In this clinical data analysis exercise, we will learn how to lay a foundation for measures development. Measures represent a relatively new concept in the healthcare data science space, and are designed to assist providers, payers, and patients with managing complex medical conditions and proactively identifying patients at risk for complications. From following basic pediatric vaccination guidelines to identifying chronic patients in need of a care gap coverage, measures provide information support aimed at improving quality of care and lowering its cost by both avoiding unnecessary care and reminders to introduce evidence backed procedures into the patient care regimen. You should review slides, recorded lecture, and supplemental materials for this unit prior to performing this exercise. If you have taken Clinical Decision Support Science (CDSS) course prior to registering for this class, you would have already seen some of the slides and lecture on the subject, albeit without actually practicing the concept.

The following National Committee for Quality Assurance (NCQA) web site provides a wealth of background and technical specifications on “real-life” HEDIS measures managed by this organization in charge of maintaining care quality standards for payers, providers, and employers.

http://www.ncqa.org/hedis-quality-measurement

1.Research a complex medical condition (CMC) of your choice, i.e. asthma, congestive heart failure. Please DO NOT select DIABETES. Focus your research on the areas of diagnosis (signs of disease and symptoms), common complications,early intervention, and recommended care. Consolidate your findings into the following Table 1:

Table 1

Symptoms

Common Complications

Intervention

Symptom 1

Symptom 2

Symptom 3

…. etc. as necessary ….

This table represents supporting evidence for your exercise. Depending on the specifics of your research, you are free to modify the format and content of this table, as you wish. A slide for disease-specific recommendations in the supplemental Unit 3 PPT lecture deck may serve as an example and additional guidance for you, using diabetes as a highlighted case. Three symptoms would be the minimum requirement for this table, yet you are encouraged to research the condition you selected in full.

2.The next step is for you to build Table 2, below, which represents your actual clinical measure, as follows:

T

Table 2

Inclusion

Exclusion

Compliance

Other Factors

If you do not have medical background to build this table, no worries; plenty of library and free web site resources provide basic facts to assist you. Sites like WebMD provide plenty of guidance for collecting basic disease information for the above table. For examples of valid medical inclusion and exclusion criteria, visit www.ClinicalTrials.gov site and review any posted study.

Hint: if you get stuck, review the slide few slides presented as part of the Unit 3 lecture. Those codes related to diabetes explain how the measure was built for that particular disease. You would approach your project in a similar way, but specific to the patient condition of your choice. 

3.The third step in this exercise is to provide sample medical documentation codes to search the electronic medical records database, in support of building a measure. Your actual measure may include a dozen or more conditions and dozens of corresponding codes. For this exercise, we want to take three conditions from your inclusion or exclusion criteria and search for useful codes. Using exclusion and/or inclusion criteria from Table 2, the knowledge about vocabularies gained in prior informatics courses or via your research of medical vocabularies for this exercise, and your research of CMC for this exercise, build Table 3, as follows. The minimum requirement for this table is three entries, so we are not building a full complete measure – unless you want to, in which case you are free to take this exercise as far as you wish. Links to basic coding systems that you can search will be provided within the learning module for this week.

T  

Table 3

Eligibility Description

Eligibility Code(s)

Compliance Description

Compliance Code(s)

For the description, you will need to identify your patient and/or, his/her symptoms, i.e. adult patient 18 to 75 years old who had at least one visit to ER during the past year. In this case, you would be looking for eligibility ICD-10 codes representing (1) ER visit and (2) diabetes (types 1 or 2). You would also control for the age boundaries, programmatically, by specifying a range in the query, so mention it for your programmers in words, not codes. As an example, let’s assume that we require patients to undergo HbA1c testing twice a year, for compliance. If so, you are at least looking for an office visit CPT code and the HbA1c LOINC code(s) to meet this compliance description. Use ICD-10 and LOINC links you utilized for the vocabulary exercise to find actual codes for Table 3. There may be multiple codes: as long as these are relatively correct codes, this is all we want to practice with. Use a lecture slide describing specific clinical measure guidelines for diabetes as your example.

There is no need to keep looking for all codes exhaustively. However, as a clinical analyst building a measure for a hospital, you would want to spend many hours finding ALL codes that build correct description, compliance, and range. We are just practicing to learn the basics. Again, if you are interested in taking a deeper dive, please feel free to do so.

4.The fourth step in this process is to show justification for your work, in terms of descriptions of the clinical codes you’ve utilized. Build a simple Table 4, as follows:

Table 4

Code Type (vocabulary)

Code Number

Code Description (from catalog)

Justification for Use (in your own words)

Healthcare data analytics is moving in the direction of not only documenting (the EMR phase of informatics development) and processing data (clinical data integration phase of informatics development), but now making sense of the data and linking disparate data that was previously deemed unrelated or impossible to relate. Understanding what data means is important, but how it can be employed in integrative ways to solve complex medical challenges, make new discoveries, and support medical providers is most important to meaningful use of  data analytics. As we know from the lecture and this exercise, clinical measures enable quality controls, care standardization and handoffs, risk management, and predictive care analytics. By understanding ways to employ data to effectively serve patient care needs, students put themselves in the position of power – by not only exhibiting an ability to obtain, move, and process the data – but understand its meaning, application, and meaningful use. 

Finally, Write an introduction describing your working strategy for the complex medical condition (CMC) you employed for building a measure  (100 words, maximum) and a conclusion (150 words minimum, 200 words maximum) describing what you learned from this exercise.