Blog Post

3 top population health questions—answered

March 1, 2018

    Across 2017, hundreds of population health leaders across the country asked our Population Health Advisor experts more than 1,100 questions on care transformation tactics and strategy. Here are answers to last year's frequently asked questions:

    1. How should we determine how many care managers we need?

    We typically see organizations take two approaches to determining staffing ratios for care managers (CMs):

    • By payer: Some providers tie hiring ratios to payer population, regardless of whether those care managers support only patient management for that payer. We recommend the following baseline ratios, then deploying staff based on risk stratified populations (e.g., 3-5% high-risk, 15-20% rising-risk, and the remainder at low-risk):
    • Commercial population to CM: 5,000:1
    • Medicare population to CM: 3,000:1
    • Medicaid population to CM: 1,500:1
    • By risk: Other providers hire staff based on risk stratified population targets. For example, the Michigan Primary Care Transformation Model, a CMS demonstration program aimed at reforming primary care payment models, recommends hiring a moderate risk and complex care manager for each of their practices at 1:5,000 patients each. However, the target caseloads for each CM vary. Moderate risk or chronic disease managers have target caseloads of 500 patients, whereas caseloads for complex care managers are much smaller, at 150 patients.

    We recommend using a hybrid approach. Determining ratios based on risk and payer type is a more nuanced and effective strategy to appropriately staff your care management team.

    To learn more about staffing for care management, download the development guide, "Advancing Ambulatory Care Management."

    Download Now

    2. How can we address ED utilization for patients with complex behavioral health challenges?

    Typically, a patient with a psychiatric emergency waits more than three times as long as a patient with non-psychiatric needs to be seen in the ED. That's bed capacity most EDs don't have—so, no wonder we get this question a lot. We recommend taking a three-pronged approach to right-size ED utilization for patients with behavioral health needs:

    • Early intervention and prevention: Conduct universal mental health screening in primary care to surface undiagnosed needs and connect patients with appropriate services.
    • Point of care intervention: Offer a calming environment and behavioral health services for patients to prevent exacerbation and optimize throughput.
    • Post-utilization education and follow-up: As part of discharge planning, provide education and connect patients with community support.

    3. What is the number of lives recommended to do downside risk contracts?

    While there is no set floor, the Medicare Shared Saving Program and the Next Generation ACO program require a minimum of 5,000 beneficiaries and 10,000 beneficiaries (7,500 for rural ACOs), respectively. While most organizations start with the 5,000 and 10,000 beneficiaries in their initial contracts, the 'more is better’ philosophy certainly stands true when it comes to risk-based contracting. The greater number of lives, the more attractive the arrangement. Here’s why:

    • Smaller sample sizes are more prone to overrun cost targets because of random variation only; the variance declines as the population grows.
    • A larger patient base entails a larger overall cost target and consequently a larger potential bonus payment.
    • Many of the investments in new care delivery systems will be largely fixed investments; spreading those costs over a larger patient base helps to improve per-patient return.

    How one organization is transforming care for the homeless

    Join us on April 4 to learn how University of Vermont Medical Center is transforming housing and health care for the homeless.

    Register Now

    Have a Question?


    Ask our experts a question on any topic in health care by visiting our member portal, AskAdvisory.