Abstract
Improving global health security will require bold action in all corners of the world, particularly in developing settings, where poverty often contributes to an increase in emerging infectious diseases. In order to mitigate the impact of emerging pandemic threats, enhanced disease surveillance is needed to improve early detection and rapid response to outbreaks. However, the technology to facilitate this surveillance is often unattainable because of high costs, software and hardware maintenance needs, limited technical competence among public health officials, and internet connectivity challenges experienced in the field. One potential solution is to leverage open source software, a concept that is unfortunately often misunderstood. This article describes the principles and characteristics of open source software and how it may be applied to solve global health security challenges.
Enhanced disease surveillance is needed to mitigate the impact of emerging pandemic threats by improving early detection and rapid response to outbreaks. But the technology to facilitate this surveillance is often unattainable because of high costs, software and hardware maintenance needs, limited technical competence among public health officials, and internet connectivity challenges experienced in the field. One potential solution is to leverage open source software. This article describes the principles and characteristics of open source software and how it may be applied to solve global health security challenges.
L
The wide collection and use of big data in many domains (eg, Facebook, Google, insurance companies, financial institutions, etc) highlight the value, power, and potential usefulness of large amounts of information. In public health, big data can and should be made useful through cost-effective and transparent methods of collection, analysis, and interpretation. How to actually put this into practice has led to the creation of the field of public health informatics.
The public health informatics field has grown exponentially in the developed world over the past decade and increasingly in developing settings as well. Use of big data is now common, and independent software development is leading to a variety of innovative solutions to manage this data. But public health systems often cannot afford the most innovative, for-profit software solutions. Open source software (OSS) may offer an opportunity to gain access to and fuel innovative solutions to global public health challenges. To fully leverage the power of OSS, a better understanding of this concept and its licensing among the public health community is needed.
What Is Open Source Software?
Open source software (OSS) refers to software that is made readily available for others to use, modify, or redistribute under a licensing agreement with very few restrictions. Anyone can use the software without having to pay royalties or negotiate a license agreement. OSS is governed by a license specifying access to the source code and terms of distribution. But an OSS license grants users more rights than a commercial license, since the user gains access to source code, has the right to change the source code, and can distribute the code under defined conditions. For this reason, open source is often confused with placing software in the public domain. If software is released into the public domain, the author surrenders the copyright, whereas an open source license still protects a copyright holder's interests even if it grants broad rights to users.
Other common misconceptions include the idea that open source is synonymous with free, or that open source licenses are unenforceable. OSS may be provided to users at no cost, although there are usually some expenses for hardware associated with its implementation, similar to commercial software. In contrast to commercial software, there are, however, limited options for prompt software support. This apparent weakness also describes a potential strength of OSS. There is an expectation with OSS that crowdsourcing of software solutions to fix “bugs” leads to further development and customization, a “wiki-type” approach that often provides rapid, albeit unofficial software support. Examples and descriptions of commonly used OSS licenses are found in Figure 1, and a glossary of related terms is found in Figure 2.
Commonly Used Open Source Licenses
Glossary of Related Terms
OSS in Public Health
The ability to easily change and share source code is a powerful tool for public health organizations adopting information technology and digital solutions. Indeed, global public health initiatives have begun to make successful use of OSS.2,3 Mobile health (mHealth) initiatives and the need for electronic processes to support healthcare (eHealth) provide particularly good examples of government use of open source software. 4 The growth of global and national mHealth and eHealth needs has spurred innovation in software development. As medical practitioners and health institutions are encouraged by the US government and other sponsors to digitize patient information in order to improve efficiency and productivity, a need for sophisticated software tools to manage these data has arisen. 5 In resource-limited areas that do not have the infrastructure for sophisticated computing tools but where cellular technology is prevalent, mHealth solutions are able to move such communities into the digital age. Monetary costs of licensing and maintaining proprietary software systems have been common challenges to these end users. OSS is used in a variety of software tools to fit various needs for mobile-based data collection. Some of the most common initiatives and the licenses used can be found in Figure 3.
