Factors Influencing NIATx Implementation: A Response

David H. Gustafson, Director, NIATx and the Center for Health Enhancement Systems Studies

Our research team at NIATX was very interested in the paper by Fields, Knudsen, and Roman on implementation of NIATx processes in substance use disorder treatment centers. (Implementation of Network for the Improvement of Addiction Treatment (NIATx) Substance Use Disorder Treatment Centers, Journal of Behavioral Health Services Research, May 2015.) A summary of the article is available in the NIDA Clinical Trials Network (CTN) Dissemination Library.

The Fields team studied the organizational factors that influenced use of the NIATx model by delving into the data they had collected for other studies involving 458 treatment centers. They found that participation in NIATx was associated with having fewer slack (available) resources as well as having more administrative resources, treating more patients, and participating in national associations and in the CTN. Surprisingly, the degree of financial pressure they found themselves under had no effect on these organizations’ implementation of NIATx.

Hindsight is 20/20. I have no idea what we would have guessed if asked prior to the study. But seeing the results stimulated a number of thoughts. Like the authors, I would have guessed that organizations that were under pressure to improve (those with fewer slack resources and those that were under financial pressure—the latter having no relationship to joining NIATx) would have been among the “joiners.” Both had a tension for change, either by being overworked, as indicated by their high patient to FTE ratio, or due to financial concerns. Yet the results were conflicting. Having more slack resources and less financial pressure increased an organization’s likelihood of implementing NIATx. Prior to this finding I would have interpreted both financial pressure and few slack resources to imply the same thing. But maybe having few slack resources really implies a well run ship; a ship with an ongoing commitment to getting better. In that case, the results make sense.

The study calls for further analyses examining interactions. For instance, if the Fields’ team looked only at organizations that were members of the CTN, would they still have found that participation in associations was related to greater likelihood of implementing NIATx? Maybe this indicates that some organizations have greater external ties, and that word of mouth may have been more likely to influence their decision to implement NIATx, whereas less connected organizations would have been less likely to hear about NIATx.

One of the advantages of having such a wonderful database to work with is that further questions can be explored. For instance, the Fields’ team collected data on reason for participation (e.g., to improve access). Did the goals interact with organizational characteristics? Were organizations with more intense management more likely to be interested in improving access? Were larger organizations more interested in retention in treatment? I hope they continue to mine the data set for further insights.

And I need to keep in mind that these are associational data, not causal. In other words, we don’t know what caused what, and we don’t know whether a deeper root cause is operating. In any case, the authors are to be congratulated for reporting such intriguing results. Furthermore, if these organizations were to be interviewed again, it would be fascinating to identify those who continue to use NIATx principles and see how they differ in size (for instance) from those who don’t.

Many thanks to the Fields, Knudsen, and Roman team for their work.