GARFIELD MEDICAL CENTER v. Belshe

80 Cal. Rptr. 2d 527, 68 Cal. App. 4th 798, 98 Cal. Daily Op. Serv. 9198, 98 Daily Journal DAR 12841, 1998 Cal. App. LEXIS 1045

• Court: California Court of Appeal
• Decided: November 19, 1998
• Docket: B115563
• Status: Published

Opinion

Opinion

VOGEL (C. S.), P. J.

Introduction

This appeal requires interpretation of a federal statute governing payment to hospitals which serve a disproportionate share of Medicaid patients. In particular, the issue is interpreting the phrase “the mean.” California, with approval from the federal government, has interpreted the phrase to permit use of a weighted mean. Appellant, on the other hand, contends that said interpretation violates federal law because Congress intended use of an arithmetic mean. The trial court resolved the issue against appellant. We affirm.

Factual and Procedural Background

In California, Medicaid patients are served by the state’s Medi-Cal program. Appellant Garfield Medical Center (Garfield) is certified to participate in the Medi-Cal program as a provider of hospital services. Garfield is reimbursed by Medi-Cal for inpatient services through a selective provider contracting program.

The Federal Disproportionate Share Hospital Program

The disproportionate share hospital supplemental payment adjustment program is a federally mandated program designed to recognize those hospitals which served a disproportionate number of Medicaid patients.

In 1987, Congress required state plans, such as Medi-Cal, to identify disproportionate share providers and to provide for supplemental payments to disproportionate share hospitals. (42 U.S.C. § 1396r-4(a)(l).) The federal statute defined a disproportionate share hospital as one in which “the hospital’s medicaid inpatient utilization rate ... is at least one standard deviation above the mean medicaid inpatient utilization rate for hospitals receiving medicaid payments.” (42 U.S.C. § 1396r-4(b)(l)(A), italics added.)

The Medicaid inpatient utilization rate for a particular hospital is a fraction expressed as a percentage. The numerator of the fraction is the total number of hospital inpatient days attributable to patients eligible for Medicaid. The denominator of the fraction is the total number of hospital inpatient days. (42 U.S.C. § 1396r-4(b)(2).) For example, if during a fiscal year a hospital has 10,000 days of total inpatient days and 3,000 of those days were for Medicaid patients, the fraction would be 3,000/10,000 resulting in an inpatient Medicaid utilization rate of 30 percent.

California Implements the Disproportionate Share Hospital Program Through Use of a Weighted Mean

To implement the federal mandate of providing supplemental payments to disproportionate share hospitals, the California Medical Assistance Commission (CMAC), the entity then responsible for determining disproportionate share hospitals, submitted a plan to the Health Care Financing Administration (HCFA) proposing use of a weighted mean. HCFA is the agency within the federal Department of Health and Human Services designated by Congress to administer the Medicaid program at the federal level. (.AMISUB (PSL) v. State of Colo. DSS (10th Cir. 1989) 879 F.2d 789, 794.) A declaration from a CMAC staff member who participated in the development and presentation of the weighted mean standard to HCFA explained:

“11. This ‘weighted average’ approach was included in the methodology after considering both a ‘weighted average’ and ‘unweighted average’ approach. I recall that the factors which I considered in comparing the two approaches were hospital size, volume of total days, the percent of such days attributable to Medi[]-Cal beneficiaries and the impact these factors had on both eligibility determination and payment amount.

“12. In an ‘unweighted approach’ each hospital, regardless of it[s] size or volume of days, is treated identically. This approach would have the effect of lowering the eligibility threshold somewhat and adding a few more hospitals. This lower eligibility threshold would create a greater relative disparity in payment rates between large and small hospital[s] because the payment methodology was tied to incremental payment increases for each percentage above the mean plus one standard deviation eligibility threshold.

“13. In using a ‘weighted approach’ size and volume impacted the eligibility determination by raising the eligibility threshold somewhat and providing less relative disparity between payment rates to large and small hospitals.

“14. The State made a determination in favor of a more equitable approach relative to allocation of disproportionate share rates of payment in California at the expense of eliminating a few hospitals from eligibility since the purpose of the federal law was augmented payments. The State employed this ‘weighted approach’ since [the federal law] was silent on this point.”

HCFA approved the use of the weighted mean in 1988.

