In the research of Hoenig,62, 63 and 65 Reker,65 and colleagues variables such as the availability of an adaptive kitchen, the number of disciplines present at chart rounds, and the physical therapist caseload were included. In Donabedian’s scheme,66 process is defined as what is actually done to or with the patient within the overall structure. It includes processes typically considered “clinical” and indirect care, “guest services,” and administrative

procedures. In the short term, structure dictates process, whereas both structure and process affect outcomes. To date, Hoenig,62, 63 and 65 Reker,65 and colleagues have not addressed process, at least not in the meaning of that term considered here. The long-standing interest of Strasser et al67, 68 and 69 in delineating characteristics of the rehabilitation CAL-101 molecular weight team and establishing their impact on patient outcomes is also focused on classifying the structure of rehabilitation. Other attempts to characterize rehabilitation services using a combination of characteristics such as location, Epacadostat molecular weight general thrust of activities, and program type have been published.70

Most of these are ad hoc efforts to impose order on the unruliness of existing services, without the benefit of (explicit) relevant theories.71, 72, 73 and 74 Structure elements and process elements other than direct care, such as chart rounds and other coordinative structures/processes, can explain changes in patient outcome only because they are necessary but not sufficient conditions for the delivery of treatments.75 One could imagine a state-of-the-art

rehabilitation facility with a well-trained staff meeting 24 hours a day busy coordinating care, with no one ever seeing a patient.75 Thus, to explain what is going on in the black box and use the information to explain outcomes, we need to do more than classify structure and the indirect categories of process. Even more recent than the work of these authors is research that has inductively (or “bottom up,” in the terminology of DeJong et al2) created classifications of the therapy process (what is actually done with, to, and for patients) as part of practice-based evidence (PBE) studies of inpatient rehabilitation. Relevant articles have been published of tuclazepam rehabilitation for stroke,76, 77 and 78 knee or hip replacement,79 and 80 spinal cord injury (SCI),81 and 82 and traumatic brain injury (TBI).83 and 84 In all of these projects, clinicians developed lists of “active ingredients” used in their practice: treatments (“activities”) that they presumed to have a significant impact on outcome, with subcategories and modifiers (“interventions”) added as appropriate. Data collection forms allowed them to characterize each treatment session in terms of the “activities” delivered, and the quantity of each, mostly in terms of minutes.