Ferranti L, Dilts D; AcademyHealth. Meeting (2003 : Nashville, Tenn.).
Abstr AcademyHealth Meet. 2003; 20: abstract no. 472.
Vanderbilt University, Management of Technology, 342 Featheringill Hall Box1518, Sta B, Nashville, TN 37235 Tel. (615) 322-3479 Fax (615) 322-7996
RESEARCH OBJECTIVE: Generate a framework that demonstrates the state-of-practice and the state-of-the-art in the realm of medical registries and highlights future needs for such data collections in real-time patient safety. STUDY DESIGN: A systematic search using MEDLINE for the years 1996-2002 to identify medical registry research. POPULATION STUDIED: Medline Indexed articles from 1996-2002 PRINCIPAL FINDINGS: Medline search identified 1,144 medical registry articles,which were reduced to 515 by excluding those referring to nurse or workforce registries, or when the registry was not the primary information source. There were 144 of such non-relevant articles, resulting in a final count of 371 articles. Medical Registries were divided into three groups. The first group is the generalized retrospective population-based data evaluated after the fact. The population-base can be either a geographic region (such as New England) or condition-based, such as asthma. Of the 371 applicable articles, 60.4% (n=224) were classified into this group.The second group is that of population / condition retrospective research based data gathered for another reason and/or clinical trials. For example, a hospital registry developed for administrative reasons and then utilized for research is an example such a registry. This group represented 32.3% (n=120) of the applicable articles.The final group was the active, prospective registry of ongoing evaluation of patient outcomes as collection continues. A representative example of this registry type is the end-stage renal disease (ESRD). Surprising, while such registries could make the most dramatic impact on immediate patient safety, they were only 7.3% (n=27) of all articles. Such a deficiency points to a critical opportunity for the use of information technology (IT) to improve the health and safety of patients.Furthering the research, we next divided the articles into those that primarily only used the registry as a data source (n=251, 67.7%) and those where the registry was the primary topic of the article (n=120, 32.3%). Of the second type, we divided them into four topics areas: 1) data quality in the registry (n=37, 30.8%, 2) registry establishment (n=29, 24.2%), 3) registry evaluation (n=27, 22.5%), and 4) other, miscellaneous uses (n=27, 22.5%). Interestingly, while a number of registries are being established, there is a dearth of research on the data quality, evaluation, and potential patient safety implications of such registries. CONCLUSIONS: Most registries are retrospective which makes it difficult, if not impossible, to provide timely outcome evaluations. Evidence illustrates that registries remain major sources of data for researchers, but that the full potential of registries are not realized in such passive studies. Although real-time science based, and prospective registries are being developed, limited research has been conducted to provide an ongoing analysis of their importance and relationship to outcome evaluation, analysis, and on-going patient safety. IMPLICATIONS FOR POLICY, DELIVERY OR PRACTICE: While there is the possibility for IT to play a central role in the redesign of healthcare systems as advocated by the IOM, NCI, and the Leapfrog group; the current use of registries does not fulfill this mandate. The basic concepts of registries and their uses must be reevaluated.
Publication Types:
Keywords:
- Delivery of Health Care
- Forecasting
- Humans
- Kidney Failure, Chronic
- New England
- Registries
- Research
- Research Design
- utilization
- hsrmtgs
Other ID:
UI: 102275447
From Meeting Abstracts