AAPC-CPMA Certified Professional Medical Auditor - Set 4 - Part 3

Correct Answer: Expanded Problem Focused

An Expanded Problem Focused examination is a limited examination of the affected body area or organ system and other symptomatic or related organ system(s). The note matches that definition exactly. Option a is incorrect because it calls for a general multi-system examination or a complete examination of a single organ system, far broader than a limited review. Option b is incorrect because it requires an extended examination of the affected areas and related systems, which goes beyond the limited examination described in this note. Option d is incorrect because it covers a limited examination of only the affected body area or organ system, with no other symptomatic or related systems included.

Correct Answer: Renal dialysis services

The regulation excepts renal dialysis services, along with certain hospital outpatient and ambulatory services and psychiatric services, from the documentation requirement. Other services, including Medicare telehealth services, must show in the medical record that the teaching physician was present. Option b is incorrect because the regulation expressly includes a Medicare telehealth service within the services whose records must document the teaching physician's presence. Option c is incorrect because the exceptions name other service categories, and surgery is subject to presence during critical portions of the procedure. Option d is incorrect because the regulation requires the teaching physician to be present during the entire viewing of an endoscopic procedure, and these procedures are not listed among the documentation exceptions.

Correct Answer: Parameter

A parameter is a numerical characteristic of the whole population that can be estimated by a statistic. A statistic comes from sample data, so the population-level value being estimated is the parameter. Option a is incorrect because this is an issue, such as poor study design or data entry mistakes, that harms the reliability of sampling data; it is not a population measure. Option b is incorrect because frequency counts how many times a value occurs in the data, so it summarizes observations and does not describe the whole population. Option d is incorrect because a statistic is a number describing a property of the sample, calculated from sample data, and it serves as the estimate rather than the quantity being estimated.

Correct Answer: well-accepted methodologies

The Medicare manual allows reviews that assist law enforcement to use sampling methodologies that differ from the standard instructions, provided they are well-accepted methodologies amongst statisticians and a complete explanation of their appropriateness is given. Option a is incorrect because the Medicare manual requires methodologies recognized among statisticians, and novelty is not a stated criterion for law-enforcement support reviews. Option b is incorrect because the manual does not describe any exemption from CMS oversight, and a full explanation of appropriateness is still required. Option c is incorrect because approval by the reviewed provider or supplier is not a requirement; acceptance is judged by the statistical community, not by the party under review.

Correct Answer: Exclude them from the sampling frame

Under the Medicare program integrity manual, claims with no payment may be in the universe but are excluded when establishing the sampling frame, because sampling units without final payment at sample selection are not included. Option a is incorrect because clustering by patient defines a possible sampling unit, but it does not change the rule that unpaid units stay out of the frame. Option b is incorrect because claims with no payment may remain in the universe; the exclusion happens only at the stage where the sampling frame is built. Option d is incorrect because sampling units with no final payment at the time of sample selection are not to be included in the sampling frame under the manual.

Correct Answer: Overpayments are fixed values, not random variables

Overpayments are fixed values, though unknown before sampling. Regardless of any correlation between sampling units, one unit's overpayment does not change based on the outcomes of other units, so the independence assumption is not defeated. Option b is incorrect because units may be correlated through the same location, provider or time period; the guidance says such correlation does not imply dependence in sampling. Option c is incorrect because the guidance says selection probabilities do not have to be equal, only known and greater than zero, so equality is not the basis. Option d is incorrect because returning units describes replacement, which is the opposite of the situation; without replacement, selection probability changes with each unit selected, creating some dependence.

Correct Answer: The value lies less than one standard deviation below the mean

The z-score is (x – μ) / σ = (1 – 5) / 6, about –0.67. The negative sign places the value below the mean, and an absolute value under 1 means it is less than one standard deviation away. Option a is incorrect because a z-score of zero occurs only when x equals the mean; here x is 1 and the mean is 5, so the difference is nonzero. Option c is incorrect because the z-score is (1 – 5) / 6, about –0.67, and its absolute value is below 1, so the distance is under one standard deviation. Option d is incorrect because a value smaller than the mean produces a negative z-score, which places it below the mean, not above it.

Correct Answer: The estimate of p, the universe error rate

In an unrestricted attribute appraisal, the sample error rate, the number of errors divided by the sample size, is the estimate of p, the error rate for the entire universe. It is a point estimate. Option a is incorrect because that figure is found by multiplying the universe size by the sample error rate, which gives a count of items rather than a proportion. Option b is incorrect because the upper limit is a separate figure computed by the RAT-STATS software, and it lies above the point estimate of the error rate. Option c is incorrect because the point estimate is not in the center of the interval, because the exact hypergeometric distribution is used rather than the normal approximation.

Correct Answer: The full stratum universe size

When the computed sample size for a stratum exceeds its universe size, RAT-STATS sets ni equal to Ni. The remaining strata are then sized with that stratum omitted from the denominator and n replaced by n minus Ni. Option a is incorrect because the stratum is omitted only from the denominator when the remaining strata are recalculated; its own size is set to Ni, so it is still sampled. Option c is incorrect because the formula allocates each stratum a share of n through its Ratio, which depends on Ni and StdDev, not an equal division across strata. Option d is incorrect because rounding up to the nearest integer applies to ordinary stratum sizes, but an oversized computed value is replaced by the universe size, not rounded.

Correct Answer: Lower limit of a one-sided 90 percent confidence interval

Medicare guidance states that in most situations the lower limit of a one-sided 90 percent confidence interval is the overpayment amount demanded. It is conservative, accounts for sampling uncertainty, and favors the provider or supplier. Option a is incorrect because is a ceiling that caps the demand when the lower bound exceeds it, not the figure normally demanded. Option b is incorrect because is demanded only as an exception, when high precision has been achieved and statistically sound reasons support it, not in most situations. Option c is incorrect because would overstate the likely true overpayment and work against the provider, whereas the guidance calls for a conservative figure that favors the provider.