Duke Univ. Health Sys. Inc. v. N.C. Dep't of Health & Hum. Servs.

• Court: Court of Appeals of North Carolina
• Decided: August 6, 2024
• Docket: 23-351
• Status: Published

Opinion

IN THE COURT OF APPEALS OF NORTH CAROLINA

No. COA23-351

Filed 6 August 2024

Office of Administrative Hearings, No. 22 DHR 02685

DUKE UNIVERSITY HEALTH SYSTEM INC., Petitioner,

v.

N.C. DEPARTMENT OF HEALTH AND HUMAN SERVICES, DIVISION OF HEALTH SERVICE REGULATION, HEALTH CARE PLANNING & CERTIFICATE OF NEED SECTION, Respondent,

and

UNIVERSITY OF NORTH CAROLINA HOSPITALS AT CHAPEL HILL AND UNIVERSITY OF NORTH CAROLINA HEALTH CARE SYSTEM, Respondent- Intervenors.

Appeal by Petitioner from final decision entered on 9 December 2022 by

Administrative Law Judge Melissa Owens Lassiter in the Office of Administrative

Hearings. Heard in the Court of Appeals 15 November 2023.

Baker, Donelson, Bearman, Caldwell & Berkowitz, a Professional Corporation, by Kenneth L. Burgess, Matthew A. Fisher, Iain M. Stauffer, and William F. Maddrey, for petitioner-appellant.

Attorney General Joshua H. Stein, by Special Deputy Attorney General Derek L. Hunter, for respondent-appellee.

Nelson Mullins Riley & Scarborough LLP, by Noah H. Huffstetler, III, Candace S. Friel, Lorin J. Lapidus, Nathaniel J. Pencook, and D. Martin Warf, for respondent-intervenor.

MURPHY, Judge.

When an appellant challenges the substantive determinations of an

administrative law judge (“ALJ”) on appeal from a contested case hearing for a DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

Opinion of the Court

certificate of need, we review the decision for substantial evidence on the whole

record. However, where our statutes dictate the proper scope of administrative

review, the ALJ may not exceed that scope. Here, although we affirm the ALJ in

almost all respects, we must remand for further findings insofar as the final decision

granting the certificate of need relied upon a site other than that presented in the

respondent’s application.

BACKGROUND

Petitioner-Appellant Duke University Health System, Inc. (“Duke”) challenges

on appeal the 9 December 2022 final decision of the ALJ to uphold the conditional

approval of a certificate of need (“CON”) granted to Respondents-Intervenors-

Appellees University of North Carolina Hospitals at Chapel Hill and University of

North Carolina Health Care System (collectively “UNC”) by the North Carolina

Department of Health and Human Services (the “Agency”).

Pursuant to N.C.G.S § 131E-183(a)(1) and chapters 5 and 6 of the 2021 State

Medical Facilities Plan (“SMFP”), the Agency determined the need to develop 40

acute care beds and four operating rooms for the Durham/Caswell County health

service areas. The “new acute care beds [and operating rooms] [could not] be

developed without a CON issued by the Agency.” On 15 April 2021, in response to

the need determinations of the SMFP, five applications to develop additional acute

care beds and operating rooms for the Durham County area were submitted to and

reviewed by the Agency. Applications were submitted by Duke and North Carolina

2 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

Specialty hospital/Southpoint Surgery Center, two Durham County health systems.

Additionally, UNC applied as a new provider in Durham County.

On 1 May 2021, the Agency independently reviewed all applications against

the statutory review criteria found in N.C.G.S. § 131E-183(a)1 and the applicable

1 In pertinent part, N.C.G.S. § 131E-183(a) provides:

(a) The Department shall review all applications utilizing the criteria outlined in this subsection and shall determine that an application is either consistent with or not in conflict with these criteria before a certificate of need for the proposed project shall be issued.

