The Emergence of Mobile-Supported National Health Information Systems in Developing Countries Ime Asangansi, Kristin Braa Department of Informatics, University of Oslo, Oslo, Norway & Health Information Systems Programme, Nigeria Abstract A major challenge for national health information systems in developing countries is their scalability and sustainability at the lowest levels where primary health care is delivered. This paper contributes to the discourse on how national health information systems can scale to the lower levels and how mobile technology is supporting the collection, handling and dissemination of data. But can mHealth go beyond the ���hype��� and visions it has come to be associated with? Using an ac- tion research methodology in a long-term action research project, the usability and then scalability of mobile solutions for large scale national health information systems are stud- ied. In this paper, initial successes and challenges with using m-Health for national public health information systems is reported and discussed. Keywords: Mobile, Health, Information system, Nigeria, India Introduction Health information systems (HIS) are having a major and on- going impact on the lives of people in both low and high re- source settings. A robust health information system is a basic foundation of public health [1]. The achievement of the health- related Millennium Development Goals (MDGs) will depend upon the effectiveness and efficiency of health systems. HIS remains the backbone for providing information to track pro- gress for improving and strengthening the different health sys- tem components and monitoring the MDG goals. On the ground, however, HIS development in developing countries has proven to be difficult due to organizational complexity, fragmentation, lack of coordinated organizational structures (that maintain disparate information systems), unrealistic am- bitions, and more generally due to the problem of sustainabil- ity. Poor availability and quality of data and a resultant poor knowledge and ���culture��� of use of information for planning and decision-making characterize HIS in many countries. Im- portantly, poor physical infrastructure has remained a serious obstacle in ensuring an efficient health information infrastruc- ture in many developing countries. The world over, both within the domains of research and prac- tice, there is an increasing recognition of the role mobile tech- nology and mobile phones can play in supporting public health systems. There is an emerging field and research domain for the application of mobile technology for health, mHealth. Mo- bile phones have a particular status across all developing coun- tries. There are 2.2 billion mobile phones in the developing world, compared to 305 million computers and only 11 million hospital beds [2]. Between the end of 2007 and the end of 2008, mobile phone subscriptions increased by approximately 1 billion [3]. Mobile technology comes with the unique poten- tial that it has already become a routine part of most peoples��� everyday lives. It is becoming increasingly affordable and ac- cessible, and has the required infrastructure (for example, the network coverage) even in villages to support its easy use and maintenance. There are over 4 billion mobile phones, 64% of which is in the developing world [4]. The majority of the mobile subscribers are now outside the major cities and wealthiest states. For ex- ample in India there are 65 times more mobile connections than Internet connections [4]. This creates opportunities to use mobile phones to capture data at the source, thus removing significant sources of data quality problems usually associated with manual transfer of data between paper reports. In addition the aim is to use mobile phones as a channel for feedback to the community health workers. We explore to what extent this technology can be used for effective data exchange and com- munication in public health how mobile phones in general can secure routine health data as well as stimulate better health provision by better communication and training. A major con- cern is how mobile phones can be coupled to and leverage district health information systems. In this paper we discuss the emergence of mobile-supported national health information systems in developing countries by describing two ongoing mobile health projects in Nigeria and India. The mobile solution will also be described and dis- cussed. mHealth Applications There are numerous mHealth projects in developing settings: the alliance of UN, Rockefeller & Vodafone in Feb 2009 formed after e-Health ideas exchanged in Rockefeller���s Bel- laio Conference (UN Foundation, m-Health Alliance) [5] Us- ing Mobile Phones and RapidSMS to Improve Child Nutrition Surveillance in Malawi (UNICEF, Govt. of Malawi and Mo- MEDINFO 2010 C. Safran et al. (Eds.) IOS Press, 2010 �� 2010 IMIA and SAHIA. All rights reserved. doi:10.3233/978-1-60750-588-4-540 540

