The response to the Ebola outbreak in West Africa shone a spotlight on the critical importance of timely and accurate data and information flows to combat the disease’s spread and deliver effective and targeted action. The majority of data and information moved by voice, radio, or paper. Where digital technologies were used they permitted critical, time-sensitive data and information to quickly scale the distances of space and time. In addition to the quantitative efficiencies gained, digital technologies also enabled important qualitative differences.
Advantages of Digitized Data and Information Exchange
Why does the digitization of data and information matter? In the context of a fast-moving disease outbreak like Ebola, the timeliness of information is critical to an effective response. The quantitative efficiencies of digitized data include the ability to more quickly collect, manage, and share data and information. Perhaps even more compelling, however, are several important distinctions that permit digital technologies to enable qualitatively different action.
Specifically, by enabling the rapid exchange of data and information across the scales of distance and time, digitization enables qualitatively different information and data exchange due to an increased:
- Plurality of actors participating in connected and real-time or near real-time information exchange, including:
–traditional response and donor organizations (such as foreign governments, NGOs, and intergovernmental organizations like UN agencies)
–frontline or extension workers (such as community health workers, contact tracers)
–“remote” responders supporting the formal response, such as the digital humanitarians, members of diaspora groups, and context experts
- Directionality of information exchange, including:
–“up,” as in case data reporting up to a centralized point
–“down,” as in SMS behavior change messages down to many decentralized points
–“up” and “down,” as in tools or programs that exchanged information in both directions
–“horizontal,” as in communications within or between peer groups active in the response
- Nodes of information exchange, whether:
–one-to-many, or many-to-one (as in SMS questions from health ministries to health workers, and responses from health workers back to the ministry)
–many-to-many (as in the case of chat platforms, such as Skype and WhatsApp, being used by response workers for informal, real-time coordination)
Implications for Decision-Making and Programming
The above qualitative distinctions in the way data and information move in a digital context create a variety of opportunities for improved decision-making. These include:
- Increased accountability, incentives, and insights:
–accountability, such as through timestamps and/or global positioning system (GPS) identification on the point of data collection, or in the real-time or near real-time availability of data for program managers, making it possible to manage people and resources in a timely way
–insights, such as sharing back contextualized data and information to the point of origin, which can enable better-informed local decision-making
–incentives, such as through horizontal, peer-to-peer information exchange that created motivation to take on difficult and at times dangerous tasks
- Increased ease of sharing data and information and devolution of decision-making, both down to many decentralized points and horizontally among peer groups, which facilitated information sharing across data silos and supported coordination
- Increased ability to create continuous two-way feedback loops through the sharing of contextualized data and information back to the point of origin
- Increased ability to implement continuous learning and adapative programming, in which a program or activity is modified and adapted in real- or near real-time, according to the right information processed at the right time by relevant decision makers
How might the Ebola outbreak have unfolded differently had stronger digital health systems been in place throughout West Africa? Ebola hit hardest in Guinea, Liberia, and Sierra Leone. But it also affected the neighboring countries of Nigeria, Senegal, and Mali, where outbreaks were quickly contained. What accounted for the difference in Ebola caseloads? A variety of factors played a role: stronger health systems, a functioning EOC, a strong health workforce and overall better preparedness, which together affected the overall response. Although data digitization by itself did not lead to the successful control of Ebola in Nigeria, a closer look at Nigeria’s success story shows, among other factors, the critical importance of preparedness plans for health emergencies, and the value of timely adaptations of existing disease outbreak protocols and existing digital health infrastructure.
Ebola made its way to Nigeria when, in July 2014, an infected Liberian man traveled to Lagos by airplane, setting off a chain of transmission that infected a total of 19 people, killing seven. According to the WHO, after the first Ebola case was confirmed health officials were able to quickly adapt existing health technologies and infrastructures to respond.
To combat the Ebola outbreak, Nigerian officials modified their existing polio outbreak system and a corresponding preparedness plan. The use of the IHR reporting meant that Nigerian officials were quickly notified of the arrival of the Ebola case and consequently were able to mobilize a timely response. Based on prior government experience using an Incident Management System to handle health crises, including a recent polio outbreak, the Nigerian government appointed an incident commander and set up an EOC that is largely credited with playing a central role in containing the Ebola outbreak.
A key aspect of the EOC’s immediate operations was the rapid tracing of contacts. Nigeria’s index patient generated a total of seven fatalities, 19 cases, and 894 contacts who were monitored for symptom development through approximately 18,500 face-to-face visits by contact tracers. Working in partnership with eHealth Africa (eHA) and other partners, the EOC put into motion a digitally supported, real-time digital workflow to increase the rapidity with which critical and time-sensitive contact tracing data would be available for decision-making. This digital approach to tracking, aggregating, and visualizing contact tracing data in the Nigerian response used a suite of tools, including Open Data Kit, FormHub, ArcGIS, and eHA’s Sense android app, which was developed during the response to capture 21-day follow up information--the critical window of time required to confirm that someone potentially exposed to the Ebola virus had not been infected.
