Consumer Driven Health Plans:
Early evidence of take-up, cost and
utilization and HSA policy implications.
Stephen T Parente, Roger Feldman, Jon B Christianson
Sponsored by the Robert Wood Johnson Foundation’s
Health Care Financing & Organization Initiative (HCFO)
February, 2005
Presentation Overview
†
†
†
†
†
†
Employer-based Analysis Overview
Policy Questions
National CDHP Take-up
Cost & Utilization Comparisons Over Time
National HSA Simulation
Policy Implications
1
Employer-based Analysis Overview
†
†
†
†
Analysis started in 2002 with six employers
Combined population drawn from 50 states
Total covered lives represented: ~250,000
Collect primarily employer HR data and
insurance claims data for all plans.
† New HCFO grant will create a study panel with
six total years of CDHP experience 2001-2006.
Policy Questions
† Do CDHPs (in the form of HRAs) have national appeal?
† What are the longer-run cost & use consequences of
CDHPs?
„ Where do they save money?
„ Where are they more expensive?
„ What is the impact on utilization of key services?
† Do HSAs have potential national appeal?
† Are HSAs a viable approach to addressing the problem
of the uninsured?
General Caveat: We are just approaching the half-way point of
our research.
2
Nearly National Appeal:
States where the study employers’ 1st year CDHP take-up was >5%
Take-up
>5%
0.1 - 5%
0%
Take-up Summary from the Study
Employers
† All states have take-up above 5% with the
exception of New York, New England States,
Indiana, California and Arizona.
† Differences may by driven by:
„ Dominance of managed care in CA, AZ
„ Insurer/provider choices in Northeast
„ Not enough data from only six employers
† Grand experiment in 2005: FEHBP
3
What is the impact of CDHPs on
cost & use?
† Study Design:
„ Reported in 2004, August, Health Services Research.
„ Look at CDHP/PPO/POS cohorts within one large
employer for employees over time to see ‘longer run’
impact of CDHP in 2001 & 2002.
„ Control for several factors to ADJUST cost & use
estimates:
† Health status/illness burden/health shocks
(cancer, catastrophic accident)
† Income
† Family size and dependents
† Age, gender
What was the ADJUSTED impact on
provider and patient payment?
2000
Mean
2001
Mean
2002
Mean
CDHP Cohort N=531
Total Expenditure
Employer Expenditure
Employee Expenditure
$ 4,396.22
$ 4,005.28
$ 416.51
$ 6,154.36
$ 5,903.61
$ 634.38
$ 8,149.26
$ 7,807.39
$ 792.01
POS Cohort N=1,551
Total Expenditure
Employer Expenditure
Employee Expenditure
$ 5,284.53
$ 4,895.75
$ 394.70
$ 6,773.62
$ 6,227.81
$ 549.32
$ 7,197.50
$ 6,428.83
$ 702.49
PPO Cohort N=1,554
Total Expenditure
Employer Expenditure
Employee Expenditure
$ 5,228.42
$ 4,688.28
$ 511.84
$ 7,050.59
$ 6,349.99
$ 657.16
$ 8,377.78
$ 7,330.94
$ 881.47
Health Plan Cohorts
One employer’s results reported in: ST Parente, R Feldman, JB Christianson. Evaluation of the
Effect of a Consumer Driven Health Plan on Medical Care Expenditures and Utilization,
Health Services Research, Vol. 39, No. 4, Part II, pp. 1189-1209, August 2004.
NOTE: These are results from a restricted continuously enrolled sample of 50% to 60% of the total employee population and are not a reflection of the
plans’ full PMPM expenditures. Also note: 1) Patient expenditures from the Personal Care Account (PCA) are included in the employer payment
category. 2) Consumer payment reflects deductibles, copayments, and coinsurance expenses.
4
What was the ADJUSTED impact on provider
& patient payment by different services?
Health Plan Cohorts
Year 2000
Year 2001
Mean
a Mean
Year 2002
Mean
CDHP Cohort N=531
Hospital Expenditure
Physician Expenditure
Pharmacy Expenditure
$
$
$
1,369.97
2,093.70
935.29
$ 1,999.25
$ 2,935.84
$ 1,103.72
$ 3,468.53
$ 3,510.83
$ 1,341.78
POS Cohort N=1,551
Hospital Expenditure
Physician Expenditure
Pharmacy Expenditure
$
$
$
1,842.80
2,381.08
1,107.64
$ 1,796.37
$ 2,959.90
$ 1,498.54
$ 1,956.83
$ 3,088.22
$ 1,640.25
PPO Cohort N=1,554
Hospital Expenditure
Physician Expenditure
Pharmacy Expenditure
$
$
$
1,779.06
2,245.22
1,007.95
$ 2,049.76
$ 2,834.32
$ 1,484.91
$ 2,367.17
$ 3,294.47
$ 1,789.26
One employer’s results reported in: ST Parente, R Feldman, JB Christianson. Evaluation of the
Effect of a Consumer Driven Health Plan on Medical Care Expenditures and Utilization,
Health Services Research, Vol. 39, No. 4, Part II, pp. 1189-1209, August 2004.