(1) The proposed project shall be consistent with applicable policies and need determinations in the State Medical Facilities Plan, the need determination of which constitutes a determinative limitation on the provision of any health service, health service facility, health service facility beds, dialysis stations, operating rooms, or home health offices that may be approved.

.....

(3) The applicant shall identify the population to be served by the proposed project, and shall demonstrate the need that this population has for the services proposed, and the extent to which all residents of the area, and, in particular, low income persons, racial and ethnic minorities, women, handicapped persons, the elderly, and other underserved groups are likely to have access to the services proposed.

....

(12) Applications involving construction shall demonstrate that the cost, design, and means of construction proposed represent the most reasonable alternative, and that the construction project will not unduly increase the costs of providing health services by the person proposing the construction project or the costs and charges to the public of providing health services by other persons, and that applicable energy saving features have been incorporated into the construction plans.

N.C.G.S. § 131E-183(a)(1), (3), (12) (2023).

3 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

regulatory review criteria found in 10A NCAC 14C. Southpoint Surgery Center

submitted an application to add four operation rooms based on the need

determination in the 2021 SMFP; UNC Hospitals submitted an application to develop

40 acute care beds and two operating rooms in the Research Triangle Park area.

Meanwhile, Duke submitted three applications: the first was to add 40 acute care

beds and two operating rooms to its existing Durham facility; the second was to

develop two operating rooms; and a final application sought to develop two more

operating rooms at its Ambulatory Surgery Center. The Agency found that

Southpoint Surgery Center failed to demonstrate financial feasibility and failed to

show that its application was not unnecessarily duplicative of existing or approved

services, among other criteria, while it found both Duke and UNCs applications

conforming to all the review criteria. As a result, the Agency denied Southpoint’s

CON application.

Since the need determination in the SMFP places limits on the number of acute

care beds that can be approved by the Agency—40 acute care beds and two other

operating rooms—accepting both the Duke and UNC applications would have

resulted in more acute care beds and operating rooms than the SMFP need

determination for Durham County allowed. The Agency therefore concluded that,

because the SMFP allowed for only 40 acute beds in the Durham County area,

granting Duke’s application would require the denial of UNC’s application and vice

versa. Pursuant to the review criteria under N.C.G.S. § 131E-183, the Agency

4 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

conducted a comparative analysis review of both Duke and UNC CON applications

for 40 acute care beds, as well as another for the two operating rooms.

On 21 September 2021, “[b]y decision and Required State Agency Findings[,]

the Agency (1) conditionally approved the UNC Hospitals-RTP Application; (2)

conditionally approved [Duke’s Ambulatory Surgery Center’s] Application [for two

additional operating rooms]; (3) denied [Duke’s] [two operating rooms] Application;

(4) denied [Duke’s acute care beds] Application; and (5) denied the Southpoint

Application [for two operating rooms].” By letter and Required State Agency

Findings dated 21 September 2021, the Agency informed Duke that its application

for 40 acute care beds and two operating rooms had been denied. Also on 21

September 2021, the Agency issued the Required State Agency Findings containing

the findings and conclusions upon which it based its decisions.

On 21 October 2021, Duke filed a petition for contested case hearing pursuant

to N.C.G.S § 150B-23 alleging that the Agency had erroneously approved the CON

application of UNC in which UNC sought to develop two operating rooms and 40

acute care beds in Durham County. On 10 November 2021, the OAH issued an order,

by consent of all parties, to grant UNC the right to intervene in the contested case

hearing. The ALJ issued a final decision in which it affirmed the Agency’s decision

finding UNC’s application to be comparatively superior to Duke’s application. Duke

appealed.