An NGO official working on digital health systems at the time of the outbreak reported, “We knew that in Lagos, where the infection was, poor mobile phone network connectivity meant that to have real-time data we need[ed] to give [contact tracers] smartphones preloaded with [contact tracing] software on it. That’s how the teams were able to respond immediately, and how everyone who needed information got it.”
The Sense tools functioned in both online and offline environments, enabling data collected by contact tracers in offline areas to automatically sync as soon as their smartphones were brought back in range of a functioning mobile network. Contact tracers were equipped with phones preloaded with credit for calls, SMS, and mobile data. Credit was topped up automatically, removing the potential barrier of lack of credit in submitting contact tracing reports. Sense tools also made it possible to generate a multitude of dashboards for various stakeholders and to auto-generate notifications to key decision-makers.
In contrast, in all three countries most affected by Ebola, health workers entered information about contacts primarily on paper, which data managers then manually transferred to electronic format. This resulted in situation reports based on data that ranged between three days and two weeks old, a critical delay in the face of a fast-moving disease.
Nigeria’s existing and more robust health systems and health information systems, as well as its strong digital infrastructure, were essential to its ability to rapidly identify and isolate cases, halting the spread of the disease before it spiraled out of control. The existence of digital tools, capacity, and systems enabled the adaptation and immediate use of integrated disease surveillance and reporting systems. Together, these resources and actions enabled officials to prevent a catastrophic outbreak in Lagos, one of Africa’s most populous cities. Nigeria’s response to Ebola highlights the critical role of appropriate technologies, adaptability, and preparedness planning in ensuring health system resilience.
These distinctions in how data and information for decision-making are gathered and shared can provide important benefits, including enabling better informed and more targeted and effective service delivery by understanding rapidly evolving needs from a variety of perspectives. Yet in the Ebola outbreak response--as is true elsewhere in development, global health, and humanitarian assistance--the full potential of digital technology to improve programs and outcomes has yet to be reached. The following recommendations elucidate practical next steps that humanitarian aid, development, and health actors can take to act upon the lessons identified in the body of this report.
Assess and Invest in Digital Connectivity Infrastructure
Timely and accurate data and information flows were most possible with digital technologies, yet in many parts of the three most-affected countries digital networks were unreliable or simply unavailable. Investments in extending the reach of digital connectivity should be seen as a fundamental component of strengthening resilience to future health and humanitarian emergencies.
“The last 15 years have seen a revolution in ICT and mobile technologies. Ebola shone a spotlight on the ineffectiveness of past health systems strengthening efforts; there is growing evidence that ICT and mobile are a vital part of the solution to build resilient health systems.” 
Suggestions for Operationalization
- Explore partnerships to create business models that work, potentially including development funding, to build out infrastructure in areas where market incentives do not otherwise exist.
- Create incentives to expand digital infrastructure. This could include public-private partnerships, subsidies, or tax-based incentives to help mobile network operators reach rural communities.
- When building out digital, consider power and seek alternative models to enable reliable power. Solar panels, for example, could be outfitted to key government ministries, prioritizing those responsible for managing critical data sets in emergencies, and to district health facilities.
- Encourage infrastructure sharing, such as base stations, to decrease the cost of network expansion. This might require incentives, such as tax breaks, to encourage sharing.
- Encourage long-term thinking and scale in network investments so that emergency-related investments in digital infrastructure last beyond the emergency phase.
- Explore alternative models to extend connectivity, such as TV white space, and balloon-, drone-, or low-orbiting satellite-based Internet connectivity to extend coverage in remote or hard-to-reach areas.
Suggestions for Operationalization
- Work with mobile network operators and Internet service providers to develop protocols for reporting data that identify connectivity “cold spots” in order to prioritize them for delivery and easy identification in an emergency scenario.
- Support the development and deployment of a baseline ICT assessment framework, and an online repository for such assessments to be shared publicly. Assessment categories might include: citizen literacy and digital technology uptake and common citizen digital use patterns; mobile and Internet network reach and capacity by geographic area; e-payments infrastructure capacity to deliver payments to frontline workers and other actors; national health information systems’ capacity to manage routine and health crisis-related data; and a mapping of other commonly used digital information systems that can be used for real-time data and e-payments management.
- Support the development, sharing, and adoption of standards to assess consumer access and the reach of digital technology. This could include emergency response protocols that enable the rapid assessment of an emergency scenario (whether conflict, disaster, and/or health related) to affect mobile and Internet network capacity, as well as the likelihood of response demands to burden existing physical infrastructure.
Suggestions for Operationalization
- Develop and share protocols for quickly assessing the ICT infrastructure in a country as an essential component of emergency preparedness and response.
