Overview of Catamount Health

 

 

Kenneth E. Thorpe

 

 

February 23, 2006

 

 

Brief Overview

 

The Catamount health proposal represents a major restructuring of Vermont’s health care system. The key elements of Catamount health include:

 

  • New efforts to enroll the 22,500 adults and children currently eligible for VHAP, Dr. Dynasaur and Medicaid but remain uninsured.
  • Creation of a modern statewide integrated delivery system designed to provide better quality care at better value to all Vermonters with a chronic illness
  • A focus on reducing the growth in private health insurance premiums—both through the use of chronic care management, but also through several of the patient safety, error reduction, and administrative streamlining that the Senate Health Committee is considering.
  • Catamount health (CH) enrollees would receive a new private health insurance plan. CH would provide enrollees a comprehensive private insurance package similar to the point-of-service plan provided to Vermont state employees today.  Under Catamount health, the state would contract with a private health plan to administer program benefits—precisely the same approach the state currently uses to select a private plan to administer health benefits for state employees.  Use of a single private insurance vendor is designed to avoid the substantial problems incurred in the Healthy New York plan that contracted with several health plans. The actual rate increases implemented in NY have been extremely volatile by county and by health plan—reflecting the fact that numerous health plans have divided up a small number of covered lives. In Healthy New York, the rise in premiums from 2004 to 2005 ranged from -1.0% to 62.0%.

 

In addition to avoiding the pitfalls found in the New York approach of using multiple health plans to administer CH, the plan also provides a new focus on preventive care by assuring all new enrollees in the program receive a health risk appraisal to fully understand the medical needs of the population, and working with primary care physicians and patients, develop a “best clinical practice” care plan individualized for the patient.

  • Applying the chronic care model to the Medicaid population
  • Substantial discounts in CH premiums to assure a broad mix of enrollees
  • Use of reinsurance to protect the state against the high costs of newly enrolled CH enrollees. The state employees plan also purchases reinsurance.

 

Estimating Assumptions

 

Estimates of the number of newly VHAP and CH enrollees depend on two key assumptions:

 

  • The number of currently uninsured eligible for each program
  • The number of uninsured enrolling in the program

 

For the first, I use a three year pooled sample of Vermont from the Current Population Survey—similar to the data presented by the Census Bureau on their web site. The pooling allows more precision with respect to the overall number of uninsured (About 65,000 in 2004). Next, I sort households reported in the Census into both Medicaid standard filing units (groups of individuals within the household that Medicaid uses to determine income eligibility) and a similar concept of health insurance units. These are both different than what the Census reports, which are generally based on households (which could include grandmom, mom, and daughter and their collective income and insurance status). I use both Medicaid standard filing units for the VHAP and Medicaid eligibles and the private insurance units (i.e. the collection of individuals in a household that would be included in a private insurance policy in the state) for Catamount health.

 

With respect to program participation, I use the consensus of published estimates of “price sensitivity” pulled together by the Robert Wood Johnson Foundation, and updated by the Congressional Budget Office. The consensus represents the empirical results over the past 20 years of program enrollment, by income, and reduction in premiums published in the health services research literature. The paper may be found at  http://www.rwjf.org/research/researchdetail.jsp?id=1058&ia=132

 

I have updated this view using subsequent work from the Congressional Budget Office that they use to estimate program participation in new programs like Catamount health. To my knowledge, these are best practice approaches.

 

Specifics

 

We start with an enrollment assumption that every 10 percent reduction in the cost of insurance results in a 5 percent rise in uninsured enrolling in the program. This is an average and varies by income (with lower enrollment) and the size of the subsidy.  This overall result is modified by two additional factors—the share of the subsidized premium in the household income, and the extent of the subsidy. The specific formula is:

 

Newly insured = ((1- (subsidized premium as a share of income) squared) * percent discount * 0.75

 

So for example:  we would assume that 85% of individuals making $25,000 per year that could purchase health insurance today for $5,000, but receives an 85% subsidy through CH  (and thus pays $750 per year) would enroll

 

So for example:

 

 

Income

% Subsidy

Premium as a share of income

Enrollment as percent of eligible uninsured

$25,000

 

$50,000

70%

 

70%

6%

 

2.7%

46.4%

 

49.7%

 

 

How long does it take to get to full enrollment?   Based on experience with other plans and the SCHIP, it generally takes about three years to hit the ultimate enrollment rate. So I have assumed in the estimates that a third of the ultimate enrollees enroll in year 1, two thirds by year 2 and all by the third year.

