A Report on the Actuarial, Marketing, and Legal Analyses of the CLASS Program. Appendix Nc: Presentation Entitled "The Long-Term Care Policy Simulator Model"

10/14/2011

The Long-Term Care Policy Simulator Model

September 22, 2010

Avalere Health LLC

Presentation Purpose and Agenda

  • The purpose of this presentation is to describe an approach for estimating the premiums for a voluntary, public long-term care insurance program.
  • Agenda
    • Provide brief project background
    • Summarize overall modeling approach
    • Highlight key issues/challenges
      • Adverse selection
      • Enrollment rates
      • Benefit qualification
    • Questions/Discussion

Description of the Long-Term Care Policy Simulator (LTC-PS)

Purpose

  • In 2009, before health reform, The SCAN Foundation funded construction of a model that would estimate average premiums for four different long-term care public insurance reform approaches

Proposals to Model

  • Mandatory or voluntary social insurance programs
  • Either a cash benefit or services benefit
  • Note: None of the proposals would allow underwriting other than age. Users could elect to require attachment to the workforce

Basic Overview

  • The LTC-PS is an Excel-based spreadsheet model
  • It has an inputs page that allows users to vary the key policy options
  • The model then calculates the premiums necessary to have an actuarially-balanced program over 75 years
LTC-PS Input Options
Choice Benefit Description Population Covered Minimum Premium Payment Period Length Of Benefit Elimination Period Cross Subsidies To Low-income Individuals1 Program Costs Funded Through Premiums2
Mandatory Cash Benefit $50 / day Workers and Their Spouses May Participate Zero Years Lifetime Zero Days 150% of the Federal Poverty Level 100%
$75 / day
5 Years
$100 / day 75%
4 Years 100% of the Federal Poverty Level
Voluntary Services Benefit No Cost Sharing All Over Age 18 May Participate3 5 Years 90 Days
3 Years
$500 Deductible. 20% Copayment 1 Year No cross subsidy 50%
Options shaded in yellow are the closest to CLASS legislative specifications but there are a number of key CLASS inputs that were not included in this model
  1. Low income individuals pay no premium. All others pay additional premiums to compensate.
  2. General revenues used to subsidize premiums in the 75% and 50% options
  3. Excludes people who are initially disabled and not working

Key Differences between the LTC-PS and the CLASS Act

  • CLASS prohibits non-working spouses from enrolling, but LTC-PS does not
  • CLASS has a minimum income and work requirement, but LTC-PS simulates that anyone at work regardless of income could enroll
  • CLASS applies the work and income requirement to low-income individuals, but LTC-PS simulates full participation by anyone below the subsidy threshold
  • CLASS has a variable ADL trigger for payment of benefits, but the LTC-PS simulates a trigger of slightly below 2 ADLs
  • CLASS has a minimal $5 premium for students and low-income individuals, but the LTC-PS has a $0 premium for low-income individuals and excludes students
  • CLASS-Medicaid dual beneficiaries retain some of their CLASS payout1, but LTC-PS simulates entire payout going to Medicaid
  • CLASS has level premiums once a person enrolls, but LTC-PS uses inflation-adjusted premiums for all enrollees
  • CLASS has the ability to require payment of premiums by enrollees receiving benefits, but LTC-PS simulates enrollees will either be paying premiums or receiving benefits, not both
  1. Public Health Service Act, §3205(c)(1)(D)(i) and (ii) specify that institutional and certain HCBS Medicaid beneficiaries retain 5 percent and 50 percent of their CLASS payouts.

