Physician's Capitation Trends-HCFA delivers its promise: Risk adjustment for capitation
Payment changes will span four years
The year 2000 marks the beginning of technical shift in Medicare capitation payment methodology that — over the next four years — will move payments toward a significant conceptual change aimed at reducing capitation's riskiest aspects.
You may not notice much difference so far this year in your Medicare capitation payments. A 5% increase is projected overall, although that varies widely by county.
But over the next four years the new payment system, which factors in principal inpatient-diagnostic cost groups, or PIP-DCGs, will pay you higher rates for patients statistically predicted to be at greater health risk. (See phase-in schedule section, p. 42.) Simultaneously, payments will be lower for HMO enrollees who do not clinically qualify for a PIP classification.
Until Jan. 1, 2000, all Medicare HMO patients were paid under the adjusted average per capita costs (AAPCC) method, which essentially amounted to an average Medicare fee-for-service rate adjusted for geographic location. Other adjustments also existed — age, gender, Medicaid eligibility and institutional status.
Each patient has an assigned risk score
Starting in the new calendar year 2000, all Medicare HMO beneficiaries were assigned a "risk score" based on their clinical history. Those who are predicted to be at the greatest risk for intense services are identified as such and assigned a risk category (PIP-DCG).
The following formula, and accompanying examples, explain how to make the new capitation calculation:
• Add Part A and Part B aged rates (basic Medicare payment levels), or the Part A and Part B disabled rates (for enrollees with disabilities); the chart on page 40 shows an example drawn from four counties in Kansas, with "rescaling factors" representing demographic adjustments.
• Multiply the corresponding re-scaling factor.
• Multiply by the risk score. (See examples of the risk scores for DCGs 5-20, for males, p. 40.)
• Apply the blend percentage (10% for the year 2000, with rising percentages over the next four years).
Here are two examples the Health Care Financing Administration (HCFA) offers in its "45-day Notice letter for 2000 Medicare+Choice Payment Rates":
Example #1. Beneficiary A was hospitalized twice during the base year (1999). The diagnoses assigned were asthma (PIP-DCG 8) and staphylococcus pneumonia (PIP-DCG 18). The highest PIP-DCG category for the beneficiary would then be PIP-DCG 18. As the sample PIP score chart indicates, PIP-DCG carries a factor of 2.656. The beneficiary would also be placed in its demographic group. In this case, the patient is male, aged 82. This age group carries an age factor of 1.077.
Also, the patient had been Medicare eligible because of a disability, which carries a factor of 0.287, but he is not eligible for Medicaid (so no Medicare factor is applied). Adding these incremental factors, the risk factor for Beneficiary A is 4.02, which indicates a high expected cost individual.
Example #2. Beneficiary B had no inpatient admissions during the base year. Thus, no specific PIP-DCG would be added. Beneficiary B would be placed in the appropriate age and sex grouping and any relevant Medicaid and disable factors would be added.
Each year, insurers are required to forward up-to-date enrollee data to HCFA for risk factor analysis. HCFA computes the risk factor for the bene- ficiary as well as the capitation payment for that patient. The risk factor is computed for each individual beneficiary for a given year. That factor follows the beneficiary regardless of which insurer the beneficiary chooses.
The current DCG-PIP classification currently has 15 capitation payment categories for 89 different diagnoses. (See listing of these 15 classifications and related diagnoses, p. 41.) HCFA experts expect that the current PIPs will represent about 12% of Medicare HMO beneficiaries. More are likely to be added in the future as the PIP-DCG system is tested over the next four years.
Under PIP, capitation payments will be based on a combination of diagnostic and demographic factors, with increasing emphasis on diagnostic factors as the phase-in moves toward its fourth full year.
In summary, here is how HCFA will roll-out the new payment system in its first year:
• Assign each Medicare HMO enrollee risk adjusters, or PIP-DCGs, based on inpatient data;
• Apply individual enrollee risk scores to determine fully capitated payments
• Utilize a prospective PIP-DCG risk adjuster to estimate relative beneficiary risk scores;
• Apply separate demographic-only factors to new Medicare enrollees for whom no diagnostic history is available.
• Use six-month old diagnostic data to assign PIP-DCG categories (as opposed to using the most recent data and making retroactive adjustments of payment rates part way through the year.
• Allow for a reconciliation after the payment year to account for late submissions of encounter data.
• Phase-in effects of risk adjustment, beginning with a blend of 90% of the demographically ad-justed payment rate and 10% of the risk-adjusted payment rate in the first year (CY 2000).
• Implement processes to collect encounter data on additional services and move incrementally to a full risk adjustment model.
One of the areas of controversy for PIP-DCGs surfaced in the clinical area of congestive heart failure. "We have received many comments raising concern about the need to reimburse plans for the outpatient management of certain chronic conditions, especially congestive heart failure (CHF)," HCFA officials explained. "As one of the most frequently billed inpatient diagnoses, CHF is unique in its prevalence and to the degree to which it can be successfully managed on an outpatient basis."
Some experts are urging HCFA to make special capitation payments for CHF treated in outpatient settings, but to date, HCFA says it needs more time to research how that could be done in the PIP model.