Pacificare of Oklahoma v. Oklahoma Health Care Authority Board

RAPP, Presiding Judge:

T1 The trial court plaintiffs, Pacificare of Oklahoma and Community Care Health Maintenance Organization, Inc. (collectively, HMO), appeal the decision in favor of the trial court defendants issued in a declaratory judgment action filed against several state agencies and individuals in their official capacities. The state agencies, which are trial court defendants, consist of Oklahoma Health Care Authority (OHCA), Oklahoma State Employees Benefits Council (EBC), Oklahoma State Employees and Education Groups Insurance Board (Board), and Department of Central Services (DCS).

BACKGROUND

T2 The State of Oklahoma provides employees with health care benefits. The employees each year may select between a health care benefits plan established by the Board or health care plans provided by private entities such as HMO. The private plans are made available annually pursuant to a bidding process conducted through DCS. The bid information and requirements are established by EBC. This dispute involves one such bid specification and is confined to the 1997 fiscal year beginning July 1, 1996, and ending June 30, 1997.

¶3 EBC eliminated a geographic risk factor for fiscal year 1997 used to adjust premiums between the Board and private health care plans in order to compensate for "adverse selection."¹ The geographic risk factor in place prior to fiscal year 1997 resulted in premium funds being deducted from Board and credited to HMO. EBC eliminated that risk adjustment factor for the fiscal year 1997.

T4 HMO maintains that EBC violated the provisions of 74 0.8. Supp.1996, § 1871(H), currently renumbered as § 1871(I), when it eliminated the geographical factor for the 1997 fiscal year, asserting it is a statutory adjustment risk factor. Specifically, HMO asserted that the elimination of the geographic risk adjustment factor was not supported by either actuarial principles or the evidence. Last, HMO argues that the elimination of that risk factor was an impermissible change to the historical interpretation of the statute. The statute provides:

Benefit plan contracts with the Board, health maintenance organizations, and other third party insurance vendors shall provide for a risk adjustment factor for adverse selection that may occur, as determined by the Council, based on generally accepted actuarial principles.

74 0.8. Supp.1996, § 1871(H), currently renumbered as § 1871(D).

15 Oklahoma offers yearly to its employees and education employees, collectively employees, a choice between either a basic or an extended private health insurance program as a part of employee benefits. The insurance is offered through the Board. In addition, private forms of insurance or health care plans are solicited in order to provide a choice to employees. EBC has the duty to negotiate and contract with qualified health maintenance organizations, such as the HMO plaintiffs here, in order to provide a choice of health insurance programs for employees and educators.

16 The insurance plan and its costs are developed by EBC. DCS then conducts the contracting process with private providers, such as HMO, using Requests for Proposals (RFP) initiated by EBC and which DCS sends to prospective bidders. The RFP contains information relating to a vendor's prospective bid, specifications, and details for the bidding process. Here, the geographical risk adjustment factor for the 1997 fiscal year RFP had been removed. HMO filed a declaratory judgment action challenging the legality of that omission.²

T 7 The detailed procedures and the historical background are cogently set forth at length in the trial court's judgment and the parties' briefs. The parties are in general agreement as to those matters. Originally, EBC's predecessor agency, Oklahoma State Educational Employees Group Insurance Board (OSEEGIB) utilized a formula whereby a risk factor adjustment would be applied so that any "adverse selection" effects that might occur due an employee's selection of the HMO or the state plan would be mitigated.³

T8 The Board, in cooperation with other health plan organizations, developed the original risk factors in 1991 and adopted age, gender, and geographic location as risk adjustment factors.⁴ The geographical risk adjustment factor developed in 1991 was .88 for rural and 1.00 for urban. Urban was defined as Oklahoma and Tulsa Counties and the remainder of the State was rural. Consequently, due to the fact that the HMO providers limited their coverages to persons living or working in the urban population centers of the state, the geographical risk factor adjustment operated to transfer premium money from Board to the private providers, such as HMO. This had the effect of rewarding the private providers for their failure to compete statewide, thereby depriving a large segment of the state's employees with a "choice" of providers.

