3M Commentary

Clinical Controversies in Outcomes Measurement

October 23, 2013 / By Norbert Goldfield, MD, Richard Fuller, MS

Although there is growing consensus that quality-based financial incentives can achieve higher quality and lower cost care, healthcare policy researchers differ on which model will bring about the most improvement. There are three overarching clinical controversies one needs to keep in mind when focusing on outcomes measures for payment adjustment:

  • Categorical vs. Statistical Regression Model
  • Case-by-case vs. Rate-Based Approach
  • Outcomes Measurement vs. Process Measurement

We have discussed two of these issues (categorical vs. statistical and outcomes vs. process measurement) in previous blogs. In this blog we will focus on the case-by-case vs. rate-based approach. By highlighting the approach the State of Maryland uses to reduce potentially preventable complications (PPCs), we will also show the differences between PPCs and other approaches to identifying complications.

Case-by-case vs. Rate-Based Adjustment at the Institutional Level

When choosing between case-by-case vs. rate-based adjustment, the issue is whether payment adjustment should be applied on a case-by-case basis or on a rate basis at an overall institutional basis. With the exception of a small number of negative outcomes that are virtually always preventable, applying judgments about quality for individual cases tends to be an inefficient and unfair way to change behavior and improve efficiency. For example, the Medicare Hospital Acquired Condition (HAC) payment policy is administered on a case-by-case basis. As a result, the only complications that are included as HACs are those that can be considered practically always preventable. This limitation greatly restricts the number of conditions that can be considered HACs, severely limiting their payment impact.

In contrast, determining the payment adjustment for quality based on overall institutional performance (for the hospital, managed care plan, etc.), avoids a key problem with the case-by-case approach. It allows the determination of an overall quality payment adjustment that is applied to all patients treated at the institution, regardless of whether the individual patient experienced the negative outcome, thereby avoiding the need to designate the care of specific patients as substandard. The payment adjustment for readmissions in the Affordable Care Act is an example of a rate-based determination of a quality related payment adjustment.

Instead of using the same model as the Medicare HACs, the Maryland Health Services Rate Setting Commission implemented a rate-based approach when it implemented the vast majority of PPC categories in 2009. A recent article published in Health Affairs summarizes the impact of the program:

The final ranking of the hospitals in Maryland’s hospital-acquired conditions program is based on overall additional resource use resulting from high or low complication rates for each hospital as a percentage of its total inpatient charges. Similar to the first program, the incentive payments are distributed based on this ranking, using a linear function in a revenue-neutral manner. The commission has gradually increased the amount of revenue at risk for penalties and rewards, reflecting more emphasis on outcome based pay-for-performance. In state fiscal year 2013, which will end June 30, 2013, the maximum penalty for the worst performing hospital was raised to 2 percent of the hospital’s total inpatient revenue, resulting in a reallocation of $16.7 million as rewards or penalties—an increase from $2.1 million in the first year.

The impact of implementing the PPCs hospital-acquired conditions program in the state of Maryland is that the PPCs declined by 15.26 percent over two years, with estimated cost savings of $110.9 million over that period. [1]

When we determine payment adjustment for quality based on overall provider system performance, we avoid the key problem found with the case-by-case approach. The overall quality payment adjustment is applied to all patients treated by the provider system. The rate-based approach includes both positive and negative patient experience outcomes, avoiding the need to designate the care of specific patients as substandard.

Negative outcomes are often the result of deficiencies in coordination and communication within a provider system. Focusing on system-wide quality is more appropriate and useful. Such a focus emphasizes that the entire health care team within the provider system has the responsibility for improving quality. The payment adjustment approach for readmissions in the Affordable Care Act is an example of a rate-based determination of a quality-related payment adjustment. In future blogs, we will cover the statistical techniques of calculating the rates as well as challenges in identifying a best practice norm.

Norbert Goldfield, M.D., is Medical Director for 3M Health Information Systems.

[1] ICalikoglu S, Murray R, Feeney D. Hospital pay-for-performance programs in Maryland produced strong results, including reduced hospital-acquired conditions. Health Aff (Millwood). 2012 Dec;31(12):2649-58.