What to Consider When Choosing Appropriate Measure Concepts 

Without evidence-based guidelines specifically directed to care of persons with MCC, best practices may remain up to the clinical judgment of the measured entities. However, certain measurable quality topics are especially pertinent to people with MCC. The MCC Measurement Framework identified these measurement concepts as having potential for high leverage in quality improvement for patients with MCC:

• Optimizing function, maintaining function, or preventing further decline in function
• Seamless transitions between multiple measured entities and sites of care
• Patient-important outcomes (includes patient-reported outcomes and relevant disease- specific outcomes)
• Avoiding inappropriate, non-beneficial care, including at the end of life
• Access to a usual source of care
• Transparency of cost (total cost)
• Shared accountability across patients, families, and measured entities
• Shared decision-making

These quality measure concepts represent cross-cutting areas with the greatest potential for reducing factors of cost, reducing disease burden, and improving well-being, all of which are highly valued by measured entities, patients, and families.
The American Geriatrics Society (AGS)/AGING Learning, Educating, And, Researching National Initiative in Geriatrics Collaborative has developed an innovative self-directed curriculum to educate emerging clinicians and translational investigators on MCC knowledge gaps and the latest research. Measure developers can access the core curriculum which discusses “Data, Measures, and Measurements”, connect with the online collaborative community to garner feedback and share learnings, and listen to webinars and podcasts discussing hot topics related to MCCs research. 

What Determines How to Address Key Issues

Guiding Principles

The MCC Measurement Framework advises measure developers to follow several principles when developing quality measures for persons with MCC. Quality measures should

• promote collaborative care among measured entities
• consider various types of measures addressing appropriateness of care
• prioritize optimum jointly established outcomes by considering patient preferences
• address shared decision-making
• assess care longitudinally
• illuminate and track gaps of care through stratification and other approaches
• use risk adjustment for comparability of outcome measures with caution, as it may obscure serious gaps in quality of care
• standardize inputs from multiple sources, particularly patient-reported data

Time Frame Issues to Consider

Measurement time frame is particularly important with chronic conditions because the nature of chronic conditions requires observation over time. Especially in the case of outcome measures for patients with MCC, it is extremely difficult to know where to attribute responsibility unless there is careful consideration and specification of the measurement time frame. Measures for this population should assess care across episodes and across measured entities and staffing using a longitudinal approach. Delta measures of improvement (or maintenance rather than decline) over extended periods are particularly relevant in this population.

Attribution Issues to Consider

Compounding the issues of attribution occur when adding the factor of MCC. Since multiple conditions usually means multiple measured entities, it becomes difficult to choose who to credit for good outcomes and which measured entity gave inadequate care when the treatment for one condition might exacerbate the other. These issues may require a more aggregated level of analysis such as at a measured entity group level or population rather than individual level. Since beneficiaries with MCC have multiple measured entities, it would be more appropriate to measure and attribute the outcomes for the population to the care provided by the team of measured entities or create a combined measured entity (e.g., primary care providers).

Methodological Issues to Consider

Measure developers should design methodological approaches to reveal and track variances in care and outcomes. The empirical link between quality processes and the outcomes of those health care processes is even more difficult to establish when dealing with MCC. Measure developers should use risk adjustment with caution in the context of MCC. Stratification may allow quality comparison across populations without masking important distinctions of access, care coordination, and other issues. The supplemental material, Risk Adjustment in Quality Measurement, provides an in-depth discussion on how to determine when risk adjustment is appropriate and how to evaluate the application of risk adjustment models.

Quality measures for this population should address quality across multiple domains. Measure developers should harmonize measures across levels of the health care system to provide a comprehensive picture of care.

Data Gathering Issues to Consider

It may be difficult for measure developers to gather data systematically for the population of persons with MCC. In particular, it may be challenging for measure developers to collect patient-reported data due to interacting conditions. For example, it might be difficult to collect fatigue data from a person with both chronic lung disease and history of stroke, because each condition may contribute to a patient’s fatigue, and it may be hard to assess the contribution of each disease to fatigue. Measure developers may need to interpret different types of data, as the data may come from multiple measured entities, multiple sources, in multiple formats, and over extended periods of time. It is important for measure developers to standardize data collection methods

What Testing and Evaluating Measures for Persons with Multiple Chronic Conditions

Evaluation methods described elsewhere in the Blueprint content on the MMS Hub, e.g.,
Measure Testing, also apply to measures of quality care for persons with MCC. In addition, MCC measures should successfully carry out the guiding principles from the MCC Measurement Framework. Measure developers should examine functional status and other outcomes using measures of change over time. If new tools and/or methods of data collection become available, measure developers should assess them carefully. Formative, or alpha, testing may be particularly important early during development, not only for new tools designed for these types of measures, but also for testing the feasibility of linking data from a variety of sources.