U.S. OFFICE OF PERSONNEL MANAGEMENT
OFFICE OF THE INSPECTOR GENERAL
OFFICE OF AUDITS
Final Audit Report
Subject:
Audit of the Federal Employees Health Benefits
Program Operations at Union Health Service, Inc.
Report No. 1C-76-00-12-006
Date: August 20, 2012
-- CAUTION --
This audit report has been distributed to Federal officials who are responsible for the administration of the audited program. This audit
report may contain proprietary data which is protected by Federal law (18 U.S.C. 1905). Therefore, while this audit report is available
under the Freedom of Information Act and made available to the public on the OIG webpage, caution needs to be exercised before
releasing the report to the general public as it may contain proprietary information that was redacted from the publicly distributed copy.
AUDIT REPORT
Federal Employees Health Benefits Program
Community-Rated Health Maintenance Organization
Union Health Service, Inc.
Contract Number CS 1571 - Plan Code 76
Chicago, Illinois
Report No. 1C-76-00-12-006 Date: ___________________
August 20, 2012
Michael R. Esser
Assistant Inspector General
for Audits
-- CAUTION --
This audit report has been distributed to Federal officials who are responsible for the administration of the audited program. This audit
report may contain proprietary data which is protected by Federal law (18 U.S.C. 1905). Therefore, while this audit report is available
under the Freedom of Information Act and made available to the public on the OIG webpage, caution needs to be exercised before
releasing the report to the general public as it may contain proprietary information that was redacted from the publicly distributed copy.
EXECUTIVE SUMMARY
Federal Employees Health Benefits Program
Community-Rated Health Maintenance Organization
Union Health Service, Inc.
Contract Number CS 1571 - Plan Code 76
Chicago, Illinois
Report No. 1C-76-00-12-006 Date: August 20, 2012
The Office of the Inspector General performed an audit of the Federal Employees Health
Benefits Program (FEHBP) operations at Union Health Service, Inc. (Plan). The audit covered
contract years 2007 through 2011, and was conducted at the Plan’s office in Chicago, Illinois.
This report questions $1,110,730 for inappropriate health benefit charges to the FEHBP in
contract years 2007 through 2011. The questioned amounts include $1,035,784 for defective
pricing, and $74,946 due the FEHBP for lost investment income, calculated through June 30,
2012.
In contract years 2007 through 2011, the Plan gave a similarly sized subscriber group (SSSG) a
discount; however, the same discount was not given to the FEHBP. Applying the SSSG
discounts to our audited rates results in overcharges to the FEHBP of $35,499; $68,307;
$270,745; $612,425; and $48,808 in 2007, 2008, 2009, 2010, and 2011, respectively. Consistent
with the FEHBP regulations and contract, the FEHBP is due $74,946 for lost investment income,
calculated through June 30, 2012, on the defective pricing findings.
i
CONTENTS
Page
EXECUTIVE SUMMARY .............................................................................................. i
I. INTRODUCTION AND BACKGROUND..................................................................... 1
II. OBJECTIVES, SCOPE, AND METHODOLOGY ......................................................... 3
III. AUDIT FINDINGS AND RECOMMENDATIONS ...................................................... 5
1. Premium Rate Review ................................................................................................. 5
2. Loss Investment Income .............................................................................................. 8
IV. MAJOR CONTRIBUTORS TO THIS REPORT .......................................................... 10
Appendix (Union Health Service’s May 21, 2012, response to the draft report)
The Plan has participated in the FEHBP since 1975 and provides health benefits to FEHBP
members in the Chicago, Illinois area. The last audit of the Plan conducted by our office was in
2006. All issues from that audit have been resolved.
2
II. OBJECTIVES, SCOPE, AND METHODOLOGY
Objectives
The primary objectives of the audit were to verify that the Plan offered market price rates to the
FEHBP and to verify that the loadings to the FEHBP rates were reasonable and equitable.
Additional tests were performed to determine whether the Plan was in compliance with the
provisions of the laws and regulations governing the FEHBP.
