Images de page

Table 6: Provider Denial Rates for Medical Necessity, 1993.”

Carrier Services with denials Percent of providers Percent of providers exceeding 90 per 1,000 with at least 1 medical receiving 50 percent of services allowed necessity denial” all medical necessity denials El Northern A0150 52.5 2.8 California A2000 65.4 6.9 Southern 78.465 58. 7 7. 9 California 92.982 19.5 2. 3 93307 53 - 0 1.5 933.20 49. 1 3 - 0 93.880 52.6 3.1 North A0010 73. 4 5. 6 Carolina A0020. 78.1 4.8 A2000 85.5 10.6 92.982 38 - 0 6.5 93549 40... 8 6.1 Illinois A2000 70.5 7.4 71020 54.2 2.1 Wisconsin A2000 56.2 6.4 92.004 4.7. 5 4.4

*Excludes South Carolina because it does not have medical necessity denial rates greater than 90 per 1,000 services allowed.

*Excludes providers that did not submit a claim for a service. Percentages are based on a 100-percent sample of 1993 claims. The method used for determining the denial median was based on the total number of denials a provider received. The percentage of allowed services accounted for by providers with 50 percent of denials was as follows: Northern California, A2000 = 29.7, A0150 = 28.2; Southern California, 78465 = 31.5, 92.982 = 3.8, 93307 = 8. 7, 93320 11.8, 93.880 = 19.6; North Carolina, A0010 = 20.8, A0020 = 8.4, A2000 29.3, 92.982 = 9.9, 93549 = 12. 3.; Illinois, A2000 = 27. 1, 71020 = 17. 1; Wisconsin, A2000 = 20.5, 92004 = 16. 6.

Our analysis suggests that a small minority of providers, between 1.5 and 10.6 percent, accounted for 50 percent of services denied for lack of medical necessity (and thus were responsible for the bulk of denials). Thus, the screens and medical policies these carriers used to determine the medical necessity of claims primarily affected a relatively small proportion of the provider community. Table 6 also shows that the proportion of providers that had at least one denial varied between 19.5 and 85.5 percent. The latter range suggests that some prepayment screens used to identify inappropriate billing patterns affected a smaller proportion of the provider population than did others.

While we cannot explain differing patterns of provider denials-for example, they may stem from unnecessary services being disproportionately offered by a few providers, differences in patient characteristics, variations in billing practices, or a number of other factors-- further examination of the reasons for them is warranted given their potential to explain substantial amounts of variation in denial rates.


The magnitude of carrier denial rates was generally low and persistent for 2 consecutive years, although rates for some services shifted across years. Medical necessity denial rates for 74 services across six carriers varied substantially. The primary reason for variation in carrier denial rates was that certain carriers used screens for specific services while others did not. Thus, carriers' selecting the services to be screened and their determining the stringency of the screen criteria probably account for a significant proportion of the variability. Further, a small proportion of the providers accounted for 50 percent of the denied claims. To a lesser degree, the varying interpretation of certain national coverage standards across carriers, differences in the way carriers treated claims with missing information, and reporting inconsistencies helped explain variation in carrier denial rates.

We did not attempt to assess whether low or high medical necessity denial rates for individual carriers were appropriate. Low denial rates are desirable from the standpoint that they imply less annoyance and inconvenience for providers and beneficiaries. However, low denial rates are desirable only insofar as providers do not bill for medically unnecessary services.

What is clear from our work is that further analysis of denial rates can provide useful insight into how effectively Medicare carriers are managing program dollars and serving beneficiaries and providers. Since funding constraints limit the number of claims carriers can examine on a prepayment basis, it is

important that they use the most effective and appropriate screens.

We believe that HCFA could improve its oversight capabilities by actively monitoring data on carrier denial rates and improving the reliability of the data that it collects. Data on denial rates are useful for identifying inconsistencies in the way that carriers assess claims for medical necessity. This information, in turn, could be used to identify the services that certain carriers have found to have billing problems. In addition, for services that are more uniformly screened by carriers, variation in denial rates could indicate that carriers are using different screen criteria, which raises issues of appropriateness and effectiveness. Finally, data on denial rates could be used to construct a profile of the subpopulation of providers that have a disproportionately large number of denials, which might suggest a solution to this problem.


We recommend that, to improve its oversight of the Medicare Part B program, HCFA

-- issue instructions to carriers on how to classify the reason for denial when reporting this information;

-- analyze intercarrier screen usage (including the stringency of screen criteria), identify effective screens, and disseminate this information to all carriers; and

-- direct carriers to profile the subpopulation of providers responsible for a disproportionate share of medical necessity denials in order to devise a strategy for addressing this problem.


At your request, we did not obtain agency comments on a draft of this report.

If you or your staff have any questions about this report or would like additional information, please call me at (202) 512-2900 or - Kwai-Cheung Chan, Director for Program Evaluation in Physical Systems Areas, at (202) 512-3092. Major contributors to this report are listed in appendix V.

Terry E. Hedrick
Assistant Comptroller General

« PrécédentContinuer »