Skip Maine state header navigation

Agencies | Online Services | Help

Skip All Navigation > PFR Home > Insurance Regulation > Dirigo Health Filing




Docket No. INS-07-900


Acting Superintendent of Insurance Eric A. Cioppa issues the following Decision and Order in this matter.


This adjudicatory proceeding was conducted by the Superintendent pursuant to 24-A M.R.S.A. § 6913(1)(C); the Maine Administrative Procedure Act, 5 M.R.S.A. chapter 375, subchapter 4; 24-A M.R.S.A. §§ 229 to 236; Bureau of Insurance Rule Chapter 350; and orders of the Superintendent in this matter.

On August 6, 2007, pursuant to 24-A M.R.S.A. § 6913(1)(B), the Board of Directors of the Dirigo Health Agency (the “Board” or “Dirigo”)1 filed its annual determination of:

the aggregate measurable cost savings, including any reduction or avoidance of bad debt and charity care costs to health care providers in this State as a result of the operation of Dirigo Health and any increased MaineCare enrollment due to an expansion in MaineCare eligibility occurring after June 30, 2004.

24-A M.R.S.A. § 6913(1)(A), amended by PL 2007, c. 1, Pt. X, § 1.

The purpose of this proceeding and hearing is for the Superintendent to review the Dirigo filing and “issue an order approving, in whole or in part, or disapproving the [Dirigo] filing.” 24-A M.R.S.A. § 6913(1)(C). The Superintendent is required to “approve the filing upon a determination that the aggregate measurable cost savings filed by the board are reasonably supported by the evidence in the record.” Id. Dirigo, as the moving party, has the burden of proving that its determination of aggregate measurable cost savings is reasonably supported by the evidence in the record.

The Superintendent previously interpreted “reasonably supported by the evidence” to refer to the totality of the evidence and not to any part of the evidence taken out of context. In re Review of Aggregate Measurable Cost Savings Determined by Dirigo Health for the First Assessment Year (“Year One Decision”), No. INS-05-700 at p. 2 (October 29, 2005). Furthermore, the Superintendent has stated that “reasonably supported” is not equivalent to a preponderance-of-the-evidence standard. Id. Dirigo does not have to prove that its chosen alternative is the best or only alternative supported by the record, nor does it have to show that its chosen alternative is the most reasonable, but rather Dirigo must show that the evidence in the record reasonably supports its alternative. Id.


The Dirigo Health Agency, through its Board of Directors, is a party to the proceeding. 24-A M.R.S.A. § 6913(1)(C). Other parties to the proceeding, pursuant to grants of intervention by the Superintendent, include the Maine Automobile Dealers Association Insurance Trust, the Maine State Chamber of Commerce, the Maine Association of Health Plans, and Consumers for Affordable Health Care.

The Maine Automobile Dealers Association Insurance Trust (the “Trust”) explained that it is a multiple-employer welfare arrangement (MEWA) that provides health benefit coverage for approximately 3,200 employee participants and 5,800 insurable lives. The Trust asserted that it employs a third-party administrator (TPA) to manage and administer its health benefit programs. Under 24-A M.R.S.A. §§ 6913(2) and 6913(3), TPAs are subject to savings offset payments that could result from an approval in this proceeding of Dirigo's determination of aggregate measurable cost savings. The Trust further asserted that any such savings offset payments would be passed on by the TPA to the Trust and, therefore, that the Trust, its members, and their participants would incur higher health insurance costs, thereby making them substantially and directly affected by this proceeding. The Trust was a party in the underlying proceeding before the Dirigo Board, as well as a party to the Dirigo review proceedings conducted by the Superintendent in Docket Nos. INS-05-700 and INS-06-900.

The Maine State Chamber of Commerce (the “Chamber”) explained that it is a statewide business association representing large and small Maine businesses. Its members include businesses that provide health coverage for their employees through self-funded plans and insured plans, and the Chamber itself has an insured plan for its own employees. The Chamber asserted that the savings offset payment determined by the Superintendent would result in an assessment made against health insurance carriers, employee benefit excess insurance carriers, and TPAs that would have a significant effect on Maine's business community because every employer in Maine that provides health care coverage to its employees (whether self-funded or insured) would be affected. Although the savings offset payment would be paid directly by health insurance carriers, TPAs, and employee excess benefit insurance carriers, the Chamber further asserted that it is Maine employers and their employees that would ultimately pay the savings offset payment because carriers would have the ability to pass the savings offset payment on to employers in their premium rates and TPAs would have the ability to pass the assessment on to self-funded plans directly. The Chamber asserted that it has a substantial and direct interest in the proceeding. The Chamber was a party in the underlying proceeding before the Dirigo Board, as well as a party to the Dirigo review proceedings conducted by the Superintendent in Docket Nos. INS-05-700 and INS-06-900.

The Maine Association of Health Plans (“MEAHP”) explained that it is an incorporated association of health plans whose members are entities licensed by the Superintendent, including health insurers, health maintenance organizations, and third-party administrators. MEAHP asserted that pursuant to 24-A M.R.S.A. §§ 6913(2), 6913(3), and 6915 each of its members is required to pay savings offset payments which may be approved in this proceeding. MEAHP further asserted that the imposition of the assessment of the savings offset payment on paid claims of customers of its member companies would necessitate an increase in prices charged by the members to customers and potential customers and may result in loss of business due to such an increase. MEAHP asserted that each of its members is substantially and directly affected by the proceeding. MEAHP was a party in the underlying proceeding before the Dirigo Board, as well as a party to the Dirigo review proceedings conducted by the Superintendent in Docket Nos. INS-05-700 and INS-06-900.

Consumers for Affordable Health Care (“CAHC”) explained that it is the State's largest consumer health coalition whose mission is to advocate for affordable, quality health care with a membership that includes individuals as well as 37 businesses and organizations, with a collective membership representing the health care and coverage interests of over 200,000 Maine citizens. CAHC asserted that its members include (i) purchasers of health insurance coverage, including DirigoChoice, and (ii) insured and underinsured individuals and small businesses, and publicly insured individuals and families, in need of affordable coverage under DirigoChoice; and that these members' health insurance rates, subsidies, and/coverage may be affected by this proceeding. CAHC was a party in the underlying proceeding before the Dirigo Board, as well as a party to the Dirigo review proceedings conducted by the Superintendent in Docket Nos. INS-05-700 and INS-06-900.


Pursuant to 24-A M.R.S.A. § 6913(1)(B), Dirigo filed with the Superintendent its 2007 determination of aggregate measurable cost savings as well as supporting materials in the form of the administrative record generated in the proceeding before the Dirigo Board. This administrative record numbers over 7,000 pages and has been made available for public inspection at the offices of the Bureau of Insurance in Gardiner, Maine throughout this proceeding. All other filings made by the parties and the Superintendent's interlocutory rulings and orders have been posted throughout the proceeding to the Bureau's web page at for public access and inspection.

On July 19, 2007, the Superintendent issued a Notice of Pending Proceeding and Hearing, among other matters setting the intervention deadline and contingent hearing dates. The July 19th Order also included initial procedures for the conduct of the proceeding. By Order Setting Actual Hearing Date issued August 6, 2007, the Superintendent established September 10, 2007, as the date for the public hearing.

On August 6, 2007, the Dirigo Board, through its counsel, Assistant Attorney General William Laubenstein, submitted the Dirigo filing with the Superintendent. The filing consists of the Board's August 3, 2007, written decision and a copy of the complete administrative record of the proceeding before Dirigo, In re Determination of Aggregate Measurable Cost Savings for the Third Assessment Year (2008).