- Identify and designate a lead agency to implement the protocol on behalf of the international community.
- Include as part of the rapid ICT assessment updates to the baseline assessment of mobile and Internet network capacity and latency by geographic area, and the operational and business viability of MNOs in an emergency. Such assessments could update the baseline ICT assessment with critical post-disruption updates, and could be automatically triggered with the activation of the deployment of the Emergency Telecommunications Cluster in a humanitarian emergency.
- Encourage aid workers to download and use applications that report the availability of mobile networks to crowdsource a picture of network connectivity. These apps could be linked to apps such as OCHA’s HumanitarianID or others designed for use in emergency situations.
Assess and Invest in Workforce Capacity
Using digital technologies does not remove the need for human capacity; it increases it. The Ebola outbreak illustrated the critical need for technological literacy and capacity, specifically that of staff and national communities of practice who were best positioned to deploy quickly and support long-term recovery efforts. The best time to build human capacity, however, is before an emergency hits. The data and information demands of the response made it difficult to build national capacity while meeting operational needs. Moreover, the volume and velocity of data and information collected and shared occasioned the need for specific expertise, including a cadre of epidemiologists, data scientists, data visualization specialists, and data analysts to manage, interpret, and render data useable.
Suggestions for Operationalization
- Assess existing staff capacity in data and digital literacy, informatics, software engineering, and other technical areas, identifying where staff capacity is thin.
- Build capacity of existing staff and retain new staff with relevant expertise to support a cadre of data and digital experts with specialized knowledge of digitized data collection, sharing, management, analysis, and use for decision-making to help quickly aggregate, manage, and interpret (digital) data. Many response actors--NGOs and donors--need to recruit for and grow their workforce’s technical capacity. This includes: technical capacity to collect, manage, and analyze data in an ethical and responsible way; methodological expertise requiring a clear understanding and awareness of what data to collect and how; the capacity to input and manage large quantities of data; and mastery of data analysis and visualization.
- Support the development of technical capacity among host country national governments and at the regional level, such as through technical associations. During the Ebola outbreak, the most-affected countries had to request and wait for technical assistance from outside experts to adapt their HMIS to track Ebola cases. Critical time, information, and ground in the fight against Ebola were lost waiting for outside experts to make these adaptations.
- Support digital literacy and the regular use of digital technologies within national governments, local organizations, and response agencies, both to strengthen health systems with routine reporting and to enable proficient use of digital technologies in the context of an emergency response.
- Deploy data managers and analysts in an emergency alongside sector experts to provide critical data capacity needed to support operations and decision-making.
- Invest in and grow niche expertise. The data demands of the Ebola outbreak response required unique and hard-to-find skill sets that encompassed global health, epidemiology, data science, and technology expertise. Although the combination of sector and technological expertise is rare, it is valuable to governments and response organizations, both in ongoing program management and during crises.
“The tech team that should have been on the ground wasn’t there. We needed someone with technology and health expertise. You can’t have a tech person who hasn’t worked in health do this. You need technology and health and development people to ask the tough questions. This wasn’t happening.”
- Address the salary competition governments face in the retention of top technical talent. Alternative models could include limited-term fellowship positions or senior executive service models with pay at slightly higher rates than normal government salaries.
Recognizing the urgent need for a cadre of technical experts to support digital health platforms in West Africa, governments and regional organizations there are partnering with USAID to improve public regional health informatics capacity. At the request of the West African Health Organization (WAHO) and the 15 members of the Economic Community of West African States (ECOWAS), USAID is supporting a team of local software engineers and informatics experts to build the organization’s health informatics expertise. The project, commencing Fall 2016, is setting up a regional team of experts with the aim of making immediate improvements in national and regional HIS that will help strengthen public health systems and build local software developer capacity to support long-term sustainability of HIS investments in the region.
The Ebola outbreak illustrated the critical need for increased institutional capacity. The volume and velocity of data and information challenged institutional capacity to effectively collect, manage, and use these data and related digital systems. Specifically, institutions--both in-country and internationally--often lacked the policies, processes, and workflows required to enable real-time or near real-time information to drive decision-making and enable an adaptive approach to programming.
Suggestions for Operationalization
- Assess existing institutional capacity to leverage digitized data and information flows, and to enable adaptive, data-driven programming, noting where deficits exist.
- Implement change management strategies to increase institutional capacity to address existing deficits. This may include organizational policies, processes, staff positions, workflows, and budgets required for implementation (see textbox below).
- Designate an internal champion to shepherd the change management strategy, and to regularly assess how expenditures need to be realigned to meet changing needs.
- Establish a national digital health committee or technical working group to guide and support the deployment of digital health information systems.