Therefore, since 1988, California has used a weighted mean rather than an arithmetic mean to calculate the percentage needed to attain disproportionate share status. Use of the weighted mean results in the statistics being weighted by total number of patient days in each hospital. The following example, using three hypothetical hospitals, illustrates the effect of the use of a weighted mean. Hospital 1 has 300 Medicaid inpatient days and 1,000 total inpatient days. This results in a Medicaid utilization rate of 30 percent. Hospital 2 has 120 Medicaid inpatient days and 600 total inpatient days, resulting in a Medicaid utilization rate of 20 percent. Hospital 3 has 960 Medicaid inpatient days and 2,400 total inpatient days, resulting in a Medicaid utilization rate of 40 percent. The arithmetic mean would be calculated by adding the three percentages together (30 percent, 20 percent, & 40 percent) and dividing that total (90 percent) by three to produce a result of 30 percent. Use of the weighted mean, on the other hand, takes into consideration the disparities in total patient days at each hospital. In other words, use of the weighted mean recognizes that each hospital is not equal in regard to its patient load. A declaration from a statistician and research analyst for the Department of Health Services (Department) explains: “By weighting the mean medicaid inpatient utilization rate by total patient days in each hospital, each hospital’s medicaid inpatient utilization rate impacts the mean medicaid inpatient utilization rate in proportion to the ratio of its total inpatient days to the total inpatient days for all hospitals, i.e., in proportion to its relative volume of inpatient hospital days. Had the arithmetic mean been used, . . . each hospital’s medicaid inpatient utilization rate would have impacted the mean equally without regard to its relative volume of inpatient days.” (Italics added.) Consequently, use of the weighted mean in this hypothetical results in a mean Medicaid utilization rate of 34.5 percent. As Department explained in one of its communications with the federal government: “A decision not to weight is the same as a decision to treat each hospital as if it were the same size.”

Commencing in 1989, Garfield was designated as a disproportionate share hospital because it accepted a higher than average number of Medi-Cal patients. This designation entitled Garfield to receive additional funds from several different programs.

Subsequent Developments in the Law

On October 31, 1991, HCFA published a notice of proposed rules in the Federal Register. One of the changes HCFA sought was to include the phrase “arithmetic mean” in the definition of a disproportionate share hospital. However, Congress delayed until September 30, 1992 (a time period after the events giving rise to this dispute) for the issuance of the new rules. In fact, the record does not disclose if the rules were, in fact, ever enacted.

At the same time, Congress amended 42 United States Code section 1396r-4 which deals with disproportionate share hospitals. In particular, subsection (b), entitled “Hospitals deemed disproportionate share,” was amended. The amendment added paragraph 4 about a state’s authority to designate hospitals as disproportionate share hospitals but the definition in paragraph 1 of that subsection, which includes the language “the mean Medicaid in-patient utilization rate,” remained unchanged. However, Congress did insert the term “arithmetic mean” in another portion of the statute: subsection (f), which addressed factors other than the definition of disproportionate share hospital. (42 U.S.C. § 1396r-4(f).) 1

In 1991, the California Legislature established a new payment program for disproportionate share hospitals which was codified in Welfare and Institutions Code section 14105.98. Subdivision (f)(4) of that section provided, in pertinent part: “To determine a hospital’s Medi-Cal inpatient utilization rate ... for purposes of disproportionate share lists, the department shall utilize the same methodology, formulae, and data sources [as used in the previous years].” Consequently, Department, which had now assumed the role previously played by CMAC, continued to use a weighted mean. For the fiscal year 1991-1992, Garfield did not qualify as a disproportionate share hospital. The threshold to qualify as a disproportionate hospital was 36.3 percent; Garfield’s percentage of Medi-Cal patients was 35.8 percent.

Garfield Sues Because It Is Not Designated a Disproportionate Share Hospital

Alleging that this decision deprived it of approximately an additional $5 million in reimbursement funds, Garfield filed a petition for writ of mandate in the superior court against Department. Garfield did so notwithstanding the fact that in consideration of its failure to make the disproportionate share hospital list for 1991-1992, California negotiated two contract amendments with Garfield to provide it with an additional revenue of approximately $3.4 million. (In the subsequent fiscal years, Garfield was again designated a disproportionate share hospital.)

Insofar as is relevant to this appeal, Garfield’s petition attacked Department’s decision to use a weighted mean rather than the arithmetic mean to compute the threshold qualification to be a disproportionate share hospital. In part, Garfield alleged that the decision to use a weighted mean violated state and federal law. The trial court sustained Department’s demurrer without leave to amend on the basis that Garfield had failed to state a cause of action.

