MMHA 6520 Week 4 Denominator Specification Example

Who Counts as a Diabetic Member? Three Denominator Definitions for Riverline Health Plan's Care Coordinator Staffing

Student Name

Walden University

MMHA 6520: Health Informatics and Population Health Analytics

Instructor Name

Month Day, Year

Who Counts as a Diabetic Member? Three Denominator Definitions for Riverline Health Plan's Care Coordinator Staffing

The Decision Behind the Definition

Riverline, the invented Medicaid plan used throughout this course, intends to hire care coordinators for adults with diabetes next year and must decide how many. The answer depends on how many members have diabetes, and that depends on how the population is defined. A definition chosen for a quality measure, a cost forecast or a staffing plan can produce very different counts from the same data. This specification compares three candidate definitions, explains what each includes and misses, and adopts the one that fits the staffing decision. Population health measurement begins with a clear definition of the group whose outcomes are being described (Kindig & Stoddart, 2003).

Three Candidate Definitions

Table 1 sets the three definitions side by side.

Table 1

Candidate Denominator Definitions

Definition A: Broad

The broad definition counts every adult enrolled for even one day with a single claim carrying a diabetes diagnosis. It produces the largest number and misses almost no one who was billed for diabetes care. But a single claim can carry a diagnosis entered to rule out a condition rather than confirm it, so the definition includes some members who do not have diabetes. It also counts members who were enrolled for a month and left, for whom no coordinator could have done anything. It is useful for estimating the total number of people with diabetes who passed through the plan but overstates the number who could be coordinated.

Definition B: Continuous Enrollment and Two Claims

The second definition requires continuous enrollment for the full year and diabetes diagnoses on two separate claims. This follows the general approach of standard health plan quality measures, which use enrollment and diagnosis rules so that a plan is judged only on members it had a fair chance to serve (National Committee for Quality Assurance, 2024). It removes most rule-out diagnoses and short-term members. It also removes many real diabetic members. Medicaid enrollment churns, and a sizable share of members have gaps in coverage during a year (Medicaid and CHIP Payment and Access Commission, 2021); every one of them drops out. Newly diagnosed members with only one claim so far also drop out, though they may need coordination most.

Definition C: Assigned to a Network Practice

The third definition takes Definition B's members and keeps only those assigned to a primary care practice in Riverline's network. It produces the smallest count. It excludes members who see clinicians outside the network, use emergency departments for care or have no regular source of primary care. Those members matter, but they are not the population a practice-based care coordinator can reach through the model Riverline plans to use, in which coordinators are embedded in or linked to specific practices.

How Churn Shapes the Counts

Enrollment churn affects Definitions B and C more than A. A member who loses coverage for two months because a renewal form was not returned, then regains it, is counted under A but not under B or C. For quality measurement, that exclusion is reasonable. For staffing, it understates the work: coordinators will serve members whose enrollment is interrupted, and those members may need more help, not less, because gaps in coverage disrupt care (Sommers et al., 2016). Any staffing estimate built on B or C should therefore be adjusted upward by the share of members with interrupted enrollment who return within the year.

The Specification Adopted

The specification adopts Definition C, adjusted for churn. The decision is how many care coordinators the network practices need, and coordinators will work through those practices, so the population should be members a practice-based coordinator could reach. Definition C reflects that boundary. The churn adjustment adds back members assigned to network practices who had a gap in enrollment but returned within the year, using last year's enrollment files to estimate their number. The final count, divided by a caseload per coordinator set by the care management director, gives the staffing estimate.

This choice has costs. It excludes members outside network practices, who may have the greatest unmet need, and it depends on the accuracy of the practice assignment file, which is updated monthly but sometimes lags behind members' actual choices. The plan should report the counts under all three definitions alongside the adopted one, so that leaders can see how many members fall outside the coordinator program and can decide separately how to reach them, for example through community health workers or telephone outreach.

Data Steps

The analyst would: pull twelve monthly enrollment files and flag continuous enrollment and returns after a gap; pull claims for the year with diabetes diagnosis codes and count distinct claim dates per member; join the practice assignment file as of the last month of the year; apply each definition's rules; and report counts by definition and by practice. Every step uses data Riverline already holds. The definitions, code lists and enrollment rules should be written down and versioned, so next year's count can be compared with this year's.

References

Kindig, D., & Stoddart, G. (2003). What is population health? American Journal of Public Health, 93(3), 380-383. https://doi.org/10.2105/AJPH.93.3.380

Medicaid and CHIP Payment and Access Commission. (2021). An updated look at rates of churn and continuity in Medicaid enrollment [Issue brief]. Medicaid and CHIP Payment and Access Commission.

National Committee for Quality Assurance. (2024). HEDIS measurement year 2024: Volume 2, technical specifications for health plans. National Committee for Quality Assurance.

Sommers, B. D., Gourevitch, R., Maylone, B., Blendon, R. J., & Epstein, A. M. (2016). Insurance churning rates for low-income adults under health reform: Lower than expected but still harmful for many. Health Affairs, 35(10), 1816-1824. https://doi.org/10.1377/hlthaff.2016.0455