Scope
FEHBP Premiums Paid to Plan
We conducted this performance audit in
accordance with generally accepted government $5
auditing standards. Those standards require that $4
Millions
we plan and perform the audit to obtain
$3
sufficient, appropriate evidence to provide a
reasonable basis for our findings and conclusions $2
based on our audit objectives. We believe that
$1
the evidence obtained provides a reasonable
basis for our findings and conclusions based on $0
2007 2008 2009 2010 2011
our audit objectives. Revenue $2.3 $2.5 $2.6 $3.4 $4.4
This performance audit covered contract years
2007 through 2011. For these contract years, the FEHBP paid approximately $15.2 million in
premiums to the Plan. The premiums paid for each contract year audited are shown on the chart
above.
OIG audits of community-rated carriers are designed to test carrier compliance with the FEHBP
contract, applicable laws and regulations, and OPM rate instructions. These audits are also
designed to provide reasonable assurance of detecting errors, irregularities, and illegal acts.
We obtained an understanding of the Plan’s internal control structure, but we did not use this
information to determine the nature, timing, and extent of our audit procedures. However, the
audit included such tests of the Plan’s rating system and such other auditing procedures
considered necessary under the circumstances. Our review of internal controls was limited to the
procedures the Plan has in place to ensure that:
• The appropriate similarly sized subscriber groups (SSSG) were selected;
• the rates charged to the FEHBP were the market price rates (i.e., equivalent to the best
rate offered to the SSSGs); and
• the loadings to the FEHBP rates were reasonable and equitable.
In conducting the audit, we relied to varying degrees on computer-generated billing, enrollment,
and claims data provided by the Plan. We did not verify the reliability of the data generated by
3
the various information systems involved. However, nothing came to our attention during our
audit testing utilizing the computer-generated data to cause us to doubt its reliability. We believe
that the available data was sufficient to achieve our audit objectives. Except as noted above, the
audit was conducted in accordance with generally accepted government auditing standards,
issued by the Comptroller General of the United States.
The audit fieldwork was performed at the Plan’s office in Chicago, Illinois, during October 2011.
Additional audit work was completed at our offices in Jacksonville, Florida and Cranberry
Township, Pennsylvania.
Methodology
We examined the Plan’s Federal rate submissions and related documents as a basis for validating
the market price rates. In addition, we examined the rate development documentation and
billings to other groups, such as the SSSGs, to determine if the market price was actually charged
to the FEHBP. Finally, we used the contract, the Federal Employees Health Benefits Acquisition
Regulations, and OPM’s Rate Instructions to Community-Rated Carriers to determine the
propriety of the FEHBP premiums and the reasonableness and acceptability of the Plan’s rating
system.
To gain an understanding of the internal controls in the Plan’s rating system, we reviewed the
Plan’s rating system policies and procedures, interviewed appropriate Plan officials, and
performed other auditing procedures necessary to meet our audit objectives.
4
reconciled rates shows that the FEHBP was overcharged $68,307 in contract year 2008 (see
Exhibit B).
2009
As in previous years, the Plan selected and as SSSGs for contract year
2009. We disagree with these selections. Again, we selected and as the
SSSGs for contract year 2009, because they were closest in size to the FEHBP.
Our analysis of the rates charged to the SSSGs shows that received a percent
discount and received a percent discount. The FEHBP received a discount
of percent for contract year 2009. Since the FEHBP is entitled to a discount equivalent
to the largest discount given to an SSSG, we recalculated the FEHBP rates using the
percent discount given to A comparison of our audited rates to the Plan’s
reconciled rates shows that the FEHBP was overcharged $270,745 in contract year 2009 (see
Exhibit B).
2010
As in previous years, the Plan selected and as SSSGs for contract year
2010. We disagree with these selections. Again, we selected and as the
SSSGs for contract year 2010, because they were closest in size to the FEHBP.
Our analysis of the rates charged to the SSSGs shows that received a percent
discount and received a percent discount. The FEHBP received a discount
of percent for contract year 2010. Since the FEHBP is entitled to a discount equivalent
to the largest discount given to an SSSG, we recalculated the FEHBP rates using the
percent discount given to A comparison of our audited rates to the Plan’s
reconciled rates shows that the FEHBP was overcharged $612,425 in contract year 2010 (see
Exhibit B).
