Part Two

    Post Thumbnail

    Model Assumptions

    In developing this model, several key assumptions were made to estimate the potential cost savings of implementing telehealth solutions across the U.S.

    1. The cost savings observed in Houston’s ETHAN system would be proportionally similar if implemented in other states.
    2. The compounded annual health inflation rate is a reasonable measure for forecasting cost savings from previous years up to 2024.
    3. The 2016 primary care-related Emergency Department (PCRED) costs for Texas—representing the average cost of seeing a primary care provider in urban hospitals—were scaled to other states using state-specific real price parities (RPP).
    4. The SEDD database, which includes emergency department visits from 31 states, provides a representative sample, though it excludes data from hospitals that do not report.

    Impact of Assumptions on Results

    While these assumptions provide a solid foundation, several factors could influence the accuracy of the results:

    1. The cost savings model is based on urban data from a single metropolitan area, which may not fully capture the challenges of rural regions, such as longer travel distances and higher transportation costs.
    2. The compounded health inflation rate may not accurately reflect future price changes in key areas like transportation and administrative expenses.
    3. Forecasting future state costs based on historical health inflation trends might not fully account for shifts in real price parity (RPP) across states.
    4. The SEDD database underreports total emergency department state visits, resulting in conservative savings estimates.

    Data Validation

    To ensure accuracy, the model’s predictions were validated against independent datasets. In Nevada, predicted emergency department visits were cross-referenced with the Center for Health Information Analysis (CHIA) database:

    1. CHIA recorded 999,579 visits.
    2. The model, using SEDD data, predicted 985,809 visits—a slight underestimation.
    3. Over the 2017-2020 period, SEDD data was, on average, 6,000 visits higher than CHIA data, reinforcing the slight underestimation. [1]

    Sensitivity Analysis

    A sensitivity analysis revealed that fluctuations in health inflation and real price levels significantly impact cost savings. Higher health inflation and rising real prices increases total cost savings for both patients and providers.

    Scalability Considerations

    The model can be adapted to other states and even international settings, provided that certain data inputs—such as total emergency department visits, PCRED costs, and telehealth cost savings per visit—are available. Supporting data, including real price parity (RPP) and health inflation rates, is typically accessible for these adaptations.

    Model Results

    Top-Performing States by Cost Savings Percentage

    The two states that achieved the highest percentage cost savings were Maine and Arizona, with over 10% savings relative to their annual health and hospital expenditures:

    1. Maine
    2. Total Savings: $394,989,745.66
    3. Percentage of Annual Health & Hospital Spend: 16.22%
      4.
    4. Arizona
    5. Total Savings: $86,009,470.27
    6. Percentage of Annual Health & Hospital Spend: 14.73%
      7.

    Key Findings on Cost Savings Distribution

    The model highlighted a strong correlation between health and hospital spend per capita and the potential benefits of a telehealth system:

  • States with per capita health and hospital spending below $700 experienced the highest cost savings. (See Figure 1)

    • Figure 1

  • States spending more than $1,000 per capita generally had lower cost savings, with the exceptions of Indiana and Missouri. (See Figure 2)

    • Figure 2

    Graphic: Deepai.org