Local Development Panel Discussion (July 19-30th, 2010)

By Neal Lesh | 07 Jul, 2010 Last edited by Joaquin Blaya, PhD on 11 Nov 2010

GHDonline.org is pleased to host its second Health IT Expert Panel,

"Local Software Development for Global eHealth."

July 19-30, 2010

Join the members and moderators of the Health IT online community in
discussing in-country coding and the development of local eHealth capacity.

With panel moderators:

Ahmed Mohammed Maawy, DataDyne

Melissa Loudon, University of Cape Town

And panelists:

Jacob Mtalitinya, ITIDO, Tanzania.

William Aviles Monterrey, Sustainable Sciences Institute, Nicaragua.

Lim Chanmann, InSTEDD iLab, Cambodia.

Ali Habib, Interactive Research and Development, Pakistan.
 
Spanish and French translations will be available on the discussion page.

The discussion will occur over posts to the Health IT community which you can access via email or through the web

For help, please email

Replies

 

Neal Lesh Replied at 1:55 PM, 13 Jul 2010

Hi everybody,

In anticipation of the panel discussion, I'm attaching the Bios of the panelist and moderators.

Looking forward to when we start this next week!

Take care,
-neal

Attached resource:
  • Local Development Panel Bios (download, 15.0 KB)

    Summary: Hi everybody,

    In anticipation of the panel discussion, I'm attaching the Bios of the panelist and moderators.

    Looking forward to when we start this next week!

    Take care,
    -neal

    Source: Dimagi

Neal Lesh Replied at 6:56 AM, 19 Jul 2010

Welcome to the second panel of GHDonline Health IT community! This will run for the next two weeks. The topic is local software development for global eHealth. This can be a difficult and delicate topic, but we are confident that we can have a constructive and open conversation. Please share your stories, questions, comments, and proposals with us! Some of the questions we’d like to explore are:

1. What is the current contribution of local developers to eHealth software used in low-income countries?

2. What would be impact of having more eHealth software developed by people from the countries they are being developed for?

3. What would be the best ways to facilitate more local development of eHealth software? Are there small steps we can take now as well as larger changes that are needed?

4. What are current strategies or movements for promoting local development?

To kick things off, we will hear prepared comments from our four panelists. I sent their bios out last week. We have a total of 3 pre-written posts, with each post having a comment from each panelist. The first post is below, in which the panelists introduce themselves and provide some background for the panel. More coming soon!

Ali Habib, Pakistan: Hi all! I’m Ali and I work with the IT team at Interactive Research and Development (IRD) in Karachi, Pakistan. My focus is primarily on software development for mHealth. IRD is involved with a number of mHealth projects but the ones I’ve been directly involved with include Pneumonia Surveillance using RFID enabled cell phones, introduction of MDR-TB DOTS using mobile clients for OpenMRS, and most recently using cell phones to monitor patient compliance using reminder/incentive mechanisms. I’ve also been working on maintaining the OpenMRS MDR-TB deployment at Indus Hospital in Karachi. Going forward we are looking at a nationwide MDR-TB program scale-up in Pakistan and plan to assist other countries in the region with similar projects.
I got involved with IRD and consequently mHealth almost purely by chance. When I returned from the US last summer post financial crisis, the only jobs available for someone with my profile were positions in the IT departments at banks. No offence to those of you in the financial industry but I’ve sworn not to work in a bank unless I’m destitute and I find that the bureaucracy of large corporate organizations drives me insane. So I kept looking until by chance a friend connected me with Aamir Khan at IRD and I found the idea of technology for health very exciting and challenging.

Jacob Mtalitinya, Tanzania: The Invention and Technological Ideas Development Organization (ITIDO) is a non-profit organization registered under NGO's Act of Tanzania with the aim of creating a collaborative environment for inventors to innovate and share ideas and processes with individuals from the academic, non-profit and public sectors. The mission of ITIDO is to encourage, support and promote technological and innovative ideas. The motive that lead to establishment of ITIDO was to create an environment that will enable innovators and inventorsk, individuals and organizations to work on their ideas that address some problems in the community. We also aim to provide skills and resources and support that might be needed to facilitate the realization of these ideas. Currently we have six staff working for ITIDO. The technical team consisting of 2 Developers and 1 Analysts/Project Manager, the remaining team is handling administrative activities.

Lim Chanmann, Cambodia: I am a software developer at the InSTEDD iLab. I joined InSTEDD with a local web-based system development background in September 2008. At InSTEDD I was first introduced to mobile application development for an Avian Influenza (Bird Flu) hotline system, and learned not only programming languages, design patterns, best practices and principles, methodologies like Agile, but also about the open-source community, designing usable software, especially technologies to foster health care innovation. For example, Geochat which allows collaboration through group communication, ReportingWheel which is a disease reporting tool, ResourceMap which lets you visualize and track your work/resource on the map and can be accessed from anywhere to mention a few.

William Avilés Monterrey, Nicaragua: Hi everyone - my name is William Avilés Monterrey, and i currently work with the Sustainable Sciences Institute (SSI) in Managua, Nicaragua. I started my work in public health in the Ministry of Health in 2000 developing and implementing information systems in various areas, mainly epidemiological surveillance and research, but also in disaster preparedness, tuberculosis and HIV / AIDS surveillance and reporting systems, among others. In 2005, I began working with SSI, mainly in observational research studies about Dengue, where I spent several years implementing a range of applications (MS Access, Pendragon forms for PDA field data collection, barcode and fingerprint ID systems for cohort participant and biological sample preparation, etc.) based on the requirements and needs in these studies. However, within SSI's small IT team, we chose software and tools without any knowledge of what was happening in similar studies in other parts of the world. We didn't realize that there were others working on some of the same problems in information management. In 2008 I participated in a conference in South Africa about open OpenMRS and a few mobile data collection applications that are part of OpenROSA. After that, our vision at SSI turned around and we started to work not only just on our studies, but also outside the institution, collaborating with other groups to learn more about open source health informatics tools. Since 2008, we've been more involved with the MoH in Nicaragua to offer support and to advise them on how to use the few resources they have for ICT in the most efficient way. Our goal is to try to help them to develop local capacity in the country to support these tools long term.

Neal Lesh Replied at 10:22 AM, 19 Jul 2010

Hi again.

Here is the 2nd round of comments from the panelist, describing some of the context and challenges around local software development. We'll post the 3rd and last comments later on, after we have some time for questions and comments from you all!



Ali Habib, Pakistan: IRD is working actively to enhance its contributions to open source mHealth software by increasing our involvement in software development. I’m of the opinion that local programmers working on projects for local use is ideal because they are better versed with the environment that the software will be used in. From a usability point of view, they understand how users will think and what would make systems easier to use. It is also a great means of supporting health development in your own country, something that attracted the first programmers who came on board at IRD. This is not to say that programmers from other countries are not passionate about their work, but the idea of doing something that helps your own country is a little bit different.

We work primarily with openXdata and OpenMRS, and are now looking at OpenClinica as well. Writing code for these projects requires programmers who are well versed with enterprise Java development and the J2ME platform. Such programmers are hard to find in Pakistan where the majority of computer science graduates move very quickly to the Microsoft .NET platform and prefer to obtain certifications for proprietary systems like Oracle instead of taking the open source route. This trend is changing slowly with open source becoming more and more popular but the allure of high paying commercial ventures using proprietary platforms and technologies has been one of our biggest hurdles when it comes to finding software developers.


Lim Chanmann, Cambodia: In a low-income country like Cambodia, where facilities, infrastructure, and technological knowledge of people are limited, the local capacity of developers to produce high quality software to meet the deadline constrains locally developed eHealth software. But outsourcing to expensive experts decreases sustainability by preventing future development, modification, and real-time support.