2011
As in previous years, the Plan selected and as SSSGs for contract year
2011. We disagree with these selections. Again, we selected and as the
SSSGs for contract year 2011, because they were closest in size to the FEHBP.
Our analysis of the rates charged to the SSSGs shows that received a percent
discount and received a percent discount. The FEHBP received a discount
of percent for contract year 2011. Since the FEHBP is entitled to a discount equivalent
to the largest discount given to an SSSG, we recalculated the FEHBP rates using the
percent discount given to A comparison of our audited rates to the Plan’s
reconciled rates shows that the FEHBP was overcharged $48,808 in contract year 2011 (see
Exhibit B).
6
The Plan’s comments regarding the term discount in the draft report do not have an effect on
the questioned cost. For the purposes of the audit, the term discount refers to any differences
between the audited rates and the Plan’s reconciled rates.
The OIG audited rates include the actual capitation rates filed with the State of Illinois. No
valid support was provided for a reconciliation adjustment.
The auditor’s calculations only go to Line 5 rates of the Attachment III of the reconciliations
that are submitted to OPM. The cost of printing the FEHBP benefit brochures is applied
after Line 5 and has no bearing on our audited rates. The audited rates correctly exclude the
printing cost of the FEHBP benefit brochures.
According to the OPM benefit brochures, the FEHBP did not receive the smoking cessation
benefit until 2011; therefore, the smoking cessation benefit should not have been charged to
the FEHBP in contract years 2007 through 2010.
We do not agree that the Plan should give the SSSGs a Medicare credit or the FEHBP a
Medicare loading. The Plan’s rating methodology does not include Medicare credits or
loadings and, therefore, neither can be added to the FEHBP’s rate or the SSSGs’ rates. Our
audited rates properly exclude any Medicare adjustments since it is not the Plan’s practice.
Recommendation 1
We recommend that the contracting officer require the Plan to return $1,035,784 to the
FEHBP for defective pricing in contract years 2007 through 2011.
2. Lost Investment Income $74,946
In accordance with the FEHBP regulations and the contract between OPM and the Plan, the
FEHBP is entitled to recover lost investment income on the defective pricing findings in
contract years 2007 through 2011. We determined that the FEHBP is due $74,946 for lost
investment income, calculated through June 30, 2012 (see Exhibit C). In addition, the
FEHBP is entitled to lost investment income for the period beginning July 1, 2012, until all
defective pricing finding amounts have been returned to the FEHBP.
Federal Employees Health Benefits Acquisition Regulation 1652.215-70 provides that, if any
rate established in connection with the FEHBP contract was increased because the carrier
furnished cost or pricing data that was not complete, accurate, or current as certified in its
Certificate of Accurate Pricing, the rate shall be reduced by the amount of the overcharge
caused by the defective data. In addition, when the rates are reduced due to defective
pricing, the regulation states that the government is entitled to a refund and simple interest on
the amount of the overcharge from the date the overcharge was paid to the carrier until the
overcharge is liquidated.
8
Our calculation of lost investment income is based on the United States Department of the
Treasury's semiannual cost of capital rates.
Plan’s Comments (see Appendix):
The Plan did not comment on this finding.
9
IV. MAJOR CONTRIBUTORS TO THIS REPORT
Community-Rated Audits Group
, Auditor-in-Charge
, Auditor
, Jr., Chief
Senior Team Leader
10
Exhibit A
Union Health Service, Inc.
Summary of Questioned Costs
Defective Pricing Questioned Costs:
Contract Year 2007 $35,499
Contract Year 2008 $68,307
Contract Year 2009 $270,745
Contract Year 2010 $612,425
Contract Year 2011 $48,808
Total Defective Pricing Questioned Costs $1,035,784
Lost Investment Income $74,946
Total Questioned Cost $1,110,730
Exhibit B
Page 1 of 2
Union Health Service, Inc.
Defective Pricing Questioned Costs
2007 Contract Year
Single Family
Plan's Reconcilied Rates
Audited Line 5 Rates
Overcharge
March 31, 2007 Enrollment
x 26 pay periods
Amount Due FEHBP in 2007 $35,499
2008 Contract Year
Single Family
Plan's Reconcilied Rates
Audited Line 5 Rates
Overcharge
March 31, 2008 Enrollment
x 26 pay periods
Amount Due FEHBP in 2008 $68,307
2009 Contract Year
Single Family
Plan's Reconcilied Rates
Audited Line 5 Rates
Overcharge
March 31, 2009 Enrollment
x 26 pay periods
Amount Due FEHBP in 2009 $270,745
Exhibit B
Page 2 of 2
Union Health Service, Inc.