 

I trend the number of eligibles each year by a population trend factor of 0.5% to allow for population growth.

 

I also assume that in year 1, the sickest of the eligibles enroll, so I assume the average cost of the newly enrolled in year 1 are 10 percent above the expected average cost of all enrollees.

 

Premiums

 

I start with the single premium from the POS plan in the state employees program in 2005  (which is $396) and trend it forward (by 8.5% per year) . Based on my analysis of 1) the prevalence of chronic disease in particular among uninsured CH eligibles, and 2)the age distribution and self-reported health status of the uninsured eligible for CH I find the expected costs are about 10% lower than the privately insured in the state. In addition, the CH plan will pay providers at Medicare plus 10% and not commercial rates, a reduction in cost of at least 10 percent. Data from Medpac from the late 1990s indicates that the average commercial payment rate to hospitals was about 35% higher than “costs” and Medicare reimbursement in Vermont. I end in 2005 with a single premium of about $310 per month. Family premiums will be much lower than the state plan since the number of children in each family policy is lower—some of these children are already enrolled in Dr. Dynasaur and other programs.

 

Demographics of the Uninsured

 

The uninsured in Vermont have a lower prevalence of chronic disease than those with private insurance (see Table 1).

 

 

 

Table 1. Prevalence of Chronic Illness among the Insured and Uninsured, 2002-2003

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

FOR UNINSURED AT LEAST 6 MONTHS DURING THE YEAR

 

 

 

Uninsured

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ages

Diabetes

High Blood Pressure

Heart Disease

Cancer

 

 

 

<=18

0.00%

1.09%

1.31%

0.00%

 

 

 

19-35

1.82%

2.02%

0.50%

1.18%

 

 

 

36-54

2.15%

7.61%

2.19%

1.44%

 

 

 

55-64

6.22%

22.90%

8.85%

0.00%

 

 

 

65+

0.00%

0.00%

0.00%

0.00%

 

 

 

Total

1.94%

5.13%

1.80%

0.95%

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

FOR THOSE WITH EMPLOYER SPONSORED INSURANCE AT LEAST 6 MONTHS

 

 

 

 

 

 

 

 

 

Ages

Diabetes

High Blood Pressure

Heart Disease

Cancer

 

 

 

<=18

0.48%

0.07%

1.37%

0.58%

 

 

 

19-35

0.51%

2.31%

1.40%

0.82%

 

 

 

36-54

3.96%

11.34%

4.15%

3.48%

 

 

 

55-64

8.01%

21.65%

13.87%

12.94%

 

 

 

65+

11.77%

48.63%

32.84%

13.73%

 

 

 

Total

3.73%

12.05%

7.17%

4.40%

 

 

 

 

 

 

 

 

 

 

 

Source: Tabulations from the 2002-3 Medical Expenditure Panel Survey, 2002-3

 

 

 

 

 

 

 

 

 

 

Table 2. Demographics of Uninsured and Privately Insured Adults in Vermont

Age

Uninsured

Employer-Sponsored Insurance

0-18

12.5%

25.8%

19-34

40.3%

19.3%

35-54

36.3%

41.7%

55-64

10.9%

13.2%

Total

100%

100%

SOURCE: Three year merged data on Vermont from the Current Population Survey for Vermont

 

 

The uninsured have a lower prevalence of chronic disease than the private insured. In addition, the uninsured population in the state is also younger than those with private insurance. Both factors would result in lower per capita spending when insured compared to the existing privately insured population in the state.

 

 

Growth in Premiums

 

I assume the per capita costs of the CH program premiums rise at 8.5% percent per year—similar to both the expected growth in Medicaid and private insurance (likely too high per capita I believe).  This is a higher rate of growth than projected by BISHCA for private health care spending over the next four years (they project a 7.6% rate of growth). CH also includes a 5% additional cost to OVHA/the state to administer the plan. The administrative costs built into the premiums are the same as the state employee plan since CH envisions a similar administrative structure.

 

Reinsurance

 

Would likely be paid on a per enrollee per month basis. It is based on the claims costs of those enrolling in the program. I assumed the state would purchase reinsurance so that the reinsurer would pay 75% of all individual claims expenses during a year above $30,000—about 10 percent of overall expected spending. I added a 10% administrative fee to this monthly premium.