The 2016 report From Principle to Practice: Implementing the Principles for Digital Development provides guidance about the institutional reform processes needed to keep pace with the changes the integration of modern ICTs present for the international development, global health, and humanitarian assistance sectors--a field some refer to as “digital development.” Although a relatively new phenomenon, the integration of tools like the mobile phone is increasingly widespread and demands the following:
- To fully mature, the field of digital development must be recognized as its own interdisciplinary field that requires professionalization and institutionalization, involving specialized and dedicated training as well as institutional reforms.
- Institutions should treat digital development as a cross-cutting and foundational field, using it to improve program delivery and development outcomes by (1) integrating digital development strategies early, and (2) tying digital development data to adaptive programming. This may require modified or new workflows.
- Institutions must have a vision and strategy for implementation that is adequately resourced, enabled by supporting policies and processes, and supported by an implementation or institutionalization strategy, with milestones identified to measure progress.
- Institutions should assess whether existing policies and processes support adherence to this strategy. Where existing policies and procedures inhibit the integration of best practice, organizations should set reforms in motion.
- Institutions should assess staff and technical capacity to implement the strategy across operations and programs, and at various stages of implementation.
- Institutions should integrate best practice by building staff capacity across sectors and geographies, and operational units, such as through trainings and supporting knowledge exchange among staff, making best practice sector- and business-process specific.
- Organizations should monitor and measure success in implementation of the strategy, building in opportunities to reward success and learn from failure, with corresponding incentives.
Assess and Advance the Enabling Policy and Regulatory Environment
The Ebola outbreak response highlighted a series of deficits related to the policy and regulatory environment to support the use of digitized data and information flows. These included tensions between data sharing needs on the one hand, and privacy and security concerns on the other, and the need for emergency preparedness protocols for data and digital information management. Although privacy protection and ethics often stand alone as separate recommendations, putting them into practice and making them meaningful requires that these considerations be integrated into policy and regulatory processes and protocols, as well as capacity-building activities. Addressing these policy and regulatory deficits requires the development of preparedness protocols in advance of an emergency that could be triggered with the declaration of a Public Health Emergency of International Concern, or a Level-3 Humanitarian Emergency.
Suggestions for Operationalization
- In advance of an emergency, negotiate a process to secure the public availability of connectivity maps of mobile network providers operating in countries affected by an emergency. (See related baseline ICT assessment recommendation.)
- In advance of an emergency, negotiate with mobile network operators to obtain short codes that can be used to support an emergency response, and those that can be used to support ongoing SMS-based communication between ministries and their remote workforce, particularly for health and social sector programs.
- Short codes should have reverse billing capacity, so that charges for messages sent over the system are borne by a government or other specialized agency, not the individuals receiving and sending responses.
- Ensure emergency short codes are given priority on mobile networks so that if network capacity is limited these messages will still be delivered.
- Designate emergency response short codes to be distributed to and accessible by response organizations in emergency settings. Some organizations could be pre-approved to reduce the vetting process in an emergency context.
- Negotiate protocols to share aggregated mobility patterns from mobile call data records to assist emergency responders.
Suggestions for Operationalization
- Design protocols for emergency data-standards development to simplify and harmonize data collection in a crisis. For example, a joint public health advisory board for a particular crisis could be tasked with agreeing upon developing common working standards (e.g., case definitions, indicators) within a period of days of the declaration of a Public Health Emergency of International Concern, or a Level-3 Humanitarian Emergency. This body should work with existing humanitarian coordination bodies, such as OCHA and the humanitarian health cluster led by WHO, to designate or create temporary standards for the specific emergency. Immediately following the announcement of public health advisory board designated standards, an associated technology advisory board would then be responsible for developing and publicly posting the associated software code to ensure interoperability of these data across commonly used data systems.
- During a crisis, emergency data standards should be reviewed on a periodic basis (such as once a month) to assess and update standards, and to push out related changes.
- Leading international health authorities, such as WHO and CDC, should publish the working emergency data standards and liaise with country governments to adopt them. This process would be facilitated by groundwork laid in advance to create awareness of and to formalize this process so that all parties are prepared to expect and quickly implement new data standards as needed. Such a process could be linked to the negotiation of data sharing protocols or the IHR. Governments, multilaterals, and donors could contractually enforce the adoption of those standards in software and data analysis related to the response.
- Assess and strengthen national health information systems, with a particular focus on interoperable, country-level digital information systems.
- Establish toll-free URLs that allow health workers and other emergency responders to access certain websites or IP addresses. For example, the website domain for the national HIS (such as DHIS2) could be toll free, allowing clinics and health workers to access the site even without data credit on their phones. Alternatively, an ODK server could be toll-free, allowing enumerators to continue to submit data and download new forms regardless of the data available on their phones. These sites could be reverse billed and paid for via a central body such as a ministry of health.
- Confer upon an established national digital health committee or technical working group a special role to advise on implementing a national digital health strategy in emergencies, and update it as needed.
