CMS Pub. 100-08, ch. 8, § 8.4.5

The results of the sampling unit reviews are used to calculate an estimate of the

overpayment amount. In most situations, the lower limit of a one-sided 90 percent

confidence interval should be used as the amount of overpayment to be demanded for

recovery from the provider/supplier. This conservative procedure incorporates the

uncertainty inherent in the sampling design and works to the financial advantage of the

provider/supplier. That is, it yields a demand amount for recovery that is very likely less

than the true amount of overpayment, and it allows a reasonable recovery without

requiring the tight precision that might be needed to support a demand for the point

estimate. However, the contractor is not precluded from demanding the point estimate

where high precision has been achieved, and when there are statistically sound reasons

for the demand.

Standard methods for calculating a one-sided 90 percent confidence interval, such as

those based on the central limit theorem or others found in standard statistics texts and

journals, are generally acceptable. It may not be feasible to guarantee 90 percent

coverage in all circumstances (i.e., that the lower bound of the 90 percent confidence

interval is below the true overpayment in 90 percent of audits) due to the use of

theoretical assumptions underlying standard statistical methods. Nonetheless, application

of these methods is generally appropriate.

In some cases, the point estimate or the lower bound of the estimate for the total

overpayment in the sampling frame may be greater than the total payment in the sampling

frame. This is expected to occur frequently when the true error rate is high. Nonetheless,

the use of the lower bound to calculate the demand amount continues to operate in

accounting for uncertainty in the estimate and providing a methodology that is generally

favorable toward the provider. If the point estimate of overpayment is greater than the

total payment in the sampling frame, but the lower bound is less than total payment, then

the lower bound may be demanded. If the lower bound of the estimated overpayment is

greater than total payment, the demand amount shall be reduced from the lower bound to

the total payment amount in the sampling frame to avoid demanding more than originally

paid.

The result of each sampling unit review shall be recorded, except that a sampling unit’s

overpayment shall be set to zero if there is a limitation on liability determination made to

waive provider/supplier liability for that sampling unit (per provisions found in section

1879 of the Social Security Act (the Act)) or there is a determination that the

provider/supplier is without fault as to that sampling unit overpayment (per provisions

found in section 1870 of the Act). Sampling units for which the requested records were

not provided are to be treated as improper payments (i.e., as overpayments). Sampling

units that are found to be underpayments, in whole or in part, are recorded as negative

overpayments and shall be used in calculating the estimated overpayment.