“My patients are sicker.”
I’ve heard that sentiment in every kind of hospital and in every part of the country. I understand the perception. Clinicians see their own patients every day and in graphic detail. On the other hand, the patients of others are only in their peripheral vision; if they are seen at all. If you have ever heard or expressed the sentiment above, this issue of the Quality Note is for you. Spoiler alert: your patients probably aren’t sicker. Today I am focusing on risk adjustment and its importance in allowing for true comparisons of patient outcomes.
Many surgeons and quality officers point to their perception that they have the sickest patients to explain suboptimal results or bolster good performance. No one admits to having healthier—or favorable risk patients. While evidence is commonly lacking to justify claims of who has the sickest patient population, there is no question that an accurate interpretation of outcomes and costs cannot be made without objective measurement of risk adjustment.
It should be no surprise that a patient with congestive heart failure and kidney disease would require additional resources to deliver appropriate care and would have a higher likelihood of adverse events. Costs and outcomes are associated with the underlying chronic disease characteristics of the patient. Failure to correctly account for patient differences may ultimately add significant costs, waste resources, and increase the chances of harming patients. Too often, however, data that has not adequately accounted for patient risk is used to make financial and quality decisions.
Risk-adjustment is the science of separating the effect of patient characteristics from the effect of care delivered on cost or clinical outcomes. The patient characteristics are present when care begins. The process requires the identification of patient specific variables that influence either the clinical or the economic outcome of the case being evaluated. A list of candidate variables must be developed, beginning with demographic variables like age and gender. Diagnosis codes that are present on admission can be used to create variables that describe a patient’s health status. If available in the data, variables can be developed for laboratory values, vital signs, and other clinical data and added to the list. The list can also be supplemented with socioeconomic information, when it can be reliably captured.
Candidate risk factors are then offered to statistical processes for building predictive models. The processes identify significant risk factors and give each a “weight” representing their influence on the specific adverse outcome (e.g., inpatient mortality) or the cost that is being studied. If properly built, carefully crafted, clinically reviewed, and validated, these models can be used to explain the portion of cost or risk of adverse outcomes associated with patient characteristics at the time care began that are beyond the control of the provider. Using these predictive models permit an “apples to apples” comparison of provider performance by removing the effect of “sicker” patients.
The essential point is risk adjustment allows hospitals to accurately benchmark outcomes and costs against the country, their region, their historical performance, or other appropriate reference groups. Areas in need of improved outcomes, or conditions where costs are greater on a risk-adjusted basis, can then be the specific focus of care redesign to achieve better effective and efficient care. If you try to make comparisons of costs or patient outcomes without the right risk adjustment, it is easy to draw the wrong conclusion.
At MPA Healthcare Solutions, we have spent decades advancing the field of risk-adjustment. We understand its importance and how to apply it properly. We can provide you with your risk-adjusted performance across a wide array of surgical and medical cases. We can benchmark your performance against hospitals across the country so that you can improve quality and reduce costs. Don’t fall for the “my patients are sicker” line. Get the facts. Take action. We’re here to help.