William Shoemaker

Hospitals that are looking for opportunities to improve performance can be stymied by uncertainty about where to begin. CMS data provide an excellent starting point.


At a Glance  

  • Publicly available data can help hospitals benchmark their experience against that of their peers to identify unexpected variations and potential opportunities for improvement.    
  • Medicare Provider Analysis and Review data and Medicare cost report data can be used to analyze a hospital's experience by medical service area and by Medicare-severity-adjusted DRG.    
  • Such an analysis provides an excellent first step for more in-depth analysis that can hone in on potential problem areas, such as inaccurate documentation or coding.    

Widespread demands for increased transparency and relentless pressures to contain costs in the face of declining government-backed reimbursement are compelling hospitals to focus on the effectiveness and efficiency of the services they provide. Publicly available data can help them in this effort. The Centers for Medicare and Medicaid Services (CMS) maintains data repositories that are powerful resources hospitals can use in benchmarking activities to identify unexpected variations and potential opportunities for improvement.

These data can be used to make detailed comparisons of one hospital's experiences with those of a selected peer group of hospitals. By such an analysis, a hospital can gain valuable insight into its utilization rates for key services as well as the relative costs of those services. Such an analysis can even help to expose deficiencies in the hospital's documentation and coding practices.

Claims data for the Medicare inpatient prospective payment system (PPS) are publicly available from CMS's Medicare Provider Analysis and Review (MedPAR) files and can also be obtained through a variety of commercial sources. These claims data, used in conjunction with concurrent Medicare cost report data, can also be used to allocate costs for specific medical services. Although the data reflect only Medicare patients, the volumes are sufficient for most services to allow for meaningful comparisons.

A Case Example

To illustrate how these data can be used, consider the following analysis comparing the experiences of a sample short-term, acute care hospital with the combined experiences of all of the sample hospital's peers in its area.a The sample hospital was arbitrarily chosen from a metropolitan area. The analysis is based on MedPAR data from federal fiscal year 2009 (FFY09) and corresponding hospital Medicare cost report data.

For the purposes of this analysis, Medicare severity-adjusted DRGs (MS-DRGs) are grouped into medical services (e.g. oncology, general surgery, etc.) rather than by major diagnostic category (MDC) to depict the information as lines of business rather than categories based on body systems. At a more detailed level, MS-DRGs have been collapsed into "base" MS-DRGs, which combine all levels of severity into a single category-that is, individual MS-DRGs within a base MS-DRG are differentiated only by the presence of a complication or comorbidity (CC) or a major CC (MCC).

Exhibit 1

f_shoemaker_exh1

In this analysis, any differences due to wages and other regional factors should be minimal because all hospitals are from a single metropolitan area. Wage adjustments might be necessary, however, in an analysis that involves hospitals in various geographic areas. For some analyses, it might also be necessary to adjust statistics according to case mix index (CMI). This adjustment may not be helpful, however, when working at a detailed level. For example, when examining costs and charges at a specific departmental detail by MS-DRG, a CMI adjustment may distort the dollar figures inappropriately. CMI is based on weights that measure resource consumption for an entire case and may not be meaningful for individual department utilization. If needed, figures can be CMI adjusted later at summary levels to address variations as deemed appropriate.

Exhibit 2

f_shoemaker_exh2

Analysis of Coding Indicators by Medical Service Area

Exhibit 1 depicts coding indicators specific to the primary hospital compared with those of all other acute care facilities in the metropolitan area for each medical service. The CC/MCC rate shows the ratio of cases coded into MS-DRGs that include comorbidities and/or complications. The MCC rate refers to the percentage of cases that fall into the category of cases with the highest level of complexity for the diagnosis group.

Although several areas warrant further investigation, this analysis focuses on the cardiology service area given its applicability to most hospitals and potential for external scrutiny due to high Medicare volumes and costs. As shown in the exhibit, the primary hospital's MCC rate-or percentage of most complex cases-in this service area is a noticeably higher percentage than the MCC rate for all of its peers. A hospital with a higher-than-average complication rate needs to understand the reasons.

Exhibit 3

f_shoemaker_exh3

There may be clinical explanations for this high rate (e.g., a high percentage of admissions from nursing homes), but it is also possible that it is due to inappropriate documentation and coding practices. This possibility should be investigated because a higher-than-average complication rate may be a red flag for recovery audit contractors. It is also interesting to note that despite having higher rates of complication, the primary hospital's CMI for cardiology services is lower than the average CMI for the other hospitals in the region, which may indicate that the other hospitals perform cardiology services that are more intensive. Because the CMI is similarly lower for other medical services at the sample hospital as well, it is important to understand the cause.