The Superintendent's Order on Intervention and Procedures, issued August 7, 2007, and Order on Intervention, issued August 10, 2007, granted intervenor applications made by the Trust represented by Bruce Gerrity, Esq.; the Chamber represented by William Stiles, Esq.; MEAHP represented by D. Michael Frink, Esq.; and CAHC represented by Joseph Ditre, Esq. The August 7th Order also included further procedures for the conduct of the proceeding in addition to those set forth in the July 19th Order.

On August 16, 2007, MEAHP filed a motion entitled Amended Party Notice to Dirigo Regarding Completeness of the Record, to which Dirigo filed a response on August 17, 2007. On August 21, 2007, the Superintendent convened a telephonic conference of counsel among his attorneys and counsel for each party to the proceeding. The Superintendent ruled on MEAHP’s August 16th motion in his Order Regarding the Record issued on August 30, 2007. In addition to ruling on MEAHP’s motion and thereby clarifying the status of certain disputed pre-filed testimonies as included in or excluded from the record, the Superintendent’s August 30th Order further identified certain documents contained in Dirigo’s August 6th filing that were not properly a part of the record because they were not offered and admitted at the Board’s proceeding, and excluded those documents from the record. By filing made on September 7, 2007, MEAHP objected to part of the Superintendent’s August 30th Order. At the hearing on September 10th the Superintendent noted MEAHP’s objections for the record. The Superintendent issued a Second Order Regarding the Record on September 12, 2007, and thereby included in the record an exhibit that was inaccurately excluded under the August 30th Order and reaffirmed the ruling in the August 30th Order on the inclusion or exclusion in the record of the disputed pre-filed testimonies.

On August 21, 2007, all intervenor parties filed separate briefs. Dirigo's brief was filed on August 29, 2007. All intervenor parties filed separate reply briefs on September 5, 2007.

On September 4, 2007, the Superintendent issued a Scheduling Order setting forth the procedure for oral argument at hearing and the order of issues to be addressed at the hearing.

The hearing was held in Augusta, Maine on September 10, 2007. The hearing was conducted entirely in public session. Counsel for each of the parties presented oral argument at the hearing. At the conclusion of the hearing, the Superintendent issued Hearing Questions for Citations to the Record, dated September 10, 2007, and established a deadline of September 12, 2007, for responsive filings by the parties. Dirigo filed its Response on September 12th as did MEAHP, the Trust, and the Chamber. CAHC did not file a Response.


The Dirigo filing attributes aggregate measurable cost savings to four identified topics, three categories of savings initiatives and another category labeled “overlap” that is intended to account for savings that are double-counted between certain of the savings initiatives. The table below identifies the four areas, the amount of savings and overlap approved by Dirigo as contained in its filing, and the amount of savings and overlap that the Superintendent finds reasonably supported by the evidence in the record.

Hospital Savings Initiatives $70.6 million $25.0 million
Uninsured / Underinsured Savings Initiatives $6.3 million2 $6.3 million
Health Care Provider Fee Savings Initiatives $5.2 million $1.5 million
(resulting in a reduction in the savings amount)
($4.0 million) $0.0 million3
TOTAL $78.1 million $32.8 million

A. Legal Issues

As explained in the Year One and Year Two Decisions, the Superintendent's statutory responsibility in these proceedings is limited to determining whether the “aggregate measurable cost savings filed by the board are reasonably supported by the evidence in the record.” 24‑A M.R.S.A. § 6913(1)(C). In making this determination, the Superintendent has the authority to “issue an order approving, in whole or in part, or disapproving the filing.” Id.

Thus, the Superintendent does not sit as an appellate tribunal with the authority to review Dirigo's interpretations of law or the conduct of the proceedings before the Board. Although the payor intervenors (MEAHP, the Trust, and the Chamber) have argued that they were denied due process by Dirigo's failure to produce its methodology and supporting data in a timely fashion, thereby denying the payor intervenors adequate opportunity to develop a record that would provide a more balanced and accurate view of the underlying facts, these issues are beyond the scope of this proceeding. As explained previously:

Certain intervenors also have asserted a variety of procedural defects or irregularities which they allege cast doubt on the reasonableness of the determination made by Dirigo. However, the Superintendent has not been granted the power by the Legislature ... to provide relief to parties who are aggrieved by such defects or irregularities. Instead, the Superintendent is tasked with a review of the evidence in the record to determine if that evidence reasonably supports the filing made by Dirigo. This limited jurisdiction over the determination made by Dirigo does not invest the Superintendent with the powers of the judicial branch, in this instance to rule on the legality of substantive and procedural decisions made by Dirigo, a separate executive agency, under its separate statutory responsibilities.

In re Review of Aggregate Measurable Cost Savings Determined by Dirigo Health for the Second Assessment Year (“Year Two Decision”), No. INS-06-900 at p. 6 (July 21, 2006).

Similarly, challenges to the admission and omission of evidence by the Dirigo Board at its proceeding are beyond the jurisdiction of the Superintendent, as explained more fully in the Superintendent's Second Order Regarding the Record issued on September 12, 2007. However, because the Superintendent does not have the authority to pass judgment on the admission of evidence before the Board, nor on the constitutionality or fairness of the process before the Board, a determination whether those allegations are of sufficient substance to disturb the Board's decision will not be decided herein.

In addition, a new legal issue has been raised by the change in Dirigo's methodology for calculating the savings due to uninsured / underinsured savings initiatives, which adds what is in effect another component of cost savings not recognized in past decisions. The benefit to the hospitals of providing health coverage to previously uninsured and underinsured individuals goes beyond merely providing compensation for services that previously went uncompensated; these new enrollees are now seeking treatment for some health problems that would previously have gone untreated entirely rather than being treated on an uncompensated basis. The net income from these services, as discussed in section IV(B)(2) below, is a source of “new money” that would allow the hospitals to reduce their charges to other payors without adverse financial effect. Whether this is properly regarded as a component of aggregate measurable cost savings is a question of law of first impression. As in previous years, the Superintendent finds that this issue is beyond his statutory jurisdiction and confines his review to determining whether the amount calculated is reasonably supported by the evidence in the record.

B. Dirigo's Determination of Aggregate Measurable Cost Savings

To assist it in developing a methodology for calculating aggregate measurable cost savings, Dirigo retained the consulting firm of schramm-raleigh Health Strategy (“srHS”).4 The recommendations by srHS were presented in a document entitled Report to the Dirigo Health Agency, Dirigo Health: Aggregate Measurable Cost Savings (AMCS) for Year 3 - Updated, dated July 20, 2007 (Dirigo record at pages 5301-5356, hereinafter “R. at ___”) (the “srHS Report”). The srHS Report determined the aggregate measurable cost savings to total $88.4 million. (R. at 5306.)5 The Board adopted all of the srHS savings initiative categories and the overlap category, but rejected aspects of the uninsured / underinsured and health care provider fee savings initiatives calculations, thereby approving a savings amount of $78.1 million. (R. at 1-10.)

1. Hospital Savings Initiatives. Dirigo determination: $70.6 million. Amount the Superintendent finds reasonably supported by the evidence: $25 million.

The hospital savings initiatives component of Dirigo's filing seeks to measure reductions in the cost of inpatient and outpatient services provided by hospitals and their subsidiaries, relative to what the cost would be in the absence of the cost containment initiatives established in connection with the Dirigo program. Although costs have continued to increase, when hospitals reduce the rate of increase in the cost of services they provide, it reduces their need to increase the rates they charge to commercial payors and results in savings to the entire health care system. The savings across hospitals and their subsidiaries were determined by srHS to be $70.6 million for state fiscal year (SFY) 2006. (R. at 5310-5313, 5321-5327, 5338-5346.) The Dirigo Board adopted this determination, approving $70.6 million. (R. at 4-7.)