ANALYSIS

5 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

On appeal, Duke challenges the ALJ’s final decision on four distinct bases, all

of which, in substance, challenge the original determinations of the Agency and only

derivatively challenge the ALJ’s final decision insofar as it did not reverse the

Agency. The bases for its challenges on appeal are (A) that the ALJ incorrectly

affirmed the Agency’s determination that UNC’s application was superior to Duke’s

with respect to geographic accessibility; (B) that the ALJ incorrectly affirmed the

Agency’s determination that UNC’s application was superior to Duke’s on the basis

of competition; (C) that the ALJ incorrectly affirmed the Agency’s finding that UNC’s

application conformed with N.C.G.S. § 131E-183(a)(3); and (D) the ALJ incorrectly

affirmed the Agency’s finding that UNC’s application conformed with N.C.G.S. §

131E-183(a)(12).

In reviewing the ALJ’s determinations, our standard of review is governed by

N.C.G.S. § 150B-51, which permits a party seeking judicial review to challenge an

ALJ’s final decision

if the substantial rights of the petitioners may have been prejudiced because the findings, inferences, conclusions, or decisions are:

(1) In violation of constitutional provisions; (2) In excess of the statutory authority or jurisdiction of the agency or administrative law judge; (3) Made upon unlawful procedure; (4) Affected by other error of law; (5) Unsupported by substantial evidence admissible under [N.C.G.S. §] 150B-29(a), [N.C.G.S. §] 150B-30, or [N.C.G.S. §] 150B-31 in view of the entire record as submitted; or (6) Arbitrary, capricious, or an abuse of discretion.

6 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

N.C.G.S. § 150B-51(b) (2023). “With regard to asserted errors pursuant to

subdivisions (5) and (6) of subsection (b) of [N.C.G.S. § 150B-51], the court shall

conduct its review of the final decision using the whole record standard of review.”

N.C.G.S. § 150B-51(c) (2023).

“In applying the whole record test, the reviewing court is required to examine

all competent evidence in order to determine whether the [final] decision is supported

by substantial evidence.” Surgical Care Affiliates, LLC v. N.C. Dep’t of Health &

Hum. Servs., 235 N.C. App. 620, 622-23 (2014) (marks omitted), disc. rev. denied, 368

N.C. 242 (2015). “Substantial evidence is such relevant evidence as a reasonable

mind might accept as adequate to support a conclusion.” Id. at 623. “This test does

not allow the reviewing court to replace the [ALJ’s] judgment as between two

reasonably conflicting views, even though the court could justifiably have reached a

different result had the matter been before it de novo.” Mills v. N. Carolina Dep’t of

Health & Hum. Servs., 251 N.C. App. 182, 189 (2016) (marks omitted).

A. Relative Geographic Accessibility

We first address whether the ALJ properly affirmed the Agency’s conclusions

as to geographic accessibility. Duke contends that the ALJ’s decision was erroneous

because the Agency had favorably evaluated the UNC application on the basis of

geographic accessibility despite being located in Research Triangle Park, a

nonresidential area of Durham, and had analyzed the geographic access factor in a

7 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

manner that lacked a coherent guiding principle and deviated from the methodology

of previous reviews. We disagree.

While analyzing the geographic access factor, the ALJ’s final decision

acknowledged many of the issues Duke raises before us and nonetheless affirmed the

Agency’s determination in favor of UNC:

420. The Agency utilized the comparative factor of Geographic Accessibility in its comparative analysis of the UNC and Duke Applications. (Jt. Ex. 1, pp. 1609, 1619).

421. In analyzing this comparative factor, the Agency looked at where each applicant proposes to place the proposed services. (Meyer, Vol. 7, p. 1299). An application placing the services at issue in a location where there are not any such services is deemed the more effective alternative under this factor. (Jt. Ex. 1, p. 253; Carter, Vol. 11, pp. 1874-75).

422. Ms. Sandlin opined that the Agency erred in its analysis of this comparative factor as having geographic dispersal of these need determined assets is not critical because Durham has less land mass than other counties in North Carolina. (Sandlin, Vol. 6, pp. 1058-67).