The Prior Appeal

Garfield appealed. This division reversed the trial court’s order in a nonpublished opinion. (Garfield Medical Center v. Belshé (Mar. 25, 1997) B093645.) 2 We held that Department’s formulation of a rule to implement Welfare and Institutions Code section 14105.98 was a quasi-legislative action subject to judicial review pursuant to Code of Civil Procedure section 1085 under the abuse of discretion standard. We wrote: “We agree with the Department that the term ‘mean’ is ambiguous in this context, and susceptible of more than one reasonable interpretation, and further agree that an agency’s interpretation of its governing statutes is entitled to ‘great weight.’ [Citation.]” (Garfield Medical Center v. Belshé, supra, B093645.) We noted: “[T]here are foundational facts essential to a proper determination of the legal questions, such as whether the Department has a policy requiring use of a weighted mean as alleged in the petition, the procedures it followed in adopting the policy, the factors it emphasizes in the weighting process, and the criteria which went into the Department’s choice of methodology. Because the trial court sustained the demurrer to [Garfield’s] petition without leave to amend, there has been no resolution of these foundational questions.” (Ib id.) We therefore concluded that the case had to be “remand[ed] to the trial court for a determination on the merits of whether the Department abused its discretion by choosing to use a weighted mean to compile its disproportionate share hospital list or whether that methodology violated federal law.” {Ibid., italics added.)

Trial Court Proceedings Following Remand

When proceedings resumed in the trial court, the parties presented evidence about both state and federal legislative intent in enacting the pertinent statutes and regulations, some of which evidence has been set forth above. The trial court, in a detailed 13-page statement of decision, rejected all of Garfield’s claims.

In particular, it concluded the decision to utilize a weighted mean was not a violation of federal law. It found: “The Department’s utilization of a weighted mean methodology for calculating the mean medicaid inpatient utilization rate in determining the threshold for disproportionate share status for fiscal year 1991-1992 under Welfare and Institutions Code section 14105.98, subdivision (e)(2)(A) and 42 United States Code section 1396r-4(b)(1)(A) did not violate federal law. As noted by the [California] Court of Appeal in its decision in [the appeal taken from the prior ruling sustaining Department’s demurrer without leave to amend], the term ‘mean’ in this case is ambiguous and susceptible of more than one reasonable interpretation and an agency’s interpretation of its governing statutes is entitled to ‘great weight.’ [Citations.] The ‘weighted mean’ methodology was selected after careful consideration of both the weighted mean, weighted by total inpatient days, and the arithmetic mean approaches and a determination that the weighted mean methodology would result in a more equitable allocation of disproportionate share rates of payment in California. After communicating with the Department on its rationale for using the weighted mean, HCFA approved use of the weighted mean for calculating the mean medicaid utilization rate for purposes of 42 U.S.C., § 1396r-4(b)(l)(A). There are many means defined in Mathematics and the Department’s choice of the weighted mean, weighted by total patient days in each hospital, approved by HCFA, was the most appropriate choice for determining the threshold for disproportionate share status. By weighting the mean medicaid utilization rate by total patient days in each hospital, each hospital’s medicaid inpatient utilization rate impacts the mean medicaid inpatient utilization rate in proportion to the ratio of its total inpatient days to the total inpatient days for all hospitals, i.e., in proportion to its relative volume of inpatient hospital days. Had the arithmetic mean been used, each hospital’s medicaid utilization rate would have impacted the mean without regard to its relative volume of inpatient days. Furthermore, Congress’ recision [ric] of a proposed rule that would have changed the word ‘mean’ in the statute to ‘arithmetic mean’ in the fall of 1991 and its amendment of the statute to use the term ‘arithmetic mean’ in subsection (f) having to do with factors other than the definitionof disproportionate share hospitals and amendment to subsection (b) dealing with the definition of disproportionate share hospitals without amending the term ‘mean’ clearly indicate the intent of Congress not to impose the requirement of an ‘arithmetic mean’ into the definition of disproportionality.”

The trial court entered judgment denying Garfield’s petition for a writ of mandate. This appeal by Garfield follows.

Discussion

On this appeal, Garfield’s sole contention is that use of the arithmetic mean is compelled by federal law. The claim lacks merit.

In a nutshell, Garfield argues: “The critical provision of the statute is clear and unambiguous. ‘The mean’ means ‘the mean,’ i.e., the arithmetic average.” To advance that argument, Garfield first characterizes as mere “dicta” our statement in the prior appeal that “the term ‘mean’ is ambiguous in this context, and susceptible of more than one reasonable interpretation.” Garfield’s characterization is incorrect.

