Blog Post on Stop the Stockouts for Health IT
Started by Farzaneh Behroozi on 24 Sep 2009
Last edited by Sophie Beauvais on 19 Oct 2009
Dear Friends,
Sophie recommended that I send this post to you for the Health IT group -
http://globalhealthideas.org/2009/09/accountability-aids-and-africa-stop-the-stockouts-financial-oversight-bemf/
Cheers, Farzaneh
Keywords: Mobile Devices
Sophie Beauvais
Thanks for sharing! Copying the beginning of your post here with link to website for members' info:
In my work in the field, I am no longer surprised to see test stockouts, essential medicines stockouts, supply stockouts, broken or missing diagnostic machines, or patients who are afraid of healthcare workers. It is a complete tragedy, and as I work to help, I think of all the people who are sick or die because of failures of the healthcare system, who cannot tell anyone their stories. For those who do not work in the health system, or haven’t had an experience of health system failure, transparency and data on implementation is practically invisible – so there’s no public awareness of the issues.
So I was thrilled to see recent developments in accountability – the Stop the Stockouts campaign: http://stopstockouts.org/, and the creation of the Budget and Expenditure Monitoring Forum in South Africa.
Stop the Stock-outs , a multi-country Africa campaign, is using text messages sent by activists and members of the public to expose stock-outs of essential medicines at public health facilities and put pressure on governments to address the issue. It was launched in Kenya, Uganda, Malawi and Zambia by Health Action International (HAI) Africa. During Pill Check week in June, facilities were surveyed, and a map of stockouts was created. It was found that many government health facilities were routinely running out of, or just not stocking essential medicines to treat common diseases such as malaria, pneumonia, diarrhoea, HIV and tuberculosis (TB).
“We were finding availability levels in rural, lower-level health facilities of 40 or 50 percent for essential medicines,” said Christa Cepuch, a pharmacist at HAI Africa.
Irin news link: http://www.irinnews.org/report.aspx?ReportId=86192#
12:26 PM, 24 Sep 2009 | Permalink
Anup Akkihal
Since traditional global health investment has been directed toward basic R&D – encouraging the creation of drugs or gadgets to fight disease in the developing world – it is refreshing to see growing interest in strengthening the supply chains that deliver these commodities to remote areas where they are needed most. Indeed, the efficacy of medicine is irrelevant if drugs are not available in the first place.
Perhaps nobody has been focused longer on the availability aspect of health system strengthening than the USAID | DELIVER PROJECT. (http://deliver.jsi.com/dhome) The mantra here begins with “The Six Rights”: the right goods, in the right quantities, in the right condition, delivered to the right place, at the right time, for the right cost. It is helpful to frame the “stockout problem” within this context.
Given global technology trends, using mobile phones to overcome logistics hurdles is an increasingly feasible approach. The Stop the Stockouts campaign (http://stopstockouts.org/) is a terrific implementation of SMS technology to bringing attention to an important problem and helping logisticians understand where supplies are needed. Indeed, knowing where & when stockouts occur is important, but it is by itself insufficient. Understanding demand patterns, visibility of inventories, and coordination of resupply are also important aspects to making reliably available medicines in remote places of the earth. In the coming months we are likely to see more and more mobile phone applications addressing supply chain challenges in low-resource settings. Even my NGO is developing a J2ME appliance to help service delivery points manage inventory & compute+schedule resupply in India. Hopefully this will also help to alleviate the stockout problems highlighted by Stock the Stockouts in Africa and the similar problems we have seen in the private sector in Indian villages.
It is a very exciting time for mobile technology to be tested in the field, and I look forward to hearing about more successful implementations such as Stop the Stockouts.
Thanks
1:59 PM, 24 Sep 2009 | Permalink
Jonathan Jackson
Anup - I'd be very keen to learn more about your J2ME appliance. Do you have any pointers to the tool? I am completely sympathetic to the need for better supply chain tools, and I'd would love to learn more about what you are doing.
3:13 PM, 24 Sep 2009 | Permalink
Anup Akkihal
Hi Jonathan,
Appreciate your interest. Our team is small, our developers are altruistic volunteers, and our budget is tight -- so we have prioritized research & product development, while de-emphasizing our web presence. We wanted to be the first to bring mobile-phone-mediated inventory control services to global health, and so all of our energy is focused on this.
The "alpha release" is scheduled for mid-November 2009. Then we'll be able to provide demonstrations, and it will be much easier to get the funding needed to support things like a project webpage, etc.
Since we are keen on widespread adoption, I would be happy to share our work via webex or telcon, especially if you can help identify new markets for the appliance.