Jacob Mtalitinya, Tanzania: Currently there is not much contribution from local developers to ongoing eHealth projects. This is due to number of reasons such as skills, funding, resources and lack of enough experience to handle such kind of projects. So far we have faced some challenges such as experts to provide mentorship, funding, acceptance to the community. Our technical team has been getting mentorship and technical assistant through this and they are now capable of developing eHealth systems and mobile applications


William Aviles Monterrey, Nicaragua: Nicaragua is the second poorest country in America; this leads to many challenges related to resource allocation in public health. In Nicaragua, the software development for public health is very limited. The Ministry of Health (MoH) of Nicaragua does not have a team of programmers dedicated to develop useful applications for health information management needs. Currently all systems development in the MoH is done through support from international cooperation or projects that involve national NGOs. Because Nicaragua is a very poor country, most of these projects use funds to directly invest in primary health care services, and there are very few projects that budget for or take into account software development itself.

Regarding to the universe of qualified programmers in Nicaragua, we can say that it is small; most qualified developers work in private companies that offer decent pay for their work, for example, in banks, transnational corporations, or the telecommunications industry. Unfortunately, the MoH and most small health NGO’s can’t offer competitive salaries for IT support staff, let alone highly qualified developers.

On a positive note, there are a growing number of professional graduates of universities that have the ability - but not the experience - to develop software related to the health field. Unfortunately, there is not a strong drive to work in the health sector (90% of which is public, i.e. government health system) for this new generation, due to the lack of competitive salaries. The public sector in Nicaragua is the lowest paid in the country making it the last choice for talented developers.

In my view, the main problem impeding the development of health informatics developers in Nicaragua is the lack of policies to encourage software development at the institutional level in the MoH. Part of this challenge has to do with very limited funds for use by the MoH for improving their health information systems. The other part of the challenge has to do with a history of bad experiences in the past with development of systems by foreign business consultants without any institutional capacity building to manage or maintain those systems.

Melissa Loudon Replied at 3:03 AM, 20 Jul 2010

Hi everyone, thanks for your comments!

To kick things off, would anyone be willing to talk a bit more about the demand for local software development? Both William and Lin mention sustainability problems in outsourced projects, and I wonder how well-recognized this is in the areas where you work. Do you think the 'history of bad experiences' is prompting government/NGOs/funders to look for alternatives in the local IT sector, and if they did, what would they find?

Steven Moore Replied at 3:33 AM, 20 Jul 2010

I think openXdata has shown us how easy form development can be, and InSTEDD has a niche with syncronization. What I wonder about is the use of standards in health app development (ie, the content). Where does one go exactly for the standards used for Gender or Blood Type or any other thing that one is likely to make into a radio button or dropdown list? I think a portal of such health data, with web or ReST services would be very helpful. Multi-lingual of course.

Nicolas Pottier Replied at 3:51 AM, 20 Jul 2010

Hi everyone,

I've had the experience of evaluating local capacity to develop RapidSMS systems in two countries, Rwanda and Madagascar, and so far I think the primary challenge is not talent but experienced leadership. If we treat the term literally, the 'capacity' or 'potential' is there, but without the guidance of someone more experienced it is going to take them a long time to be effective.

I think there is a general feeling in software that if you can figure out how to code then you have it figured out. That combined with the rarity of individuals who have reached that individual contributor level makes it so that local organizations often have a much higher assessment of their own skills than is really the case. The problem is not potential here, simply that learning a good dev process and how to work as a team without any guidance is simply impossible.

So my recommendation for any companies trying to do this is to make sure to hire some experienced developers to help mentor the local talent. You need at least one person who has built production software, understands the processes needed and who can draw on that experience, to be effective. The need for that seems to be greatly under appreciated.

We ourselves are doing this in Rwanda, and even here with experienced developers, I expect it to be challenging. But that said, I think the demand and desire to use local developers is very very high.

Nic Pottier
Nyaruka Ltd.

Melissa Loudon Replied at 4:15 AM, 20 Jul 2010

Hi Nic, thanks for a great point.

RapidSMS, like many of the other open source systems mentioned in the intro posts, is an international community with an established core of (mostly) Western developers. I'm very interested in how local groups access mentorship within these communities. Do you think there are structures in place that make this possible?

Panelists (or others!), it would be great to hear your experiences of working with open source communities.

Ahmed Maawy Replied at 4:16 AM, 20 Jul 2010

Hi,

So far we have discussed the following important points:

1. Skill - in both open source and closed source systems. We have seen that supported technologies with commercial backing (Such as Microsoft .NET) receives more attention and has more skilled developers compared to open source equivalents. Without proper skill in Open Source technologies projects can get expensive during implementation. What can facilitate the development of open source skilled talent?
2. Open Source has been a major player in the global health arena (OpenXData, OpemMRS, ...). Proprietary systems can get very expensive especially when the public health project's budget is limited. What can be done to help solve this problem and grow open source technologies?
3. Leadership - Even when the skilled developers are there, experienced and talented leadership is lacking. How to close this gap?
4. The will from local authorities to boost local capacity through supporting the growth of local talent. What can be done to encourage this?

Caroline Mbindyo Replied at 6:24 AM, 20 Jul 2010

Hi
We have been working on various e/mhealth applications here at AMREF. One challenge I face with apps developed outside Africa (or by none Africans) is are not not generally developed with the proper context in mind. This is certainly a generalization and should be interpreted as such.

Second, although I know that there are many organisations implementing e/mhealth programs in Africa, it is not easy to find out who is doing what and sometime find myself re-inventing the wheel so to speak.

With regard to local expertise, I find that there is no shortage of capable young developers (in Kenya at least). One thing I hear repeatedly from this group is that they are "expert" in developoing eSolutions but not in identifying challenges that may be mitigated by an eSolution. I see the Apps4Africa competition (apps4africa.org)as one way of bridging this gap where civil society organizations who have a particular technology need can post them on the Apps4Africa ideas board and ask for help from African developers who might develop these tools for the competition.

Gayo Mhila Replied at 8:32 AM, 20 Jul 2010

Hi guys
I’m working as an implementer supporting different projects in Tanzania. We develop and provide mobile tools to support community health workers. In the past few years I have spent a lot of time in the field with users to collect user requirements before software development. I also spend some time with users during testing and refining of the application.
After the work I have done with helping community health workers in the field, I see a huge value of supporting local software development.
Here are my few points regarding local software development.

1. It is better and a bit easier if the person who does software development understands the phrase, culture and the language of the person who documented the requirement specification. It is even more helpful if the software developer understands the language of the users too.
2. Strengthening the local software development can minimize the number of unnecessary misunderstanding between users and developers. It make easier for the implementers during refining process/testing phase. As many implementers, my experience has been no matter how well you document and understand the existing system and involve the users, there are still some changes and suggestions that need the developer’s attention during the testing phase. There are some changes that need to be made while in field which make it very difficult if developer is not locally available. Example, in one of my project, I gathered all the requirements from users. I wrote down a document, I printed it and went back to discussed the documentation with the users to make sure I had understood them correctly. After the discussion we agreed that I had written that right but when I went back with the electronic application and the users could actually test it, they made many new suggestions for things to add. Because the changes they requested were minor and I had the ability to implement those changes right in the field, it allowed us to much more quickly get to a system they wanted to use.
3. Software develop locally can help new technology penetrate and more easily be adopted by local leaders. At this point I am talking the chance of penetrating and sharing the technology with local authorities. Very often people can accept and trust something made locally especially if they meet with developers. As in many places around the world, if there is any new technology, the authorities can hesitate to adapt it quickly. We can reduce the amount of doubt if the local authorities know that they can visit and talk to people who developed it. They can also be proud to see that their local people have worked on the tools. If the software is developed away from home, people can say what if the developer loses interest on the technology. What will happen, where we can go if we have a problem?
4. To conclude this post, I think it is a good idea whenever we do software development, we should think how to empower and strength the local developers not only for sustainability purpose but also for the effectiveness and success of our projects both to users and wining the authorities.
We can achieve this if governments, NGO’s and funders can look at alternatives and support local IT sectors

William Aviles Monterrey Replied at 10:58 AM, 20 Jul 2010

Hi Melissa

Thank you for the post,

I think that the needs in Nicaragua regarding public health software development is high. Unfortunately, I think MoH's plans aren't consistent with this. Reality shows that outsourced non-local projects aren't the answer due to many factors and authorities have realized this. There's a lot of potential for creating development projects using local developers but there's not an elaborated plan to follow. I think that despite this potential and the local capacity, there's no resources to bring more experienced developers aboard in such projects. Funding is somenthing missing, but if the funders look for alternatives, there are a couple of local companies and NGOs dedicated to develop software that would be very interested in this kind of projects. The challenge is make it sustainable, and here is where training play its big role.