T 9 Subsequently, after development of the 1991 methodology, EBC assumed administrative responsibility. All of the risk adjustment factors remained in place until the 1997 fiscal year, when EBC changed the methodology by increasing the geographic factor to 1 from .88, thereby announcing that there is no difference in costs between urban and rural areas and effectively neutralizing the geographic risk factor. This neutralizing of the geographic risk factor had the effect of eliminating it as a method of additional payment to HMO.

T 10 At trial, the EBC stipulated that it did not speak with an actuary when it eliminated the geographical risk factor.⁵ EBC justified its position as follows:

-_ EBC had no evidence to justify continuance of the risk factor.

-- Major changes in the health care delivery system have occurred since 1991 and previous distinctions between urban and rural areas have substantially diminished so that distinctions in costs experienced by employees in the respective areas no longer exist.

- Claims experienced by Board and EBC reflect that costs related to physician and hospital out-patient care were equalized between the urban and rural areas and that prescriptions were also approximately equal although slightly higher in rural areas. Differences of in-patient hospital care costs could be attributed to the higher capital costs incurred by the more sophisticated facilities in urban areas. Moreover, patients from rural areas who required sophisticated care received that care in the urban hospital.

¶ 11 The HMO presented testimony of two witnesses, one a marketing salesperson, and one an actuary. The actuary testified that in fact, according to his analysis, the cost differences between those employees in urban areas and those in rural areas justified continuation of the geographical risk factor adjustment at the previous level.⁶ This witness used the data base developed from 1996 Medicare data and not the population base of this State. He did not obtain data from the EBC.⁷ He also observed that other insurance providers and other actuarial conclusions supported his result, without stating the population base, upon which these conclusions were based.

112 The trial court made findings and rulings in favor of the State agencies. In addition, the trial court observed that historically the urban areas were deemed to be Tulsa County and Oklahoma County. However, under the evidence, the private health care plans, such as HMO, were providing some service outside those counties. The trial court further found that HMO had failed to produce evidence from which a determination of what is "urban" and what is "rural" could be made, even if an adjustment would be in order based upon geographical area. HMO appeals.⁸

STANDARD OF REVIEW

113 Declaratory judgment actions are reviewed in the same manner as other judgments. 12 0.$.1991, § 1654. The appellate court has the plenary, independent, and nondeferential authority to reexamine a trial court's legal rulings. Neil Acquisition, L.L.C. v. Wingrod Investment Corp., 1996 OK 125, 932 P.2d 1100, fn. 1. Matters involving legislative intent present questions of law which are examined independently and without deference to the trial court's ruling. Salve Regina College v. Russell, 499 U.S. 225, 111 S.Ct. 1217, 113 L.Ed.2d 190 (1991); Keizor v. Sand Springs Ry. Co., 1993 OK CIV APP 98, ¶ 5, 861 P.2d 326, 328. The trial court's findings of fact will not be disturbed unless they appear clearly to be against the weight of the evidence. Storck v. Cities Service Gas Co., 1977 OK 227, ¶ 29, 575 P.2d 1364, 1369. Thus, in actions of equitable cognizance, the judgment made by the trial court will be reversed if it is clearly contrary to the weight of the evidence or contrary to accepted principles of equity or rules of law. In re Estate of Eversole, 1994 OK 114, ¶ 7, 885 P.2d 657, 661.

