The Importance of Medicaid Data To Improve Diversity in Clinical Trials

By Blair Mohney & Michael Underwood
Data Analysis By Anna Wang

June 2, 2023

Paralleling many conversations across the current socio-political landscape, life sciences have been striving to fix the lack of diversity in clinical trials. This has taken a few different angles ranging from access to representative results.

While these conversations have been ongoing, two recent updates prompted more urgent action: Medicaid became responsible for the costs of clinical trials for Medicaid patients,1 and the FDA issued guidance requiring clinical trial diversity plans.2

Using real world evidence in the design of a clinical trial allows life science organizations to ensure they meet diversity requirements in advance of beginning the trial, thus reducing the need for subsequent iterations and using resources most efficiently.

Often, conversations around diverse real world evidence data focus on social determinants of health (SDoH), which are critical data points. However, it is also important to start with a diverse data set. Many CareJourney members utilize the 100% CMS Medicaid data set for this reason.

Overall Medicaid enrollment is public, yet for clinical trials it’s important to get as granular as possible, for example:

  • Distilling granular geographic levels, like county-level trends
  • Profiling Medicaid patient demographics (income, age, race, etc)
  • Highlighting indication prevalence
  • Assessing linked and longitudinal data to understand more accurate outcomes
  • Combining Medicare Advantage, Medicaid, and Medicare FFS cohorts of patients

As an example of what we can learn from the Medicaid data set, see the profile below of Jefferson County, KY, outside the Louisville area, with high overall Medicaid enrollment.

At a high level, we can understand the proportion of the populations across Medicaid FFS, as well as Medicaid Managed Care Organizations. We can also see the breakdown across key Medicaid cohorts.

Examining Medicaid and CHIP enrollment lends insights into a vast population (over 93M lives in January 2023) including children, parents, people who are pregnant, elderly people with certain incomes, and people with disabilities. Medicaid patients can help diversify the recruitment pool of study participants thus increasing the efficacy of clinical trials and potential impact of new and emerging therapies.

2019 2020
MCO Member Count FFS Member Count MCO Member Count FFS Member Count
TOTAL 240,831 (88%) 33,929 (12%) 241,149 (84%) 47,057 (16%)
Temporary Assistance for Needy Families (TANF) 200,562 (73%) 13,217 (5%) 189,190 (66%) 25,295 (9%)
Aged, Blind, and Disabled (ABD) 27,816 (10%) 14,937 (5%) 27,703 (10%) 15,213 (5%)
Maternity 3,591 (1%) 757 (0%) 14,491 (5%) 1,035 (0%)
Child 8,587 (3%) 286 (0%) 9,591 (3%) 372 (0%)
Expansion 43 (0%) 18 (0%) 39 (0%) 21 (0%)
Home and Community Based Services (HCBS) 14 (0%) 24 (0%) 13 (0%) 38 (0%)
Institutional 198 (0%) 4,078 (1%) 122 (0%) 4,184 (1%)
Missing 20 (0%) 598 (0%) 899 (0%)
Other 14 (0%)
Programs of All-Inclusive Care for the Elderly (PACE)

We are able to profile the population even further to understand the age spans, gender breakouts, and income distributions.

TANF ABD Maternity Child
MCO FFS MCO FFS MCO FFS MCO FFS
Member Count 189,190 25,295 27,703 15,213 14,491 1,035 9,591 372
Age Between
15-18
8.08% 3.14% 5.00% 2.14% 0.69% 1.93% 18.68% 14.52%
Age Between
19-20
3.78% 3.22% 2.03% 1.05% 1.27% 6.09% 2.07%
Age Between
21-44
39.79% 53.69% 20.58% 15.53% 12.90% 83.19% 0.00%
Age Between
45-64
16.07% 26.86% 37.22% 37.16% 0.00% 0.00% 0.00%
Age Between
65-74
0.22% 1.47% 14.02% 26.52% 0.00% 0.00% 0.00% 0.00%
Age Between
75-84
0.00% 0.29% 4.39% 10.19% 0.00% 0.00% 0.00% 0.00%
Age Between
85-125
0.00% 0.15% .41% 3.67% 0.00% 0.00% 0.00% 0.00%
Female 54.11% 49.29% 50.47% 57.02% 57.32% 95.94% 49.66% >48.66%
Male 45.88% 50.71% 49.53% 42.98% 42.68% 4.06% 50.34% 51.34%
State-defined family income is from 0 to 100% of the federal poverty level (FPL) 74.57% 35.92% .22% 51.77% 78.86% 59.23% 9.01% 4.03%
State-defined family income is from 101 to 133% of the FPL 13.56% 5.58% 1.01% 22.77% 11.11% 2.51% 3.93%
State-defined family income is from 134 to 150% of the FPL 3.23% 1.75% 0.62% 3.53% 2.25%
State-defined family income is from 151 to 200% of the FPL 3.66% 1.45% 0.39% 4.06% 1.55% 23.63% 17.47%
Missing Income Data 3.47% 54.69% 89.72% 24.42% 1.06% 35.27% 1.29%

Note: Percentages will not equal 100% as rows of data have been removed for brevity.

From this example, we can observe that nearly 40% of the Medicaid enrollment in this area are members under Medicaid Managed Care Organizations between the age of 21-44 and specifically in the Temporary Assistance for Needy Families program. It is important to understand the patient population in a given market when selecting sites for clinical trials to understand if your target population will be properly addressed. New drug applications require results to be stratified by factors like age group, gender, and race. Profiling the population and diversity in a market can help launch clinical trials efficiently with a greater likelihood of success, increasing the speed to market for new therapies.

Building on these demographic profiles, life science organizations can further profile the patient population across other SDoH factors, such as Distressed Communities Indicators, Transportation Access and Food Access, as well as disease states to ensure the Medicaid population is proportionally represented in their trials.

How CareJourney Can Help

Curious about a specific indication? With access to 100% of the full Medicare Fee For Service data asset, 100% access to the full Medicaid data asset, 100% access to the full Medicare Advantage data asset data, as well as rich commercial data, CareJourney regularly supports life sciences organizations with:

  • Understanding Burden of Illness
  • Informing Clinical Trial Site Selection
  • Evaluating Effectiveness and Safety
  • Supporting Commercialization
  • Assessing Markets for Education

Gain critical insights across key indications, populations, geographies, and therapeutic areas.