Examples of OSS Initiatives Used in Public Health
Just as OSS is a decentralized way of developing software, the more recent emergence of open application programming interfaces (APIs) decentralizes how programmers get software to communicate and can further enhance access to information for public health uses. Open APIs are freely available for use by anyone and can integrate data from one site into another without having to negotiate agreements or engage in a cumbersome process of sorting through all available data to find elements that are relevant. For example, an open application programming interface for Ebola data was created that allows users' sites to access some or all of the Humanitarian Data Exchange (HDX) Ebola data through a web query (the user site communicates with HDX using the API), and HDX returns the requested information in a machine-readable format. 6 Without an open API, the users would have to download and parse all the HDX data for the information they want and manually update information in their own application for use on their site. Another example is the OpenStreetMap API, which gives users access to all the raw OpenStreetMap data or smaller, more manageable portions of data. 7 Associated OpenStreetMap APIs also allow querying, which can be useful for certain applications. As will be discussed below, mapping data is particularly useful during the management of humanitarian crises.
OSS in Low-Resource Settings
Open source software has played a significant role in resource-limited, austere settings, namely in postdisaster areas such as Haiti and the Philippines and, most recently, the Ebola epidemic in West Africa. In these situations, one issue that can emerge is that the data are not “big” enough—that is, there is too little data or its accuracy cannot be verified. With Ebola, several OSS tools were used to track the outbreak, but the amount and accuracy of the information collected by workers on the ground was inconsistent. This problem was partly because of the tumultuous conditions in the field and partly because of underreporting out of fear of repercussions in the local community. 8 The Centers for Disease Control and Prevention (CDC) uses Epi Info VHF, a tool that helps with contact tracing and locating individuals exposed to the specific disease. But this and other similar OSS tools require data, and “underfunded public health infrastructures in the affected regions make it hard to get good data.” 8 While the number of tools used to help track and manage the outbreak can cause confusion, overall, OSS has been critical in creating a system of data collection that improves communication between health workers and the affected communities. 9
OpenStreetMap is a crowd-sourced mapping application that provided a detailed map of the areas hit by typhoon Haiyan within 3 days of landfall.7,10 OpenStreetMap uses satellite data in combination with an image of the infrastructure captured on the OpenStreetMap and relies on volunteers to trace the path of roads and the location of buildings using a visual editor within the application. It has more than 1 million volunteers who have helped map disasters and other incidents worldwide. 7 The tracing allows relief workers to see where buildings are no longer standing or where roads have been damaged or destroyed. The Red Cross now uses OSS in all of its projects, OpenStreetMap being one example, citing that OSS reduces or eliminates sustainment costs of software after the organization leaves an area. 10 In addition, any software or data created by the Red Cross are released as open source, making it available to the broader community.
The use of OpenStreetMap by an organization such as the Red Cross is significant, in part because it speaks to the credibility of the application, but also because similar efforts in earlier disasters could have benefited from large nongovernmental organization (NGO) use and endorsement. In the aftermath of the Haiti earthquake in 2010, several crowdsourcing applications like OpenStreetMap were used to map the damage. Ushahidi, a company that develops OSS for information collection and interactive mapping, was one of the main volunteer efforts to produce a crisis map. 11 In an independent evaluation of the use of Ushahidi technology tools in Haiti, the report noted that Ushahidi's mapping effort provided critical situational awareness that influenced operational and tactical decisions and saved lives. 12 The report urged that stronger support from the NGO community would be useful in making the application more widely used in the response community, something OpenStreetMap was able to do through its relationship with the Red Cross in the Philippines. It is worth mentioning that Ushahidi worked collaboratively with OpenStreetMap, a partnership that was critical to its effectiveness. The ability to share information and provide near real-time updates was facilitated by the fact that all the technology used OSS and relied on volunteer input to improve the information's accuracy, underscoring the value of a “wiki-type” approach to data collection and management.
