Blue Cross and Blue Shield of Kansas City (Blue KC) and Nymbl Study

Study Design

This cost savings study was conducted by the Blue Cross and Blue Shield of Kansas City (Blue KC) Medical Economics team. It used Propensity Score Matching (PSM), a statistical technique used to reduce selection bias in observational studies where a random assignment of treatment (or intervention) is not feasible. The goal is to simulate a randomized control trial by matching units with similar characteristics (propensity scores) across treatment and control groups. 

Specifically, the study used the Optimal method for propensity matching.

Study Parameters

Treatment Group: 1,396 Medicare Advantage members who enrolled in Nymbl in 2023.

Control Group: 1,396 matched out of 27,763 Medicare Advantage members who did not enroll in Nymbl in 2023. To match the treatment group, the control group was limited to claims starting in 2023 allowing for at least 3-months of runout on claims.

Exclusion Criteria:

• Members without 6 full months of coverage before and after program referral. 

• Members whose total allowed amount in the six months following their program referral exceeded the 75th percentile plus three times the value of the interquartile range.

• Members who had $0 in allowed claims post-enrollment or after the earliest active month

Propensity Matching Process

The optimal method uses 1:1 matching without replacement where each treated unit is matched to the control unit with the closest propensity score. Each control unit is used only once, minimizing bias.

Matching Criteria:

• Age

• Gender

• Metro

• Risk Category (1-5)

• Avg Prsp MARA - 6mo

• PCP Visits - 12mo

• Admits - 6mo

• ED - 6 mo

• High Risk Medications - 6mo

• Neuro-Conditions

• Diabetes with Neuropathy

• Incontinence
  
  

After matching, an assessment was conducted as to whether the covariates are balanced between the treatment and control groups. The Balance was calculated to be 0.0794, meaning that the groups are very well balanced and the difference in the outcome can be attributed to the treatment rather than to covariates.

Study Results

The Treatment Group showed a significant reduction in Inpatient (IP) and Emergency Department (ED) costs relative to the Control Group.

• IP costs decreased by $26 per member per month (pmpm)

• ED costs decreased by $9 per member per month (pmpm)

• Total acute care cost savings of $35 per member per month (pmpm)

Implications

Deploying Nymbl saved Blue KC $35 per enrolled member per month (pepm) for a total annual savings of more than $1.1M and a 3.2x ROI.

Nymbl is willing to put fees at-risk to guarantee these outcomes with Medicare Advantage (MA) health plans and at-risk providers.

Follow us on LinkedIn | Learn more at Nymblscience.com