423. Mr. Meyer opined that this factor is important because it is related to access, a foundational principle of the CON Law. The CON Law seeks to avoid geographic maldistribution of services, and North Carolina has a “compelling interest in helping to ensure that all North Carolinians have access to [. . .] healthcare services[.]” (Meyer, Vol. 7, p. 1299).

424. In the acute care beds review, the Agency noted there were 1,388 existing and approved acute care beds in the Durham/Caswell County service area, all of which are located in the central area of Durham County, illustrated by the following table:

8 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

Facility Total AC Address Location Beds Duke 1,048 2301 Central University Erwin Rd, Durham Hospital Durham County 27710 Duke 316 3643 N. Central Regional Roxboro Durham Hospital Rd, County Durham 27704 North 24 3916 Ben Central Carolina Franklin Durham Specialty Blvd, County Hospital Durham 27704

(Jt. Ex. 1, p. 1609; see also Meyer, Vol. 7, p. 1300).

425. Similarly, in the ORs review, the Agency noted that there were 93 existing and approved ORs in Durham County, the vast majority of which were concentrated in the central area of Durham County, illustrated by the following table:

Facility Type Durham Total Address Location SA OR ORs System NCSH Exiting NCSH 4 3916 Ben Central Hospital Franklin Durham Blvd, County Durham 27704 DUH Exiting Duke 66 2301 Central Hospital Erwin Rd, Durham Durham County 27710

9 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

DRH Exiting Duke 13 3643 N. Central Hospital Roxboro Durham Rd, County Durham 2704 DASC Existing Duke 4 2400 Pratt Central ASF St, Durham Durham County 2704 Arringdon Existing Duke 4 5601 South ASF Arringdon Durham, Park Dr, near I540 Morrisville at I40 27560 SSC Approved NCSH 2 7810 NC South ASF Hwy 751, Durham, Durham near Hwy 27713 147 UNC-RTP Proposed UNC 2 Parcels in South Hospital [RTP] Durham, 27709 just below I40

(Jt. Ex. 1, p. 1620).

426. For both the acute care beds and ORs comparative analyses, the Agency determined that the UNC Application was the more effective alternative, and Duke’s Applications were the less effective alternatives for geographic accessibility. (Jt. Ex. 1, pp. 1609, 1620; Hale, Vol. 1, p. 188).

427. UNC proposed placing the acute care beds in this Review in the southern area of Durham County, where there were no existing acute care beds, while Duke proposed placing additional beds at DUH where there were already over one thousand existing or approved acute care beds. (Jt. Ex. 1, p. 1609; Hale, Vol. 1, p. 188). The Agency also found UNC Hospitals-RTP, Duke Arringdon, and Southpoint Surgery Center to be more effective because

10 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

they “propose to develop ORs in South Durham County where there are currently only six of 93 existing/approved Durham County ORs[,]” as opposed to the Duke ORs Application which proposed placing additional ORs at DUH where there were already sixty-six existing and approved ORs. (Jt. Ex. 1, p. 1620).

428. Mr. Meyer agreed with the Agency’s analysis of this comparative factor. (Meyer, Vol. 7, pp. 1299-1300, 1330- 31). In the beds analysis, the existing facilities in Durham are concentrated in the center of the county. (Jt. Ex. 97, p. 11; Meyer, Vol. 7, p. 1301). Mr. Meyer analyzed the locations of hospitals in certain populous counties in North Carolina, including Wake, Mecklenburg, Guilford, and Forsyth counties, all of which have hospitals in the perimeter of the county and generally have good geographic dispersal of hospitals. (Jt. Ex. 103; Meyer, Vol. 7, pp. 1302-1305). His analysis showed that compared to these highly populated counties, Durham County as another highly populated county, “does not have an acute care hospital that’s located anywhere but in the center of the county,” (Meyer, Vol. 7, p. 1305).

429. Similarly, both Mr. Meyer and Mr. Carter observed that both the UNC Application and the Duke Arringdon application proposed to place ORs in south Durham County, and both were deemed the more effective alternative as to this comparative factor, which they agree was the correct decision. (Meyer, Vol. 7, pp. 1330-31; Carter, Vol. 11, pp. 1886-87).