Thanks.
3:39 PM, 24 Sep 2009 | Permalink
Jonathan Jackson
Sounds great and I understand the difficulty of spending time on a website while you are trying to do programming quite well! A few things that might be of interest:
RapidSMS - the is an open-source python system written by a collection of developers (both paid and pro-bono) that was been used to track food supply, bednets, malnutrition, and polio vaccines in a variety of countries over SMS. It might be useful to your project, particularly for responding to submitted data via SMS.
JavaRosa - this is an open-source J2ME based mobile-project for data collection. This might be useful but probably not from how far along it sounds like you are.
3:56 PM, 24 Sep 2009 | Permalink
Anup Akkihal
RapidSMS is also an impressive implementation that works without http protocol -- a major advantage when deploying in places without internet connectivity of any kind.
OpenRosa is also really cool, especially because of the integration with analysis & reporting functions. Believe this requires http & associated infrastructure. Just saw a presentation on DataDyne's EpiSurveyor tool at USAID and it was nice to see such a clean application. However, my understanding is that bridging the gap from status quo functionality to a logistics management system is a sizable effort. Yet many features (like integration with R, and export to MS office tools) are attractive value-adds. My NGO team is therefore considering rebuilding on this open-source platform after we've proven the concept later this year. Again, lots of things we'd like to do, but capacity only to do a few of those things.
4:18 PM, 24 Sep 2009 | Permalink
Hamish Fraser, MBChB, MRCP, MSc
It is great to see some progress in this area, as the supply chain deficiencies in developing countries are having a disastrous impact on patient care. A report from JSI on stock-out levels in Tanzania gave similarly alarming figures to the Kenya results with 40% stocked out of oral rehydration therapy for example. In addition to the use of cell phone communication between sites in the supply chain, there is an urgent need to create good tools to manage stock in clinics and smaller pharmacies. These need quick, intuitive user interfaces and the ability to create local reports for the team using the system. Up till now the very large investment by US government and other agencies in drug supply management has mostly focused at district and national level, and/or on vertical programs such as HIV. So it is not surprising that essential drugs in smaller clinics are badly managed.
Hamish
4:24 PM, 24 Sep 2009 | Permalink
Isaac Holeman
Hi Everyone,
Another useful open source tool for managing supply chains is FrontlineSMS (http://www.frontlinesms.com). My organization (FrontlineSMS:Medic) is adapting FrontlineSMS for use in medical settings, including supply chain management with several organizations in Malawi and Bangladesh. A few key features for this use case include:
- automatic installation on a $300 netbook pc takes 5 minutes for a non-developer - key for the small clinics Hamish mentioned.
- $25 java enabled phones can send & recieve forms with short text fields, numeric fields, large text fields, and check boxes. The system can use GRPS, but can also compress forms and send using SMS only. In Bangladesh our 42 field form is compressed into a single text message.
- Within a month FrontlineSMS is releasing a modules platform that will allow other open source developers to extend the functionality of the platform without having to understand/branch FrontlineSMS core. (my group has been developing on this platform all summer).
If you're interested in this system, either as a developer or a potential user, don't hesitate to contact me at isaac [at] medic.frontlinesms.com.
3:09 AM, 25 Sep 2009 | Permalink
Isaac Holeman
ps it's worth mentioning that the stop the stockouts project used FrontlineSMS to receive the stock out reports via SMS. FrontlineSMS is integrated with the Ushahidi mapping engine (the mapping software used for stop the stockouts, read more at http://is.gd/3Eyfr). Both of these tools are open source and free for the public to use. We're working with Ushahidi to do mapping in a more organizational context (e.g. where are my org's community health workers sending messages from? What villages have they been finding and referring cases of diarrhea from?).
3:16 AM, 25 Sep 2009 | Permalink
Mike McKay
Ushahidi is a fantastic system with an amazing set of developers, most of whom live in Africa. It can and has already been used to monitor disease outbreaks as well as doing election monitoring and other very needful tasks. I was really excited to hear that it is being used for stockouts.
One question though. Everyone says that FrontlineSMS is open source, but the source code is not freely available for download on the web. Using the term "open source" seems like false advertising to me. Even downloading the software requires filling out a request form and waiting for them to clear you and email you the software. What's up with that?
