It's elementary: Collecting data that unite research efforts
- It's elementary: Collecting data that unite research efforts
By Caroline M. Tanner, MD, PhD and Karl Kieburtz, MD, MPH
Special to The Movement Disorder Society
During the last year, experts in Parkinson's disease (PD) identified the instruments most likely to be used in clinical research studies of PD in order to allow the development of common methods for recording information collected using these instruments, called common data elements (CDEs), as explained in the accompanying article. This work has taken advantage of the MDS Task Force rating scale critiques and many MDS members have provided their expertise to this effort. In this edition of the website, the NINDS announces a request for comment and review of the proposed CDEs.
Why is the development of common data elements important? Although this effort may at first blush seem uninspired, in fact the establishment of common methods for recording data can facilitate PD research in many ways as discussed below.
a. Convenience and efficiency: The use of a standard method for recording key information will simplify the development of data-collection instruments and study databases. Investigators will be able to use established templates for standardized instruments. This efficiency will shorten start-up time, allowing more study resources to be used to address important scientific questions rather than infrastructure development.
b. Facilitate comparisons among studies and meta-analyses: Findings in a single study require replication. When study variables differ, comparisons can be challenging. In extreme cases, it may be difficult to determine whether differences between studies are due to differences in study methods, or to true differences in results. Using common data-collection tools with standardized documentation will promote uniform methods and minimize such uncertainties in data interpretation.
In addition, the use of CDEs will facilitate combining data collected in different studies (Thurmond et al 2010). Because PD is relatively rare, study populations are typically not large, and many critical research questions cannot be addressed. For example, investigations of gene-environment interactions typically require large population sizes. Because PD is in many regards heterogeneous, too few members of a particular subgroup may be present in a single study population (e.g., impulse control disorders, dyskinesias). Combined analysis of information collected on disease subgroups in multiple populations will be needed to study many PD subgroups. The use of CDEs to record data will simplify and facilitate these analytic approaches.
Furthermore, the use of CDEs will enable investigators to more easily conduct meta-analyses in which the statistical synthesis of results from a series of studies may allow for the examination of a body of evidence (Chalmers et al 2002). Additionally, because CDEs will be uniform not only for PD research, but, when applicable, in other neurologic disease areas, novel research hypotheses may be able to make use of data from studies of other conditions.
c. Facilitate international studies: As the global burden of PD is increasing and the international distribution of the disease burden is changing (Dorsey et al 2007), uniform exposure-assessment methods that can be shared across language and cultural groups are becoming increasingly important. As many of the instruments identified by the topic-specific subgroups are available in multiple languages, using the CDEs can be expected to facilitate comparisons of data collected in different languages.
d. Provide uniform descriptive data for banked specimens: Identification of biomarkers of disease susceptibility, diagnosis and progression is a critical need in PD research. The use of well-annotated biologic samples from tissue repositories is a useful approach to biomarker investigation (Patel et al. 2005). For PD, early efforts are under way (Ravina et al, 2009; http://www.ppmi-info.org/), and such efforts are expected to expand, as in the proposed NINDS-sponsored Parkinson's Disease Biomarkers Identification Network. Annotation of biological samples using CDEs will facilitate use of specimens from different populations, in turn improving generalizability and allowing more efficient investigation, including subgroup analyses.
The ultimate goal in implementing standardized methods such as the CDEs is to provide comprehensive evidence that will facilitate many areas of research and, in turn, result in an enhanced understanding of PD and improved treatments.
Chalmers I, Hedges LV, Cooper H. A brief history of research synthesis. Eval Health Prof. 2002;25(1):12-37.
Dorsey ER, Constantinescu R, Thompson JP, et al. Projected number of people with Parkinson disease in the most populous nations, 2005 through 2030. Neurology. 2007;68(5):384-6.
Patel AA, Kajdacsy-Balla A, Berman JJ, et al. The development of common data elements for a multi-institute prostate cancer tissue bank: the Cooperative Prostate Cancer Tissue Resource (CPCTR) experience. BMC Cancer. 2005;5:108.
Ravina B, Tanner C, Dieuliis D, et al. A longitudinal program for biomarker development in Parkinson's disease: a feasibility study. Mov Disord. 2009;24(14): 2081-90.
Thurmond VA, Hicks R, Gleason T, Miller AC, et al. Advancing integrated research in psychological health and traumatic brain injury: common data elements. Arch Phys Med Rehabil. 2010;91(11):1633-6.