The methodology employed by srHS evaluates hospital savings using estimates of the average cost per hospital discharge over time, as adjusted for the hospital's inpatient case mix and outpatient activity. Using this series of annual cost per case-mix adjusted discharge (cost per “CMAD”) estimates, the actual value for the assessment period was compared to a projected value for the same period to determine the rate of estimated savings per CMAD, which was then multiplied by the estimated volume of adjusted discharges. For this reason, the calculation is often referred to as the “CMAD” calculation.

According to srHS, it attempted to address the issues raised by the Superintendent in the Year Two Decision by adding SFY 2004 to the base period, by including only the net portion of the hospital tax, and by adjusting available hospital cost savings to remove savings related to the portion of hospital activity reimbursed on a cost basis. (R. at 5305-5306.) Costs per CMAD were calculated for years 1999-2006 across all hospitals. The savings were then calculated by comparing an estimate of the actual 2006 cost per CMAD to an estimate of what the 2006 cost per CMAD would have been in the absence of the Dirigo-related initiatives, obtained by starting with the 2004 cost per CMAD, adding back the $33.7 million hospital savings approved by the Superintendent in the Year One Decision, and projecting that figure forward to 2006. This projection was calculated by applying an inflation factor based on the Hospital Market Basket Index (HMBI) and a factor reflecting the degree to which the growth in cost per CMAD in Maine had historically exceeded the HMBI. The same Medicare cost report data used in Years One and Two were utilized, with the addition of 2006 data. For the Year Three calculation, discharge and case-mix data were obtained from the Maine Health Data Organization.

During the hearing before the Dirigo Board, the payor intervenors raised a number of objections to srHS's hospital savings calculation, the implications of which would reduce or eliminate any aggregate measurable cost savings associated with the hospital savings initiative. The Board's decision cites its belief that the report's attempts to address these issues “is reasonable and the data supports the results reached.” (R. at 6.) The Superintendent's review of the record includes consideration of the srHS Report, the debate about the objections raised by the payor intervenors, and the Board's consideration of these objections in approving $70.6 million, as well as subsequent arguments made before the Superintendent.

Payor intervenors asserted that the approval of $70.6 million for this year's savings amount was clearly inconsistent with the Board's decision last year to approve $14.5 million when the cost growth as measured by the srHS Report was very similar during SFY 2005 and SFY 2006. It is true that both the SFY 2004 to SFY 2005 and the SFY 2005 to SFY 2006 cost per CMAD growth figures are within a half a percentage point of each other (using the unadjusted SFY 2006 value). However, as noted in section I(A) above, the standard of “reasonably supported by the evidence in the record” only requires that the amount approved by the Board be reasonably supported, and the Superintendent is not required to determine that it is the most reasonable amount. As such, direct comparison between approved savings figures is not straightforward, especially when the judgment as to reasonable support for each of these figures is based on the information in the record during each respective proceeding. Nevertheless, the magnitude of the difference between the figures approved by the Board for Year Two and Year Three would make it unreasonable to accept the Year Three figure without an understanding of why these estimates are so different.

The Pre-Dirigo Base Period

The srHS Report moved the base period forward to include SFY 2004. According to srHS, the use of more recent data addressed one of the major shortcomings identified by the Superintendent in the Year Two Decision and responded to the guidance offered by the Superintendent. The report specifically quoted the following from the Superintendent's Year Two Decision: “...the further removed the year being measured from the base period, the more tenuous the connection and the more questionable the assumption that all subsequent changes are related to Dirigo.” (R. at 2990-2991.) The extension of the base period was also intended to address concerns that cost growth during SFY 2002 was abnormally high, and therefore an outlier value that should be excluded in some way or given limited effect to prevent distortion of relative cost growth pre-Dirigo. Those concerns were one reason the Board took the conservative step of substituting the median rate of growth in place of the mean in its Year Two cost projections, which resulted in a significantly lower hospital cost savings estimate than the one proposed by its consultant. According to srHS, adding additional data points and basing the projection on a longer time frame avoids giving undue weight to the 2002 growth spurt.

Payor intervenors objected that this choice did not draw on the most current data, and that if the base period also included SFY 2005, then a lower savings figure - zero by one estimate - would result from using the same methodology. Both srHS and the payor intervenors, however, overlook a fundamental flaw in the effort to extend the pre-Dirigo base period. There are not and cannot be any “observed pre-Dirigo” 2004 or 2005 data available to use for this purpose. The “new” data points proffered by srHS and by the payor intervenors are themselves nothing more than projections from the base period ending in 2003, based on trends observed during that base period. This is true because these data points were determined from actual costs by adding back the hospital savings approved by the Superintendent, which were in turn determined as the difference between actual costs and projections from the base period ending in 2003. Purporting to extend the base period by its own bootstraps by using additional projected data points adds no meaningful information. The pre-Dirigo base period ended in 2003 and cannot be prolonged. The tenuous connection cited in the Superintendent's Year Two Decision between historic and current cost per CMAD is a basic characteristic of the methodology. That connection becomes more tenuous each year due to the combination of the passage of time and the CMAD methodology's lack of control for factors unrelated to Dirigo. Time both diminishes the relevance of the available pre-Dirigo historical data and assigns an increasingly disproportionate dollar value to small variations in the trend rate chosen to project forward from 2003.

Furthermore, applying savings estimates from prior decisions to observed data, even if the appropriate adjustments had been made,6 is not a reasonable way to project what the costs for 2004 or 2005 would have been in the absence of the Dirigo initiatives. An amount approved by the Superintendent in a prior year that was judged to meet the legal standard of reasonable support is one among a range of estimates that could be reasonably supported, and in no sense should be judged as the best estimate analytically. As such, it is not a point estimate that could be added to actual measured cost to arrive at “cost that would have been” to be used in subsequent calculations. Thus, the projection methodology adopted by Dirigo is not reasonably supported by the evidence in the record, neither as proposed by srHS nor with the modifications recommended by the payor intervenors.

Effect of Cost-Based Reimbursement

The projection methodology is not the only significant flaw in the hospital savings estimate. In prior years' proceedings, the payor intervenors pointed out that some portions of reimbursement to Maine hospitals are cost-based, limiting the degree to which reductions in hospital expenses would be available to commercial payors in the form of reduced prices. The evidence in the record indicates that these components are (i) all Medicare and MaineCare costs for critical access hospitals (CAH), and (ii) MaineCare hospital outpatient costs in non-critical access hospitals. The failure to take this effect into account was cited by the Dirigo Board as one of several reasons to approve a lower hospital savings estimate than Mercer had proposed in the Year Two proceeding, and by the Superintendent in finding the Board's recommendation to make such a reduction was reasonably supported by the evidence in the record. The srHS Report submitted for this year's proceeding made an adjustment intended to take this effect into account. Costs, revenues, and discharges for the cost-based components of hospital activity were identified, and calculations were performed to remove this activity from the 2006 figures used for the cost/CMAD savings calculation.