Model Overview

Model Description

  • We use an incidence and continuance model
    • Track enrollees by age
    • Model incidence and continuance of disability to determine when a person becomes disabled and how long he or she remains disabled

Data Sources

  • Point-in-time surveys for prevalence of disability in the community (Survey of Income and Program Participation, American Community Survey, Current Population Survey) and in nursing homes (National Nursing Home Survey)
  • Longitudinal survey for continuance rates among elderly aged 65+ (National Long Term Care Survey) and actuarial data for continuance rates among disabled aged 18 to 65
  • Data Issues
    • No national, longitudinal data for disability across age spectrum
    • Aggregation of data from multiple surveys
    • No single accepted method to estimate adverse selection

Model Overview

Flow Chart: Total U.S. Population leads to Enrolled Population. Leads to Population Receiving Benefit (includes 1. Disability and 2. Vesting). Leads to Value of Benefit (includes Avg. Services Spending and Cash Daily Amt.). Leads to Program Payments. Leads to Must Be Equal Over Estimating Period. Enrolled Population also leads to Population Not Receiving Benefit. Leads to Premium Payments. Leads to Program Income. Also leads to Must Be Equal Over Estimating Period.

Modeling Enrollment: Population and Program Eligibility

  • We use Social Security estimates of the total population by age through 2085.
  • Eligibility can be extended to:
    • All workers: we estimate attachment to workforce from American Community Survey.
    • All over age 18: we exclude people currently disabled unless they are currently working (regardless of reported income).
      • We estimate 5 to 7 percent of people with 2+ ADL disabilities in the community setting are currently working (approximately 400,000 people).

Modeling Enrollment: Participation

  • We modeled participation using a points system
    • We constructed a points system based on plausible upper and lower bounds for participation.
    • Options that reduce cost, like adding a deductible or elimination period, or reducing benefit amount or length, increase participation
    • Typical enrollment rates for CLASS like program: 12 to 18 percent
    • We age-adjust participation rates
      • Participation at age 50 is same as overall estimate
      • Participation increases at a 2 percent growth rate for individuals aged over 50
      • Participation decreases at a 1 percent growth rate for individuals aged under 50

We plan on refining the participation methodology for ASPE using assumptions about employer adoption and demand elasticity


Participation by Age*

Line Chart 1.

* Assumes 10 percent average enrollment


Modeling Enrollment: Vesting

  • We estimate compliance with the 5 year vesting period using SSDI vesting as an analog
  • We do not model the effect of lapses
    • In our Excel-based model, we only need to know the percent of people in any given year that would be eligible to receive benefits
    • Ineligibility could be related to vesting or lapses

Line Chart 2.

Modeling Disability: Prevalence

  • We estimated prevalence from:
    • 2004 SIPP for community setting
    • 2004 NNHS for institutional setting
  • We collected data on percent of individuals in each setting with:
    • Only 1 ADL
    • 2 or more ADLs
    • 3 or more ADLs
  • We assumed 50 percent of individuals with only 1 ADL would become eligible for the program
    • Any individual in a nursing home with only 1 ADL would be eligible
    • 48 percent of individuals in the community with only 1 ADL would be eligible
  • We also adjust the SIPP data to account for individuals in an assisted-living facility
    • Only the 65+ population
    • We add these people to the ‘institutional’ estimates
    • Shifts approximately 700k people from the community to institution estimates

Modeling Disability: Continuance

  • To estimate continuance, or how long someone remains severely disabled, we used two data sets
    • Over age 65: transition matrices from National Long Term Care Survey1
    • Under age 65: continuance tables from IDEC survey2
  • Non-continuance can be caused by two factors: mortality or improvement in condition
    • Tend to see improvement at younger ages: these individuals are returned to the population eligible to pay premiums
    • Mortality is higher for all ages of disabled individuals compared to non-disabled individuals
    • We required non-continuance to always be at least as high as age-specific mortality from SSA
  1. Stallard, E. and Tee, R.K.W. 1999. “Non-insured Home- and Community- Based Long-Term Care Incidence and Continuance Tables.” Society of Actuaries
  2. Society of Actuaries. 2005. “Experience Studies in Individual Disability.”