William Aviles Monterrey Replied at 11:04 AM, 20 Jul 2010

Hi Nicolas

Great post,

I completely agree with you. One approach for this is working directly with the developers while they're studying. In this way they could gain experience while learning and in the future they would be invaluable.

Organizations that really believe in local capacity should be involved in this process, and they should commit in using their resources to build talented young local developers within our countries.

William Aviles Monterrey Replied at 11:17 AM, 20 Jul 2010

For now, in our organization we have experience working with the OpenMRS community. We organized (with eHealth Systems in Chile) the first IMeCa (Informatica Medica de Codigo Abierto - Open Source Health Informactics) meeting for latin america last year, where we had a developer course with some of the core developers from the OpenMRS community. This year for the second meeting, we'll have a training developer course for mobile opensource projects in November. The whole idea is to establish local developers as an active part of these communities.

Gerald Mutuhu Replied at 1:29 PM, 20 Jul 2010

I agree with most of you that one of the biggest challenge for local
software development is retaining skilled developers. In the
developing world ,software development is not well developed as a
career. Most developers leave software development after working for
3-5 years.

Some ideas on how local software development and implementation can
strengthened:

1.More collaboration and coordination among NGOs , government and
other institution in ehealth initiatives. A good example of how this
can lead to better systems is the OpenMRS implementation by AMPATH ,
Indiana University and USAID in parts of Riftvalley province in Kenya.
2.Better terms for developers .Most software development jobs in the
health sector are usually short contracts.
3.Training developers and other health workers on open source health
information systems.
4.Establishing standard bodies to ensure that the information systems
developed are interoperable.

--
Regards,
Gerald Mutuhu
Skype:geraldmutuhu

john wesonga Replied at 4:06 PM, 20 Jul 2010

The largest software implementation projects should generally leave the biggest footprint in the areas where they have been implemented but in most cases this is not what happens. We find many instances where software development teams have come in from outside the country done their work, often lasting months, and left with little or no effort on knowledge and skill transfer. When this knowledge is retained within a select few it becomes an expensive commodity that most local organizations don't have the resources to attain. Lets see a major focus on identifying local organizations, building capacity and equipping them to grow and manage software projects,

Nicolas Pottier Replied at 4:29 PM, 20 Jul 2010

Thanks for the insightful comments everyone.

Someone asked, how do we encourage capacity building via the partnership of experienced developers and local talent. I think there are two critical things to consider here.

1) Paying what's fair. I've seen a tendency by organizations, large and small alike, to severely discount the rate they will pay local developers, only because they are local. While it is true that the differences in cost of living must be accounted for, it also prevents the local startups from being truly sustainable. The payment should be based on quality, with cost of living considered to some extent, but not overriding it entirely.

2) Local development organizations need to recognize their limits. In Rwanda, the government actually has an active bias AGAINST using local developers, because they have been burned too many times previously. This is primarily a problem of overcommitting and lack of experience. But it must be recognized that when you fail, it is a judgement put on the community at large, not just you.

My last point is simply that we must balance the desire to build local capacity with the project at hand. Although I absolutely agree that local developers bring about a perspective that is required, that should not supersede the ability to deliver the final product. This is a hard thing to judge in practice, as the result of projects is not binary, it is qualitative. But I think every organization needs to consider long and hard as to whether their true mission (separate from building IT capacity) is being compromised by using local developers.

I feel strongly that when that all other things being equal, the bias should be towards local developers, but never at the cost of the project as a whole. Supporting the growth of IT is important, but it is an entirely different mission than most of these projects are aiming for, we must not sacrifice the final result in implementation.

Nic Pottier
Nyaruka Ltd

Ali Habib Replied at 4:55 PM, 20 Jul 2010

Gayo, those are some great reasons why it's important for local developers to get involved in health IT projects in their own countries. However, as Nicolas rightly points out, the problem is often the absence of guidance.

Williams's point about connecting with developers while they're still at school is one potential solution. Another way to deal with that is to seek mentorship within open source communities. The OpenMRS community, for example, has a mentorship program that is designed to help those relatively new to the community make enhancements to the system. In the absence of experienced individuals at local sites this is one way to help create that expertise over the long term. In the short term though, the need often remains unmet due to the absence of funding to get the right person on board.

Regarding proprietary platforms, one hurdle is that open source is still seen by many as a sort of brave new world. Many are simply unaware of how stable and usable these platforms are and so they shrink from using them. If open source platforms are pushed by governments, and people are made more aware, that could lead to a push towards open source being more widespread and consequently a surge in open source developers. The Pakistan Software Export Board began such an initiative a few years ago, by sponsoring Linux, PHP, and MySQL courses in various cities in Pakistan to push open source as viable alternative to proprietary solutions. I would say that more such efforts are needed with cooperation from local organizations.

john wesonga Replied at 5:23 PM, 20 Jul 2010

Nicolas, thank you for pointing out the issue of paying a fair wage for the work. I see this happening many times where local and sometimes very skilled developers are paid very low rates compared to foreign developers. This leads to many developers especially those working for NGOs/NPOs being disillusioned with the work itself.

William Aviles Monterrey Replied at 6:00 PM, 20 Jul 2010

Hi

Mentorship is a good way to do this, but, like some others metion before,
Organizations have to pay what's fair for the local people. In order to keep
trained developers working in Health projects, as Gerald mentioned, we have
to offer "better terms for developers .Most software development jobs in the
health sector are usually short contracts". I think this is crucial, or
else we'll lose any trained personel for these reasons.

Kieran Sharpey-Schafer Replied at 7:24 PM, 20 Jul 2010

Thanks for a fantastic discussion, and the many valid observations of the current status of this challenge.

Touching on Nic's point about leadership & experience, I think this is especially crucial and have observed this gap a number of times.

One addition to mentoring local developers, would be to specifically look to support & mentor the non-development roles in local tech teams. Specifically, I suspect we should also be looking to build up team leaders to manage all the human aspects of tech projects, such that the developers can focus on their functional area. Sometimes we see the developers trying to cover all roles, but if there was more invested in the team infrastructure around them it would perhaps improve their capacity.

Perhaps this can be summarised by:
1. Invest in local leaders
2. Invest in local developers

I think there may be some successful precedents for this (datadyne? OpenXdata? DTree) and Unicef are trying it in the recent project in Zambia.

Obviously, the challenges mentioned of short-term contracts etc can be a critical barrier to such efforts.

Ahmed Maawy Replied at 4:36 AM, 21 Jul 2010

Some great points indeed. So far we see the following major suggestions to improving local software development initiatives (Especially in the open source arena):

1. Investment into local leaders
2. Investment into developers
3. Mentorship - Developers and leaders
4. Better terms for contracts offered to local talent
5. Incentivizing and prioritizing local talent by local authorities

I do see a challenge in the implementation of these issues. In order to find a resolution we may ask ourselves the following questions:

1. How can we facilitate the implementation of these issues and make an impact in the shortest time-frame possible?
2. Which parties may get involved and how can these parties push for these issues to get implemented?

I would love to see the comments on these two questions :)

Peter Lubambi Replied at 5:54 AM, 21 Jul 2010

Hi all,
thanks everybody for the fantastic points.
Commenting on Ahmed's questions,to facilitate the mentor-ship of Developers and leaders the following can be involved
1. Universities,encouraging them establish open Source Units/courses under departments or Students ICT Associations by sponsoring these Initiative
2. Local private ICT companies, encourage them get involved in the initiative by establishing Open Source Units by sponsoring them and provide trainings.

