Year: 2016 Source: Annals of Internal Medicine.(2016). Published online 4 October 2016. doi:10.7326/M16-1281 SIEC No: 20160474

Background: Linking national, state, and community data systems to data from prevention programs could allow for longer-term assessment of outcomes and evaluation of interventions to prevent suicide.
Purpose: To identify and describe data systems that can be linked to data from prevention studies to advance youth suicide prevention research.
Data Sources: A systematic review, an environmental scan, and a targeted search were conducted to identify prevention studies and potentially linkable external data systems with suicide outcomes from January 1990 through December 2015.
Study Selection: Studies and data systems had to be U.S.-based and include persons aged 25 years or younger. Data systems also had to include data on suicide, suicide attempt, or suicidal ideation.
Data Extraction: Information about participants, intervention type, suicide outcomes, primary analytic method used for linkage, statistical approach, analyses performed, and characteristics of data systems was abstracted by 2 reviewers.
Data Synthesis: Of 47 studies (described in 59 articles) identified in the systematic review, only 6 were already linked to data systems. A total of 153 unique and potentially linkable data systems were identified, but only 66 were classified as “fairly accessible” and had data dictionaries available. Of the data systems identified, 19% were established primarily for research, 11% for clinical care or operations, 29% for administrative services (such as billing), and 52% for surveillance. About one third (37%) provided national data, 12% provided regional data, 63% provided state data, and 41% provided data below the state level (some provided coverage for >1 geographic unit).
Limitation: Only U.S.-based studies published in English were included.
Conclusion: There is untapped potential to evaluate and enhance suicide prevention efforts by linking suicide prevention data with existing data systems. However, sparse availability of data dictionaries and lack of adherence to standard data elements limit this potential.

Contact us for a copy of this article, or view online at