- In an emergency, conduct country-level rapid assessments of available digital platforms and identify those that should serve as primary tools to support the response. Ensure these are widely available to responders, together with guidelines and supporting standard operating procedures guiding digital platform use.
- Develop or adapt existing standards related to unique identifiers for an emergency outbreak. The lack of robust unique identifiers for patients represented a significant hindrance to data integration across data sets and systems. Having pre-negotiated guidance in establishing unique identifiers could have mitigated this problem.
- Integrate “disease surveillance and reporting” data and systems with national health information systems so that disease outbreak data can be readily collected alongside and compared to routine health data.
- In routine and outbreak disease surveillance reporting, ensure that missing data are reported as missing and not as zero cases. In the early stages of the outbreak in Liberia, some counties were unable to report their cases. These were counted as zero cases, leading to fluctuations in case data reporting.
- In building new health information systems, adapting existing systems, and linking existing systems, support and leverage global public goods (including open and reusable frameworks, processes, systems, and tools) to minimize duplication of effort and wasted resources.
- Support revisions to the IHR to expand WHO Member State required reporting to facilitate infectious disease data sharing. Specifically, global notification of domestic disease outbreaks should be expanded from those for the plague, cholera, and yellow fever to a more comprehensive list of infectious diseases, such as Ebola, of importance to international public health. Notifications of disease outbreaks should be routinely published in a format that is machine readable and available to the public.
- Develop country-specific emergency outbreak protocols, including the use of digital technologies. These protocols should identify existing forms and platforms, outline standard operating procedures, and make them available for use by responders.
Suggestions for Operationalization
- Develop processes and protocols that respect individual data privacy and facilitate data sharing. Integrate privacy risk and ethical analysis into processes for aligning data collection and use from the beginning.
- Promote policies that encourage responsible data sharing and ownership for different types of response data, and the circumstances under which special processes would apply.
- Adapt policies and processes to include risk and benefit analysis for sharing different types of data (e.g., data owned by an MNO versus a government versus an NGO) among actors in an emergency.
- Invest in resources for and capacity building to enable responsible collection, use, and management of data, including the necessary information security tools, policies, and human resources. (See recommendations above related to “Institutional Capacity” and “Workforce Capacity.”)
- Negotiate a protocol to share case data, with full protection of personally identifiable information, to pre-approved actors (e.g., academics, response agencies) in order facilitate disease modeling upon declaration of a Public Health Emergency of International Concern.
- Develop a methodology to assess risks and benefits of data use (collecting, storing, managing, sharing) in emergencies that could be tailored by emergency type (conflict, natural disaster, health emergency).
Advance Data ‘Infostructure’ and Standards
The response demonstrated a direct correlation between strong and interoperable information systems, and the ability to deliver a targeted and sustainable response. A proliferation of platforms for data management and use--many of which could not easily be linked to or used with other platforms--fragmented a unified picture of health and humanitarian needs. Greater awareness of, attention to, and investments in interoperability, including both through technical and standards development, is critical to unlocking the full value of digitized data flows.
Suggestions for Operationalization
- Understand common barriers to paper-based data and information flows, since these are likely to impact digital information and data flows as well.
- Support the mapping of public infrastructure, such as hospitals, clinics, or schools. Include these maps as part of the “common operational datasets” that are available to response actors at the beginning of an emergency situation.
- Integrate GIS into preparedness protocols related to data standardization and data collection.
- Collect only what is needed. In adopting digitized data collection and use, the types of data collected should first match the information needs of those collecting and using the data.
- Identify and agree upon data standards, including harmonized disease case definitions and reporting formats in preparation for potential future outbreaks, and to decrease confusion among responders, facilitate data aggregation over the course of an outbreak, and minimize the data burden on frontline responders. Where these data standards have not yet been agreed upon, see the recommendation regarding the creation of emergency data standards.
- Convene discussions about data standards that cut across sector silos (e.g., health, WASH, education, logistics) and skill sets (e.g., technologists/developers, operational humanitarians, development practitioners, researchers). Such discussions could eliminate some issues that arose from non-aligned data standards and interoperability challenges.
- In advance of or at the outset of an emergency, gather relevant stakeholders to develop minimum data collection standards, particularly in the early phases of the emergency. Defining the minimum viable product and baseline data, including definitions and standards for data collection, can minimize the data burden and help increase the likelihood that data collected are of higher quality (due to reduced competing demands for data collection) and of the broadest possible utility to response and other actors.
"In the early phases of the response, operational responders needed fast and shallow but broad operational data combined with statistically relevant sentinel site data for epidemiological purposes. We tried to get both all of the time and got neither. A more rationalized approach would be to design an information strategy based on minimal indicators for action and invest in key places for deeper information." 
- Governments and multilateral partners in this space should strongly encourage and, ideally, contractually require funded organizations to adopt and implement harmonized data standards for both paper and digital technology-based systems.
