[This is a preview of some of the exciting mHealth research being presented at the upcoming Medicine 2.0 Congress in September. This abstract and others are candidates for the iMedicalApps-Medicine 2.0 mHealth Research Award]

By: Nadi Kaonga, iMedicalApps-Medicine 2.0 Award Finalist

Background: The Importance of the Social Network

Ever wonder what sort of impact the social structure of an organization could have on its overall performance?  The network structure of an organization can influence how well or poorly an organization communicates and manages/utilizes its resources, including information.  Within organizations, especially those involved in public health and policy, information flow is usually critical to the success of programs.

In the Millennium Villages Project (MVP), in particular, this has been well-noted, especially amongst the local health teams.  There are fourteen local health teams working across ten Sub-Saharan African countries, supported by regional team members with additional counsel from multi-disciplinary researchers and practitioners based at Columbia University.

The typical local health team is quite extensive as well, as it includes community health workers, facility-based nurses, program-specific facilitators and a health team coordinator.

To enhance intra-site communication, a mobile phone closed user group (CUG) was set-up at all Millennium Villages Project  sites. This closed user group allows members of a specific site’s health team to hold voice conversations, via mobile phone, with one another at no cost. 

So, what sort of an impact would the use of mobile phones in this manner have on how the local health teams communicate and manage their information?

Objective of Research: Evaluating the Social Network

To date, no analysis of the closed user group use and utility has been conducted. Consequently, it was unclear whether the flow of information had improved within the health teams or what kind of impact the closed user group-related information flows were having on day-to-day activities of the health workers or on health-related outcomes.

Therefore, the researchers sought to assess if and how a mobile phone closed user group (CUG) ‘disrupts’ the traditional social/communication structure of one of the MVP local health teams—the Bonsaaso, Ghana, MVP Health Team.

The researchers identified social network analysis (SNA) methods as an innovative tool to explore the closed user group social network structure juxtaposed to the traditional social structure (or organogram/organizational chart) of the same network without the closed user group.  In public health, the ability to visualize and analyse a social network provide an added element to understanding the environment and impact that the structure may have on the success of an intervention.

Methods: Social Network Analyses

Social network analysis methods were used to assess if the mobile phone closed user group disrupted the traditional, hierarchical communication structure of the organization and/or the efficiency of information flow within that system. The methods were also used to identify central actors in information flow within and outside the closed user group.

The analysis leveraged the rich call data from the telecommunications provider.  De-identified call data, spanning from March 2011 through September 2011, for all members of the closed user group was used to run analyses that would help identify central actors within the network, the attributes of the network and create sociograms (or visual representations) of the network for observational analyses.

The call data was complemented by a qualitative component that included interviews with key informants of the Bonsaaso MVP Health Team and prospective call journals kept by key informants for a two-week period.

Results: The Network and What It Means

The Bonsaaso closed user group consists of 79 users.  The Bonsaaso MVP Health Team and key local staff from the Ghana Ministry of Health and Ghana Health Service make up the closed user group.  The Bonsaaso MVP Health Team includes the following:

  • The Health Coordinator, who reports to the Bonsaaso Team Leader/Science Coordinator, leads the MVP health team.
  • Midwives, who are based at facilities, oversee both community health nurses (CHNs) and the community health extension workers (CHEWs) supervised by the CHNs.
  • Health facilitators oversee the midwives, CHNs and CHEWs.
  • The ambulatory staff is responsible for providing emergency transportation from the community to the nearest referral hospital.

The results of the analysis show that members of the closed user group conversed with MVP Health Team members within and outside of their geographical area.  Midwives had the most intra-site communication.  The sociograms (see Figures 1, 2 and 3) illustrate how in the CUG, the Health Team Management (i.e., Science Coordinator/Team Leader, the Health Coordinator and health facilitators) bypassed the traditional chain of command (see Figure 4) and conversations took place between those at the bottom and the top of the Health Team hierarchy, without passing through intermediate levels.

While CHNs were identified as the central actors in the traditional network, members of the Health Team Management were the most central actors in the CUG network.  These results were consistent over time (March 2011 through September 2011). High rates of personal call use were expected on weekends but the data documented high levels of personal use during all periods, not just weekends. 

Conclusion: The case for Social Network Analysis

As evidenced by the social network analysis and qualitative data, the mobile phone closed user group (and use of mobile phones) creates constructively disruptive communications channels between members of the Bonsaaso MVP Health Team, subverting traditional hierarchies of authority and information control. This may, in part, explain the central role of the Health Team Management in the social network findings.

Implications of this include a heavier burden on the Health Team Management as they deal with increased communication in addition to fulfilling their daily responsibilities.  However, the unlikely connections may allow for issues to be resolved quickly, increased capacity (i.e. knowledge base) at all levels and information to be channeled more quickly and directly to relevant persons.

This study also shows how social network analysis methods and de-identified mobile phone data can be used to improve the performance of health teams.  We believe that the methodology can be adapted for use in the assessment of other closed user group or mobile phone networks as a management and public health tool.  Furthermore, such findings from network analyses could be used to contribute to the evidence base for mHealth and help inform business and policy cases on CUGs.

Blog Post Author

Nadi Nina Kaonga recently received an MHS in the International Health Department’s Global Disease Epidemiology and Control Program at the Johns Hopkins Bloomberg School of Public Health.  She currently works for the Department of Biomedical Informatics and the Earth Institute at Columbia University on eHealth initiatives.

Authors on Medicine 2.0 Abstract (from which this article is derived)

Nadi Nina Kaonga, Alain Labrique, Patricia Mechael, Eric Akosah, Seth Ohemeng-Dapaah, Richmond Kodie, Andrew S. Kanter, Orin Levine