Payor intervenors asserted in their arguments that an error was made in this calculation, and that correcting the error would eliminate any savings in cost per CMAD. (Chamber Brief, August 21, 2007, p. 19.) A review of the srHS Report, the applicable spreadsheet, the payor intervenors' briefs and the associated testimony suggest that there is an issue with the handling of the non-CAH outpatient MaineCare portion of the calculation, and that addressing the issue would in fact significantly reduce the estimated savings generated by the “virtual hospital model” if the error cited by the payor intervenors were corrected in isolation. However, a more basic issue stems from a fundamental flaw with the approach taken to cost-based reimbursement adjustment, which invalidates both the adjustment in the srHS Report and the elimination of savings asserted by the payor intervenors related to the alleged error in the adjustment. By adjusting only 2006 data, and therefore comparing base-period expenses containing cost-based reimbursement activity against SFY 2006 data with these expenses removed, the result is an “apples and oranges” comparison of costs. The record does not contain all the data required to adjust the base period for any years other than 2006 to remove the cost-based activity. If such data were available, incorporation of it into the calculations would be a complicated endeavor, which would be complicated further by the significant increase in the number of hospitals obtaining CAH designation over the period included in the srHS data. These issues highlight the problematic nature of the method chosen by Dirigo for measuring hospital savings.

The spreadsheet setting forth srHS's hospital savings calculation contains information that suggests that the fraction of hospital expenses associated with patient care that is reimbursed on a cost basis in SFY 2006 is approximately 12%. This amount suggests that the size of adjustment made for the cost-based reimbursement effect (estimated to be approximately $22 million by backing the adjustment out of the spreadsheet), is larger than the proportion of costs represented by cost-based reimbursement, which is approximately $11 million in 2006 (12% of the sum of $70.6 million and $22 million). Thus, although the adjustment was not methodologically sound, it embodies an estimate of the effect of cost-based reimbursement that is sufficiently conservative to be reasonably supported by the evidence in the record.

Recoverability of Savings

Payor intervenors argued that it is unreasonable to assume that hospitals that are losing money or have weak operating margins can be expected to provide rate relief to commercial payors when the benefits of cost reductions are needed to improve their poor financial health. The Superintendent notes that the Board treated the issue of whether savings are recoverable differently for physician fees than it did for hospital savings, and differently than it did in the prior years' proceedings. It is therefore appropriate to consider the extent to which the calculated savings are attributable to hospitals with negative or poor margins. In both 2004 and 2005, the percentage of hospital expenses in hospitals with margins below 1% was 16%. (R. at 5876.) It is reasonable to assume that hospitals with margins below 1% could not be expected to generate recoverable savings, and reasonable to apply the estimated 16% to the Board's approved savings of $70 million. This reduction for lack of recoverability would yield an approximate $11 million reduction to savings. While this is not the only possible amount that could be used, since arriving at an objective standard for minimum financial health would be difficult, it is important to note that any reasonable allowance for this consideration would reduce the estimated savings figure significantly.

Effects of Outpatient Charges

Another argument raised by payor intervenors is related to the distortion in the cost per CMAD calculation caused by its dependence on inpatient and outpatient “charges,” where charges refer to gross revenue, or the total amounts charged to payors based on list prices. The denominator in the cost per CMAD measure includes an “outpatient discharge equivalent” (ODE) that is calculated as [outpatient charges / (inpatient charges/inpatient discharges)]. In effect, it assumes that the outpatient activity with charges in an amount equal to the charges for one inpatient discharge is hospital output equivalent to one hospital discharge. As a result, increasing hospital outpatient prices over time, other things being equal, would increase the measured outpatient activity and associated inpatient discharge equivalents, therefore inflating the denominator in cost per CMAD and understating cost per CMAD and therefore overstating savings.

Payor intervenors pointed to the effect of outpatient charge increases on ODE and cost per CMAD, and to the increases in hospital outpatient charges in the data in the record, as proof that this growth distorts the cost per CMAD measure. However, they referred to total hospital outpatient charges, which increase for one of three reasons, only one of which produces a clear distortion in the ODE part of the cost per CMAD denominator. Outpatient charges can increase because of (i) outpatient volume increases; (ii) outpatient service intensity increases (e.g., outpatient angioplasty); and (iii) increases in the list price of outpatient services. The first two of these factors measure valid increases in hospital output; only the third causes the distortion noted by the payor intervenors. Increases in total charges reveal nothing about the way in which hospital price levels may be changing, particularly in an environment in which both hospital outpatient activities and intensity levels are understood to be increasing.7 Furthermore, other things being equal, hospital price changes will only distort the cost per CMAD measure if outpatient prices change at a different rate than inpatient prices. If both go up by the same percentage, the CMADs are unchanged. Only if outpatient prices increase more quickly than inpatient prices will the CMAD be inflated and thus the cost per CMAD deflated. There is no evidence in the record that allows the Superintendent to understand how Maine hospital inpatient and outpatient prices have changed. Therefore, this threat to the validity of the CMAD measure may be only theoretical and not actual.

Testimony during the Board hearing indicated that hospitals may be increasing hospital outpatient prices faster than inpatient charges. (R. at 240-242.) However, this testimony was based on anecdotal evidence from other states and conjecture about practices in Maine. At the same time, the change in the rate of growth in CMAD would be significantly affected if outpatient prices grew more quickly than inpatient prices. The effect on the annual rate of change in CMAD would be slightly less than half the relative increase in outpatient price levels. For example, if outpatient prices increased by 2 percentage points more than inpatient price levels, the rate of change in CMAD would be affected by slightly less than one percentage point. The size of this margin is very relevant given that the difference in pre-Dirigo and post-Dirigo growth relative to the HMBI is approximately 2.3%. The ambiguities in the output measures are a significant concern, ameliorated to some degree by the fact that the pattern of growth in absolute hospital costs (without dividing by a discharge or output measure) displays a similar slowing pattern pre- and post-Dirigo. (R. at 5342.)8 The potential effect of this measurement issue on the cost per CMAD calculation encourages conservatism in estimating savings.

MaineCare Reimbursement Cuts

As in past years' proceedings, payor intervenors objected to the failure of the cost per CMAD methodology to take into account other potential factors that influence cost growth in Maine hospitals. The method's failure to do so was cited as a reason why savings related to hospital cost growth should be disapproved in their entirety. As a general matter, the passage of time makes the failure to control for other factors affecting costs increasingly problematic for the method, and at the same time makes other methods which control for these factors, such as multivariate multi-state analyses, more feasible. The cost per CMAD method will soon reach the point at which its drawbacks prevent it from producing reasonably supported findings. While small random fluctuations in cost will occur in any year, the more serious threats to the persuasive power of this method would be raised by specific, additional credible explanations for the pattern of slower cost growth post-Dirigo.

The most compelling allegation that a significant factor was ignored by the method arises from a revenue reduction caused by cuts in MaineCare reimbursements to hospitals in SFY 2004 and SFY 2005. These cuts included a cut in the MaineCare outpatient reimbursement from 100% of costs to 77% of costs, and the elimination of inflation increases for inpatient payments, which were estimated to amount to $24.5 million and $33.7 million respectively. (Chamber Brief, August 21, 2007, p. 15.) The Chamber asserted that hospitals would respond either by reducing costs (thus affecting cost per CMAD directly) or by increasing charges. (Id. at 15-16.) The latter was asserted to reduce cost per CMAD by increasing the “outpatient equivalent discharge” portion of the denominator. As discussed above, increases in charges can influence this measure either upward or downward, depending on the relative change to inpatient vs. outpatient price levels, and an across the board percentage price increase would not affect the cost per CMAD measure at all.