NOTE: These are results from a restricted continuously enrolled sample of 50% to 60% of the total employee
population and are not a reflection of the plans’ full PMPM expenditures.
Was ADJUSTED service use different for
CDHPs?
Health Plan Cohorts
2000
Mean
2001
Mean
2002
Mean
CDHP Cohort N=531
Hospital Admission Rate
Physician Visits
Prescriptions Filled
0.05
5.74
18.89
0.10
7.49
22.23
0.16
7.15
25.25
POS Cohort N=1,551
Hospital Admission Rate
Physician Visits
Prescriptions Filled
0.07
6.75
22.23
0.06
7.56
22.59
0.09
7.29
30.89
PPO Cohort N=1,554
Hospital Admission Rate
Physician Visits
Prescriptions Filled
0.07
5.78
20.63
0.07
6.54
23.79
0.11
6.95
24.50
One employer’s results reported in: ST Parente, R Feldman, JB Christianson. Evaluation of the
Effect of a Consumer Driven Health Plan on Medical Care Expenditures and Utilization,
Health Services Research, Vol. 39, No. 4, Part II, pp. 1189-1209, August 2004.
NOTE: These are results from a restricted continuously enrolled sample of 50% to 60% of the total employee
population and are not a reflection of the plans’ full admissions and prescription drug experience.
5
Results Summary
† CDHP plan did not have the lowest cost and utilization
across all plans.
† CDHP best (lowest) cost result was for pharmacy.
† CDHP worse (highest) cost result was for hospital
admissions – partially explained by pent-up demand for
elective procedures & provider pricing differences across
years.
† Utilization results have no dramatic differences across
plan types for pharmacy and physician services. Obvious
access to care problem not apparent.
† CDHP hospital admissions dramatically higher by 2nd
year.
Using HRA Results to Explore HSA
Policy Questions
† What is the expected take-up rate of HSAs in the
individual market?
† What is the likely impact of the Administration’s
proposed HSA subsidies?
„ Take-up rate of HSAs with subsidies
„ Reduction in the number of uninsured
„ Cost of the subsidy
† What is the impact of other possible subsidy
designs?
6
Analysis Design
Data Sources
Model
Estimation
MEPS
CDHPs
eHealthinsurance
Estimate plan offerings
using insurance data
Merge employer
data
Estimate hedonic
premium regression
Estimate plan
choice
regression
Assign plan choices
to full MEPS sample
Choice set
Assignment/ Use parameter estimates
Prediction
to predict plan choice
Define HSA
plan design &
premium
probabilities for MEPS
Policy
Simulation
Calibrate take-up rates
Simulate impact of
proposed policies
Status Quo* Impact of MMA 2003
Preliminary Results
Plan Choice
INDIVIDUAL
MARKET
EMPLOYER
INSURANCE
OFFERED
MARKET
Status Quo
Population %
Status Quo
Project Pop.
HSA-Full Price
PPO_High $$
PPO_Low $$
PPO_Medium $$
Uninsured
16%
23%
1%
6%
54%
9,877,798
14,344,204
819,464
3,880,982
34,313,001
HMO
HRA
HSA-Shared Prem
HSA-Full Price
PPO_High $$
PPO_Low $$
PPO_Medium $$
Turned Down
30%
2%
1%
1%
7%
2%
43%
14%
25,564,398
1,757,424
507,307
597,724
6,194,696
1,598,877
36,252,356
12,246,802
NOTE: Population is 18-64, non public insurance
* Status Quo population is defined as 2005 population facing no tax subsidies.
7
Diminishing Subsidy Returns
16,000,000
14,005,598
Previously Uninsured
14,000,000
Sim #2
12,000,000
Sim #3
10,841,110
10,000,000
8,000,000
6,000,000
Sim #1
4,191,441
4,000,000
2,000,000
0
0
$0
$10,000,0 $20,000,0 $30,000,0 $40,000,0 $50,000,0 $60,000,0 $70,000,0 $80,000,0
00,000
00,000
00,000
00,000
00,000
00,000
00,000
00,000
Subsidy Cost
HSA Simulation Implications
† Without any subsidies, the 2003 MMA HSAs could have a
take-up of ~10 million.
† Hypothetical tax subsidies for HSAs could increase
coverage among the uninsured from 4 to 14 million.
† Offering a free premium HSA to the non-working
population covered by public programs reduces the
uninsured, but less efficiently than income targeted
subsidies.
8
Summary
† For study employers, almost all states had some CDHP take-up
– many with 5% or more take-up in the first year offered.
† Cost results are mixed. Lower costs initially – but rapid rise in
expenditures, adjusting for case-mix and demographics.
† Utilization results are not extraordinarily different in pharmacy
and physician services, but significantly higher for admissions.
† HSA take-up, based on HRA experience and actual 2005 HSA
premiums, could be quite substantial.
† Administration proposals to use HSAs as a mechanism to
substantially reduce the number of uninsured may be viable
depending on the level of subsidy provided.
† All future results will be VERY dependent on benefit design:
premiums, account/deductible gap, coinsurance.
9