Playing smallball: Approaches to evaluating pilot health information exchange systems (HIE)
Review of: Playing smallball: Approaches to evaluating pilot health information exchange systems (HIE). Kevin B.Johnson, Cynthia Gadd Journal of Biomedical Informatics 40 (2007) S21–S26.
This article succinctly summarizes the importance of HIT applied at the larger or largest level of organization of people, for which the technology applies-the community. Health information exchange is defined as the mobilization of healthcare information electronically across organizations within a region or community, and perhaps the more common acronym RHIO; regional health information organization better describes this area being discussed, because by definition the reader sees the complexity involved in the name. This is the point of the authors for this paper—how do you study the effects of a HIE. Not just how, but that it is complex. They start by paraphrasing Lorenzi’s observation, that HIE is intricately related to the technologies involved at the level of software, hardware, and policies/content. Implied here is organizational behavior of the people involved. Examples include the space needed for the hardware themselves. The need to have specifically written software for the clinic and sub-specialty. Then the need for actual written guidelines of people involved with the HIT themselves, vary from the bottom up, all affected by essentially psychosocial behaviors, they themselves, evaluable and unpredictable. The authors were careful to describe politics and competitive posturing and behaviors which are known to occur, and rightly so, because politics within powerful groups of physicians or clinics, is a known maker or breaker of a RHIO. That there is a dynamic or fluidity with the evolution of HIE, therefore the authors warn and suggest, taking evaluations and studies at various levels and phases of the development, “smallball: instead of the classic powerball randomized clinical trials studies of a powerball.” One example they give is that you cannot always do a RCT to evaluate an HIE, e.g. for ethical considerations, nor can you answer the question with such a design. The authors suggest frequent and smaller study and evaluation of an HIE project as it evolves, from problem definition, through phases 1-5, (pre-implementation analysis, usability, use, impact, to routine use). Pressures from developers or backers of the HIE to perform, cost effective outcomes studies, are not good reasons to be prudent and evaluate at the proper times.
Dan Dalan, for BMI512 Clinfowiki review 3#.