ANALYSIS AND REVIEW

T14 EBC concedes that if cost of medical care differences exist based upon where the insured employee lives, then such cost differences must be equalized under Section 1871(H) for the 1997 fiscal year. Moreover, EBC does not dispute that the geographic risk factor as used in prior years would be appropriate if the costs differences are as contended by HMO. Last, EBC admits that it did not employ actuarial expertise in the form of a certified actuary to arrive at its decision to exelude the geographical risk factor for the 1997 fiscal year. EBC chose instead to rely on historical facts it developed, which in turn are consistent with section five of the Actuarial Standard of Practice No. 12, Concerning Risk Classgification, Actuarial Standards Board, adopted October 12, 1989.⁹

¶ 15 According to EBC, the HMO did not present evidence in this action to establish that such cost differences exist. The trial court agreed. Therefore, this Court must decide whether the trial court's decision is against the clear weight of the evidence and contrary to law. In that regard, the trial court was presented with conflicting evidence regarding the existence of cost differences between urban and rural medical care costs based upon geographic location. The trial court's decision to accept the evidence presented by the State agencies is not against the clear weight of the evidence.

116 Nevertheless, the legal premise underlying the competing positions requires this Court to determine whether the decision is contrary to law. HMO argues that EBC was statutorily required by Section 1871(H) to employ actuarial principles to determine whether cost differences exist.¹⁰ HMO's evidence was consistent with its position on the statutory actuarial requirements. Thus, the question of whether the administrative procedure, or absence thereof, employed by EBC in its decision-making process empowered EBC to eliminate the geographical risk factor in the 1997 fiscal year bid specifications is not before this Court for review.

[ 17 While EBC admits that it did not use an actuary, it correctly argues that the determination of the existence or non-existence of health cost differences in urban and rural areas is not a required actuarial determination. HMO defines the actuary's role as a mathematician, whose function is to develop, analyze, and evaluate formulas which determine probabilities enabling the actuary to compensate for and equate such cost differences as may have been found to exist in the population based upon an unbiased database.

[ 18 This Court holds that the State agen-cles have not violated Section 1871(H).

119 First, Section 1371(H) does not confine EBC to actuarial methods in order to learn whether cost differentials exist. The statute merely refers to "adverse selection that may occur, as determined by the Council consistent with generally accepted actuarial principles." It does not direct that the EBC is to use a particular means to ascertain the underlying issue of cost differences. Thus, EBC, as did HMO at trial, could perform and use any simple arithmetical analysis which is mathematically consistent with actuarial principles, to ascertain whether cost differences occur based upon geography, and utilize the result in its decision-making. Further, EBC may also, consistent with Section 1371(H), use its own real-life historical database, and that of other similar State agencies, to determine whether cost differences did exist in the different areas in the process of reaching a decision. The evidence presented by the State agencies reflects that such data, along with demonstrable changes in the provision of health care delivery, did persuade the agency to eliminate the risk factor because geographical cost differences did not, in fact, then exist based on their analysis.

120 Moreover, the Actuarial Standards of Practice (Standards) are not inconsistent with that conclusion.¹¹ Here, the concern is with a class of possible risk, that is, one flowing from geographical differences among the total employee population. Risks are defined by the Standards as: "Uncertainty arising from the possible occurrence of given event." The risks are classified, or grouped, according to similar risk characteristics. The groups then can be compared in order to ascertain whether cost differences may exist.

121 The Standards explain methods to demonstrate cost differences. The Standards explicitly provide that actual, or reasonably anticipated, experience and information from reliable sources may be used to demonstrate a relationship between a risk characteristic and cost. It is self-evident that one result of that exercise would be the possibility that historical experience and reliable information could yield factual data that demonstrates the non-existence of cost differences. The EBC here has presented its case that it essentially looked at the risk classifications based upon geography and, using its own historical experience and data, concluded that cost differences between urban and rural populations do not exist. This factual data developed by EBC based on its experience of providing health care coverage to approximately seventy percent (70%) of the covered population of this State was not disputed by HMO.

22 HMO also maintains that EBC could not change its policy in the absence of some reason supporting that change. HMO's reliance on cases such as Oral Roberts Univ. v. Oklahoma Tax Commission, 1985 OK 97, 714 P.2d 1013, is misplaced. There, the agency had interpreted and applied a statute. Such is not the case here.

