Before we can achieve the vision for smartphones to broadly improve MCH, there are significant challenges that need to be addressed. First, privacy
concerns and ethical safeguards for the access, utilization, and sharing of personal data are crucial. Procedures to protect personal health data, similar to those currently used with electronic medical records, need to be integrated with any electronic health system. Structural and procedural mechanisms that will enable us to address the simultaneous needs of scalable personalized healthcare and personal privacy accountability include firewall, encryption, and password protections on computers, phones, and web-based systems; data-sharing policies and tiered access for different user groups such as field workers and supervisors; creation of an audit trail; and ongoing monitoring of security threats. Given these protections, there is evidence that electronic records are, in fact, more secure than paper records and that many patients prefer electronic records. For example, Curioso [61
] found that sex worker patients highly preferred the mHealth system when rolled out in their clinic because they were more concerned about privacy violations from their paper charts being left on desks, while the electronic records were perceived to be more secure.
Second, attention and sensitivity to cultural issues and local contexts can be critical in the effectiveness of mobile phone technology implementation. For example, a feasibility study in Sierra Leone [62
] found that individuals often share a single phone with other family members, or even with neighbours, due to limited resources. Some women, when sharing phones with their husbands, must ask for permission to use the phone, limiting privacy and the type of health information that women found acceptable to be communicated by healthcare providers via mobile phone. In some instances, husbands became jealous and physically abusive if they suspected that the text messages that their wives were receiving were evidence of infidelity. Thus, awareness of the impact of local customs, practices, and resources on the use of technology (and healthcare decision making) should inform implementation design.
Third, significant investment of resources in developing and integrating current technology into a cohesive, standardized, yet flexible system is needed. To date, there has not been a concerted effort in creating standardized interfaces, data collection strategies, or programming platforms. Innovative user interfaces
are needed to make data collection as easy as using an iPod and as seamless and clear as using a thermometer. Data analytics and visualization
, with the possibility of integration over multiple time points and varying levels of data aggregation, require further innovation and research around human-computer interfaces. An open platform
(i.e., similar to the web) with well-defined, standard interfaces must be defined early, and in detail, in order to develop a broad architectural framework of interoperable and portable services, rather than individual stovepipes based on proprietary solutions [63
]. In an initial review of websites that report federally funded studies, a University of Maryland researcher counted 486 mobile phone projects, with only 29 of those being actual mobile health interventions [64
]. There are now hundreds of pilot programs on mobile health applications with numbers and investment increasing rapidly over the past several years. While the cumulative investment is substantial, each project recreates about 80% of the programming with different programming languages, interfaces, and is highly tailored to the specific research study [63
]. The ubiquitous distribution of mobile phones, the relatively inexpensive costs for programming adaptions to open-source platforms (on the order of $50,000 USD in our experience with several developers and projects), and their massive scalability, all support great potential for cost-effective returns on investment.
Further, most mobile health demonstrations have not integrated mobile data with web-based data sets, which would allow us to infer activities and monitor environmental hazards or exposures—a huge leap in the types and scope of inferences possible by linking personal GPS date to Geographic Information System (GIS) data. Finally, the feature set implemented for these mobile health projects have not been able to take full advantage of smartphone capabilities. Investing now in the software and methodological infrastructure will allow us to harness the potential of smartphones by increasing the efficiency and scalability for healthcare delivery, and our ability to meet the MDG, over the long-term.
The development of standardized function programming will facilitate cost-efficient creation of optimal strategies for capturing and sharing data for different types of information, diseases, and purposes (e.g., epidemiological observations, prevention, treatment, and self-management). Finally, the ease with which mobile applications can be programmed and updated creates tremendous opportunities for relatively quick, broad, and inexpensive diffusion of health behaviour innovations.
Given these challenges and the current national and global climate to harness science and technology to improve public health, we encourage substantial investment to create mobile health platforms that serve the public good, by promoting health service innovations while attending to the need for the individual to control access and sharing of their personal data stream. The widespread wireless mobile phone infrastructure in LMIC is an untapped network that can facilitate low-cost, scalable delivery of, and access to, healthcare via smartphones and smart proxy systems to achieve Millennium Development Goals.