Dicta consists of observations and statements unnecessary to the appellate court’s resolution of the case. (9 Witkin, Cal. Procedure (4th ed. 1997) Appeal, § 945, pp. 986-988.) Our statement was not dicta because it was responsive to the issues raised on appeal and was intended to guide the parties and the trial court in resolving the matter following our remand. (United Steelworkers of America v. Board of Education (1984) 162 Cal.App.3d 823, 834-835 [209 Cal.Rptr. 16].)

Furthermore, our characterization of the term “mean” as ambiguous was correct. “Ambiguity exists if reasonable persons can find different meanings in a statute . . . [citation]; when good arguments can be made for either of two contrary positions as to a meaning of a term . . . [citation].” (Black’s Law Diet. (6th ed. 1990) p. 79, col. 2.) The phrase “the mean” is ambiguous because it is susceptible to more than one reasonable interpretation. A declaration from Department, executed by an individual holding a master’s degree in statistics from the University of California at Berkeley who was then the president of the Sacramento chapter of the American Statistical Association, explained: “There are many means defined in Mathematics. Among the most commonly used means are the arithmetic mean (also known as the arithmetic average), the weighted mean (also known as the weighted average), the geometric mean, and the harmonic mean. The appropriateness of a particular mean as a statistical tool depends upon its purpose.” Garfield therefore errs in arguing that when Congress employed the phrase “the mean” in 42 United States Code section 1396r-4(b)(l)(A), it intended to require the use of the arithmetic mean as opposed to any other mean. In fact, Congress’s actions point to a contrary conclusion.

In 1991, Congress amended the statutory scheme to include the phrase “the arithmetic mean” in another part of the statute but made no such amendment to the subdivision in question. When “ ‘. . . Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion.’ [Citation.]” (Brown v. Gardner (1994) 513 U.S. 115, 120 [115 S.Ct. 552, 556, 130 L.Ed.2d 462].) Consequently, Congress’s decision to include the phrase “the arithmetic mean” in subdivision (f) (see fn. 1, ante), but to not amend subdivision (b) to change it to read “the arithmetic mean” instead of “the mean” presumptively demonstrates Congress did not intend to require use of the arithmetic mean in defining a disproportionate share hospital. This conclusion is fortified by Congress’s decision not to implement the rule change which would have amended the definition of disproportionate share hospital to include the phrase “arithmetic mean.” In other words, if Congress intended “the mean” to require use of the arithmetic mean, it would have said so in the statute. It did not do so.

This takes us to the next issue: what meaning should be attached to the phrase “the mean” in the context of this statutory scheme? Decisional law supplies a guidepost. The United States Supreme Court has held: “If the intent of Congress [in using a particular phrase in a statute] is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress. If, however, the court determines Congress has not directly addressed the precise question at issue, the court does not simply impose its own construction on the statute, as would be necessary in the absence of an administrative interpretation. Rather, if the statute is silent or ambiguous with respect to the specific issue, the question for the court is whether the agency’s answer is based on a permissible construction of the statute. fl[| . . . Sometimes the legislative delegation to an agency on a particular question is implicit rather than explicit. In such a case, a court may not substitute its own construction of a statutory provision for a reasonable interpretation made by the administrator of an agency. HQ We have long recognized that considerable weight should be accorded to an executive department’s construction of a statutory scheme it is entrusted to administer, and the principle of deference to administrative interpretations ‘has been consistently followed by this Court whenever decision as to the meaning or reach of a statute has involved reconciling conflicting policies, and a full understanding of the force of the statutory policy in the given situation has depended upon more than ordinary knowledge respecting the matters subjected to agency regulations. [Citations.] . . . If this choice represents a reasonable accommodation of conflicting policies that were committed to the agency’s care by the statute, we should not disturb it unless it appears from the statute or legislative history that the accommodation is not one that Congress would have sanctioned.’ [Citation.]” (Chevron, U.S.A. v. Natural Res. Def. Council (1984) 467 U.S. 837, 842-845 [104 S.Ct. 2778, 2781-2783, 81 L.Ed.2d 694], italics added, fns. omitted.)

Garfield goes to great lengths to point out that because this case involves administration of a federal program, the pertinent administrative agency is a federal agency (HCFA), not a state agency (Department). We have no quarrel with that proposition. (See, e.g., Orthopaedic Hosp. v. Belshe (9th Cir. 1997) 103 F.3d 1491, 1495-1496.) Nonetheless, the relevant point is that HCFA—a federal agency—on several occasions approved the plans submitted by California, each of which stated that the weighted mean would be used to define disproportionate share hospital. HCFA’s approval necessarily required it to interpret 42 United States Code section 1396r-4(b)(l)(A) and to decide that a proper interpretation of the statute permits use of the weighted mean. We must defer to that decision unless it is palpably erroneous. It is not. Absent use of a weighted mean, each hospital would be treated as if it were the same size when, in fact, that is not true.