Defective Pricing Questioned Costs
2010 Contract Year
Single Family
Plan's Reconcilied Rates
Audited Line 5 Rates
Overcharge
March 31, 2010 Enrollment
x 26 pay periods
Amount Due FEHBP in 2010 $612,425
2011 Contract Year
Single Family
Plan's Reconcilied Rates
Audited Line 5 Rates
Overcharge
March 31, 2011 Enrollment
x 26 pay periods
Amount Due FEHBP in 2011 $48,808
Total Defective Pricing Questioned Cost $1,035,784
Exhibit C
Union Health Service, Inc.
Lost Investment Income
Year 2007 2008 2009 2010 2011 Jun-12 Total
Audit Findings:
Defective Pricing $35,499 $68,307 $270,745 $612,425 $48,808 $0 $1,035,784
Totals (per year): $35,499 $68,307 $270,745 $612,425 $48,808 $0 $1,035,784
Cumulative Totals: $35,499 $103,806 $374,551 $986,976 $1,035,784 $1,035,784 $1,035,784
Average Annual Interest Rate: 5.500% 4.938% 5.250% 3.188% 2.563% 2.000%
Interest on Prior Years Findings: $0 $1,753 $5,450 $11,939 $25,291 $10,358 $54,791
Current Years Interest: $976 $1,686 $7,107 $9,761 $625 $0 $20,155
Total Cumulative Interest $976 $3,439 $12,557 $21,700 $25,916 $10,358 $74,946
Through June 30, 2012
Appendix
May 21, 2012
Chief
Community-Rated Audits Group
U.S. Office of Personnel Management
Office of the Inspector General
800 Cranberry Woods Drive, Suite 270
Cranberry Township, Pennsylvania 16066
Re: Response to draft audit report of Union Health Service, Inc., Chicago, Illinois,
Report No. 1C-76-00-12-006
Note: This document references supporting information accessible via hyperlinks. It is recommended
that the document be viewed from a computer file opened from the accompanying CD or from a
hard drive in which all of the accompanying files are stored in the same folder.
Dear :
Deleted by OIG – Not Relevant to the Final Report
The above paragraphs quote statements by OIG in 2007 and I agree with them. Our circumstances have
not changed since then regarding the selection of SSSGs.
The various observations described in the draft report are not significantly challenged in this response, but
the conclusion is. I agree with most of the numbers although a few differences will be mentioned and I
will identify some items that were apparently overlooked. But in general, I believe the reported
observations are insufficient to support the draft recommendations. My main interest now is to present
additional information that I think is essential for correctly assessing our overall compliance with OPM’s
rating guidelines and objectives. This response will focus on the following areas that I believe are most
relevant.
• The suitability of the selected SSSGs
• Observations with the draft calculations and reported findings
• Other relevant information for supporting a conclusion
• Our conclusion and recommended actions
Appendix
Selected SSSGs
We were surprised that OIG selected and We initially expressed our reservations with the
auditor-in-charge and he encouraged us to not be overly concerned at that point because the issues we
identified should be addressed at another level in the process. I inferred that he was basically saying, in
so many words, “I will proceed with the general audit template we routinely use, and at a later stage you
can work with others to determine how the findings should be interpreted and applied”. I was
comfortable with that approach believing that we should reach the same conclusions regardless of what
was selected for SSSGs because all premiums are calculated from common base capitation rates
(applicable to either a “clinic-only” or a “clinic and non-clinical” benefit package). However, I now see
two developments that I did not expect —
• We made two administrative mistakes over the years; (1) our billing department failed to bill the
newly calculated rates for one of the selected SSSGs in one year, and (2) I erred in adjusting the
step-up factors for one selected SSSG when we eventually started combining the two small
groups for rating purposes.
• I did not foresee OIG extrapolating such heavy conclusions from the limited observations.