This can help implement the issues in long-term frame.

Neal Lesh Replied at 5:54 AM, 21 Jul 2010

It’s terrific to see such an interesting and illuminating conversation! I’m very much looking forward to more. I’ll take Ahmed’s questions as a good opportunity to post the last of the panelists pre-written statements which focus on how to address some of the barriers to local development, to help start the discussion:

William Aviles Monterrey, Nicaragua: A better use of resources in my opinion would be to develop eHealth tools using local developers and with buy-in and ownership of the MoH throughout the process. In order to achieve this, we should encourage investment of resources in the eHealth projects involving socially committed young programmers, ensuring appropriate remuneration for their work.

Also, it’s essential that young students receive a social vision when they are in universities. We need to strengthen curricula to include optional projects related to actual implementations of eHealth systems locally. In this way, students will gain both professional skills in applied project work (not just theoretical training) and they will also gain skills in the development and implementation of health information systems specifically.

In SSI, we are working towards reducing these problems. We are developing software for eHealth using recently graduated students from universities; we are in discussions with the authorities of the universities to help build interdisciplinary curriculum including both public health information management and software development, and we are developing networks of cooperation and collaboration in communities in Nicaragua and abroad. We firmly believe that training of local resources is the real way to promote local development and implementation of eHealth tools.

Jacob Mtalitinya, Tanzania: Code in Country is one of the approaches which can be used to strengthen capacity because it allows interactions of local developers and international developers which enable easier knowledge and skill transfer. It also provides room for young local developers to participate in big and complex projects which will ensure sustainability, ownership and local support of eHealth projects in future.

Lim Chanmann, Cambodia: Because the InSTEDD iLab adopted the coded in country initiative, the ResourceMap is fully under local development here in Cambodia and benefits from InSTEDD’s vision of local capacity building, product innovation, sustainability, and having local people to make the decision and evolve.

Ali Habib, Pakistan: One strategy we used to bring programmers on board was a Java training session that we held in Karachi and used that as a pipeline to find programmers. The intensive 10 day training program was taught by a professor from the University of Bergen in Norway. Because most of the course participants were fresh graduates who knew little or no Java, the course was a basic introduction to Java and proved very useful in finding programmers who were interested in Java and were attracted by the idea of health IT. These were junior programmers however and while they have proven very eager and have made great progress, there is a somewhat steep learning curve that is still being negotiated. We plan to follow up with more advanced courses in the near future to try and attract more experienced individuals to the field. We also have videos about our work posted to Youtube and these have proved very useful in getting through to prospective candidates about the scope and impact of health IT both in Pakistan and globally.

Caroline Mbindyo Replied at 9:08 AM, 21 Jul 2010

Nicolas Pottier makes the point that "Supporting the growth of IT is important, but it is an entirely different mission than most of these projects are aiming for, we must not sacrifice the final result in implementation". Working in eHealth with a health organisation I must say that developing inhouse IT skills has been critical to the success (and sustainability) of our eHealth programs. Developing these skills is a costly venture and one which is not necessarily "highly regarded" by many donors. An area that needs further discussion?

Gerald Mutuhu's observations on the career path for many developers (Most developers leave software development after working for 3-5 years)suggests that entrepreneurship skills to assist developers monitize their applications is critical.

Regards,
Shakwei Mbindyo
Twitter: @shakwei

Joaquin Blaya, PhD Moderator Replied at 1:04 PM, 21 Jul 2010

This has been a really interesting discussion so far. I worked at the non-profit Partners in Health in Boston and am part of the OpenMRS community, and in seeing the need for local development in my home country of Chile, I moved back to Santiago and have started a local company (named eHealth Systems) to provide open source health IT solutions (mostly OpenMRS and OpenROSA) to the public and private health sectors. We started the company Sept. 2009, and what we are finding difficult is penetrating the market with a solution. This involves
a) Figuring out what the local needs are that can be answered using health IT and seeing if there is a market for that solution i.e. enough organizations willing to pay for it
b) Seeing if that health IT solution is within the reach of our budget. Even though we are using open source platforms, there is still customization that is required and that customization may require too much time/money for us. For example, we are looking to build a primary care EMR for Chilean public health centers using OpenMRS, but customizing it (like creating large reports required by the Ministry of Health) and creating new functionality for local needs (such as an agenda for patient scheduling, a pharmacy inventory system) require months of programming time and an initial capital that we don't have yet. We are now looking to Chilean government programs, after looking for funding from international organizations.
c) Getting your first customer/contract. We have found this to be difficult, partly because the public health system is very conservative i.e. they prefer larger companies and don't want to be the first to try a program and also they have limited funding at hand and it takes them months to be able to designate funding for a project.

Our case is perhaps somewhat different than others in that
a) Chile has a large pool of programmers and in that pool those that know open source software, so we've been able to hire two programmers who knew Java before starting
b) There is little international funding to Chile because it is a middle income country

But even then I think the lessons we've learned apply to any new startup in health IT.

On a slightly different note, we as an organization are actively promoting the use of open source health IT in Latin American and the development of local capacity through IMeCA (the open source community and conference that William mentioned), and also actively looking for partners to work on projects together (like for example with SSI), so if any one is interested in either the conference (which will occur in Nicaragua in November) or possible collaborations in Latin America, please contact me at <mailto:>. Along with SSI, and other organizations, we are actively looking to promote other NGOs and companies in Latin America to work with OpenMRS, OpenROSA and other open source platforms.

Warm regards,

Joaquin


___________________________________________________________________
Founder, eHealth Systems Chile
Research Fellow, Harvard Medical School/Partners In Health
Moderator, GHDOnline.org<http://www.GHDOnline.org>

juan bru Replied at 1:42 PM, 21 Jul 2010

Item C pointed by Joaquin Blaya is a real issue.
Capacity Building in eHealth is one of the current motos for both developed and developing countries.

But usually, Contract and Vendor management is not included in this Capacity Building, and it should be a priority that governemnt officials in charge of procurement receive real training in these, despite of OMS recommends eProcurement policy development. Big blue chips are 10 times more expensive than real price for ICT in Health, and in fact they act as lobby.

Jonathan Payne Replied at 8:40 PM, 21 Jul 2010

Great discussion. I would like to hear more about the motivation behind in country coding. It seems like there is a split between ideologically driven capacity building and business-driven focus on efficiency and quality. These approaches affect both HOW software is spec'd, developed, implemented, and maintained, and the WHAT the product is (including meets specifications at some level of quality). Most organizations seeking software development will need to find a balance between the two. They must weigh available resources against desired level of quality and support, timeline, understanding of target audience, and cost of expertise.

From the business standpoint, the decision to do development locally or out of the country is really not that different from decisions faced by profit-seeking companies in the US. Add in a motivation to build local capacity, it may steer a company towards favoring local development, but a company can only sacrifice project requirements so much in favor of something they believe in principle.

In summary, I'm trying to say that I think this is an important discussion, and it's covering a lot of details that I glossed over. In the end, though, I think the decision to code in one place or another is a decision that cannot be removed from its usual business context. This group and others like it are just adding in a much needed priority on capacity building.

Gayo Mhila Replied at 1:28 AM, 22 Jul 2010

Jonathan Payne
Thank you for seeing the split between ideologically driven capacity building and business-driven that focuses on efficiency and quality. It has been and it will continue to be a challenge to find a balance between the two. Because very often when software is developed in business driven, people are often facing time constraints and quality issues. These two make very difficult for young developers to get a chance in doing the actual work of coding. Because of that I don’t see business driven become on the frontline to mentor and support code in country however they can do so by setting aside some funds to support small academic institutions such as colleges and small NGOs which very often have the time and energy. I talk of small institutions because I believe that’s where we can strengthen code in country rather than big universities which very often have a lot of things going on.
In short I see code in country grow if funds and few dedicated experts from developed software business will be made available in developing countries. I see many retired people from developed countries become missionaries in developing countries. So the same spirit should be encouraged for retired software developers who are well experienced should come to support code in country specifically in the following areas:
1. Small institutions such as colleges
2. Small NGOs also can play a great role on code in country since very often they do small projects
3. It would be even better if Institutions and NGOs could work together because very often NGO’s get some projects which give them some money and Institution have the resources such as students and supervisors who can make it happen
Sometimes I see undergraduate students from developed countries are motivated to work in developing countries during summer time. That’s a good idea for them to share and gain experience with other people and especially if they work with other students. It is good and I real like it but when we talk of strengthen code in country let’s talk of having experienced software developers who have done software development for several years and have a better understanding of what is a good and stable software.