Garfield advances several arguments as to why HCFA’s approval of California’s plan should be disregarded. None is persuasive.

The first argument is predicated upon the claim that a March 1989 letter from Department to HCFA contains inaccurate information. The purported inaccuracy is the implicit representation that use of a weighted mean would only affect small rural hospitals. Noting that it is a large urban hospital, Garfield essentially argues that HCFA was not accurately informed about the ramifications of using a weighted mean. This argument is not persuasive. For one thing, Garfield’s argument that the letter contains such a misrepresentation is based upon a strained reading of the document, the contents of which are set forth below in footnote 3. 3 For another thing, this letter was generated in response to a question posed by HCFA in 1988 about California’s 1988 plan. Nothing in the record even suggests that HCFA relied upon this letter when, in 1991, it again approved California’s use of a weighted mean.

Garfield next argues that HCFA’s approval of California’s plan violates guidelines contained in HCFA’s Medicaid manual provisions. Not so. The guidelines contain the same definition of disproportionate share hospital as is found in 42 United States Code section 1396r-4(b)(l)(A): a “Medicaid inpatient utilization rate at least one standard deviation above the mean Medicaid inpatient utilization rate . . . .” (Italics added.) The guidelines then provide that a state may use another definition of disproportionate share hospital but “the definition must include at least those hospitals which meet” the statutory definition which employs the phrase “the mean.” These guidelines do not aid Garfield. The question still boils down to the meaning of the phrase “the mean,” be that the phrase found in the statute or the guidelines, each of which must be interpreted identically.

Lastly, Garfield argues that we should simply disregard HCFA’s approval of the California plan because HCFA often approves such plans without a thorough review. Putting aside the fact that we must presume official duty was properly performed (Evid. Code, § 664), the cases Garfield relies upon for the proposition that HCFA gives only a cursory review to state plans are clearly distinguishable. Each involved an action filed in federal court by a group of health care providers which established (as opposed to merely speculating) that a state plan approved by HCFA was deficient. (Illinois Health Care Ass’n v. Bradley (7th Cir. 1993) 983 F.2d 1460 [state health care association and nursing home obtained declaratory judgment invalidating Illinois’s Medicaid reimbursement program for nursing homes]; AMISUB (PSL) v. State of Colo. DDS, supra, 879 F.2d 789 [hospitals successfully challenged Colorado’s system for reimbursement of Medicaid costs for inpatient services]; and California Hosp. Ass’n v. Schweiker (C.D.Cal. 1982) 559 F.Supp. 110 [hospitals and hospital associations successfully challenge California’s plan for reimbursing hospitals for inpatient services].)

In sum, the trial court properly determined that Department was not in violation of federal law in using a weighted mean to define disproportionate share hospital.

Disposition

The judgment is affirmed.

Hastings, J., and Curry, J., concurred.

1 Subsection (f) was entitled: “Denial of Federal financial participation for payments in excess of certain limits.” The subsection set forth the limited circumstances in which money would be given for payment adjustments. One such situation was if the payment adjustments had been made in accord with a state plan which “designates only disproportionate share hospitals with a medicaid or low-income utilization percentage at or above the Statewide arithmetic mean . . . .” (42 U.S.C. § 1396r-4(f); italics added.) This provision was repealed in 1997.

2 We have granted Department’s request to take judicial notice of our file in that appeal.

3 The pertinent portion of the letter merely states: “The impact of a decision to use an unweighted ‘average of the averages’ would be to add a small number of hospitals [to the list of disproportionate share hospitals]. The statute requiring the disproportionate share adjustment does not require that any additional consideration be made for small or rural hospitals. Other parts of the State’s plan approved under Title XIX address adjustments for these types of hospitals. [H] Another, and larger impact would be to increase the amount of payments to those hospitals that qualified for an adjustment under the weighted method. This is because the impact of not weighting is to lower the mean plus standard deviation threshold. Since the choice of methods does not affect any individual hospital’s percentage, hospitals which met this threshold continue to meet the unweighted criteria, but by a larger amount. This larger amount then causes the amount of the disproportionate share payment adjustment to be higher. The increase to this original group of hospitals exceeds the increase in funds given to small or rural hospitals by a significant amount.”