All Times ET
Program is Subject to Change
Linking Facility-Level and Client-Level Substance Use Survey Data: A Cautionary Tale (307976)Cathie Alderks, Center for Behavioral Health Statistics and Quality/SAMHSA
Haslyn Hunte, Eagle Technologies
*Kenona Southwell, Eagle Technologies
Doren Walker, Eagle Technologies
Keywords: N-SSATS, TEDS, datasets, linking datasets
This presentation will explore the feasibility of linking the N-SSATS and TEDS data sets. The presence of N-SSATS, a national substance use treatment facility-level data set, and TEDS, an episode-level data set, has led to the assumption that linking the two data sets would provide a better understanding of substance use treatment in the United States. Throughout the past approximately 20 years, several attempts have been made to link the Treatment Episode Data Set and the National Survey of Substance Abuse Treatment Services (N-SSATS), all with limited success. The results from a comprehensive attempt to link the TEDS and N-SSATS data will be described in this presentation. Two methods were explored: (1) linking using the Inventory of Behavioral Health Services identifier, or I-BHS ID; and (2) linking using the Federal Information Processing Standard state codes (State FIPS) and State Provider IDs. Results show that few facilities in TEDS and N-SSATS can be linked. In fact, six to 100 percent of the TEDS data would be lost in the linking process depending on the state of interest. Due to the limitations in combining these data sets, any inferences made between the two data sets with regards to client characteristics in TEDS and/or facility characteristics in N-SSATS will be unreliable.