- Data sharing and use agreements for intra- and international exchange (such as between and among national governments and international agencies) should be automatically triggered with the declaration of a Public Health Emergency of International Concern or a Level-3 Emergency.
- Contribute to periodic reports documenting the maturity of digital and information systems, such as the ITU’s annual eHealth survey or the Demographic and Health Survey program, to enable continuously updated indexing of the capability and reach of country-level “infostructure.”
- To enable open sharing and to facilitate re-use of data, ensure published data are machine readable. If data are published in a non-machine-readable format, such as many PDF documents, release the same data simultaneously in a machine-readable format such as .csv-compatible spreadsheets.
- Identify, modify, and implement policies that support sharing of line-list case data or machine-readable data, ideally by default.
- When creating new data policies or practices, build on existing international standards.
- For datasets relevant to emergency response, use the Humanitarian Exchange Language (HXL) as a starting point for terminology and taxonomies to enable data sharing. To institutionalize its use, donors could require its adoption as part of standard data collection in grants and contracts for emergency operations. HXL was developed collaboratively by and for humanitarians to simplify the aggregation of diverse datasets. The HXL hashtag-based approach is broadly relevant and should be expanded for use across other sectors and datasets.
- Set up an easily accessible website for standardized forms (with version numbers) and key messages.
- Data standards must proactively address the question of who “owns” data, and not only the products that result from use or analysis of data (e.g., research papers or reports).
- Publish data early and often, enabling others to cross-check and confirm data (e.g., in the case of GIS locations of health facilities).
“Technology must be interoperable and future-focused. Experience indicates that interoperability between systems is critical. Increased use of Application Program Interfaces (APIs) is needed to connect systems (it’s not about seeking to create one large system at a country level); sustainable solutions are not tied to quickly obsolete technologies. For example, the technology underpinning future health systems must be able to cope with the innovation that new phones and other mobile devices enable. This approach requires thinking about the underpinning ‘infostructure’ where there is less emphasis on sharing specific data and more on connecting the systems that support data to be shared.”
Increase Coordination of Digital Health Investments
To date, disparate interest groups have driven digital health platforms and strategies to meet their own needs rather than approaching investments from a systems-level perspective. To derive the greatest possible value from investments in digital systems, more unified and structured funding approaches are needed.
Suggestions for Operationalization
- Prioritize investments in interoperable platforms and systems to increase the ability of country governments and other actors to readily compare and share data that originate from difference sources.
- Build upon existing open, adaptable processes, standards, tools, and platforms, whenever possible. Particularly during emergencies, the most useful innovations frequently are those that make incremental changes to existing tools, processes, and operations. Where broader changes and breakthrough innovations do occur in a crisis response, often they grow to be implemented at scale only during the recovery phase of the response or thereafter.
- Invest in digital health knowledge-sharing systems, tools, and processes that can be accessed by a variety of global health, humanitarian, and development partners. This could include repositories of commonly used, open source tools, maturity indexes of national digital health ecosystems, documentation of processes and planning, decision-support tools (such as checklists), guiding policies, and frameworks, such as enterprise architecture frameworks that can bring greater coherence to the proliferation of platforms and tools in use in many countries.
- Invest in and provide other needed support to intra- and inter-donor coordination around digital health technologies to promote aligned policies and actions, such as through technical working groups within donor organizations and among donors, as in the example of Health Data Collaborative’s Digital Health and Interoperability Working Group.
- Create funding mechanisms and models that enable co-funding among donors and both build and sustain digital health commons so that related platforms, systems, frameworks, and tools are more sustainable than the current program, sector, and/or disease vertical-oriented funding streams may allow.
- Create review boards for spending on digital health through collaborative funding mechanisms to provide input, feedback, and guidance on digital health investments and deployments. Members might include representation from donors, governments, technical experts, and civil society groups who together offer a cross-sector insights.
- Support more collaborative, participatory design and investment, and build processes in donor-funded development work to reduce parallel investments, such as through co-design among local and global development partners, and mechanisms that enable pooling of financial resources and technical expertise.
- Ensure that funded efforts build on national systems, reuse existing tools, and align with emergent local standards whenever possible.
- Encourage sustained donor coordination around the use of data and data systems and platforms, not to support a single system but to build in policies and processes that work toward openness and interoperability and reduce fragmentation and duplication related to ICT.
- Integrate explicit guidance that adheres to established best practice, such as the Principles for Digital Development, in requests for proposals and other development funding application processes. In the reviewing proposals, award technical points to proposals that adhere to best practice.
Understand and Use Digital Technologies in Context
Use digital technologies appropriately in context. Oftentimes this may mean using digital technologies alongside voice, paper-based, or analog channels. It always means using digital in a manner that is appropriate given the local sociocultural environment and end user needs.
Suggestions for Operationalization
- Incorporate human-centered design processes into the deployment of digital technologies in humanitarian and development contexts, to ensure that the technology is accessible and any data and information it relays are appropriate to the context.
