|
APPENDIX I
|
|
Premium Estimates for Substance Abuse Parity Provisions for Commercial Health Insurance Products
|
Assumptions and Limitations
This section describes key assumptions and sources for our premium rate estimates, including cautions about how to interpret and use the estimates.
We estimated costs for the currently-insured commercial population. This does not include individuals covered by Medicare or Medicaid or uninsured individuals. We used standard Milliman & Robertson, Inc. demographic assumptions, intended to represent the age and sex mix of a typical employee group with the demographics of the US labor force population.
We estimated per capita costs for a mix of "typical" benefit plans in today's marketplace. Based on a 1995/96 survey report on employee benefits completed by Watson Wyatt Data Services, we assumed the following plan type distributions:
|
Fee-for-Service Plans |
25% |
|
PPO/POS Plans |
53% |
|
HMO/EPO Plans |
22% |
The plan provisions that we used for these "typical" benefit plans are summarized in Appendix II.
The starting point for our premium rate estimates for the commercially insured population is the M&R Health Cost Guidelines (HCGs) (July 1997 edition). The HCGs are M&R's proprietary information base that show how the components of per-capita medical claim costs vary with benefit design, demography, location, provider reimbursement arrangements, degree of managed care delivery, and other factors. In most instances, cost assumptions are based on our evaluation of several data sources and are not specifically attainable to a single source. The HCGs are used by client insurance companies, HMOs, and other organizations for, primarily, pricing and evaluating insurance products.
We incorporated estimates of managed care delivery effects in our premium rate estimates. We utilized the M&R Healthcare Management Guidelines (HMGs) (November 1996 edition) in developing our utilization and service intensity assumptions for a managed care scenario. The HMGs describe three managed care scenarios -- loosely managed delivery (typical of most fee-for-service, indemnity plans which use some pre-admission certification, concurrent review, and hospital audit), moderately managed delivery (typical of plans which use some inpatient care management protocols and standards with moderate conformity to those standards), and well managed delivery (very close conformity to inpatient care management standards such as those contained in the HMGs). We assumed that the level of managed care in our "typical" benefit plans as:
|
Benefit Plan |
Degree of Managed Care Assumed |
|
Fee-for-Service |
Loosely Managed |
|
PPO/POS |
Moderately Managed -- In-Network benefits
Loosely Managed -- Out-of-Network benefits |
|
HMO/EPO |
Well Managed |
Discounted fees are common in HMO/EPO plans and PPO/POS plans for in-network healthcare providers. We assumed that the health plan could negotiate physician reimbursement consistent with 120% of RBRVS. Hospital per diems were assumed to be $1,100, except for mental health and substance abuse, for which we assume a $500 per diem. We also assumed a 30% discount from national average charges for hospital outpatient and all other services. We assumed that no discount would be obtained for any out-of-network services provided in the PPO/POS plans or for fee-for-service plans.
In our premium rate estimates, we considered the following factors and benefit features as appropriate:
- The maximum number of inpatient days and outpatient visits for treatment for substance abuse disorders
- Deductible, copay, and coinsurance adjustments appropriate to various benefits
- An administrative expense load of 20% of claim costs
- Increases in utilization by service category due to benefit richness and induced demand
We did not consider the following factors:
- Regional variations in costs
- Effects on Medicare or Medicaid costs
- Self-selection. In an environment where employees may have a multi-option choice, people who suspect they will need benefits for mental illness or substance abuse will likely choose the plans with lower cost sharing or with less managed care for such benefits. This would likely increase the costs for the low cost sharing and low managed care plans and decrease the costs for the high cost sharing and high managed care plans, but the decrease will be less than the increase. In addition, the presence or absence of managed care carve-out programs could affect which plan an individual or family chooses.
- Adverse selection. The high cost plans may face an adverse selection spiral. Self selection may increase the costs of certain plans which would cause the premiums in these plans to increase. This may encourage low-risk individuals to seek the low cost options, which will further increase average costs of the high cost plans.
|