OSS and IHR Implementation
The International Health Regulations (IHR) were designed to increase member states' ability to detect and respond to public health emergencies with potential global impact. 13 However, many of the core capacities—such as disease surveillance, laboratory capacity, risk communication, and outbreak response—require digital resources and infrastructure that many member states lack. For instance, the World Health Organization (WHO) Africa Region faces a number of threats from epidemic- and pandemic-prone diseases, yet no member states had fully implemented their national IHR plans by the June 2012 deadline. 14 Some of the reasons cited for the missed deadline include inadequate financial and human resources, insufficient communication infrastructure for reporting public health emergencies to IHR focal points, and the lack of necessary equipment and supplies for detecting, reporting, and responding to public health events. 14 There are likely similar rates of compliance for all regions of the world. The recent and largely unchecked epidemic of Ebola in Guinea, Sierra Leone, and Liberia clearly highlight some of these deficiencies.
To date, 102 countries have been unable to implement their national IHR plans and have requested a 2-year extension, many as a result of financial and human resource limitations. 15 The adoption more broadly of OSS has the potential to improve the efficiency of IHR implementation and, therefore, global public health initiatives in general. OSS can provide free, modifiable software options that can be altered to meet specific requirements for data collection, outbreak management, and reporting to the WHO that are so critical to controlling an epidemic. These can later be shared and adapted for larger local initiatives that may require more extensive computing and information-sharing capabilities. Importantly, OSS tools can be used by local governments, NGOs, and outbreak response teams in tandem without the need for each to purchase a license for a proprietary software product. Thus, through supporting the implementation of the IHR, use of OSS has the potential to improve global public health security.
Benefits and Challenges
While not all public health challenges can be solved with an information technology solution, many can. To the extent this type of solution requires software infrastructure, OSS can be used to reduce costs and make accessible software that would otherwise be too costly to procure. Sustainability of surveillance systems is always an issue in under-resourced, developing settings; often a sponsor will commit to supporting a system for a fixed period of time, but the “tail” costs after the sponsor leaves can be prohibitive. Thus, through keeping software open source, costs can be minimized and improvements can be readily shared among a larger group of users in developing settings.
Moreover, the ability to modify the code to a specific purpose is also an opportunity to train local information technology experts so human capacity is fostered and sustained. This “wiki-type” approach has been very effective in creating cadres of technically competent software engineers who require little supervision to modify existing products to fit diverse public health needs. 2 Other potential benefits of OSS are listed in Figure 4. 16
Recognized Benefits of Using OSS
Improving public health security through timely disease surveillance and information sharing will not be easy, even with OSS. Whenever data sharing is entertained, mistrust is frequently encountered because of uncertainties about the reliability of data, how data will be used, individual privacy issues, and how data might be monetized by others. The concept of data sharing has to be replaced by that of information sharing, where information = processed data. There is little need to provide raw data in most public health circumstances. There are many barriers to data sharing, 17 some of which may be addressed by using OSS. For example, programming code transparency via OSS sets the stage for sharing aggregated information, which then has the potential to foster trust and open up a broader resource pool. And with improved trust among the many stakeholders in this field, a consolidated effort to address the global common issue of public health security has a chance to succeed. As far as the legality of OSS licenses, laws and regulations do vary by country but, in general, are enforceable under standard contract and copyright law.
Conclusions
Open source software in the global public health field is a rapidly evolving space that drives innovation and brings needed tools into the hands of those often disadvantaged by a lack of financial assets. OSS can have a positive impact on major global health initiatives, like IHR implementation, by facilitating compliance with the core capacity requirements for developing countries, especially relating to data collection and timely reporting. Better understanding of OSS principles and how they apply to global public health will facilitate the broader use of these tools. Enabling access to easily attainable, high-quality, and well-documented OSS tools with supportive hosting environments has the potential to improve global health security.