Modeling Disability: Incidence

  • Incidence can be computed once we have estimated prevalence and continuance
  • Prevalence T2 = Prevalence T1 + Incidence T2 - Non Continuance T2
  • We constructed a single cohort of individuals at all ages and tracked them for 100 years to develop incidence rates
    • Population as of 2000, according to SSA
    • Used age-specific prevalence and continuance
    • Applied age-specific mortality estimates from SSA to non-disabled population
  • After computing incidence by age, we accounted for an expected decline in prevalence through 2025
    • We modeled a 0.5% decline in age-specific prevalence until 2025, at which point we hold prevalence constant
    • We also hold continuance constant, which results in a decline in age-specific incidence
    • A debatable proposition

Modeling Disability: Adverse Selection

  • We increased incidence of participants in the LTC-PS to account for adverse selection
    • Enrolled population in voluntary program has higher disability than general population
  • Under the extreme scenario, every individual who would develop disability within 5 years would enroll -- this is the “perfect knowledge” scenario
  • For the LTC-PS, we assumed enrollment in the initial years was weighted 75% to perfect knowledge scenario
    • This declines to 25% weighting within 10 years
    • Mimics pent up initial demand with continuing adverse selection
  • Impact of adverse selection much higher for low-enrollment options

2010 Incidence Curve Adjusted for Adverse Selection*

Line Chart 3.

* Assumes 10 percent average enrollment


Modeling Costs: Medicaid Interactions

  • We model the impact on Medicaid based on an assumption about participation by people who would eventually become Medicaid enrollees and the low-income subsidy.
  • We model of Medicaid baseline using data from SIPP and NNHS, supplemented by information published by Brian Burwell and Josh Wiener.
  • Even with a low-income subsidy, some future Medicaid beneficiaries would still be unlikely to enroll
    • Not all future Medicaid beneficiaries are currently below the Federal Poverty Limit (FPL)
  • The table below shows our estimated participation rates by people who would eventually become Medicaid beneficiaries by the different low-income subsidy levels.
  • We apply these participation rates to our Medicaid baseline to develop estimates of Medicaid savings.
  Premiums     None     100% FPL     150% FPL  
<$50 25% 50% 75%
$50-80 20% 45% 70%
$81-100 15% 40% 65%
$101-120 10% 35% 60%
$121-150 5% 30% 55%
>$150 0% 25% 50%

Limitations of the Model

  • Disability estimates: The data sets used have somewhat different ways of measuring disability
  • Participation rates: Estimating participation for such a novel program is complex. Participation will be driven by many factors, premiums being a large one. Our points system is an approximation.
  • Adverse selection: Estimating adverse selection is complex. We followed a method used by actuaries and the CBO. However, there is considerable debate among researchers.
  • Impact to Federal budget: Interactions with the federal budget, specifically around the tax implications of the program, are beyond the current scope of the model.

Questions and Answers

View full report

Preview
Download

"index.pdf" (pdf, 350.5Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appA.pdf" (pdf, 150.08Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appB.pdf" (pdf, 548.14Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appC.pdf" (pdf, 150.56Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appD.pdf" (pdf, 318.45Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appE.pdf" (pdf, 319.19Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appF.pdf" (pdf, 155.03Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appG.pdf" (pdf, 181.05Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appGa.pdf" (pdf, 151.02Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appGb.pdf" (pdf, 123.12Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appH.pdf" (pdf, 256.35Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appI.pdf" (pdf, 358.34Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appJ.pdf" (pdf, 1.21Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appJa.pdf" (pdf, 396.06Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appJb.pdf" (pdf, 313.17Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appJc.pdf" (pdf, 252.71Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appJd.pdf" (pdf, 261.97Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appK.pdf" (pdf, 186.22Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appL.pdf" (pdf, 788.88Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appM.pdf" (pdf, 249.45Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appN.pdf" (pdf, 7.72Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appNa.pdf" (pdf, 208.04Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appNb.pdf" (pdf, 6.96Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appNc.pdf" (pdf, 622.02Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appNd.pdf" (pdf, 211.36Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appO.pdf" (pdf, 2.01Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appP.pdf" (pdf, 9.15Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appPa.pdf" (pdf, 233.71Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appPb.pdf" (pdf, 253Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appPc.pdf" (pdf, 546.97Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appPd.pdf" (pdf, 505.4Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appPe.pdf" (pdf, 462.21Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appQ.pdf" (pdf, 285.57Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®

View full report

Preview
Download

"appR.pdf" (pdf, 797.76Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®