Analysis of Coding Indicators by Cardiology MS-DRG

Exhibit 2 shows the same service line in detail by base MS-DRG group, this time with corresponding payment and cost data. The base MS-DRGs 310-309-308 and 293-292-291 ("Cardiac arrhythmia and conduction disorders" and "Heart failure and shock," respectively) stand out as the likely cause of the higher CC/MCC rate for cardiology as a whole.

The analysis should investigate the reasons for these comparatively higher rates of complication. Are there possible problems with the care being provided? Are proper documentation and coding practices being followed? Such questions need to be identified and explored internally to determine whether there is need for corrective actions. Failure to manage such issues internally increases the risk that they will be targeted by outside auditors.

In a different analysis, a facility might find that its CC/MCC rates are significantly lower than those of its peers. Such results also warrant deeper investigation to make sure the hospital is not losing revenue due to inaccurately documented and coded claims.

Analysis of Payment and Cost by Cardiology MS-DRG

The same group of MS-DRGs for the cardiology service line are examined in exhibit 3 with a focus on the financial aspects of the services: charges, costs, and payment (excluding capital pass-through and organ acquisition amounts). This type of analysis can be useful to identify how efficiently peer hospitals provide similar services in relation to how they are paid. Interestingly, there appears to be a need for the hospital to focus on the "heart failure and shock" MS-DRG grouping once again. Although other base MS-DRGs show even greater variation from the peer group, the smaller change in the MS-DRG 293-292-291 group is greatly amplified by its volume and resulting effect on the bottom line, because the cases in this group both cost more and are paid less than are corresponding cases in the peer group. Despite the tendency toward a higher case mix for this base MS-DRG (i.e., a higher percentage of cases with CCs and MCCs), the likelihood of favorable reimbursement would be negated by this facility's high cost structure. To determine the root cause of higher cost, the hospital should conduct a departmental cost analysis for the MS-DRG group.

Exhibit 4

f_shoemaker_exh4

Analysis of Accommodation Costs and Length of Stay (LOS)

Accommodation costs and LOS for base MS-DRG 293-292-291 are shown in exhibit 4. These cost figures offer a glimpse at how efficiently the hospital is offering one of its most common services. In this instance, the hospital should take note of the higher utilization of the more costly accommodations (i.e., intensive care unit [ICU] and coronary care unit [CCU] utilization).

Exhibit 5

f_shoemaker_exh5

The sample hospital has costs per day in each setting that are quite comparable with those of its peers. But the average LOS in the CCU for this base MS-DRG is significantly higher for this provider. The hospital should conduct an immediate review of utilization practices for this unit to examine the use of the higher cost setting. It is important for hospitals to reduce unnecessarily high CCU utilization; doing so could yield significant savings. It is also worth noting that the overall LOS for this MS-DRG is also high by comparison.

Analysis of Ancillary and Variable Costs

Exhibit 5  shows how ancillary and variable costs are allocated in the care for base MS-DRG 293-292-291. Again, most of the hospital's costs are comparable with those of peer hospitals, with one notable exception: medical and surgical supplies. As with the utilization of the CCU, the hospital should review these supply costs. In this instance, the utilization patterns of a single department could stand behind a $1,000-per-case cost reduction in a high-volume service. (It is important to note, however, that such a comparison can sometimes be skewed by differences in cost reporting for surgical supplies among hospitals.)

A Critical First Step

Increasing scrutiny of hospital delivery systems seems inevitable. It will be difficult to know which new issues will capture the attention of regulators next, but keeping an eye on how patterns of care are distributed among peers may help identify potential issues before they become subject to regulatory scrutiny. At a minimum, such analyses can help administrators identify, address, and explain variations.

The analytical steps described here represent only a beginning of an in-depth review of delivery systems, but they offer the advantage of using readily available data. Further analysis and process review will almost certainly be necessary to determine whether the potential exists for costs savings or whether coding and documentation processes must be adjusted. But hospitals can find it daunting just to identify a starting point and then gather the tools to get started. That's where a hospital can benefit from using these basic analytical templates.


William Shoemaker is senior vice president, American Hospital Directory, Louisville, Ky., and a member of HFMA's Kentucky Chapter (wshoemaker@ahd.com).


 

Footnote

a. This peer group was also chosen for illustrative purposes only; in actual practice, the individual hospital's experiences would be compared with those of a carefully selected set of peers (e.g., hospitals in a competitive market area, teaching hospitals of a certain size, and hospitals with strength in a particular specialty).
 

Publication Date: Wednesday, June 01, 2011

Login Required

If you are an existing member, please log in below. Username and password are required.

Username:

Password:

Forgot User Name?
Forgot Password?







Close

If you are not an HFMA member and would like to access portions of our content for 30 days, please fill out the following.

First Name:

Last Name:

Email:

   Become an HFMA member instead