This article originally appeared in Becker’s Health IT and CIO Review on March 3, 2016.
As doctors, we’re trained to treat patients one at a time. Our research and treatments are individually focused, and so are the care outcomes. So you can imagine how bad data or misinformation (the wrong lab or radiology report, or a wrong medication list) can have serious, negative implications on the care and outcomes for a particular patient.
As doctors, we’re also riding the wave into a new care delivery system. Data now affects our bottom line as well as our patient care. While bad data may harm a single patient, it can grievously injure or kill your accountable care organization (ACO). Several factors, such as quality measures, physician satisfaction and medical utilization rates, depend on accurate data.
Before an ACO can share in generated savings, it must demonstrate that it met the quality performance measures for that year—a measurement function that is almost entirely dependent on technology.
When doctors provide care that can be tied to a quality measure, that information makes its way back to a claim, and then back to the ACO in an organized fashion so that a physician knows where he or she stands on that measure. This information is collected and then analyzed on an ongoing basis to benchmark physicians on quality measures, and then benchmark the ACO’s success as a whole. If an organization’s data is compromised, leadership will have the wrong understanding of the ACO’s success with its quality measures. Overestimating or underestimating performance can have major implications for how an ACO allocates resources and earns surplus.
Physician satisfaction and alignment
More often than not, providers are initially skeptical at the outset of an ACO undertaking. In fact, a 2014 poll from KPMG found that 33% of providers believe participating in ACOs will reduce profits and disrupt the revenue cycle.
The introduction of bad data can create even more skepticism among providers. If leadership makes a poor decision based on bad data, physicians may receive incorrect messages or instructions. This creates an additional level of mistrust between leadership and providers within the ACO.
Physicians also know that in order to achieve good care outcomes, data must be actionable. Bad data is not actionable and creates bad outcomes. To combat this, ACO leadership should have the technology necessary to handle and provide the level of analysis that doctors require and deserve.
It is vital that ACOs use resources (TME) appropriately to manage and stay within budgets. There are several areas of utilization that ACOs look at, including hospitalizations, observation stays, pharmacy, radiology procedures, ER visits, and specialty procedures. These areas of utilization are 100% dependent on data to inform if they are being under- or over-utilized against existing benchmarks.
Because ACO leadership teams apply utilization data to their business decisions, bad data will lead to flawed resource allocation and other misguided decisions downstream. For example, every ACO should be looking at rates of utilization for radiology (MRI, CT), especially for certain ailments. When tracking utilization rates for these modalities, leadership assumes that the data they’re working off of is correct. If there are errors in the data, it might appear as though a group is over- or under-utilizing the technology, sending the wrong message to doctors.
Another example is with patient stratification. For particular patient events, such as home visits, accurate data and lists are particularly critical. An ACO reserves home visits for the top percentage of the sickest patients, and will often use data to find who these patients are. If the data is flawed, however, it might over-identify patients who don’t need these services, wasting money and resources on those who don’t actually need it most, and leaving those who do require it to be under-served.
Another area where over- or under-identifying patients is detrimental is disease management. As ACOs develop registries for certain illnesses such as heart failure, diabetes and COPD, good technology and data is required to identify the patients in need of treatment. Bad data will identify the wrong patients, which will result in poor care outcomes while reducing resources for those who needed treatment the most.
Making bad data good again
Those involved in the inner workings of an ACO do their daily work under the assumption that the numbers are correct. Blindly relying on data—especially the wrong data—can be dangerous not only for patient safety, but for the ultimate success of the ACO.
To ensure the highest quality data, ACOs must implement technologies capable of data aggregation and analysis across multiple EHR systems, and the IT infrastructure must be able to organize both clinical and claims data. Combining clinical and claims data ensures that ACOs are working off of a larger, and more complete data set, reducing the chances of bad data causing harm to either patient care or your ACO.