MMHA 6520 Week 8 Care Gap Analysis Example

Two Emergency Visits and No Primary Care: A Care Gap Analysis of Riverline's Assigned Adult Members

Student Name

Walden University

MMHA 6520: Health Informatics and Population Health Analytics

Instructor Name

Month Day, Year

Two Emergency Visits and No Primary Care: A Care Gap Analysis of Riverline's Assigned Adult Members

Defining the Gap

This analysis defines a care gap among Riverline's adult members as follows: adults aged 18 to 64, attributed to one of the plan's network primary care practices, enrolled for the full calendar year, with two or more emergency department visits and no primary care visit during the year. Emergency and primary care visits are identified from claims; practice assignment comes from the plan's monthly assignment file, using the assignment in effect at year end. Members who were enrolled for the full year are used so that every member counted had a full year in which to see their assigned practice. The definition describes members who are using the emergency department while their assigned primary care relationship exists only on paper. Because population health concerns how outcomes are distributed within a group, not only their average (Kindig & Stoddart, 2003), the analysis looks at where and among whom the gap concentrates. All figures below are illustrative.

Where Members in the Gap Live

Table 1 shows the distribution of members in the gap across the plan's three counties.

Table 1

Members in the Care Gap by County (Illustrative)

How Long Members in the Gap Have Been Enrolled

Table 2 shows members in the gap by length of continuous enrollment with Riverline as of the start of the year.

Table 2

Members in the Care Gap by Prior Enrollment (Illustrative)

The Pattern

Two patterns stand out. First, the gap is concentrated in the urban county, where the share of assigned adults in the gap is nearly twice that of the suburban county. The urban county has the most network practices, so distance to a practice does not explain the difference. Second, and more surprising, most members in the gap are not newcomers. About a third had been enrolled for less than a year, which might suggest they had not yet established care, but the larger share had been enrolled for more than a year, and a third for more than three. These members have been assigned to a practice for years without visiting it.

Social needs data are too sparse to explain the pattern. Z codes appear on very few of these members' claims, consistent with national evidence that such codes are rarely recorded (Truong et al., 2020), and fewer than a tenth of members in the gap have a care manager assessment. The analysis therefore cannot say whether housing instability, transportation or other barriers drive the pattern; it can only say where the members are and how long they have been with the plan.

Possible Interventions

Telephone outreach plan-wide. Care coordinators would call every member in the gap, offer help scheduling a primary care visit and ask about barriers. This is inexpensive per member and reaches all three counties, but telephone outreach to Medicaid members often fails to connect, because phone numbers change frequently.

Practice-based outreach. Riverline would send each network practice a list of its assigned members in the gap and pay a small incentive for each completed visit. This uses the relationship members are supposed to have, but the members in the gap have not used that relationship, sometimes for years, so a call from the practice may be no more successful than one from the plan.

Community health workers in the urban county. Riverline would place community health workers, who often share members' neighborhoods and backgrounds, to find members in person, learn what keeps them from primary care and help them get there. This costs more per member and covers fewer people, but it targets the county where the gap is concentrated and can discover the reasons the data cannot.

Recommendation

The analysis supports community health workers in the urban county, where most of the gap lies, and telephone outreach elsewhere. The urban concentration means a focused investment reaches the largest number of members, and the long enrollment of many members in the gap suggests that the barriers are persistent and will not be solved by a reminder call. Community health workers can also record what they learn in care manager assessments, which would begin to fill the gap in social needs data. The practice incentive is worth testing with one or two high-volume urban practices, but it should not be the main strategy.

The recommendation also has a sequencing logic. Community health workers take time to recruit and train, so telephone outreach should begin in all three counties at once, including the urban county, while the community health worker program is built. Members in the urban county whom telephone outreach reaches and connects to primary care can be removed from the community health workers' list, which lets those workers concentrate on members who could not be reached by phone. After six months, the plan can compare the share of members in the gap who completed a primary care visit under each approach, using the comparison design described below, and decide whether to extend community health workers to the other counties.

Any program should be evaluated with a comparison group. Members selected because they had two or more emergency visits this year will tend to have fewer next year even without help, because high use in one year often falls back toward the average (Barnett et al., 2005). A program that compares members' use before and after outreach, without a comparison group, will overstate its effect.

Limits

The analysis relies on claims, so it misses primary care visits that were not billed to Riverline, such as visits to free clinics, and emergency visits to hospitals that bill late. It uses year-end practice assignment, which may differ from the assignment a member had during the year. And it describes the gap without explaining it. The community health worker program, if adopted, would be the first source of evidence about why these members do not use the primary care they have been assigned.

References

Barnett, A. G., van der Pols, J. C., & Dobson, A. J. (2005). Regression to the mean: What it is and how to deal with it. International Journal of Epidemiology, 34(1), 215-220. https://doi.org/10.1093/ije/dyh299

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

Truong, H. P., Luke, A. A., Hammond, G., Wadhera, R. K., Reidhead, M., & Joynt Maddox, K. E. (2020). Utilization of social determinants of health ICD-10 Z-codes among hospitalized patients in the United States, 2016-2017. Medical Care, 58(12), 1037-1043. https://doi.org/10.1097/MLR.0000000000001418