Suitability of the selected SSSGs:
Deleted by OIG – Not Relevant to the Final Report
The selected SSSGs violate requirements of OPM’s guidelines.
• Neither group ensures that “OPM receives an equitable and reasonable market-based rate”. The
groups do not shop their premium; they have no negotiating clout; there is no employer-paid
premium contribution for and the current active employees are excluded from
joining the Plan (retirees only are enrolled with us).
• The members are not employees and, further, would satisfy OPM’s definition of a
“Purchasing Alliance” (requiring exclusion of the group as an SSSG). OPM’s guidelines exclude
Purchasing Alliances with less than 100 enrollees from eligibility as an SSSG.
In addition to OPM’s published guidelines, other reasons to exclude and as eligible SSSGs
include:
• New members are not being added to these groups. The groups have been reorganized; was
in receivership and is now administered by another union. Although we have maintained
the "group" structure for administrative purposes (list billing), the remaining members are, in
substance, individuals that otherwise we would be obligated to cover with the same benefits under
state required individual conversion privileges (one member has a dependent with a serious pre
existing condition).
• Our underwriting guidelines, like most carriers, will not allow us to enroll a group of less than 10
employees without individual underwriting.
• Most of the subscribers have Medicare coverage.
• Extrapolation of measurements from an SSSG (e.g., SSSG variance values) to estimate
corresponding amounts in a target population (e.g., FEHBP corresponding values) requires that
the SSSG has a statistically credibility database (this is not the case for very small groups). The
implications are extreme. The sensitivity of this issue can be seen by going to an Excel
Appendix
spreadsheet that shows the change in premium for resulting from a slight change in the
demographics. Change the value in cell C27 from 1 to zero (eliminating the one family of three
members). The calculated premium yield PMPM (capitation) shown in cell H30 drops from
to , a decrease of more than 20%. The annual premium of the FEHBP is roughly
$5 million dollars. A refund of 20% of the FEHBP premium for a single year would be
approximately $1 million (20% of $5 million). The combined annual premium of both SSSGs is
roughly $65,000. Clearly something is wrong with an approach that could cause an adjustment
equal to fifteen times the SSSG’s annual premium if the SSSG merely loses one family
subscriber.
Calculations and reported findings
Discount
Text from the draft report’s executive summary states,
“For contract years 2007, 2008, 2009, 2010, and 2011, we determined that the FEHBP's
rates were overstated by $35,499, $68,307, $270,745, $612,425, and $48,808, respectively.
More specifically, the Plan did not apply a similarly sized subscriber group (SSSG) discount
to the FEHBP's rates.”
In the Audit Findings and Recommendations section of the draft report, the first three sentences
of the findings paragraph for each year of the audit are based on the issue of discounts. The text
is:
“Our analysis of the rates charged to the SSSGs shows that received a ___ percent
discount and received a ___ percent discount. The FEHBP received a discount of
___ percent for contract year ____. Since the FEHBP is entitled to a discount equivalent
to the largest discount given to an SSSG, we recalculated the FEHBP rates using the ____
percent discount given to Local ___”.
OIG is using the term “discount” in the draft report as a catch-all word related to any and
all observable differences. But that is not the meaning constructed from OPM’s guidelines
and reconciliation instructions.
The context from OPM’s guidelines and instructions consistently implies that the term
“discount” is intended to have a common usage meaning that essentially indicates a willful
price mark-down made to create an economic benefit. The meaning of “discount” does not
include such things as routine errors (e.g., a procedural mistake within an organization to
properly bill a communicated price change) or normal statistical variation (e.g., sudden
enrollment changes in very small groups).
Evidence of what “discount” means within the context of OPM’s guidelines includes:
• The format of the OPM reconciliation form shows that many factors affecting rates
are distinct from what is labeled “discount” (e.g., benefit loadings, standard
loadings, brochure printing, and reconciliation adjustments from the previous year).
• The “Example of TCR/ACR Comparison Sheet” within “OPM’s Rate Reconciliation
Instructions” shows that the first-level step-up factor and the family/self ratio are
clearly distinct from what is labeled “Total Discount”.
