@Stepen - OpenMRS-JR was a demo app built by the awesome team at Cell-Life - it's just a prototype for demo purposes to show that the javaROSA library can be used to build an app that connects to openMRS. When I last spoke to them about it in Cape Town in September it wasn't a functioning tool and I don't think they were developing it further - this might have changed. If you've seen a working implementation I'd love to know.
When we came to actually implement an openMRS mobile system, we couldn't find any J2ME apps that were actually being used in the field.
Because there had already been work with the xForms module and the openXdata mobile client we decided to take this and build on it for our pilot. We also like the fact that the openXdata client supports lower end phones. However, as we are using 2710NE's in this case it's a null point.
We actually work closely with the Cell Life team on openXdata, but these projects aren't related.
mOpenMRS was developed by Omar and the informatics team at Interactive Research & Development who are based in Karachi, Pakistan. We're an independent non-profit who do both pubic health research and community outreach & treatment. Our website is really, really bad at the moment (we're working on it) but we do have a few videos posted at
youtube.com/irdresearch. Our informatics group exists mostly to develop tools and support the needs of the many different projects we run (both research & service). MDRTB treatment has been a big part of our program team's focus lately and this is where the request for this software came from. We have a couple of hundred MDR-TB patients on treatment at the moment, and we want to improve our treatment management tools before we continue to scale this number.
@Alvin - the connection is made over GPRS not SMS so no short codes necessary. We give the field workers phones so we just pay the bill ourselves.
Sincerely,