11:31 AM, 28 Sep 2009 | Permalink
Isaac Holeman
Hi Mike,
The rumor that FrontlineSMS is not open source has sadly been going around for ages, but the code has been on source forge for an awfully long time (the 1,087th commit was a few days ago). Before that Ken would personally share the code with people who were interested in collaborating.
http://sourceforge.net/projects/frontlinesms/
I think the confusion comes down to how you define the "open" in open source. Ken made the unconventional decision to build a tool that was first and foremost useful to end users, and to not invest *any* resources (time or money) in fostering developer collaboration until the tool had an active and committed user community. This rubs a lot of developers the wrong way, but for the end user it means the tool is *much* more polished than it would be if Ken had dedicated more time to cultivating developers.
Not spending time cultivating the developer community made the project less open to software developers than it could have been - there is a difference between available code and open code (likewise, you could argue that even a rockstar open source community like OpenMRS isn't as open to nubes as it could be because the documentation isn't as thorough as it could be).
Now the FSMS has the largest and most active user community revolving around a software tool for mobile phones in development, so the next stage for the project is to launch a true OS developer community. Ken has been working with Alen Gunn of Aspiration Tech (who also helped get the OpenMRS community going) to set up tools for collaboration among developers, including good documentation. In the coming weeks they will also be releasing the modules platform, which will function similarly to Wordpress, Firefox, or OpenMRS in that it will allow anyone to write a module that will extend FSMS without digging into the core code (my group has been developing on this platform all summer).
Sorry this is such a long-winded response, I just want to put the rest the recurring charges that FrontlineSMS isn't open source, or that it's begrudgingly open, or that the people behind it don't "get" OS. The bottom line is that they explicitly prioritize end users over developers, and will continue to do so even while they begin to invest time in fostering developer community tools. A lot of OS communities would do well to learn from this hard nosed focus on end users.
(full disclosure: I'm not part of the FrontlineSMS team per se, but my org is building modules on top of the FSMS platform).
ps. It annoys me too that you have to fill out a request form to download the application (not the source code). It's better now that the download happens immediately after you fill out the form, rather than waiting for an email. The request form is still there because being able to get basic info about the people who are downloading it is HUGE when you start talking to funders about proof that your tool is useful.
5:06 AM, 29 Sep 2009 | Permalink
Mike McKay
Thanks Isaac. I think in this case "perception is reality", and the perception has stuck with FrontlineSMS for a long time. Glad to hear that you guys are trying to change that.
I don't accept the argument that keeping something closed at the beginning makes a project more open in the long run, but we can agree to disagree on that.
A few suggestions:
There should be a link from the FrontlineSMS website to sourceforge - anybody could put their project on sourceforge and only point trusted people to the code. It is good news that the download is now available immediately, but I still perceive funny business. You must "request" a download and there is text saying "if you're a non-profit organization..." implying that it is only available for non-profits, yet there are thousands of for profit businesses (health clinics) in developing countries that would benefit from FrontlineSMS. There is even a captcha! The entire page should be replaced with a link to sourceforge, and a download link that is only activated after you enter your details with an explanation about why that is necessary.
I hate the annoying "that's not open source" trolls as much as anybody, so it sucks to have to be that guy in this case. Yet, I think FronlineSMS is an important piece of software that needs to embrace the open source community and indeed the international development community (not just non-profits) more effectively. The open source initiative is considered the authority on what open source software is, and according to http://www.opensource.org/docs/definition.php, FrontlineSMS seems to be stretching the rules.
9:45 AM, 30 Sep 2009 | Permalink
Isaac Holeman
I actually appreciate your critiques because you're making them in an open forum - here it can function as positive peer pressure - very different from when people make the same critiques via email or at conferences without giving voice to anyone familiar with FSMS.
I agree about the link to source code on teh website, and it's on the way. I think the "if you're a non-profit" thing is a non-issue though. I work at a "for profit" hospital that has been one of the most prominent FSMS users for more than a year. It's just an old string of words that no one deemed it necessary to change.
OSI has obviously set the bar very high, excluding the myriad groups that participate in the spirit of open source but do not have the legal resources to declare a specific license appropriate for their project. This would include Baobab, but Baobab is still open source, right? Like FrontlineSMS, Baobab's source is available on a public repository, but it's not linked to from the website, no license meeting OSI's criteria has been declared, and when I first went looking around for the source, it took me months to find it (I wasn't familiar with github at the time).
The main point though is that positive peer pressure in a public form is good, and I know FSMS is working hard to respond.
3:23 AM, 1 Oct 2009 | Permalink
Kenolisa Onwueme
Anup, it's glad to hear you are near an alpha release. Looking forward to a demo sometime. Tools like that obviously have tremendous potential for horizontal application in a the myriad flavors of supply chain fracture (beyond the intended usual focus (malaria/hiv/tb). kco
11:51 PM, 2 Oct 2009 | Permalink