Dirigo asserts testimony this year and last year shows that payment cuts were offset by payment increases. (Dirigo Response, September 12, 2007, p. 1.) Testimony from the same portion of the hearing (R. at 2086-2089), however, indicates that the payment increases cited by Dirigo have already been attributed to the Dirigo initiative in prior years' proceedings. The cited testimony also asserts that payment cuts must be a product of the Dirigo initiative itself to be relevant to estimates of Dirigo initiative savings. On the contrary, any factor affecting costs to hospitals statewide that is unusual (as opposed to the normal fluctuations at individual hospitals from year to year) and significant in size relative to the estimated cost savings has the potential to undermine the assumptions of the cost-per-CMAD method. The Superintendent's Year Two Decision stated that this issue was a factor in the evaluation of the savings estimates for Year Two: “In particular, the $58.2 million reduction from the Mercer estimate redresses...the failure to consider potential Medicaid payment cuts...” (R. at 2983.) To the extent that MaineCare payment reductions resulted in expense reductions, they would directly influence the cost per CMAD measure downward. To the extent that they were responded to by increases in price levels, the effect may or may not have been to depress cost per CMAD. Furthermore, there is no evidence in the record about the degree to which hospitals responded with price changes vs. cost reductions vs. reductions in profit margin. At the same time, the payment reductions were sizable and pose a plausible alternative explanation for part of the slower observed cost growth. If one assumed that the entire $25 to $30 million payment cut per year was fully reflected in a cost per CMAD reduction, the $70 million figure approved by the Board would be reduced by a comparable amount. The actual size of this effect cannot be determined from the record, but cost per CMAD reductions related to the reimbursement cuts are plausible and would be substantial even if they were a fraction of the payment cuts.

Observed Savings

The discussion above highlights severe shortcomings in srHS's methodology. Nevertheless, sufficient post-Dirigo experience has now developed to identify a significant change in cost growth trends pre- and post-Dirigo. This year's cost savings calculation has added a cost per CMAD estimate for 1999, allowing calculation of the 1999 to 2000 growth rate, which was 6.8% (from $4,572 to $4,882). Calculating the geometric mean growth rate from 1999 to 2003 results in an average rate of growth in cost per CMAD of 5.9% (from $4,527 to $5,739), while calculating the same rate of growth from 2003 to 2006 results in an average rate of growth of 4.1% (from $5,739 to $6,481; 3.7% if the cost-based reimbursement adjustment is removed). The addition of the 1999 year, if the basis of the cost per CMAD calculation is sound, adds credibility to the argument that cost growth pre-Dirigo was higher than the cost growth post-Dirigo, which would lend credibility to the conclusion that savings have been generated by the hospital savings initiative, even if they do not appear to be of the magnitude asserted by srHS.


The Superintendent's Year Two Decision approved the “rough justice” approach taken by the Dirigo Board in its reduction to the SFY 2006 hospital savings proposed in the Mercer report. This year the Dirigo Board judged that the modifications to the hospital savings methodology adequately addressed the shortcomings in the method, and approved $70.6 million, a number similar to that put forward by Mercer in Year Two. The Superintendent finds this amount is not reasonably supported by the evidence in the record.

At the same time, evidence in the record supports a finding that some savings are related to the hospital initiative. The simple trend in hospital costs, especially with the addition of the 1999 data, indicates that cost growth pre- and post-Dirigo has been different, and has been lower relative to the HMBI after Dirigo. Furthermore, two significant shortcomings in the Year Two analysis (treatment of hospital tax and cost-based reimbursement effect) have been adequately adjusted for in the analysis presented to the Dirigo Board. Thus, a total disallowance of savings is not warranted.

Nonetheless, there are serious problems with the cost per CMAD methodology that compel interpretation of the evidence with a great degree of conservatism to allow for the potential measurement error. While some shortcomings in the methodology have been rectified, a review of the record shows that others have not been addressed and have become worse with the passage of time. Also, some shortcomings dismissed in the Year Two Decision are now supported by additional information in this year's record. Specifically, the Board's determination does not consider the potential effect of hospitals with negative margins (the estimate above indicated $11 million as an effect of one assumption about this issue), the effect of cuts in MaineCare (up to $30 million), and the difficult to quantify, but potentially important, issue of rising hospital outpatient prices artificially deflating the cost per CMAD figures over time. In addition, it does not reflect the almost $4 million overlap between hospital and periodic interim payment (PIP) savings and the unquantified overlap between hospital and uninsured / underinsured savings, both of which are discussed in section IV(B)(4) below.

Because the record continues to support a finding that Dirigo's Year Two decision was conservative, especially since the reported cost per CMAD figures for SFY 2005 have turned out to be lower than the preliminary figures used in developing the Year Two estimate (R. at 39-40, 5326-5327), the Superintendent finds that an amount of approximately $15 million represents a lower bound on the savings in this category. On the other hand, the Superintendent concludes that the three major issues cited above (MaineCare cuts, negative hospital margins, ODE-related distortions to the cost per CMAD) average at least $10 million per issue. A roughly $30 million reduction, along with the overlap correction, would produce an upper bound of approximately $35 million in savings. The Superintendent finds that neither end of this range is reasonably supported by the evidence in the record. The Superintendent further finds that an estimate of savings in the middle of this range appropriately balances these issues and accordingly finds that $25 million in hospital savings is reasonably supported by the evidence in the record.

The critical sources of concern in the methodology, which have been pointed out since Year One, have not been redressed sufficiently. The Superintendent recommends that a new analytical approach be taken in any future proceedings. In particular, given that any proceeding assessing SFY 2007 would allow assessment with four years of post-Dirigo data, it would be both more feasible and more desirable in the future to assess Dirigo's effect with a multivariate statistical model which incorporates cost and other data from both Maine and from other states.

2. Uninsured / Underinsured Savings Initiatives, including new money from populations now enrolled in DirigoChoice and from MaineCare Parents Expansion. Dirigo determination: $6.3 million. Amount the Superintendent finds reasonably supported by the evidence: $6.3 million.

The uninsured / underinsured savings initiatives component of Dirigo's filing seeks to quantify the amount of new money in the health care system coming from persons who were previously uninsured or underinsured and are now enrolled in DirigoChoice or the MaineCare Parents Expansion. Once insured, the previously uninsured and underinsured receive more health care coverage and providers are reimbursed for their existing services as well as the additional services. This new money to the health care system reduces the amount of bad debt and charity care in the system and adds new dollars to the system, both of which reduce the need for cost-shifting to private payors, resulting in savings available to the system. The savings were determined by srHS to be $10.3 million for new money from populations now enrolled in DirigoChoice and $3.7 million for new money from the MaineCare Parents Expansion, for a total of $14.0 million. (R. at 5314-5315, 5328-5331, 5347-5354.) The Dirigo Board adopted this determination in part, and approved $6.3 million. (R. at 7-8.)

In Years One and Two, Dirigo's consultant developed estimates of this savings initiative based on a process that used the total amount of bad debt and charity care reported by Maine hospitals as a starting point, and then developed various assumptions to estimate the reduction in this amount that could be attributed to enrollment of a portion of the uninsured and underinsured population into DirigoChoice and MaineCare. In Year Three, srHS revised this approach and derived its estimate of this savings from the actual claim costs of members enrolled in DirigoChoice and MaineCare Parents Expansions. They characterized the result of this approach as an estimate of “new money to the healthcare system for those previously uninsured and underinsured that are now enrolled in DirigoChoice or in the MaineCare Parents Expansion programs.” (R. at 3296.)