bile Development Solutions) [6] SIMPill ��� embedded mobile phone chip in medicine bottle to remind patients in South Af- rica (2007). 90% took TB medicine while earlier only 22%- 60% took it mHealth for Development paper, 2009) [7] Text- to-Change ��� Sends HIV awareness messages in Uganda [8] Cell-Preven. Health workers use mobile phones to send SMS messages with real-time data on symptoms experienced by clinical trial participants, enabling immediate response to ad- verse symptoms [9] Frontline-SMS ��� a bulk SMS solution [10] OpenRosa [11]. Generally, m-Health projects can be broadly discussed accord- ing to: technology, domain area of application, the hierarchical level in the data flow and data handling processes it is used for. By technology, m-Health applications can involve SMS (Short Message Service) or ���texting���, voice services and other packet data services such as WAP, GPRS, etc. These can be simple typed SMS from any phone coded SMS texts (following some predefined logic (which gets interpreted at a central server) or SMS-based data transfer from applications (J2ME, Android, etc) installed on the phone. WAP and GPRS have been little used in the settings we are concerned with in this paper for the reason of unreliability (at this time). By domain, m-Health has been applied to various health areas including maternal and child health, community health volun- teering support, immunization, general emergencies, monitor- ing of patients with illnesses such as HIV/AIDS, supporting the control of diseases such as malaria, etc. In this study, the focus is on mobile systems feeding routine data from the lowest levels where they are produced such as communities and health facilities, through different levels up to the national trunk. In this scenario, facility/community level datasets are transmitted to upper levels through the state to the country warehouse, with health information defining a whole range of data elements spanning from utilization data, maternal and child health data, mortalities, nutrition and disease surveil- lance data. In developing countries, the collection of such data has historically proven to be intractable. Most mHealth applications are in the piloting phase. A com- munity of practice for mHealth is still being developed and mHealth standards are yet to be developed. This paper thus reports on a significant phenomenon, the emergence and early institutionalization of mobile systems for routine data collec- tion, with data flowing from the lowest levels to the national level. Materials and Methods Our research approach is action-oriented and interpretative and characterized as a ���network of action��� methodology. The network of action approach is based on the principle of creat- ing learning and innovation through multiple sites of action and use, and sharing these experiences vertically and horizon- tally in the network [12, 13]. It is premised on collective action where connected research units are able to share experiences and learning. The cases presented here are derived from units (or nodes) within the Health Information System Programme (HISP) network of action. The authors are actively engaged in these units (HISP-Nigeria and HISP-India) which are, with their respective partners, the principal development partners for this system in their respective countries. The pilot in India started in February 2009 and is ongoing. While that in Nigeria started in July 2009 and is ongoing. However, these pilots are converting into full-blown deployments and installations as the rate of adoption has been tremendous. Data sources for this study have been primary and secondary. Primary sources have included notes from participant observations, performing train- ing, and formal interviews with health workers at different levels as well as administrative and technical personnel. The authors have also been involved in the iterative development of the solution. Secondary sources included formal reports from the projects highlighted. Cases Here two cases are presented which are significant by virtue of being based in the national system of two very populous and complex countries in Africa and Asia ��� Nigeria and India re- spectively. The Nigerian Case With a population of over 148 million people [14], Nigeria is the largest country in Africa and accounts for about half of West Africa���s population [15]. Health service delivery is largely a government function and as in all countries, the es- tablishment of a robust national information system is a prior- ity. Though the HMIS framework was articulated (in 1992) and implementation commenced (in 1997) in a number of states, the HMIS is only recently (2003) beginning to be institutionalized [16]. This recent strengthening efforts (mainly donor-led) can be attributed to increased demand to show progress towards attaining the MDGs. Since 2003, a free and open source data warehouse solution, the District Health Information System (DHIS) [17] has been implemented at the national and state levels. However, a recent situational analysis has revealed the very low base from which the HMIS is being developed. Computer equipment is usually either in short supply or poorly maintained where it exists. Power supply is very poor and transportation through long distances and from hard-to-reach areas is difficult. Data use is almost non-existent at all levels of the system. Reports are submitted late and data quality is poor in the HMIS. It is in this premise and following from the observation that the mobile networks have greatly improved that this study is set. The application of mobile technology has a huge potential for circumventing the aforementioned challenges and improving data reporting. At the time of independence in 1960, Nigeria had a population of about 45 million people with 18,724 func- tioning fixed telephone lines - a tele-density ratio of 0.04 tele- phones per 100 people [18]. At the commencement of mobile telephony in 2001, there were only a few thousand lines avail- able from the operators and services were too expensive for the average Nigerian. By 2002, the number of mobile sub- scribers stood at 1.5 million and prices fell [19]. By the end of 2004, the GSM operators had recorded well over seven mil- I. Asangansi and K. Braa / The Emergence of Mobile-Supported National HIS in Developing Countries 541