Date Published: February 26, 2008
Publisher: Public Library of Science
Author(s): Daniel Chandramohan, Kenji Shibuya, Philip Setel, Sandy Cairncross, Alan D Lopez, Christopher J. L Murray, Basia Żaba, Robert W Snow, Fred Binka
Abstract: Background to the DebateDemographic surveillance—the process of monitoring births, deaths, causes of deaths, and migration in a population over time—is one of the cornerstones of public health research, particularly in investigating and tackling health disparities. An international network of demographic surveillance systems (DSS) now operates, mostly in sub-Saharan Africa and Asia. Thirty-eight DSS sites are coordinated by the International Network for the Continuous Demographic Evaluation of Populations and Their Health (INDEPTH). In this debate, Daniel Chandramohan and colleagues argue that DSS data in the INDEPTH database should be made available to all researchers worldwide, not just to those within the INDEPTH Network. Basia Żaba and colleagues argue that the major obstacles to DSS sites sharing data are technical, managerial, and financial rather than proprietorial concerns about analysis and publication. This debate is further discussed in this month’s Editorial.
Partial Text: Vital statistics on births, deaths, and causes of death are essential for guiding policy, planning, and evaluation of development programmes in all sectors, and particularly in the health sector. Health development is facilitated by critical and incisive analyses of a population’s disease burden and determinants of health. But policy and programme planning in many developing countries is severely hampered by the lack of representative vital statistics, data on disease burden, and critical analyses of the limited data that are available. This state of relative ignorance about disease burden and determinants of health is regrettable and unacceptable when there is a network of demographic surveillance systems (DSS) operating throughout sub-Saharan Africa and Asia.
Wider public access to primary data has become a clarion call in the bio-medical field, since the highly successful example of the 1996 Bermuda Principles developed by the International Human Genome Sequencing Consortium . These principles called for the automatic, rapid release of sequence assemblies of 1–2 kb or greater to the public domain. Funding agencies such as the Wellcome Trust are extending the requirement for researchers to provide data sharing plans to those working in public health epidemiology, including demographic and behavioural sciences . In their Viewpoint above, Daniel Chandramohan and colleagues fail to point out that many scientists working in DSS actively welcome the new interest in data sharing and the opportunities presented for pursuing new research collaborations to enhance the value of their data. Our Viewpoint addresses the nature of the real obstacles to data sharing by DSS sites.
Basia Żaba and colleagues appear to be in broad agreement with our Viewpoint about the need for easier access to public health data, including DSS data. They raise three main concerns with respect to providing wider and more rapid access to DSS data: (1) building local capacity for analysis and dissemination of data; (2) making the structure and content of data amenable for meta-analysis; and (3) protecting data ownership of DSS investigators. Clearly these concerns have to be addressed to make optimum use of DSS data without demotivating DSS sites.
We are pleased to note a subtle change in the stance of Chandramohan and colleagues in their response above. The discussion has now shifted to the need for easier access to all public health data while building research capacity and recognition of DSS sites and their scientists.