According to srHS, this change in methodology relied in part on deliberations by the Bad Debt/Charity Care (BDCC) working group. This working group included representation from Dirigo, the Chamber, MEAHP, and CAHC. (R. at 383, 6456.) Apparently, there was general acquiescence by the working group that measurement of claims could be an appropriate methodology for calculating savings. There is some dispute about the context of that acquiescence, but for the purposes of the AMCS determination,9 the method and number are closely enough related to this category of savings that the Board could find this method reasonable and could rely on the payor intervenors' participation in the development of the methodology as an indicator of its reasonableness, despite the difference in context of those discussions.

According to srHS, the savings estimate was developed using data and assumptions about the following (R. at 3297):

1. Average PMPM Claim Costs for DirigoChoice and MaineCare Parents Expansion members.

2. The percentage of the new enrollment that was previously uninsured or underinsured.

3. The turnover (“churn”) of the population of Maine from insured status to uninsured status and back over time.

4. The portion of the claims that represents “new money to the health care system,” i.e., claims that will be paid by the member or the plan that would not have been paid in the absence of Dirigo because of the members' uninsured or underinsured status.

5. Medical trend to 2007.

6. Projected membership during 2007.

MEAHP's consultant Jack Burke, FSA, MAAA, of Milliman, identified several issues and concerns about the new srHS methodology in a report included in his pre-filed testimony. (R. at 6768.) In this report, Mr. Burke offered an alternative analysis, preserving many of the srHS assumptions and methodologies and revising others. These revisions included the following:

1. An adjustment for induced utilization.

2. An adjustment to reflect the claim level of previously uninsured and underinsured members of DirigoChoice and MaineCare Parents Expansion relative to the entire Dirigo-related enrollment of those programs.

3. An adjustment for other potential payors of claims of the uninsured, including charities and other insurance (homeowners, auto, worker's compensation).

4. An adjustment for the variable costs to providers for the incremental health care received by these members due to the fact that they are now insured rather than uninsured.

These assumptions resulted in a revised estimated savings of $7.9 million. (R. at 6788.) Based on the testimony during the Board hearing, Mr. Burke further revised his estimate to $6.3 million, based on reducing the assumption about the portion of DirigoChoice and MaineCare Parents Expansion enrollment that was previously uninsured. (R. at 467, 6958-6961.)

During the Board hearing, the Trust raised additional concerns about the method srHS used to estimate the number of enrollees who were previously underinsured, given the number of people who failed to respond to the relevant question in the survey upon which srHS relied. (R. at 441.)

The Board adopted Mr. Burke's estimate of $6.3 million, which addressed MEAHP's stated concerns before the Board.

In briefs submitted in the Superintendent's proceeding, the payor intervenors continue to raise several challenges to this determination. The Chamber continued to allege that the definition of uninsured was overly broad and also asserted that the methodology employed by srHS failed to comply with the “plain language” of the Dirigo statute. (Chamber Brief, August 21, 2007, p. 22.)

Regarding the issue of the definition of uninsured, there is some evidence in the record that it may have been more appropriate to assume that 36.5% of the 2005 DirigoChoice enrollment was previously uninsured rather than the 44% assumed by the srHS calculation before it was modified by Mr. Burke. (R. at 6253.) However, Mr. Burke clearly adjusted for this when he derived the estimate of $6.3 million. (R. at 467, 6958.) It is also possible that there may be a degree of conservatism in other assumptions, such as the “churn” factor employed by srHS, that adequately offsets any overstatement of savings due to the definition of uninsured used by srHS. The Chamber's brief does not provide any citation pointing to specific evidence that the effect of the srHS definition of uninsured was not adequately adjusted in Mr. Burke's estimate.

Furthermore, even if such evidence existed, the Superintendent would be unable to conclude that the definition of uninsured, as modified by Mr. Burke's adjustment, was unreasonable without an analysis of the effect of a revised definition of uninsured. This analysis would include citations of the evidence and supporting data. All parties to the proceeding would have an opportunity to review and comment on this analysis. All parties agree that there is no such analysis in the record. (Party Responses, September 12, 2007.) The record is not entirely consistent on this point, but the evidence suggests that an appropriate adjustment to correct for the definition of uninsured was part of Mr. Burke's $6.3 million estimate.

To measure the effect of the Dirigo initiatives on bad debt and charity care, one does not have to start from aggregate bad debt and charity care data. The statute requires AMCS to include an estimate of bad debt and charity care that is avoided, but it does not require that total bad debt and charity care reported in hospital financial statements be the starting point for deriving this estimate. The basic assumption in the srHS Report, as modified by Mr. Burke, is that a portion of claims for services that are not paid by either the member or the plan or other payors will become bad debt. The Board adopted this methodology. The Superintendent finds this approach to be reasonably supported by the evidence in the record.10

In calculating the percentage of new enrollees who were formerly underinsured, srHS assumed there was not a statistically significant difference between the prior deductible level of the people who responded to the deductible question and those who did not respond to that question. The Trust challenged this assumption, arguing that it was not reasonable to do so and would be reasonable instead not to count those who did not respond to this question, effectively making the opposite assumption. The Superintendent is unpersuaded by the Trust's argument, finding srHS's approach to be reasonable.

The estimate formulated by srHS was based on a combination of different studies of the behavior of different populations over different time frames. The review by the payor intervenors led to the revisions summarized in Mr. Burke's report, which built upon the srHS estimate and improved it.

One final test of the overall reasonableness of this result is that it is not inconsistent with the $5.5 million amount found reasonably supported in Year Two, when adjusted for growth in enrollment during the intervening year and for the addition of a new category of saving within this initiative.

The Superintendent finds that Dirigo's approval of $6.3 million for uninsured / underinsured savings is reasonably supported by the evidence in the record.

3. Health Care Provider Fee Savings Initiatives, including hospital fee initiatives and physician fee initiatives. Dirigo determination: $5.2 million. Amount the Superintendent finds reasonably supported by the evidence: $1.5 million.

The health care provider fee initiatives measure cost savings from reductions in cost-shifting that result from increased funding to hospitals and other providers. As srHS observes, hospitals and other health care providers meet their annual financial requirements using a variety of funding sources. Over the long term, differences between financial requirements and payments by various payors may be shifted to private sector payors, whose rates are negotiable (unlike the public sector - Medicare and MaineCare - where rates are determined by the public payors), resulting in higher rate increases to private payors. The State will make payments early or additional payments to hospitals and physicians as a result of the Dirigo Health Reform Act and its related initiatives, to recognize differences identified by the Maine Hospital Commission in its review of the funding of the MaineCare (Medicaid) program. Thus, the need for cost increases to other payors will be reduced when this additional cash is received or received earlier than planned by hospitals and physician providers, resulting in savings to the health care system. The savings were determined by srHS to be $3.7 million for hospital prospective interim payment (PIP) increases and $4.1 million for physician fee increases, for a total of $7.8 million. (R. at 5316-5317, 5332-5333, 5355.) The Dirigo Board adopted this determination in part, and approved $5.2 million. (R. at 8-9.)

(a) Periodic Interim Payments.

The Mercer reports for Years One and Two estimated savings attributable to the accelerated receipt of PIP payments for the MaineCare program. According to those reports, the budgets for SFY 2006 and SFY 2007 included significant increases for these PIP payments relative to the state budget for SFY 2005. PIP amounts were found to be reasonable based on the record of the proceedings in Years One and Two. Based on evidence in the records of those proceedings, these increases were Dirigo savings and represented the increase in the present value of PIP payments as a result of making them three years earlier than they would have been in the absence of Dirigo.