Melissa Loudon Replied at 7:48 AM, 22 Jul 2010

Hi everyone,

Perhaps leading on from discussion on how to attract local developers to government/NGO-led projects, would anyone like to comment about women's involvement in the projects you work on? Is there a gender imbalance in eHealth, because there is definitely a gender imbalance in this discussion ;-)

Jokes aside, I imagine this varies a lot for different countries, and reflects the gender balance of university courses and the local IT sector. Feel free to discuss these as well - personal experiences welcome, too!

melissa

Ali Habib Replied at 8:06 AM, 22 Jul 2010

In response to Melissa's question, we have one female programmer in our team of four. This is reflective of the general trend here in Pakistan where women generally do not gravitate towards programming. However our CTO is female and she does a lot of the work that's not directly related to programming but is more about usability, and systems and process development. She's not Pakistani though. Traditionally the software development arena has been a man's world here.

I noticed a similar gender gap at the University of Bergen's recent summer course on mobile technologies for health. The gender ratio was massively tipped towards men but there were people from some 20 countries there. I'm not sure if this was representative of the trends in other countries though.

On the other hand, when we look at the people who actually run the studies and head up programs that are using eHealth/mHealth, there it's overwhelmingly women at least in our group... this is again reflective of the trends in Pakistan.

Rowan Seymour Replied at 11:48 AM, 22 Jul 2010

Hi everyone
I'm currently managing a training course for E-Health software developers in Rwanda, run by Partners in Health. This is the second year of our course and we have 12 students on a 9 month course where they learn everything necessary to develop modules for the OpenMRS EMR. We take computer science graduates, but we've found that for programming we pretty much need to start from scratch. Most of the students know some theory but very little coding experience.

Most of the students who graduated last year are now working for the Ministry of Health and are currently developing modules for OpenMRS. In this capacity I think they have been fairly successful (an OpenMRS module is usually a fairly small project), but I would agree with earlier commentors that there is definitely a problem of a lack of experienced software developers.

Re the question above about gender - we have 3 female students this year and had two last year. We see this is as a reflection of the gender balance of Comp. Sci. courses at the university. We've contemplated some form of positive discrimination but haven't used any yet.

Heather Zornetzer Replied at 4:53 PM, 22 Jul 2010

Hi all,

Great discussion - excited to see this set of topics get so much attention. It's been really interesting and useful to hear from so many folks working in different regions.

I work with William Aviles (who's one of the panelists) and the Sustainable Sciences Institute in Nicaragua. I wanted to pipe up about 2 points in particular that have been raised these last few days.

First, on the issue of training and capacity building around health IT project management, leadership for IT teams and the "business end" of contracts and service work with various users/clients for health IT solutions -- this is a critical point. Similar to what Keiran and others referenced from their experiences, in our experience as well, we've learned there not only has to be a strong focus on actively engaging the partners involved in solution development and implementations, but also a significant focus on managing the non-tech parts of team development projects. This includes everything from managing expectations/timelines/product releases, etc. with end-users, but also managing the communications with those clients along the way -- budgets, mechanisms for reporting and checking in, mechanisms for iterative design and adaptation changes, internal team management, etc. We've learned by mistake at times, but we've also benefitted from being able to access a range of more experienced development groups that we've met and worked with over the last few years - Cell Life, Dimagi, D-Tree, DataDyne, the ODK group, and others - who have generously shared their anecdotal knowledge with us about what to watch out for, etc. As we (SSI) work hard to help connect more groups within this space, especially in the Latin America region, we're trying to pull together some "lessons learned" materials around this set of topics. If any of you are interested, it would be great to take some of the details of that offline and see if we can't come up with some shared resources for addressing some of this non-tech capacity building knowledge.

Second, on the issue of gender -- although in many ways the Latin America region has intense gender barriers for professionals in some sectors, it seems to be less of "a man's world" ironically when it comes to developers. More than half of the students in the computer science programs that we've been in touch with in Nicaragua and other countries in the region are women. Not the professors mind you, but the students. Likewise, there's great female representation in the more recent graduating classes in the health fields, including at the MD clinic/hospital admin levels. This last bit is likely more true in Nicaragua than other parts of the region for historical and political regions, but definitely we're seeing more highly trained and qualified women both health and tech fields in the region than i believe is the case in other parts of the world with similar socio-economic profiles.

In our IT team of 12 full time staff and one part time high-level coder (all Nicaraguans, not including me - an American), we've got 4 guys and 8 women. The women in our team are the data entry and junior developers and 3 of the 4 guys guys are the more senior developers with deeper skill sets on the programming end. I think in our case this has mostly to do with the fact that our team is mostly in their 30s. Consistent with what we're seeing in the computer science departments here, this may be changing. Our newest hire is a young woman developer (quickly moving up from junior developer to senior developer in terms of tech skills) right out of school, after she interned with us for her thesis year as an undergraduate. In any case, I think that we're seeing a different landscape in the LAC region to some extent, although there is certainly still machismo and inequality in terms of equal salary/wage rates for women and men in the tech sector in general.

looking forward to more great discussions here this week and next!

cheers,
heather

--
Heather Zornetzer
Instituto de Ciencias Sostenibles | Sustainable Sciences Institute
Managua, Nicaragua

Patricia Mechael Replied at 11:13 AM, 25 Jul 2010

What a great discussion so far! I wanted to chime in on a couple of points-but more from a macro level as the policy environment in which e and mHealth rest is increasingly becoming mainstreamed into the broader public health agenda of many countries. This shift from donor-funded ICT for health projects to government-sponsored initiatives to capitalize on increased access to ICT infrastructure is starting to increase demand for skilled programmers- many of which will be contracted directly by Ministries of Health and/or partner agencies. In Ghana for example, they recently launched their eHealth strategy. As a next step, they will begin to assess the capacity gaps in the ability of health workers to use technology as well as in the availability of skilled managers and programmers to design, develop, test, and deploy systems at scale. For this there are short-term training initiatives and longer term curiculum review activities that will need to come into play as the anticipated demand for such skills will only increase in the next 5-10 year horizon. This should and can only come from within any given country. So in this paradigm shift, what would be the most beneficial role for donors and/or external stakeholders to play?

At the same time, there is a significant gender gap in the programming environment- especially in the countries in Africa where I tend to work. I was quite pleased to read Heather's post that LAC is different. So here is one observation on this- many if not the majority of health workers in developing countries are women- from doctores to nurses to midwives to CHWs-- but many of the applications that we expect them to use are being designed and implemented by men and oftentimes- in non-coded in country environments- men from North America and Europe. With the increased attention towards maternal and newborn health- the issues become even more apparent as developers begin to engage in direct to client applications and systems for pregnant women and their support networks.

Doesn't this strike anyone as a bit strange? Having studied the sociology of gender and technology and growing up with two female programmers (my mom and my sister) and one male programmer (my dad)- my sense is that there are engendered ways of approaching programming that potentially yield different and equally valuable results. What has been people's experience and what can we do to ensure that the opportunities are there to leverage the strengths that women have to offer in programming initiatives? It is not enough to meet a quota or increase representation- we need to truly assess the value of having female programmers and then support a process that creates an enabling environment to increase their professional development.

Any light you can shed is appreciated-
Patty

Melissa Loudon Replied at 4:58 AM, 26 Jul 2010

Thanks to everyone who commented on gender balance. One of the things the responses had in common was some kind of structured internship/training programme to support new developers. Do you think this is something we should do more of? As a couple of people mentioned earlier, a developer (or a business analyst, or project manager) straight out of school still has a lot to learn. Creating a less intimidating environment for this learning to take place might be one way to attract a more diverse group to eHealth projects.