430. While Durham County has relatively small land mass compared to other counties, Durham County is the third most densely populated county in the state, and such density leads to traffic congestion that can make geographic dispersion of healthcare facilities more important. (Meyer, Vol. 7, pp. 1306-07, 1309-10).

431. Ms. Sandlin produced two maps showing different amounts of population density in Durham County. In Sandlin’s initial expert report, the map showing population

11 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

density illustrated that UNC Hospitals-RTP would be located in a densely-populated area of the county where there are no existing hospitals. (Jt. Ex. 54, p. 12; Meyer, Vol. 7, p. 1309). However, in Sandlin’s rebuttal report, the map showing population density illustrated there is no population in the zip code where UNC Hospitals-RTP would be located, but still showed that the surrounding zip codes are densely populated. (Jt. Ex. 212; Meyer, Vol. 7, pp. 1307-09).[] [A footnote affixed to this finding in the original text reads as follows: “Similarly, there is no population in the zip code that comprises DUH. (Jt. Ex. 4, p. 242; Sandlin, Vol. 7, p. 1201; Carter, Vol. 11, p. 1903).”]

432. Mr. Meyer opined that despite the lack of population in UNC Hospitals-RTP’s zip code, UNC’s primary site is easily accessible by “the largest, most significant traffic arteries in that part of the county” such that residents in densely-populated southern Durham County would have easy access. (Meyer, Vol. 7, pp. 1308-09).

433. Mr. Carter likewise explained that the UNC Application illustrated that UNC Hospitals-RTP is located along prominent roadways in addition to being located near the heavily populated southern Durham zip codes. (Carter, Vol. 10, p. 1703; see also Jt. Ex. 4, pp. 51-58).

434. Ms. Sandlin also opined that UNC Hospitals-RTP is not near a majority of Durham County zip codes and that this does not improve geographic access for the majority of the service area zip codes. (Sandlin, Vol. 6, p. 1061).

435. In contradiction, Mr. Meyer noted that it is more important for a healthcare facility to be proximate to more people, rather than more zip codes. (Meyer, Vol. 7, p. 1310). The zip codes in southern Durham County which are near UNC Hospitals-RTP “comprise more than half of the population of Durham County.” (Jt. Ex. 4, p. 55; Meyer, Vol. 7, p. 1310; Sandlin, Vol. 7, pp. 1205-06).

12 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

436. When looking at population rather than zip codes, UNC Hospitals-RTP was proximate to over half of the population of Durham County. (Meyer, Vol. 7, p. 1311-12).

437. Mr. Carter added that UNC Hospitals-RTP’s primary site is “on the border of RTP” and is “near where a lot of people live.” (Carter, Vol. 11, pp. 1904-05). He further opined that UNC Hospitals-RTP’s location being in the southern region of Durham County improves access by providing another option for those residents. While some of those residents may still choose one of the existing facilities, they have another option that may be closer to where they live. (Carter, Vol. 10, p. 1733). Furthermore, compared to DUH, UNC Hospitals-RTP would be easier to find parking and navigate as a smaller facility. (Id. at pp. 1733-34).

438. The fact that DUH may be closer to some residents in Caswell County and northern Durham County does not change the Agency’s analysis that UNC Hospitals-RTP enhances geographic accessibility. In Mr. Meyer’s opinion:

[R]esidents of northern Durham County are not going to be disadvantaged by this proposal. They will continue to have the same access to any of those existing acute care hospitals that they do currently. This doesn’t take away from their access.

(Meyer, Vol. 7, pp. 1313-14). Instead, UNC’s proposal “enhances access for south Durham County residents,” which is where the greatest need exists for these services due to the population growth in that area. (Id. at p. 1314).