The present value of the early PIP payments in the SFY 2006 budget was split evenly between calendar year (CY) 2005 and CY 2006, since SFY 2006 spans those two calendar years. In the same manner, the present value of early PIP payments in SFY 2007 was split evenly between CY 2006 and CY 2007. So, for example, the amount of savings measured for Year Two was half of the amount for SFY 2006 and half of the amounts for SFY 2007, which are the two fiscal years that cover CY 2006, the Year Two measurement period for this savings component.

In Year Two, Mercer erroneously computed a future value as of the end of the three year period rather than a present value as of the beginning of the three year period. This error was corrected in the Superintendent's Year Two Decision. With this one correction, the Superintendent found that the amount calculated by Mercer was reasonably supported. (R. at 2989.)

In order for these to be valid savings that can be recognized as AMCS, two conditions must be met: first, the increase must be other than normal PIP growth, and second, the savings must not be captured in the hospital savings initiative category. The Superintendent, for the reasons described below, finds that neither of these conditions is met based on the evidence in this year's record.

In Years One and Two, there was no evidence in the record to contradict Mr. Schramm's imbedded assumption that the increases in these PIP payments in the SFY 2006 and SFY 2007 budgets were Dirigo savings absent any other plausible explanation. Also, there was insufficient evidence presented to contradict the expectation that the financial effect of these early payments would most likely be recorded as investment income in the hospital financial statements and accordingly, not captured in any other savings component.

For Year Three, srHS has estimated savings for early payment of the PIP attributable to the second half of SFY 2007. The principal amount for SFY 2007 was the same as used last year, but the interest rate was updated to reflect the interest rate on a 36-month Federal treasury instrument as of the beginning of April 2007. The srHS calculation did not include any equivalent amount for the half of SFY 2008 that covers the second half of CY 2007. The amount computed by srHS was $3.7 million, which was adopted by the Dirigo Board. (R. at 3300, 3322.)

The Chamber introduced Exhibit 10B summarizing PIP payments over six years of state budgets, both pre- and post-Dirigo. This exhibit contains evidence not on the record before the Superintendent in prior years. Exhibit 10B demonstrates that these increases reflect the normal growth of PIP over time; growth that is in line with the overall growth in hospital expenses due to increases in volume and unit costs over time. For example, PIP growth from SFY 2002 to SFY 2004, a pre-Dirigo period, was 39%, while PIP growth from the SFY 2005 to SFY 2007 period that was analyzed by srHS was only 24%. In these calculations, PIP growth is based on net PIP revenue, taking into account the effect of the tax component of the “tax and match” program.

There was some discussion by the payor intervenors of the contribution of the tax portion of “tax and match” to the growth in PIP measured by srHS from SFY 2005 to SFY 2007. However, Chamber 10B shows that this component of PIP was represented in all three SFYs 2005 - 2007 at a mature and fairly constant level. The significant growth in this item was from SFY 2004 to SFY 2005 and did not contribute to the growth measured by srHS as implied by the payor intervenors.

Another valid criticism of this category was raised by Roland Mercier, a consultant for the Chamber and a principal at Baker Newman and Noyes. Before the Board he provided testimony describing how the financial effect of this increased present value would be reflected in hospital financial statements. (R. at 520-522.) According to Mr. Mercier, any investment income earned by hospitals would first be used to offset interest expense. Since interest expense is a component of the expenses measured in the CMAD calculation, any reduction in this expense would be ultimately reflected as a lower rate of growth in costs per CMAD and would be captured as savings by the measurement of that component. According to Mr. Mercier, this would represent double-counting.

The payor intervenors also asserted that the Superintendent's Year Two Decision failed to recognize this potential for double-counting when PIP savings of $6.1 million were approved. They assert that $3 million should be deducted from any CMAD finding in this year's review to avoid double-counting of the same savings. (R. at 5764.) This issue is addressed in section IV(B)(4)(b) below.

There is insufficient evidence in the record to support any conclusion other than that the growth of PIP payments from SFY 2005 to SFY 2007 is the normal growth in hospital spending due to cost and volume increases over time. In addition, Mr. Mercier's testimony means that only incremental investment income that was in excess of any interest expense could be counted as a true savings on top of the savings already captured by the hospital savings initiative. Any investment income that offset interest expense would have the effect of lowering CMAD growth and should not be captured again as PIP savings.

Accordingly, the Superintendent finds that the recognition of any savings from the early payment of PIP in Year Three is not reasonably supported by the evidence in the record.

(b) Physician Fees.

The Mercer reports for Years One and Two estimated savings attributable to increased physician fee payments for the MaineCare program. According to those reports, the budgets for SFY 2006 and SFY 2007 included increases of $8.2 million per year for these fee payments relative to the state budget for SFY 2005. Based on evidence in the records of those proceedings, these increases were Dirigo savings. According to Mercer, these fee increases would reduce the need for physicians to shift costs to commercial payors.

The Superintendent found that $4.1 million was reasonably supported for Year One (the portion in SFY 2006 that is attributable to CY 2005), and that $8.2 million was reasonably supported for Year Two (the remaining $4.1 million from SFY 2006 and half of the $8.2 million for SFY 2007).

In Year Three, srHS estimated savings due to increased physician fees of $4.1 million, the remaining 50% of the SFY 2007 increase. (R. at 3300.) At the Board hearing, MEAHP witness Jack Burke discussed the likelihood that these additional physician fee payments would result in savings to payors. (R. at 527-528.) Mr. Burke testified that physicians will tend to retain fee increases in areas where they feel they are not adequately reimbursed. (Id.) In addition, Mr. Burke provided a report for review by the Board that included unrefuted information that some providers serve a largely Medicaid population that does not include sufficient commercial payors and thus those providers are precluded from passing on savings. (R. at 6782-6783.) Mr. Burke estimated that perhaps $1 - $2 million of the $4.1 million might be passed on to payors. (Id.)

The Dirigo Board adopted the midpoint of Mr. Burke's range or $1.5 million. (R. at 607-610.) There is minimal discussion of this issue in the post-hearing briefs. MEAHP offers a qualified endorsement of the Board's findings. (MEAHP Brief, August 21, 2007, p. 39.)

The Superintendent finds the Dirigo Board's adoption of savings of $1.5 million attributable to physician fee increases is reasonably supported by the evidence in the record.

4. Overlap. Dirigo determination: $4.0 million. Amount the Superintendent finds reasonably supported by the evidence: Incorporated in the Superintendent's Hospital Savings determination.

(a) Hospital / Uninsured Savings.

The srHS “overlap” calculation is intended to measure the potential double-counting of savings within the various initiatives. An overlap was acknowledged by srHS between its savings calculations from the hospital savings initiatives and the uninsured / underinsured savings initiatives because costs and charges used in the hospital savings calculation would include those previously uninsured and underinsured now enrolled in DirigoChoice or MaineCare Parents Expansion. The amount of overlap was determined by srHS to be $4.0 million. (R. at 5318-5320, 5334-5336, 5356.) The Dirigo Board adopted this determination, approving $4.0 million. (R. at 9-10.)

The overlap calculated by srHS was based on hospital savings for the year ending June 30, 2006 (SFY 2006) and uninsured / underinsured savings for CY 2007. To adjust for the difference in time periods, srHS used a trend factor to discount the uninsured / underinsured savings by 18 months. The Board reduced the uninsured / underinsured savings but did not make any corresponding adjustment to the overlap. CAHC asserts that an adjustment should be made. (CAHC Brief, August 21, 2007, p. 7.) Dirigo agreed that there would be a reduction in the overlap but stated that there is no evidence in the record as to what that should be. The Board therefore concluded that it would be reasonable not to make any adjustment. (Dirigo Brief, August 29, 2007, p. 10.) At the Superintendent's hearing, Dirigo stated they had done some calculations and the overlap should probably be $1.3 to $1.4 million. However, Dirigo maintained that it may be reasonable not to make any adjustment “at this time.” The payor intervenors argued that a simple pro rata adjustment would not be appropriate and that there was no evidence in the record to support any change to the calculation.