Nicolas Pottier Replied at 5:33 AM, 26 Jul 2010

Just a quick note that on the gender issues this is a problem that is larger than just in the developing world. At least in the states, it is still fairly rare to see female software engineers, especially experienced ones. Most seem to get routed to other positions fairly early in their careers for whatever reason.

My evidence is fairly anecdotal, from my 15 years or so in companies large and small on the West coast, but I would be surprised if it wasn't fairly representative. My point is only that this isn't an issue only when building capacity but rather everywhere, so we should expand our scope when considering causes and solutions.

-Nic

Peter Lubambi Replied at 5:56 AM, 26 Jul 2010

Hi all,
Almost everything has posted by other commenter and panelist. Just to Emphasize on importance of investing on local developer. I am working on a Maternal and Child Care SMS system here in Tanzania under the Computer Science Unit- University of Dar-es- Salaam Tanzania, I came to realize that end user(Health Workers)from whom we collect requirement they are cooperative and give real requirement when they see a working system and most of Health center/Hospitals in Tanzania doesn't have IT specialist.So it needs closeness between developers and Health Workers during development and implementation otherwise the software is likely to be rejected. If the software will be developed by the foreign developers but within the country it will incur very high cost of development and maintenance.

I think no body is against that, one thing to brainstorm is, how are we going to implement the "coded in country"?

Ahmed Maawy Replied at 8:28 AM, 26 Jul 2010

An interesting point. Regarding close bonding between the developers and health workers, what sort of structure can be put in place to facilitate proper communication between the two parties as well as suit the needs of both of there parties. This is because Software Expers visualize the entire process in Object Oriented models, and Health Workers may not understand the technical aspects of the entire problem.

Any suggestions?

Chaitali Sinha Replied at 8:39 AM, 26 Jul 2010

Hello all,

I am thoroughly enjoying the thoughtful posts and discussions on this discussion group. Many thanks to the organizers and to the contributors. I work at Canada's International Development Research Centre (IDRC), where my portfolio of work focuses on learning how best to (and not to) strengthen health information systems in developing country contexts. The comments that follow are based both on my professional and personal experiences. Although much of what is written below is shared by many of my colleagues at IDRC, I wanted to be explicit that these opinions do not represent a formal organizational stance at IDRC.

Before joining IDRC I had a brief stint as a software developer in Canada. Although interested in finding solutions to problems, I found the work environment I was in to be disconnected from the problems I was interested in solving. My subsequent studies and experience led me to work at IDRC. With a mandate to empower through knowledge and research, IDRC has spent several years supporting different open source developer networks such as OpenMRS, OpenROSA, and, more recently, the upcoming IMeCA conference. IDRC also supports projects to inform demand-driven curricula, mentorship arrangements, applied research, and evaluations of how capacity building initiatives are translated into changes in behaviour and state.

My experiences has shown that strengthening the capacities of health IT professionals requires a systems perspective, with strong emphases on understanding incentives, power dynamics, and the nuances of the local context. Each of these elements consist of cross-cutting issues of class, langage and gender. My sense is that we are at a pivotal moment - where a critical mass of individuals, products, and credibility has accrued in many parts of the world for locally developed eHealth solutions. This is no small feat, as this trend has flown in the face of blackbox engineering, centralization of power and knowledge, and long-term vendor lock-in and other forms of dependencies.

This movement has demonstrated plurality in the world of sound software development. A world that should consist of developers, implementers, users and managers that are listening to the real needs and responding with flexible, affordable and sustainable solutions. To echo some earlier posts, I agree that local IT individuals and organizaitons should be compensated appropriately for the work they do. This addresses the issue of incentive and helps develop a local software industry that can meet the changing needs for years to come. Again, thinking about this as a system, there is a need to strengthen all components to achieve favourable outcomes. This includes supporting a representative workforce with regards to sex, age, language, culture, etc. I believe the absence of such a workforce undermines the software that is developed, which in turn mitigates the achievement of beneficial health outcomes This, in my opinion, is yet another way innovation in the 'developing world' can pave the way for a much-needed revision of thought in the 'developed world' on how, and by whom, software solutions are developed. I feel the best way to do this is to continue demonstrating proven practices, credible solutions, supportive networks, robust functionality, and the necessary flexibility to continue growing to meet changing needs.

- Chaitali

Om G Replied at 10:00 AM, 26 Jul 2010

Object oriented thinkers can communicate with the rest of the world!

I'd suggest using very primitive objects and some committed time to have some fun.

Literally a 'Play Session' with blocks, markers and some paper can assure that creativity and simple understanding can take place.

Solid objects help non technical personnel understand the concepts involved and can accommodate a wide variety of technical understanding.

Sometimes it is really informative to know when certain end users become 'saturated' and the "toys" can be used to extend understanding as time goes on.

Effective cross pollination is time consuming, but reaps huge rewards later on.

Also, I think attention to this will help keep the larger goals in perspective while improving morale.

Ali Habib Replied at 10:38 AM, 26 Jul 2010

Interesting discussion on what is essentially a sort of communication gap
between technical personnel and non-technical personnel. At IRD we see this
everyday and while it can often be frustrating for folks on both sides of
the spectrum, we tend to chug along (while often admittedly pausing to tear
our hair out). As technical folks it is very easy to say that "they just
don't get it" but it's probably just as frustrating for the other side. I'm
intrigued by Om's idea of using fun ways to get around this and will see if
I can get people back at the office to give it a shot. For the most part
though, and this is easier said than done, it's often useful to make no
assumptions about what the other side (for lack of a better term) knows. In
terms of structures that can be put in place, I think it is useful to
involve people from both ends of the health-tech spectrum in joint internal
training sessions. For example, I gave a basic but still somewhat
technical presentation on database design to health researchers in our group
and have also attended training sessions on SPSS, presentations on research
results, field experiences etc. I find that that sort of cross-pollination
within a group helps to gain perspective about the work done by others in
the group and a sense of ownership and involvement in that work. It also
helps to blur the boundaries of terminology across both sides and can help
lead to a common language.

Ali

Joaquin Blaya, PhD Moderator Replied at 1:06 PM, 26 Jul 2010

I agree completely with Ali that giving basic talks or overviews about technical matters to clinical or administrative personnel can be a good view of having cross-pollination to both the presenter and the listeners, I especially think that this could be of use when looking to implement any soft of health IT system. There are usually many myths about what software will do and giving a basic overview of the health IT area can dispel some of those myths. For example, I've seen systems not work because the secretary or administrator thought that they would be replaced by the electronic medical record (EMR) that was being implemented or doctors refuse to use an EMR because it was going to "tell them what to do" and in their opinions machines are not capable of reasoning.
This is not only a developing country problem, though I haven't worked much in the US or other developed countries, I know from colleagues that similar issues exist there. But I do believe that in these countries there are more individuals with the combination of health and IT knowledge/experience, mostly due to the fact that this field has had more funding and has existed longer in those countries. In contrast, I also do see, at least in Peru, Mozambique, and Chile, that informatics and clinicians are more inflexible in their way of thinking and have a much harder time communicating in a way that the other can understand. Hence even more of a reason to have local development and provide that training to those organizations and individuals who are hopefully going to build and implement systems, but who at least are going to have to maintain them.

Joaquin

___________________________________________________________________
Founder, eHealth Systems Chile
Research Fellow, Harvard Medical School/Partners In Health
Moderator, GHDOnline.org<http://www.GHDOnline.org>

Neal Lesh Replied at 11:18 AM, 27 Jul 2010

Hi everybody,

I'm learning so much from this conversation-- new ideas, new perspectives, and new questions to ask. I'd love to also get feedback and invite participation around an idea that was started a while back of trying to promote a brand for software, similar to Fair Trade for merchandise, called Coded in Country (CIC). We've taken this panel conversation as impetus to launch a Coded in Country website. John Wesonga has been busy putting it together--it's still under construction, but you can see it at cic.supportlocalsolutions.org.

