12VAC30-70-381. DRG relative weights and hospital case-mix indices.
A. For the purposes of calculating DRG relative weights and hospital case-mix indices, base year claims data for all groupable cases shall be used. Base year claims data for ungroupable cases and per diem cases shall not be used. In calculating the DRG relative weights, a transfer case shall be counted as a fraction of a case based on the ratio of its length of stay to the arithmetic mean length of stay for cases assigned to the same DRG as the transfer case.
B. Using the data elements identified in subsection E of 12VAC30-70-221, the following methodology shall be used to calculate the DRG relative weights:
1. The operating costs for each groupable case shall be calculated by multiplying the per diems and ancillary cost-to-charge ratios from each hospital's cost report ending in the state fiscal year used as the base year to the corresponding days and ancillary charges by revenue code for each hospital's groupable cases.
2. The standardized operating costs for each groupable case shall be calculated as follows:
a. The operating costs shall be multiplied by the statewide average labor portion of operating costs, yielding the labor portion of operating costs. Hence, the nonlabor portion of operating costs shall constitute one minus the statewide average labor portion of operating costs times the operating costs.
b. The labor portion of operating costs shall be divided by the hospital's Medicare wage index, yielding the standardized labor portion of operating costs.
c. The standardized labor portion of operating costs shall be added to the nonlabor portion of operating costs, yielding the standardized operating costs.
3. The average standardized cost per DRG shall be calculated by dividing the standardized operating costs for all groupable cases in the DRG by the number of groupable cases classified in the DRG.
4. The average standardized cost per case shall be calculated by dividing the standardized operating costs for all groupable cases by the total number of groupable cases.
5. The average standardized cost per DRG shall be divided by the average standardized cost per case to determine the DRG relative weight.
C. Statistical outliers shall be eliminated from the calculation of the DRG relative weights. Within each DRG, cases shall be eliminated if (i) their standardized costs per case are outside of 3.0 standard deviations of the mean of the log distribution of the standardized costs per case and (ii) their standardized costs per day are outside of 3.0 standard deviations of the mean of the log distribution of the standardized costs per day. To eliminate a case, both conditions must be satisfied.
D. In calculating the DRG relative weights, a threshold of five cases shall be set as the minimum number of cases required to calculate a reasonable DRG relative weight. In those instances where there are five or fewer cases, the department's Medicaid claims data shall be supplemented with Medicaid claims data from another state or other available sources. The DRG relative weights calculated according to this methodology will result in an average case weight that is different from the average case weight before the supplemental claims data was added. Therefore, the DRG relative weights shall be normalized by an adjustment factor so that the average case weight after the supplemental claims data were added is equal to the average case weight before the supplemental claims data were added.
E. The DRG relative weights shall be used to calculate a case-mix index for each hospital. The case-mix index for a hospital is calculated by summing, across all DRGs, the product of the number of groupable cases in each DRG and the relative weight for each DRG and dividing this amount by the total number of groupable cases occurring at the hospital.
Statutory Authority
§ 32.1-325 of the Code of Virginia; 42 USC § 1396 et seq.
Historical Notes
Derived from Virginia Register Volume 16, Issue 18, eff. July 1, 2000; amended, Virginia Register Volume 34, Issue 11, eff. February 21, 2018.