Since the purpose of the overlap adjustment is to avoid double-counting of savings, it is not reasonable to base the adjustment on savings from different time periods. Applying a trend factor does not make it reasonable. The hospital savings in this year's filing overlap not with the uninsured / underinsured savings in this year's filing but with the uninsured / underinsured savings in Years One and Two. It would have been more appropriate to calculate the overlap based on uninsured / underinsured savings from parts of Years One and Two that correspond to SFY 2006. However, since the hospital savings found reasonably supported this year differs from that determined by srHS, the numbers from that determination are no longer appropriate to use in the overlap calculation. Since there are no corresponding numbers in the Superintendent's determination of the reasonably supported hospital savings, it is not possible to quantify the overlap between uninsured / underinsured savings in Years One and Two and the hospital savings found reasonably supported this year. However, as noted in section IV(B)(1) above, the overlap is reflected in the conservatism of the hospital savings found to be reasonably supported.

(b) Hospital / PIP Savings.

The srHS Report stated that there is no overlap between hospital savings and PIP savings based on the Superintendent's Year Two Decision. However, payor intervenors argued that the PIP savings methodology double-counts hospital savings. They concluded that not only should PIP savings be disallowed this year, but the hospital savings should be reduced by half of the CY 2006 PIP savings. (Chamber Brief, August 21, 2007, p. 27.)

As discussed in section IV(B)(3)(a) above, new evidence in this year's record leads to a different conclusion on this issue. Although PIP savings are not reasonably supported in this year's decision, the PIP savings that would overlap with the 2006 hospital savings in this year's filing are not the CY 2007 PIP savings in this year's filing but rather parts of the CY 2005 and CY 2006 PIP savings allowed in the previous two years. Although the payor intervenors have only argued that hospital savings should be reduced by part of the CY 2006 PIP savings, the same argument applies equally to CY 2005. Also, the payor intervenors argue that the appropriate portion of the CY 2006 PIP savings to offset against SFY 2006 hospital savings is one-half. However, the Year Two Mercer report indicates that less than half of the CY 2006 PIP savings was due to the SFY 2006 PIP increase. The portion of last year's $6.1 million PIP savings that reflects SFY 2006 is $2.2 million.

The Superintendent's Year One Decision indicated that all of the $1.7 million allowed PIP savings related to increased PIP payments in the first half of SFY 2006. Therefore, if all of the PIP savings are reflected in hospital savings initiatives, the appropriate offset to this year's hospital savings is $3.9 million ($2.2 million plus $1.7 million). As noted in section IV(B)(3)(a) above, PIP savings double-count hospital savings if the PIP savings result in reduced interest expense or if they result in investment income that is offset against interest expense in hospital financial statements. While it is possible that some hospitals have investment income in excess of interest, which would reduce the overlap, there is no evidence of this in the record, so it is likely that all or nearly all of those previously recognized PIP savings represent overlap. The Superintendent finds that a hospital savings initiative figure that does not account for this overlap in some manner is not reasonably supported by the evidence in the record. This conclusion is reflected in the Superintendent's decision regarding the hospital savings initiative above in section IV(B)(1).


By reason of the foregoing, the Superintendent ORDERS that the Dirigo Board's determination of aggregate measurable cost savings is APPROVED IN PART and that $32.8 million of aggregate measurable cost savings approved by the Dirigo Board is found by the Superintendent to be reasonably supported by the evidence in the record.


This Decision and Order is final agency action of the Superintendent of Insurance within the meaning of the Maine Administrative Procedure Act. Any party may appeal this Decision and Order to the Superior Court as provided by 24-A M.R.S.A. § 236, 5 M.R.S.A. § 11001, et seq., and M.R. Civ. P. 80C. Any such party must initiate an appeal within thirty days after receiving this notice. Any aggrieved non-party whose interests are substantially and directly affected by this Decision and Order may initiate an appeal within forty days after the issuance of this decision. There is no automatic stay pending appeal; application for stay may be made as provided in 5 M.R.S.A. § 11004.


Dated: September 17, 2007 ________________________________
Acting Superintendent of Insurance


1 At the time of the agency's decision, the members of the Dirigo Board were: Robert McAfee, M.D., Chair; Jonathan Beal, Esq.; Edward David, M.D.; Trish Riley ex officio, Director of the Governor's Office of Health Policy & Finance; Rebecca Wyke ex officio, Commissioner of the Maine Department of Administrative & Financial Services; and Anne Head ex officio, then Acting Commissioner of the Maine Department of Professional & Financial Regulation. The ex officio Board members do not have voting power. Ms. Riley and Ms. Wyke recused themselves and did not participate in the deliberations. The full complement of the Board consists of 5 voting members and 3 ex officio nonvoting members, but 2 of the 5 voting positions are vacant at this time. 24-A M.R.S.A. § 6904(1). Pursuant to PL 2007, c. 447, § 4, the Board will be expanded to consist of 9 voting positions and 4 ex officio nonvoting members as of September 20, 2007.

2 The actual amount adopted by Dirigo for this savings initiative was $6,343,400. The Superintendent finds that the use of five significant figures reflects a degree of precision that is not reasonably supported by the evidence, and is not consistent with the rounding conventions used by Dirigo elsewhere in its Year Three decision and by the Superintendent in his Year One and Year Two Decisions.

3 As explained in section IV(B)(4)(a) below, the amount of the overlap recognized by the Superintendent is included in the adjustment made to the determination of hospital savings.

4 Throughout the tenure of Dirigo, the agency has employed the consultant Steven Schramm. During Years One and Two Mr. Schramm was employed by consulting firm Mercer Government Human Services Consulting (“Mercer”), but in Year Three Mr. Schramm was a principal of schramm-raleigh Health Strategies.

5 The $88.4 million of aggregate measurable cost savings determined by the srHS Report consists of Hospital Initiatives of $70.6 million, Uninsured / Underinsured Initiatives of $14.0 million, Health Care Provider Fee Initiatives of $7.8 million, less a reduction from Overlap in the amount of $4.0 million.

6 They were not. Neither srHS nor the payor intervenors gave any consideration to the adjustments to the savings estimates that would have been necessary in order to make them consistent with the updated observed data or the revisions chosen to the trending methodology.

7 These discussions are complicated by a terminology problem: a hospital “increasing its charges” is the expression commonly used to describe a hospital price increase, while “increasing hospital outpatient charges” is used to refer to the increase in gross revenue that all three factors produce.

8 If one deflates the overall total cost line by the HMBI trend and calculates the geometric mean rate of growth from 1999 to 2003 and compares it to the geometric mean rate of growth from 2003 to 2006, the difference in rate of growth is similar to the similarly calculated difference in the rate of growth of the cost per CMAD including newborns with previous uninsureds.

9 The BDCC working group was part of a process intended to find an alternative to the Savings Offset Payment as a means of financing Dirigo. The payor intervenors assert that their acceptance of the “new money” approach was only for that limited purpose.

10 It should be noted, however, that in the form adopted by the Board, as discussed above in section IV(A), this methodology captures additional income from the new enrollees above and beyond the compensation for services that would have otherwise been provided on an uncompensated basis.

Last Updated: September 27, 2010