¶23 Here, EBC did not interpret and then change the interpretation. EBC applied the statute and determined, based on historical data and experience obtained from its population base, that a cost difference rooted upon geography did not exist-the statute does not say that geographical cost differences exist in fact, so when EBC adjusted for that risk factor in the past it was not interpreting the statute. When the purpose of ageney action is some form of regulation, as distinguished from applying law or policy to past facts, an agency needs to have flexibility, at all times, to take such steps as may be proper in light of the present cireum-stances and facts irrespective of its past decisions. Rosebud Enterprises, Inc. v. Idaho P.U.C., 128 Idaho 609, 917 P.2d 766, 775; Washington Water Power Co. v. Idaho P.U.C., 101 Idaho 567, 617 P.2d 1242, 1254 (1980); Reaveley v. P.S.C., 20 Utah 2d 237, 436 P.2d 797, 799 (1968).

CONCLUSION

124 The trial court's decision correctly analyzed the matter presented and its judgment is neither contrary to law nor against the clear weight of the evidence. Therefore, the decision is affirmed in all respects.

1 25 AFFIRMED.

126 TAYLOR and JONES, JJ. (sitting by designation), concur.

1 . The effects of "adverse selection" occur because employees select the State plan or the HMO based upon individual choices, which include the cost of the coverage and physical location within the State. This cost is predetermined before the time the employees make their selection and, because of factors such as age, gender, or location, either the State plan or the HMO plans could be burdened with a disproportionate group of employees with higher medical costs attributable to one or more of these "factors." However, the HMO will not accept employees outside of its service area, which is basically Oklahoma City and Tulsa and a few surrounding areas. Rate adjustment mitigation takes place because, when "adverse selection" occurs under the statute, the risk factors are used to establish an adjustment so that when premiums are collected by EBC some money may be withheld from the "advantaged" plan and paid over to the "disadvantaged" plan. Historically, this adjustment has resulted in premiums being withheld from the Board and paid to the HMO private providers.

2 . The action also sought injunctive and other relief. The trial court issued other rulings and denied permanent injunctive relief, but HMO has not appealed that aspect of the case.

3 . The original risk factors were ascertained after the agreed methodology was established by the Oklahoma State and Education Employees Group Insurance Board and the HMO's for the period 1991. This methodology was not, in accord with the record, based on good data.

4 , See fn. 3.

5 . All agencies' positions were identical. EBC was the "lead" agency.

6 . EBC's witnesses described their use of historical records obtained from their seventy percent (70%) coverage of the State's healthcare population. These records allowed EBC to arrive at its determination that a difference between Urban and Rural areas did not exist. HMO'S assertion that this approach did not meet actuarial principles is apparently grounded on EBC'S admitted fact that EBC used, as stated, factual historical data, which was not reviewed by an actuary. However, this data is in itself factual and acceptable in statistical and actuarial usage. While HMO complains of EBC's methodology, it is of interest to note here that HMO, in arriving at its values before the court, used simple arithmetic. HMO did not present or display any mathematical methodology in the course of trial not available to the average high school graduate.

7 . This witness also admitted he used rate data obtained from public record filings of four companies in New York State. He also admitted that the "scientific basis" for selection of this data and its use was solely because they were made available to him by his company.

8 . HMO's Motion to Supplement the Record is denied.

9 . HMO's exhibit 32.

10 . It must be noted that HMO did not present any testimony establishing how EBC failed to employ actuarial principles in reaching its decision. Moreover, the record fails to demonstrate how HMO adhered to such principles in advancing its argument. While it did, at one point, present a simple three-element formula, it did not establish or provide the basis for derivation of any of the formula's elements. The problem here is that HMO's witness and HMO assumed a set of facts not in evidence; for example, usage of the utilization factor "U" in the witness' formula, PU=C, did not provide an explanation or evidence of how it was derived, or state what its component parts were, or even provide that any element of that formula is a constant or a variable.

11 . The relevant standard was admitted in evidence here as Exhibit 32. "Actuarial Standard of Practice No. 12, Concerning Risk Classification", Actuarial Standards Board (October 12, 1989).