DataDyne and Dimagi started CIC around what we now call the CIC-50 rule, that a project can qualify as CIC if over half their software development budget on local software developers. This relates to an earlier point in this discussion about business vs. ideology motivations. I don't see this as a choice we have to make, though I do think there is a tradeoff between short and long term results. That is, I think that investing in local capacity for eHealth developing projects makes great business sense-- that this investment will pay off in a small number of years. At that point, you'll have better projects that are cheaper to extend and maintain.

Getting back to "Coded to Country", we've found the name to be catchy and to convey a core idea that resonates with a lot of our colleagues. We also realize that it means different things to different people and want to use the CIC website to rally a larger effort that encourages the principles that we've been discussing on this thread including capacity strengthening, local ownership, context appropriate systems, and gender equity. We'd like the website to be a place that people can share ideas and resources, as well as self-certify that they meet the CIC-50% rule.

Looking forward to your feedback-- what you think of the idea and where we should take it from here. And we'd be grateful if you could help us collect content for the website-- including training resources, pointers to upcoming events, etc. Please send pictures too, that we can display on the website (ideally with a short caption and the name of who we should credit for the picture). Feel free to send the photos or other specifics to . Many thanks in advance!

Take care,
neal

Dr Carola Hullin Replied at 5:59 PM, 27 Jul 2010

Dear All,
It is extremely helpful the discussion. My own contribution to the discussion based on my twenty years as a clinician/health informatician is that the technology or developers are not the drivers. The drivers in the health system is PEOPLE. Clinical leadership is critical for getting the right solution in this context. Capacity building is important for software developer,however, this strategy is a long term one since education is critical for sustainable interventions. Due to my own time constraints to answer all the dimensions mentioned here. I can share with you that THE GAP is large between the drivers you described here for health information systems and clinical/operational workflow. In developing countries, developing in-house software is attractive but due to the lack of skills and knowledge this approach may work against the lack of resources, since teaching people how to program is hard in the industry.Firstly because a computer is a rare thing for them, and educational framework are different in developing countries. Gender difference in the health sector is a well recognized issue and workforce strategies around the world are trying to cope with these challenges.

In summary, my view is the your perspectives need a more clinical focus for providing solution, the drivers for improving health outcomes within a health system are PEOPLE, PROCESS and TECHNOLOGY. The technology is not the ISSUE anymore, it is the process you are trying to capture if this exist and the people you need to involve in it..................... this we call it Clinical Transformation or sometimes Healthcare Transformation.

It is a pleasure and a privilige to share this with you

Cheers
Carol

Dr Hullin
(PhD,Hons,Nursing)

William Aviles Monterrey Replied at 11:10 AM, 28 Jul 2010

I am very excited; we are sharing so many ideas in this panel. It’s been
very helpful in better understanding the challenges to local development.

One point I would highlight as a limitation is that health system in
Nicaragua doesn’t trust completely in opensource projects. It is well-known
that there are some myths in this type of project and in my country for lack
of knowledge are very active in the mind of decision makers in the MOH. Can
you share experiences of how to approach this challenge? Has anyone else had
to deal with this problem? Are there any resources available to sell the
idea of opensource in public health?

On the other hand, I would like to say that we are working on SSI to reduce
the gap between IT people and public health persons. The best way is to
create new courses in universities to train people with expertise in both
areas. Anyone experienced doing this?

Joaquin Blaya, PhD Moderator Replied at 3:44 PM, 28 Jul 2010

Hi William,
I think your point about decision makers not knowing much about open source and hence not "trusting" it is true in many countries. What I have found is that
a) individuals think that open source means a group of volunteers who do this in their spare time
b) that there are no organizations or companies providing support or that they could hire to use these systems
c) open source systems are new and they don't want to take the risk of being the first to try it

I think there are two ways to approach this (perhaps more). The first is to try and educate decision makers. The California Healthcare Foundation has written several good reports about open source software used in the US and also a primer for health care leaders that you can find here http://www.chcf.org/search?query=open%20source. Also, the Chilean government did a study on the use of open source within the government (http://www.estrategiadigital.gob.cl/node/386) which shows that in some sectors e.g. the more technological there is some use of open source, but almost everywhere else it's closed source.

The way my organization in Chile has decided to go is to promote and support open source projects by using them and giving code back to the community, but not using the words open source as a selling point to decision makers. Instead we list the benefits of open source without ever mentioning those exact words i.e. we use OpenMRS and therefore we mentioned that it's been implemented in 23 countries, that there are over 45 organizations using the system, that it is modular and has been connected to other systems via HL7.

Joaquin

___________________________________________________________________
Founder, eHealth Systems Chile
Research Fellow, Harvard Medical School/Partners In Health
Moderator, GHDOnline.org<http://www.GHDOnline.org>

Ali Habib Replied at 3:51 PM, 28 Jul 2010

I think CIC is a very interesting initiative. Kudos to the folks who have
put this together and I hope it spurs groups to invest in local
development.

Regarding William's question about selling open source, it can be difficult
to dislodge those misconceptions. It may be useful to research and then
highlight open source deployments in other countries with similar
infrastructure and technology availability that are running well as examples
of the success and viability of open source in public health. I think that
one of the things that is often a factor is availability of technical
support for open source projects. While the community around such projects
is often a brilliant source of support, this can seem very alien to those
who are used to calling a support hotline and getting the help they need. If
you can demonstrate that support is available and that projects won't be
left hanging if something goes wrong it becomes that much easier to convince
people to use open source. This is admittedly easier said than done and is
yet another reason to support open source local software development for
health IT since that would contribute to a more robust support structure
within the country and make it more apparent that open source is viable in
public health. The underlying assumption in my response however, is that, in
these situations, the initial support for such initiatives would have to
come from sources other than the health ministry.

Ali

Sanjay Kumar Replied at 1:53 AM, 29 Jul 2010

I am an IT professional working in public health for the last 10 years. I would like to contribute at my level best if any open source IT utility in any public health is required to be developed. I would encourage the idea of having a repository of open-source IT utility for public health at one common location which should fully tested. Please suggest something may require be done from my end.

Sanjay

Om G Replied at 6:25 AM, 29 Jul 2010

I think osgeo is a great example of how an industry can promote open source, best practices, and educate the public while presenting a professional 'face'

Melissa Loudon Replied at 6:54 AM, 29 Jul 2010

I see I'm falling behind! I was just about to ask for examples of organisations promoting or providing services around locally-supported open source software, with the goal of presenting a more 'legitimate' alternative to proprietary options. For these and OpenMRS and OSGEO, which have been mentioned, what activities/programmes do you think are particularly useful?

melissa

Peter Lubambi Replied at 12:26 PM, 29 Jul 2010

"'Health IT' for better Health Care Delivery, 'Open Source Software' for better Heath IT implementation". Hi all!
Melissa,One activity I can suggest,For organisations promoting Open Sources Software is to sponsor Open Sources courses in local Universities. This will assure the availability of long term support for Open Source Softwares at low cost.

Joaquin Blaya, PhD Moderator Replied at 3:28 PM, 29 Jul 2010

I think one of the key factors for the promotion of open source is organizations and companies that are implementing those systems and providing technical support and services. What I've seen in Latin America reflects what others have commented.
1. Decision makers want someone or an organization to be responsible for maintaining the system and making when something goes wrong that they know who to call.
2. When deciding on what type of system to choose, the decision is usually made based on who's promoting or selling the systems rather than on what is the best technology.

This to me means that the way that open source health IT is going to continue expanding is if there are organizations and companies (hopefully local) that use these open source platforms and create systems that meet local needs and are on the ground promoting/selling them, just like the companies who use closed source products. That's they reason why I started a company in Chile to use OpenMRS and OpenROSA systems here, and are promoting for other organizations to both use these systems and created a open source health IT community (IMeCA, http://www.informaticamedica-lac.org/) in Latin America. There is also a similar project going on in Africa with the OASIS project and the non-profit Jembi.