• OPM’s Community Rating Guidelines indicate that rates cannot be changed once
they are proposed, but discounts can later be offered. It follows that discounts are
distinct from the other factors that go into establishing rates.
Appendix
Proper understanding of the term “discount” is not intended to be an argument of
semantics. It is central to an auditor’s effort to fairly apply marketplace accountability. It
would be an issue of fairness (or lack of) if there is a practice of denying the FEHBP an
advantage that is applied to benefit other groups; however, that is not what we’re dealing
with in these audit calculations. I believe no group except the FEHBP ever received a real
discount. Further, even if OIG takes the position that one or both of the two small groups
received a larger “discount”, it remains true that the vast majority of our members (from
) did not have a discount at all, according to OIG’s usage of the term.
FEHBP received favorable treatment relative to the marketplace.
Prior/Post Reconciliation Adjustments
The auditor’s work papers supporting the draft report include worksheets for calculating the difference
between the Plan’s rates and the reconciled rates established by the audit. The amounts shown for the
Plan's rates do not include the Plan’s reconciliation adjustments from either the prior or subsequent year.
Therefore, recognized differences that have been properly corrected via the annual reconciliation process
are ignored in the draft audit report.
Perhaps OIG is assuming that reconciliation adjustments offset one another from year to year and, thus,
can be ignored; in other words, reconciliation adjustments are mere timing differences. That assumption
might be reasonable (although not precise) if we were attempting to calculate the accumulated effect of
errors and corrections over a span of time. However, that is not what the auditor is attempting to
calculate. The auditor is attempting to independently define the rate components of each respective year,
add them together, compare the sum to the amount charged by the Plan, and define the difference as a
discount. Aside from my earlier discussion of the meaning of "discount", the effect of prior and
subsequent reconciliation adjustments must be included to recognize corrections already made by the
Plan.
Overlooked or incorrectly calculated items
The calculated rates per the audit do not include the cost of the smoking cessation benefits ($ per
member-per-month).
The calculated rates per the audit do not include the cost of printing the FEHBP brochure. The costs vary
slightly from year to year; it was $11,966 in 2010.
Deleted by OIG – Not Relevant to the Final Report
Prescription drugs benefit variance
The premium rates developed for all of the years covered by this audit were prepared without any loading
for the prescription drugs benefit. Prior to 2011, we did not have group-specific data showing the cost of
the group’s prescription drugs benefit for the FEHBP. During the years covered by the audit we did not
have groups with a prescription drugs benefit except (1) the comprehensive benefits offered to the
FEHBP, and and (2) a minimal benefit offered to covering only limited items on a
Appendix
restricted formulary of low-cost generic drugs. For the FEHBP we did not even have any claims data; all
of the prescriptions were dispensed directly from our clinic’s internal pharmacy, without generating a
claim. In other words, our total book of business for prescription drugs benefits was predominately
minimal generic program, the FEHBP’s comprehensive program, and the comprehensive benefits for
the two small groups that, together, were too small to produce any credible data.
We knew that the aggregate cost of all prescription drugs benefits was about in 2011 (even less in
the earlier years when the formulary was more limited). We also knew that the cost for the FEHBP
should be more because it covers brand-name drugs as well as many popular generics that were not on
ormulary, but we had no idea that the costs would be times more. We now know that the
prescription drugs benefit cost for the comprehensive program (FEHBP’s) was per-member-per
month, a figure OIG confirmed in the audit. This knowledge was learned only after we started
contracting with Walgreens to provide pharmacy benefit management (PBM) services. The PMB reports
provided the group specific data and we were surprised at what we saw. With the help of the physicians
on our Pharmaceutical and Therapeutics Committee, and with consulting input from Walgreens, we
reviewed the differences in the experience of (with the limited benefits) compared to the FEHBP
(with the comprehensive benefits). We now understand the FEHBP cost is not abnormal; commercial
groups with comprehensive benefits and conventional copayments commonly have expenses in the range
of per-member-per-month. The cost for the benefits is not comparable.
Question: What does all of this pharmacy history have to do with the FEHBP reconciliation?
Answer: Connecting this history with an understanding of how the baseline premium capitation
rates were developed shows that and (the two large groups not considered by
OIG as SSSGs) have been heavily subsidizing the FEHBP.
