Welcome back. In the final lesson of this module, let's move on to commercial groupers. Conceptually, these often perform the same tasks as open source grouper systems. These grouper tools, however, usually must be licensed for fee. It may be well worth the cost to invest in such tools, given that these are often more sophisticated in the open source versions. Thus similar to our open source group or lessons, I will discuss the benefits and costs associated with using commercial groupers. After the lesson, you will be able to prepare a value statement for various commercial groupers to inform analytic teams what benefits they can gain from these commercial tools in comparison to the licensing and implementation costs. Let's get started. Since we finished the open source section with pharmacy groupers RxNorm, let's look at a few commercial drug groupers. There are a few licensed sources that are commonly purchased by health organizations, by companies such as First Databank or Medi-Span. These vendors have NDC feeds from numerous sources and most assert that their list are authoritative. Of course given the uncertainty of what NDCs actually exist, each vendor has counts that differ widely from others. These counts sometimes differ by thousands or tens of thousands. What are some of the advantages of commercial drug groupers? First, there are fewer input formats as compared to public domain groupers such as RxNorm. Thus, these are easier to build and maintain. Second, there are monthly updates and good documentation. Third, commercial groupers for NDCs are used by major players in health care such as large health plans. Finally, vendors say that they have proprietary insights that others such as RxNorm may lack. Okay, so what are the disadvantages? Cost is probably the biggest factor. From my experience, these vendors often charge a minimum of $50,000 per year to license the proprietary group or system. Also, these paid vendors do offer some support. Thus in contrast to some other systems such as RxNorm where there's less support, you may get more value because you're getting more high quality support from the vendor. Now, let's move back to the groupers that have the purpose of measuring expected cost. A big innovation in this area was the creation of the diagnosis related groups or the DRGs. Initially the DRG system was developed by Yale University in the 1980s. A newer all patient DRG system was developed more recently by the 3M company. The purpose of the DRG system is narrow in scope. It functions to categorize hospital patients by their expected cost. Thus when using the DRGs, administrators can reimburse hospitals by the case mix of patient illness rather than for specific treatments that are administered. For example, if a patient met all the demographic diagnoses and treatment criteria for specific DRGs, then the hospital will be paid on an agreed upon rate for that DRG rather than for the exact cost of the specific services that that patient had received. The input data are mainly demographic factors such as age, gender and then the diagnosis codes is defined by the ICD terminology system. Concerning the actual groups, the DRGs produced 23 major diagnostic categories and then there are specific subdivisions within those groups. Next, the Adjusted Clinical Groups or the ACG's are conceptually similar to the DRGs, given that these are groupers for payments. Researchers at the Johns Hopkins University in Baltimore Maryland created the ACGs. Their purpose was to create a so-called case-mix adjuster for ambulatory populations, in other words, similar to the DRG groups these are formed based on ICD codes and demographic characteristics. Then payment amounts can be assigned to each group based on predictive modeling and actuarial valuation. Okay, so how are these ACGs constructed? Diagnosis codes are clustered into groups with similar condition, severity and likelihood of persistence through time. As a result, diagnosis codes are assigned to one of 32 Aggregated Diagnosis Groups or ADGs. Then a 102 discrete clusters are assigned based on age, sex and ADG. Okay, great. Now, let me move on to another commercial group or system designed for financial monitoring, quality improvement and other analytic tasks. Episode Treatment Groups or ETGs were created by a company called Symmetry. This was then purchased by Optum. Optum is a service company associated with UnitedHealthcare. Episode treatment groups are illness classification methodology. ETG is routinely used collected claims data as an input. The claims data includes professional, inpatient, outpatient, ancillary, and pharmaceutical data. The ETG software captures the relevant services provided during the course of a member's treatments and organizes these claims or encounters into meaningful episodes of care. The result is the accurate identification of clinically homogeneous episodes of care, regardless of the treatment location or duration. Overall, the ETGs use diagnoses, procedures, and prescriptions to group claims and encounters into meaningful episodes of care. Possibly, most important of all, the episodes of care are created to be clinically homogeneous. For example, imagine a person breaks their arm in a bike accident. The episode of care would include the patient's ER visit, possibly three outpatient visits associated with the accident and then any pharmacy prescriptions in any durable medical equipment that was needed for the person during this episode. The ETG condition classification system is comprised of approximately 524 clinically homogeneous and statistically stable groups. These groups are referred to as ETG base classes representing the underlying disease or condition. Some examples include diabetes, asthma and hypertension. Once again, ETGs are designed to be clinically homogeneous. Thus each member's illness and severity are medically consistent with others belonging to the same ETG. This is important because physicians and other types of providers can understand these illnesses and groupings. In addition, observed differences in treatment patterns within an ETG leads to clinical homogeneity. These groups can be used for risk stratification to identify providers vary with respect to how sick their patients are. ETGs have been developed to account for differences in member severity including variations and complicating conditions, co-morbidities and major surgeries. Once broken out from the base class code, adding and complications, co-morbidities and treatment categories, this leads to over 1400 ETG groups. The ETG base class is a six digit number where the first four digits are the condition class and the fifth and sixth digits are the body location. For example, the ETG base class number for diabetes is 163000. The ECG base class for a major joint inflammation of the hand, wrist and forearm is 7111904. The ETG number is a nine digit number that concludes a sixth digit ETG base class code, plus the last three digits on the right, which identifies the complication, the co-morbidity and the treatment indicator codes. Episodes are evaluated to determine if they have any complicating factors or if there are any co-morbidities associated with the episodes condition and if the activity within the episode contains any treatment indicators. This information is reflected in the episode treatment groups or the ETG number. Thus allowing to see specific characteristics of each episode. Now, let's briefly review the process of creating episode treatment groups. Anchor records start the process because an anchor records starts an episode. Anchor records represent services by clinician engaging in the direct evaluation or treatment of a patient. These might include office visits, surgeries or various therapies. The identification of an anchor record is significant because it represents that a clinician is evaluated a member and is decided on types of services required to identify and treat the member's condition. Anchor records are identified by a combination of the provider and the procedure codes. There are three types of anchor records, medical procedure by clinician, surgical procedure, and facility records showing room and board codes. Next, incidental services that are associated with the episode or added. These are types of services that include pharmaceutical prescriptions, lab work, radiology or physical therapy. Third, clusters of records are combined in episodes based on temporal overlap within the clean period. Fourth, complication, co-morbidity and treatment flags are added based on this information. Fourth, the episode is final if no additional episodes are found within the ETG clean period. After the ETG system builds the episode, it has all of the claims information gathered for that episode. The system is now ready to finalize the episode. This process consists of three steps. Step one, the algorithm determines the completeness of the episode. Step two, it determines if the episode has an outlier status and step three, it assigns responsible provider to the episode. In summary, why are the episode treatment groups useful? Well, they can clearly be used for many tasks. But here's a list of a few things that I consider to be important. ETGs can be used for tracking patients or members through the course of their illnesses. They can be used for provider profiling, which includes analysis of adherence to drug guidelines or clinical protocols. ETGs can also be used for reimbursement purposes, disease management, clinical benchmarking, case and outcome management, clinical protocol development and finally, for fraud detection. That concludes our overview of commercial groupers commonly used in healthcare. It also concludes this module. Thank you very much and we'll see you soon
