Make Risk Adjustment Fair and Accurate for Quality Measurement

Each set of patients, and each individual patient, presents a different challenge. No two patients will be the same. Finding accurate methods to adjust for risk is becoming more and more crucial for physicians. Public oversight of physician and practice performance via the CMS Physician Compare website is changing private insurance and Medicare reimbursement strategies.

Academic medical centers see more difficult cases than single physician practices. Improving the quality of healthcare is more challenging for the single or multi physician practices because they often lack the resources enjoyed by large medical centers. For healthcare to improve for all patients, practical quality measurement is critical across all medical centers and physician practices to determine where improvements can be made both financially and clinically. Risk adjustment for differences in patients and the care they receive from different physicians provides standards with which physicians can be gauged against their peers.

Accurate Risk Adjustment Challenges

Risk adjustment calculations are only as good as the measurements used to calculate the modification. You cannot adjust for risks you are unable to measure. Whether a risk can be measured or not is subjective. For example, an obese patient is likely to experience complications after surgery. If the patient was not diagnosed as obese, the surgeon will be graded based on the expected outcome of a patient at ideal weight. Patients with height and weight that indicate obesity are often not coded with a diagnosis of obesity. This scenario is more common than you may realize.

Several factors can influence outcomes and indicate the patient will require more attention, but are not coded. These factors will be missing from risk adjustment calculations. The question becomes, why aren’t they coded? If an undetermined diagnosis is reimbursed in the same way as one that includes complications, it’s easy to assume saving time on the coding is better for the practice. This assumption fails to recognize that additional coding will ensure that providers are scored correctly and their score will show an accurate picture of their patient population.

Challenges on a program level

Many of the traditional risk adjustment protocols were developed in hospitals and large medical centers. However, the PQRS traditionally included quality metrics focused primarily on the ambulatory side. That wasn’t a problem when the program’s chief focus was on reporting, but with revenues linked to comparative provider performance through the Value Modifier, risk adjustment is not widely utilized or defined. In PQRS and VM this means that not everyone holds the high ground or equal footing in the arena. The challenge is increased by the introduction of care coordination metrics, which includes, but is not limited to, measures of communication, accessibility and quite a bit more. These areas lack established criteria for risk adjustment calculations.

PQRS measures, and proposed measures, include an abundance of process measures to determine if an action was performed. These are often adjusted by including reasons why the action or procedure was not performed so providers are not penalized.

At least one outcomes measure is necessary for quality reporting under CMS’s final rule, which leaves some providers in a challenging position. For example, an existing outcome measure looks at patients who have hypertension and if the patient’s blood pressure is controlled. Currently, there is no risk adjustment applied to this measure; there is no accounting for socio-economic status, co-morbidity, race or other factors. Since patient populations can vary wildly, some providers may find adjusting for risk easier than other providers. To further complicate matters, there are more process measures than outcome measures. This can force some providers to report a negative measure and deal with the consequences.

Improving Risk Adjustment

Quality measurement is crucial, but they must be reliable quality measurements. Providers and QCDRs must work together to develop accurate procedures to adjust for risk. For these risk adjustment strategies to be meaningful, trusted tools for refining outcomes, each side has responsibilities:

For Providers:

Coding should be accurate, complete and specific so it mirrors what you see in your office. Your patients will vary, but unless that is coded, you are the only one who knows how wide that variance may be. The Value Modifier gives providers an opportunity to earn an additional incentive if their patients are considered at high risk. If you are not coding for risk factors, you are sabotaging yourself by lowering your patients’ adjusted risk, which, in turn, makes it more difficult for you to compare in a positive way to your peers when being scored on quality or rate.

Strive for cooperation between your health information technology vendors and your clinically integrated network, as well as for operational data for their patients. Do not rely on patient responses if possible; you can’t be sure that patient received treatment or underwent a procedure at another medical center or practice. To accurately risk adjust, you need claims data, so if you aren’t aware or are unsure of a patient’s history, treatment may be different from what those applying risk adjustment anticipated. Failing to treat a symptom, condition or perform a procedure may have dire consequences. In other words, risk adjustment for a patient may favor you, but if you aren’t treating all the patient’s conditions, you’ll score poorly when compared to your peers.

For Qualified Clinical Data Registries (QCDR):

When developing a risk adjustment system, don’t stop with claims data. Non-billed clinical data, such as height, weight, addictions or tobacco use and the number of medications a patient takes can be used to determine which patients are more likely to incur higher costs.

By using information gathered from outside your network of services, QCDRs can provide a better footing for providers by giving them access to data typically only available to health plans. Knowing which providers a patient has seen, what procedures they have undergone or what other providers have treated the patients for crucial to making well-informed clinical choices.

As QCDRs continue to develop measures, they must determine how to score providers accurately. There may be incalculable data that providers wish to incorporate into personalized risk scores, or to develop clinical registries to enable care delivery for an at-risk patient population.

What results can you obtain by comparing patients in varying levels of risk to utilization type and/or volume?

Scoring providers on patients’ outcomes is not always going to produce absolute and fair results. Clinicians can’t control everything and many are challenged to measure processes, much less outcomes. Medicare is going to use value based care to measure quality rather than outcomes for reimbursement.  To remain competitive and to help identify at-risk patients, both providers and QCDRs must adapt and learn new ways to identify and adjust for risk.