Understanding Your Hospital’s Case Mix

Understanding Your Hospital’s Case Mix

By Minnette Terlep

‘Why does my hospital’s case mix have month-to-month peaks and valleys?”

“How much does coding really impact our case mix?”

“Our chief financial officer (CFO) questions whether we are optimizing coding because our case mix is lower than our competitor’s. How can variances in case mix from hospital to hospital be explained?”

These questions are often asked by health information management (HIM) directors, CFOs and coding supervisors. The answers are simple when one understands which DRGs drive the majority of the hospital’s case mix.

Understanding Medical vs. Surgical DRG Case Weights

The first step is to understand the differences in case weight between all current surgical DRGs as compared to the medical DRGs.

Figure 1 represents all current DRGs excluding the seven neo-nate DRGs. The 262 medical DRGs are displayed in the lower line of the graph (represented by open circles) beginning with the highest weighted medical DRG at 4.0299 and the lowest at .0777.

The 221 surgical DRGs are displayed in the upper line of the graph (represented by asterisks) with the highest case weight at 16.0413 and the lowest at .1504. (For purposes of case mix analysis, DRG 475, Respiratory System Diagnosis with Ventilator Support, and DRGs 482 and 483, Tracheostomies, are classified as surgical DRGs.)

Note that 213 of the 221 surgical DRGs are higher weighted than the corresponding medical DRG when each group’s DRGs are displayed in descending order by DRG case weight.

There are 99 surgical DRGs with a case weight of 1.5000 or higher. In contrast, there are only 11 medical DRGs with a case weight of 1.5000 or more. Of these 11, only two are common medical DRGs–DRG 79, Respiratory infections, and DRG 121, AMI with cardiovascular complication.

This comparison of medical and surgical DRG case weights shows that surgical DRGs are a major part of a hospital’s case mix. However, to quantify the impact, one must remove these cases from the case mix and re-calculate.

Calculating Case Mix

The Health Care Financing Administration (HCFA) FY1997 DRGs have a case mix index of 1.3577. This index is calculated by summing the weights of the 490 current DRGs and dividing by 490. HCFA has determined that for the average patient consuming the average amount of resources and staying in the hospital the average number of days, the case will be assigned to a DRG with a weight of 1.3577.

For purposes of case-mix analysis, focus only on major surgical DRG cases. Major surgical DRGs are those with a HCFA case weight of 1.5000 and higher (including DRGs 475, 482 and 483).

Experience with more than 400 hospitals’ Medicare case mix has shown that if only these major surgical DRG cases are removed from a hospital’s case mix and then the case mix is recalculated, the hospital’s case mix generally drops to the range of 1.0000-1.1000, significantly below the DRG case mix index of 1.3577.

The significant peaks and valleys that occur in monthly Medicare case mix do not occur when these major surgical DRGs are excluded from the calculation, as demonstrated in Figure 2.

When the volume of monthly cases in each of the 99 major surgical DRGs is monitored, the underlying cause of peaks and valleys can easily be identified. Most common causes of these peaks and valleys include:

* increase, decrease in vents/trachs;

* surgeon vacations or meetings that cause a decrease in major surgery volume; and

* decrease in scheduled major surgery due to holidays or season.

When monthly case mix trends upward or downward over a period of months there has usually been either a sustained growth or decline in one or more major surgery services, or sustained growth or decrease in the percent of medical and/or minor surgery cases as compared to volume of major surgery cases.

Optimizing Coding and DRG Assignment

Generally, coding and DRG optimization occur in one of several ways:

* adding a complication/comorbidity (CC) diagnosis to change DRG from no CC DRG to CC DRG (for medical or surgical DRGs); and

* selecting a more resource-intensive alternate principal diagnosis (for medical DRGs) that is assigned to a higher-weight medical DRG.

While there are a few optimization techniques related to surgical DRGs, the majority of ethical coding/DRG optimization comes from adding, changing or re-sequencing diagnoses.

To ensure that coding optimization is occurring with CC and no CC DRGs, the hospital should perform coding quality checks on all no CC DRG cases prior to billing.

CC DRG percentage should also be calculated and trended monthly to ensure that coding/DRG optimization is continuing to be done. CC DRG percentage is calculated by summing all CC and no CC DRG cases and dividing by the number of CC DRG cases.

In the author’s experience, Medi-care CC DRG percentage has generally been found to range from 75 percent to 85 percent. However, to avoid potential unethical assignment of additional CC diagnoses, it is recommended that the hospital not establish specific percentage goals.

If the hospital finds that its CC DRG percentage ranges from 90 percent to 100 percent, coding quality should be conducted on a sampling of CC DRG cases to ensure that all additional CC diagnoses have been assigned within official coding guidelines.

Hospital CC DRG percentage will vary somewhat from hospital to hospital, simply due to differences in complexity of illness in different mixes of patients. If the hospital has determined its own CC DRG percentage and has ensured that the coders are performing ethical optimization, then the hospital should accept that percentage as their benchmark and continue monitoring on a monthly basis.

Smaller hospitals with fewer Medicare discharges will experience more fluctuation in the CC DRG percentage number from month to month than larger hospitals with greater volumes.

Other coding optimization quality reviews that should be done include:

* All cases in DRGs 89 and 90, Simple pneumonia with and without CC, should be reviewed to determine whether documentation in the record ethically supports assignment of probable gram-negative pneumonia, 482.89. This would cause the case to be re-assigned to the more resource-intensive DRGs 79 and 80.

* Cases in DRG 122, AMI without additional cardiovascular complicating diagnosis, should also be reviewed to verify whether such a diagnosis exists but has not been coded.

These DRGs have been selected because when optimized they generally assign to the two common medical DRGs with a weight of 1.5000 and higher, DRGs 79 and 121.

An easy way to measure potential effects of coding optimization is to concentrate on optimizing no CC DRGs. Using three month’s billing data, the hospital can convert all billed no CC DRGs to CC DRGs as shown in Table 1. Remember, however, that 100 percent CC DRGs is not a realistic goal.

Recommendations for Managing Case Mix

To solve the long-standing puzzle of case mix analysis and coding optimization, the following steps should be followed:

1. Identify the hospital’s volume of major surgery DRGs.

2. Recalculate case mix excluding these cases.

3. Calculate and trend the hospital’s monthly percentage of major surgery cases.

4. Monitor monthly fluctuations in each major surgery DRG to explain peaks and valleys.

5. Calculate and trend the hospital’s CC DRG percentage.

6. Perform coding optimization quality review checks prior to billing.

These simple steps will provide ongoing answers to the case mix questions most often asked by HIM directors, CFOs and coding supervisors.

* About the author: Minnette Terlep is founder and president of MT Associates, Woodruff, WI.

Table 1–Hospital Medicare Case Mix: 1.4010

600 cases billed

240 of 600 in CC/no CC DRGs

CC DRG Percentage 80% (192 in CC DRGs, 48 in “no CC” DRGs)

Hospital Medicare blended rate = $4,000

Case wt. Optimized to Case wt.
10 cases in DRG 97 .6035 *10 = 6.0350 10 cases in DRG 96 .8272 *10 = 8.2700
14 cases in DRG 139 .4971 *14 = 6.9594 14 cases in DRG 138 .8008 *14 = 11.2112
10 cases in DRG 147 1.6018 *10 = 16.0180 10 cases in DRG 146 2.6363 *10 = 26.3630
14 cases in DRG 175 .5485 *14 = 7.6790 14 cases in DRG 174 .9952 *14 = 13.9328

About The Author