I would love to hear of other organizations in this space or ideas of how to promote these kinds of organizations.

Joaquin

___________________________________________________________________
Founder, eHealth Systems Chile
Research Fellow, Harvard Medical School/Partners In Health
Moderator, GHDOnline.org<http://www.GHDOnline.org>

Lim Chanmann Replied at 12:56 PM, 30 Jul 2010

Great discussion, I see I have learned a lot so far from this conversation.


I agree that in order to have sustainability in the long-run, capacity
building is the-most important for not only eHealth but software development
in general. To be able to push facilitating the implementation on local
leader, and developer investment, better term for contract, incentivizing
and prioritizing local talent by local authorities, local authorities itself
play a major role. In addition while depending on other parties there is
nothing stop us from doing a startup. I and some colleague and developer
friends have formed a group called “ShareVision” to start spreading
technology, we goes to universities to share experience, we conduct
DeveloperCamp to target improving software development quality for free of
charge.


For women in technology, there is no different about other countries as
well as Cambodia, more men than women enter software industry, in fact to be
successful in the industry it also need product development including user
requirement gathering, usability design, user behavior study and analysis,
and more which give equal opportunity for both men and women. Currently
InSTEDD iLab product manager is women.


From developer point of view, there are some methodologies we could use to
bring the gap between technical and non-technical people who is willing to
have final product such as Kanban which is a lean software development
methodology to early find out the blockages during development and to be
able to release early and often so people can decide what they think the
product would look like, what is important to have, and what is the best to
have. Further more in conjunction with BDD methodology which facilitate
communication between developer, designer, tester, and end user plus weekly
demo during which we can both show user the progress of the development and
educate them technology.


Selling open source is a bit hard to people who don't trust technology
could help them. But if they do, people often don't care what platform you
are using in order to get his job done quicker, easier, and fun plus
securing their data, and fast support response when they need it. It's the
developer choice for using or not using open source, and the same reason for
developer if open source technology can make him more productive he will
consider implementing it plus encouragement from friends, management level,
community, local authorities.


Very appreciate for the useful ideas during this discussion.


Cheers,


Lim Chanmann

Software Developer *InSTEDD iLab*, Cambodia

Om G Replied at 2:57 PM, 30 Jul 2010

Great points! 1 thing i haven't seen addressed is what we can do as developers to ease the pain of transition is at least come to some level of agreement at a platform level. The way ubuntu packaged a 'one click' install and a module picker/update did a lot to win people over.

The analogy isn't that clean, but people aren't hesitant just because they are ignorant of what open source 'could' be, I think it has more to do with the feeling of being alone in development.

Selling an extensible, "polished" solution isn't that hard.

I think polished enough can still leave a ton of room for development, customization and future growth.

Vincent Njoroge Replied at 4:43 PM, 30 Jul 2010

Hi all,
This is a very interesting discussion especially from my perspective as an
implementer. Some of the concerns expressed about the lack of experienced
developers and the lack of higher level project guidance resonate with me.
My partners and I are hoping to bring our global systems integration
experience working with global corporations to health and education projects
in East Africa. As locals returning, we are working with local coders and
finding success by adding governance and object oriented and agile
methodologies to our projects. We insist on addressing governance, because
when new processes are introduced you will need to obtain a basis for
legitimacy, define decision and management processes at both the strategic
and operational levels and plan communications to focus on building momentum
by aligning support and to manage the resistance that comes with process
change. Not all projects will do this, but successful ones will tend to
consider this seriously in the planning.

best regards,

Vincent Njoroge
www.anthillpartners.net

Neal Lesh Replied at 8:48 PM, 30 Jul 2010

Hi everybody!

As we reach the designated time to officially close the panel, I'd like to take this opportunity to thank GHDonline for their support, our amazing moderators and panelists Ahmed, Melissa, William, Ali, Mann, and Jacob, and of course all of you for your participation and attention.

The conversation of course must continue in many forums, both internal and external. Of course feel free to start new related discussion threads on GHDonline, e.g., by using a new subject line if you reply by email (which would actually be preferred over extending this impressively long discussion.)

Looking forward to more! Thanks again.

-neal

Hamish Fraser, MBChB, MRCP, MSc Moderator Emeritus Replied at 1:52 PM, 31 Jul 2010

I know this panel has officially closed but as a moderator of the site I would like to thank everyone for such an extensive and thoughtful discussion. I would also like to summarize some observations from this topic from my experience in setting up a training program in Rwanda, described by Rowan Seymour who directs the course.
We in the PIH informatics team tend to be very pragmatic about the projects we do, responding directly to the requests of our local colleagues and users. In Rwanda as we developed and deployed some of the earliest OpenMRS based EMR systems it was quickly clear that the benefits of open source would not be realized without local programmers who could build, implement and support the system. It was obvious that there were two deficits, hands on programming skills in Java, and an understanding medical information management and healthcare. So the EHSDI course seeks to address both of those gaps with a very strong practical/vocational approach.
This has engaged different constituencies including the development community interested in local capacity building and software industry, and the healthcare community looking to get useful and sustainable tools working on the ground.
Software has an unfortunate "winner takes all" tendency which leave less developed countries very vulnerable to vendor lock-in, license fees and lack of choice. But in healthcare IT another force is at work, the huge amount of expensive customization that occurs with EMR systems in the US and other countries. So to be successful a county like Rwanda or Cambodia or Nicaragua *has* to be able to do their own local customization both to match the real problems on the ground and to have a hope of affording this extra programming work.
We need to build the cross over skills for programmers to better understand the challenges of healthcare and for healthcare workers to understand the potential and limitations of IT systems. In most successful projects that I have seen there are close partnerships and friendships between individuals from the two communities who increasingly learn each other’s craft. We also need to determine the best working environments for local programmers that provide a supportive environment and good project management which may be *local* companies.
Open source is not the only way to move forward here but it has several major advantages. Without access to source code and data models it is very hard to teach effective skills and the work of students and graduates cannot be easy used in the real world. There is also a much stronger tendency to fragmentation, if you can’t build on the shoulders of others you usually start each project a-new. Until now open source medical software has attracted little of the funding flowing into information systems in global health, except from a couple of far-sighted funders, IDRC and Rockefeller and some thought leaders at CDC. We hope that the increasing success of such projects and the CIC initiative will ensure that it is seen as a viable option to the big funders who have spend more than $500 Million (!) on proprietary health information systems in developing countries in recent years.

Joaquin Blaya, PhD Moderator Replied at 7:00 PM, 1 Aug 2010

We at GHDonline also wanted to thank all who participated, moderators, panelists and members, as well as send a special thank you to Neal Lesh who has done an amazing job preparing and coordinating the panel.

Joaquin

Joaquin Blaya, PhD Moderator Replied at 8:08 PM, 5 Aug 2010

Sanjay,
To answer your question a few days late, one of the easiest ways for you to help in health IT software is to become a programmer within one of the open source systems that were mentioned during this panel (OpenMRS, Osgeo, OpenROSA) or perhaps others. I personally know that OpenMRS is always looking for good programmers to improve the system and fix bugs.

Sanjay Kumar Replied at 12:31 AM, 6 Aug 2010

I am fine with it or any other IT support required.

Joaquin Blaya, PhD Moderator Replied at 4:16 PM, 6 Aug 2010

OK, could you email me personally at <mailto:> so we can email more directly and not bother the group.

Thanks,

Joaquin

___________________________________________________________________
Founder, eHealth Systems Chile
Research Fellow, Harvard Medical School/Partners In Health
Moderator, GHDOnline.org<http://www.GHDOnline.org>

Joaquin Blaya, PhD Moderator Replied at 5:58 PM, 1 Sep 2010

Hi everyone,
Along with thanking again all of the participants of the Local Software Development Panel, the GHDonline team wanted to announce that we've posted a Discussion Brief on this panel. It's a 2 page summary that can be found at:

http://www.ghdonline.org/tech/discussion/local-development-panel-discussion-j...

Joaquin