Our rating methodology builds the FEHBP premium from two baseline capitation rates, one rate for
services generally rendered in our clinics (e.g., physician services) and another for non-clinical benefits
(e.g., hospital services). The “clinical” rate is derived from a financial forecast prepared each fall that
determines the January renewal premiums charged to the large groups, and . All known or
estimated factors are loaded into the forecast model before locking in on the clinical rate in the final step.
The clinical rate is a plugged amount that will yield the desired level of profitability. If values loaded into
the forecast model for the FEHBP premium (earlier steps in the forecast) are inadequate because the cost
of prescription drugs is understated, the understatement will cause an offsetting increase to the capitation
rate for clinical services. Yes, that offset is charged to the FEHBP, too; nevertheless, the large groups
have borne the vast majority of the burden.
Medicare offset
OPM recognizes that it is important to adjust the premium for the impact of Medicare benefits offsetting
risk that otherwise would be payable by the Plan. Line 4 (c) of OPM's form for premium reconciliation is
entitled “Medicare loading”. Most commercial employers would not have retirees included in their group
with Medicare benefits (at least not to the extent as within the FEHBP group). However, as stated above
in the section entitled “Suitability of the selected SSSGs”, most of the subscribers in and
have Medicare coverage.
Medicare impacts the net claims cost with a very large reduction for the SSSGs ( & relative
to the FEHBP. Our analysis shows the combined SSSGs should have a Medicare loading, relative to the
FEHBP, of a negative PMPM; that is, the SSSGs risk to the Plan is PMPM less than the
risk of FEHBP due to the SSSGs having more help from Medicare. However, since the loading was not
built into the SSSG's premium, the SSSGs were effectively overcharged (all else being equal)
Appendix
relative to the FEHBP. A reconciliation of FEHBP and SSSGs should apply this observation as either an
increase in the SSSG's premium or a reduction of the FEHBP premium to make them comparable.
Premium stability and step-up factors
OPM understands the importance of price stability. And OPM understands the need to build safeguards
into the calculations that will protect against price spikes from year to year. This is why OPM’s Office of
Actuaries negotiates and administers contingency reserves with the plans. Establishment of a contingency
reserve is not feasible for groups such as and but at least we can avoid abrupt major changes
in demographic assumptions (e.g., caused by the routine enrollment or disenrollment of a few members,
perhaps only one family) and, thus, avoid whiplashing the premium up and down.
In most circumstances, it is important to use a group’s actual demographic data in developing step-up
factors. Otherwise, the blended premium per member can be distorted (unwittingly or by willful
manipulation). For traditional community rated premium, OPM's guidelines require the usage of group
specific data for the SSSGs if group specific data is used for the FEHBP. If the FEHBP and the SSSGs
were all groups large enough to have credible demographic data (a reasonably predictable distribution of
contract types), the guidelines would work as expected. The FEHBP is large enough, but the SSSGs are
not. We need to consider the implications as well as the intent of the guidelines and balance the
competing objectives of price stability and usage of actual data.
Unfortunately, I caused a complication by incorrectly entering a wrong value for the conversion factor in
2011. The auditor is correct that the capitation to single premium conversion factor used in 2011 must be
greater than if the ratio of single to family premium in a three-tiered structure is less than I
greatly regret that error; however, we must ask, “What now is the correct response to that mistake?” I
believe the auditor should either ignore the error with consideration to the entire context or should fully
adjust for all of the differences between the assumed and actual demographics for all years.
My preferred choice is to accept the error and move on rather than make adjustments for all prior years: I
say that because the demographic data is not sufficiently reliable with such small groups to extrapolate
aggregate premium adjustments to the much larger FEHBP group. Otherwise, the calculations for all of
the prior years in which this error was not made would show that that was significantly
overcharged (i.e., or PMPM overcharged in 2009). But that’s not my interpretation; I do
not view this as an over-charge to the SSSG because a very slight change membership could swing the
numbers radically in the other direction. Nevertheless, that would indeed be the picture if we look solely
to the actual demographics of a very small group each year. Further, looking solely to the actual
demographics each year would make it impossible to have any semblance of price stability for the small
groups. I think it makes much more sense to assure the demographics are reasonable with respect to the
FEHBP, they are generally consistent for the small SSSGs, and the charged premium is derived from a
common base capitation.