- Consider rates of literacy, phone ownership, and access to power among intended audiences when designing digital information programming, including SMS-based communications. (See related baseline ICT assessment recommendation.)
- Use hybrid communication approaches (e.g., digital in combination with print, radio, television) that reflect and are appropriate to the country and cultural context, in a way that reinforces messages across multiple channels.
- When using digital approaches, ensure that they work in both online and offline environments, such as the use of mobile data collection programs that automatically sync data collected in offline environments once reconnected to WiFi or a mobile signal.
- When using mobile applications that require phone numbers, such as mHero or uReport, incorporate a process to regularly update users’ phone numbers.
- When developing digital systems, consider barriers to paper-based information flows (e.g., lack of roads, rainy season) since they are likely to impact digitized data and information flows as well.
- In designing digitally supported programs, draw on available information about consumer use patterns, literacy, and numeracy. Design digital programs with the understanding that digital tools are not a panacea and reflect the information environment in which they are used.
- When building data collection systems, design with a degree of flexibility to enable adaptation based upon circumstances and the specific requirements of a particular outbreak.
- Analyze what national systems are in place to handle information and data, and what capacity exists to act on it. Understand the capacity and limitations of the existing digital ecosystem and design digital programming accordingly.
- Whether using digital or paper-based tools, frame messaging according to the local cultural context and leverage existing trust networks to maximize impact. It is important not to forget empathy in developing messaging, especially when sent through digital channels.
- Employ digital technologies to support psychosocial needs where face-to-face contact is not possible. In a number of circumstances, organizations operating treatment centers facilitated digital connections (e.g., Skype or video-conferencing) between family members who were unable to meet face to face. These virtual connections helped to address the emotional needs of patients.
- In building new digital technology systems, adapting existing systems, and linking existing systems, support and leverage global public goods (including open and reusable frameworks, processes, systems, and tools) to minimize duplication of effort and wasted resources.
Employ Feedback Loops and Adaptive Programming
Where digital technologies are used they introduce an opportunity to leverage real-time or near real-time insights to support decision-making. The Principles for Digital Development, reflecting broader trends within both the humanitarian and development sectors, call for “data-driven development” or “evidence-based decision making” that designs projects so that evidence can be measured at key milestones to evaluate impact, and uses the availability of timely data to inform agile management decisions at all levels.
Suggestions for Operationalization
- Treat information and information sharing as an essential activity in an emergency response, as pivotal as providing food, water, or shelter in emergency response, recovery, and in longer-term resilience planning. This requires ensuring affected communities have regular access to vital and up-to-date information about the crisis and response, conveyed in culturally relevant and appropriate formats, from the beginning of an emergency.
- Design programs to create bidirectional feedback loops from the outset. Digital data flows can support bidirectional communications and feedback loops. By providing data collectors with the assurance that the data they collect will be returned to them with contextualized information that can support informed decision-making at the point of data origin, these digitized, bidirectional feedback loops can help to create incentives for regular and high quality data collection. Feedback loops also promote accountability and generate new insights, such as in data and information flows both “up” and “down” between government ministries and their remote workforce, and/or between local communities and response organizations.
- Use digital data flows to support a plurality of communications and feedback between responders, such as those among peer groups or health workers from across the range response actors. Such communications can support a variety of functions, including community needs and actions, routine disease surveillance, and health information systems strengthening.
- Design and implement flexible programs that allow faster feedback and proactive iteration throughout the program cycle.
 An article published in the Public Library of Science (PLoS) highlights the need for ICT in Ebola case surveillance, including case investigation, management, and strategic planning, and the use of open software (Open Data Kit and FormHub) in the Nigerian outbreak. The authors conclude: “A remarkable improvement was recorded in the reporting of daily follow-up of contacts after the deployment of the integrated real time technology." Daniel Tom-Aba et al. “Innovative Technological Approach to Ebola Virus Disease Outbreak Response in Nigeria Using the Open Data Kit and Form Hub Technology” PLoS One. 2015; 10(6): e0131000. Published online 2015 Jun 26. doi: 10.1371/journal.pone.0131000
 Here we refer to nodes and networks as in the sense of social network analysis, which maps, measures, and analyzes the patterns of information flows between people or groups.
 Analyzing the role of digital technologies in providing varying levels of complexity and directionality in communications within and between various actor groups is not new. The 2009 report New Technologies in Emergencies and Conflicts notes, "The report also makes distinctions in terms of the format of the communications. While the underlying communications technology may be radio, mobile, or internet, it is useful to [note] how the information is conveyed, whether: one-to-many (broadcast—radio, television, web, mobile applications (apps) and services, short message service (SMS) broadcast); one-to-one (mobile voice and SMS); or many-to-many, such as social networks (online or mobile internet, mapping, and crowdsourcing)." D. Coyle and P. Meier. New Technologies in Emergencies and Conflicts, 6.
