In response to these issues, ASPE funded a national project to develop quality measures for post-acute care that could be used in future research. The project strategy was to first specify quality measures that are most relevant to episodes of post-acute care provided in all three settings, based on extensive literature review and expert consensus.
In this project, we identified a set of quality measures for Medicare post-acute care services for four tracer conditions that are prevalent in the Medicare post-acute population: stroke, CHF, pneumonia, and back and neck conditions. Eight conditions were initially selected based on policy considerations,31 such as volume and Medicare utilization criteria (i.e., hospital re-admission rates). These eight conditions were later narrowed to the four tracer conditions based on a framework of clinical criteria,32 the goal of which was to represent different types of illness (chronic vs. acute), different post-acute care settings and services, and different domains of outcome measures. The methods for how the initial eight conditions were selected and how these were narrowed down to the four conditions included in this study are discussed in Appendix A.
Throughout this report, we frequently make reference to the terms quality indicator and quality measure; we often reference these constructs as they relate to outcomes or processes of care. The following definitions clarify the distinction between an indicator and a measure, and between a process and an outcome of care.
Outcome indicator -- Change (e.g., improvement, decline, recovery) in a health status attribute that reflects a patient outcome, but does not require precise quantification of the outcome. Examples of outcome indicators include improvement in global function, decline in pain, and recovery of ambulation/mobility. Use of certain services can also be considered as outcomes.
Outcome measure -- A precise quantification of an outcome indicator. For example, the outcome indicator of improvement in global function can be translated into an outcome measure when a scale such as the SF-36 is used at baseline and follow-up time points. The outcome measure might be the change in total score between baseline and follow-up.
Process indicator -- An attribute of care at the patient level, without specific quantification of the attribute. Examples of process indicators include evaluation of depression, patient education, and use of sedatives/hypnotics.
Process measure -- A precise quantification of a process indicator. For example, the process indicator of evaluation of depression can be translated into a process measure by establishing a set of criteria for determining the circumstances under which a depression evaluation should be conducted, defining the essential features of the evaluation, and rating the extent to which these are met.
Based on initial review of the medical literature and clinical experience, we generated comprehensive lists of potentially useful quality indicators relevant to post-acute care for each of the four conditions. To provide a basic structure for reviewing the indicators, we organized the indicator lists into domains (e.g., functional outcomes, utilization, process). We then convened four panels of health care providers (i.e., one for each of the four targeted conditions) to rate the importance and feasibility of including each indicator on a post-acute care quality assessment instrument. Based on this first round of panel meetings and the quality measures identified by the panel, draft longitudinal quality measurement tools were assembled for each condition using validated measures wherever possible. These draft instruments were then presented to a second expert panel comprised of experts in the field of post-acute care, research methodologists, and researchers with expertise relevant to each condition. Instruments were revised based on this meeting and subsequently pilot tested for feasibility of administration and time studies. Feasibility testing was limited in scope, taking place in two waves using post-acute care facilities recruited in the Denver metropolitan area. Revisions were made to the instruments based on qualitative experience with the instruments gained during the first wave of testing and input from the second expert panel meeting. A second wave of feasibility testing was then conducted, and final revisions were made to the instruments. The final instruments and users manuals produced as the products of this project are the result of the processes described above.
The following sections (Sections B through D) discuss the indicators selected, and instrument development and refinement.