Other relevant information
Deleted by OIG – Not Relevant to the Final Report
Appendix
Deleted by OIG – Not Relevant to the Final Report
Conclusion and Recommended Action
I believe the audit observations and other supporting information indicate compliance with the rating
guidelines; the FEHBP has not been charged more than any other group for reconciled benefits. Although
I have argued that and are not groups eligible to be considered SSSGs, a corrected
reconciliation on a per-member-per-month basis would, nevertheless, show a pattern of charging less to
the FEHBP. Another approach for making a corrected reconciliation would be to retain the current
structure used by OIG but make additional adjustments as described in this response. Considerations for
these two approaches include:
1. A reconciliation of premium per member per month eliminates distortions caused by
demographics and calculated step-up factors: it exposes the main issue — that is, the actual
amount paid by and on behalf of each member.
2. Reconciliation with the current structure adopted by OIG would add adjustment(s) as described in
this response, but not every adjustment necessarily has to be developed. As I see it, the process
can stop once the conclusion is reached that the FEHBP rates do not exceed the SSSG rates for
reconciled benefits and no other foreseeable adjustment remains that could push the conclusion in
the other direction. I suggest starting with the most powerful and most persuasive adjustments;
such as, the described adjustments for Medicare loading or the costs of prescription drugs
subsidized by the large groups. Either of those issues, alone, should successfully complete the
Appendix
needed reconciliation. If not , I suggest taking the next items that are the easiest to address; such
as, inclusion of brochure co st. smoking cessation benefits, and correcting for a miscalculation of
the blended discount for 2010. After that if needed . I suggest applying the annua l reconciliation
adju stments from prio r years built into the rate proposals.
~reference for resolvin g tltis audit would take an approach that does not rely solely on" and.
• Although they do provide some minimal enlightenme nt they do not provid e sufficient grounds ~
DIG ' s draft conclusions. There are other approaches from reaclting the correc t conclusion. I do not have
a preference on tile general approach OIG should take. but I can identi fy the followin g options:
I . Abandon the curre ur structure and continue with tile SSSG selection u sed in OIG ' s prior audit
using the larger group s _ and _ _ Althoug h the benefit variances for these ~
would make the rec onc~more~ it is. neverthe less. feasible. _ and_
are valid group s and have the marketplace accountability pro perties that a~d for an SSSG.
2. Exp and the current structure of tile draft audit and. as previou sly done by OIG. use all four groups
as SSSGs ~ _ _ _ TIlis approach might show some rough spots
(dep ending~~i~op ted from tltis respon se). but the prevalent pattern will
be clear - favorable of tile FEHBP.
3. Conclude that a conventional SSSG audit is not the best means of assessing complianc e unde r tile
circumstances found for tins Plan . In rhis case. alrernarive tests should be given primacy. OIG
might fmd that a review of the Plan ' s historical rates relative to competitors' , or its filed medi cal
loss ratio . establish the best support for a conclusion.
• • •
My conclusion is significantly different from what the draft report shows. Yet mo st of our ob servatio ns
(not tho se focused on in tltis docume nt) are the same. TIle on-site auditors were courteous. professional .
and worked very hard on the report; for that I am grareful. But I think much of tile story here is that OIG
set out on a course that appeared to be relatively simple and straightforward, ration al. and likely to show
an outcome consistent with our long-renn track record. That is what I expected . too. However, we found
a few things we didn't expect (the phannacy variance. the conversion factor error with one group in 20 11.
and an administrative oversight in corre ctly billing the_ premium in 20 10). These issues required us
to look deepe r: and. as a result. we have now cOllllecte~ots on issues that were not previo usly so
apparent (disqualification of the selected SSS Gs. distinction between a discount and a differe nce for
another reason. Medi care loading. and MLR performance).
I hope this respon se enable s us improve our un derstandings.
Very trul y yours.
Audit of the Federal Employees Health Benefits Program Operations at Union Health Service, Inc.
Published by the Office of Personnel Management, Office of Inspector General on 2012-08-20.
Below is a raw (and likely hideous) rendition of the original report. (PDF)