 Other critical factors included the existence of a preparedness plan and a well-functioning Emergency Operations Center. Nigeria case example compiled from interviews with multiple health officials, January, May, June, and September 2015.
 See World Health Organization, “WHO Declares End of Ebola Outbreak in Nigeria,” Statement, October 20, 2014, accessed May 28, 2016, http://www.who.int/mediacentre/news/statements/2014/nigeria-ends-ebola/en/.
 The IHR are legal instruments, signed by 196 countries, designed to detect and prevent public health crises. See “International Health Regulations,” World Health Organization, accessed May 28, 2016, http://www.who.int/topics/international_health_regulations/en/.
 Faisal Shuaib et al., “Ebola Virus Disease Outbreak--Nigeria, July September 2014,” Morbidity and Mortality Weekly Report (MMWR) 63, no.39 (2014): 867-872, http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6339a5.htm?s_cid=mm6339a5_w.
 Tom-Aba D, Olaleye A, Olayinka AT, et al., “Innovative Technological Approach to Ebola Virus Disease Outbreak Response in Nigeria Using the Open Data Kit and Form Hub Technology,” Harper DM, ed. PLoS ONE. 2015;10(6):e0131000. doi:10.1371/journal.pone.0131000.
 Email correspondence with NGO official, October 2016.
 Interview with NGO official, January 2015.
 Wilton Park/mPowering Frontline Health Workers, "(Re)Building Health Systems in West Africa: What ROle for ICT and Mobile Technologies? Monday 15 - Wednesday 17 June 2015 | WP1409," July 2015, 2, http://www.wiltonpark.org.uk/wp-content/uploads/WP1409-Report.pdf.
 An example of such a partnership is that announced in September 2016 among the Government of Liberia, Google, and USAID to bring high-speed communications infrastructure to Monrovia, Liberia. See “Liberia To Work With Google, USAID To: Increase Broadband Access In Monrovia,” Daily Observer, September 22, 2016, accessed November 8, 2016, http://www.liberianobserver.com/business/liberia-work-google-usaid-increase-broadband-access-monrovia.
 The regulatory environment plays an important role in facilitating these incentives, especially in terms of mobile base stations policies, extending fiber access, and regulating spectrum to facilitate rural access to Internet and broadband services.
 See “The Emergency Telecommunications Cluster (ETC),” Inter-Agency Standing Committee, accessed November 8, 2016, https://www.etcluster.org/.
 For more information, see “Welcome to Humanitarian ID,” Office for the Coordination of Humanitarian Affairs (OCHA), accessed November 8, 2016, http://www.humanitarian.id.
 Interview with international health official, May 2015.
 The Senior Executive Service is a cadre of government positions below top presidential appointees who “possess well-honed executive skills” and are remunerated at rates of pay that tend to be higher than other federal government service employees. For more information, see https://www.opm.gov/policy-data-oversight/senior-executive-service/.
 Adapted from Waugaman, Adele. From Principle to Practice: Implementing the Principles for Digital Development: Perspectives and Recommendations from the Practitioner Community. Washington, DC: The Principles for Digital Development Working Group, January 2016, p. 62.
 Interviews with national and international officials familiar with Liberia case data collection, December 2014, April 2015, and February 2016.
 For more information on ongoing revisions to the IHR, see World Health Organization, “Global infectious disease surveillance,” Fact sheet N200, http://www.who.int/mediacentre/factsheets/fs200/en/.
 For more information, see: United Nations Office for the Coordination of Humanitarian Affairs (UNOCHA), Building Data Responsibility into Humanitarian Action, OCHA Policy and Studies Series, May 2016.
 Correspondence with USG official, July 2016.
 More information about the DHS is available at http://www.dhsprogram.com.
 For example, the International Organization for Standarization (ISO) has developed internationally recognized standards for everything from risk management and occupational safety, to country codes and medical devices: http://www.iso.org/iso/home.html. Also the International Aid Transparency Initiative standards on aid spending (http://www.aidtransparency.net/).
 “Humanitarian Exchange Language,” HXL Standard, accessed June 8, 2016, http://hxlstandard.org/standard/1_0final/.
 Wilton Park/mPowering Frontline Health Workers, "(Re)Building Health Systems in West Africa: What ROle for ICT and Mobile Technologies? Monday 15 - Wednesday 17 June 2015 | WP1409," July 2015, 3, http://www.wiltonpark.org.uk/wp-content/uploads/WP1409-Report.pdf.
 For more information, see http://www.healthdatacollaborative.org/news/article/taming-the-wild-west-of-digital-health-data-linking-systems-to-strengthen-global-health-outcomes-76/.
 For more information see http://www.digitalprinciples.org.
 See https://www.globalinnovationexchange.org/learning-adapt.
 See http://digitalprinciples.org/be-data-driven/.