439. As a small hospital, “the intent is not to serve each and every patient within Durham County,” because UNC Hospitals-RTP does not “have the capacity to do that.” (Carter, Vol. 10, pp. 1703-04).

440. Ms. Sandlin testified that the Agency’s analysis of this comparative factor was inconsistent with the way the Agency analyzed it in prior reviews. (Sandlin, Vol. 6,

13 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

pp. 1045-46).

441. Mr. Meyer disagreed with Ms. Sandlin of the Agency’s prior reviews. While he interpreted Ms. Sandlin’s testimony as opining that the Agency needs to analyze geographic accessibility based on municipalities, Mr. Meyer noted that there is no rule requiring that. Moreover, analyzing geographic accessibility based on municipalities is impractical in Durham County, where there is only one incorporated municipality, the City of Durham. (Meyer, Vol. 7, pp. 1314-15). More importantly, the geographic accessibility comparative factor should look at where people live compared to the existing and proposed services. (Id. at 1315-16).

442. Likewise, Mr. Carter disagreed with Ms. Sandlin. In his opinion, the 2020 Forsyth Acute Care Beds Review mentioned by Ms. Sandlin was an inapt comparison, where the existing hospitals were more dispersed than the existing facilities within Durham that are contained in a five-mile radius. (Carter, Vol. 11, p. 1877)

443. Ms. Sandlin testified that UNC’s analysis splitting Durham into different regions based on zip codes “seemed manufactured and illogical.” (Sandlin, Vol. 6, p. 1017).

444. However, Ms. Sandlin’s testimony ignores the fact that Duke itself, assisted by Keystone Planning while Ms. Sandlin was still with that company, analyzed geographic accessibility in this same “manufactured” manner in its 2018 application to develop the Duke Arringdon facility. In its 2018 application, Duke described the same four zip codes (27703, 27709, 27707 and 27713) as “South Durham” that UNC described as south Durham in its application in this Review. (Compare Jt. Ex. 106, p. 30 with Jt. Ex. 4, p. 54; see also Meyer, Vol. 7, pp. 1317-18; Sandlin, Vol. 6, pp. 1120-22).

445. Mr. Carter explained the process by which UNC determined to split Durham County into regions and

14 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

concluded that UNC divided Durham County into three regions by zip codes so it could analyze where in the county a new hospital should be located, which the SMFP does not discuss in any detail. (Carter, Vol. 10, pp. 1704-06). Mr. Carter further opined that not all patients within the City of Durham were equally served by the existing hospitals due to the lack of available facilities in southern Durham. In other words, “there aren’t enough facilities to serve residents in Durham County notwithstanding the fact that the municipality of Durham may go well into the southern part of the county.” (Id. at p. 1708).

446. Ultimately, Mr. Meyer agreed with the Agency’s analysis of this comparative factor, describing it as “an easy call for the Agency.” (Meyer, Vol. 7, p. 1318).

447. Mr. Carter agreed that the Agency was correct in determining the UNC was the more effective alternative, and that it was consistent with other findings he has seen. (Carter, Vol. 11, pp. 1874, 1886). Mr. Carter further opined that he did not believe “the Agency’s analysis or conclusions would have been any different if UNC had proposed a different site really anywhere else in the county that was not within five miles of another hospital.” (Id. at p. 1877).

Reviewing the record for substantial evidence, see Surgical Care Affiliates, 235 N.C.

App. at 622-23, we affirm the ALJ’s decision with respect to this factor.

At the threshold, we note that Duke has primarily framed its arguments as

though our task on appeal were to review the determinations of the Agency rather

than the ALJ. However, this is incorrect. While the statute governing judicial review

of administrative decisions, N.C.G.S. § 150B-51, used to contemplate direct judicial

review of Agency determinations, revisions by our General Assembly in 2011 have

refocused our substantive review on the final decision of the ALJ:

15 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

In 2011, the General Assembly amended the Administrative Procedure Act (“APA”), conferring upon administrative law judges the authority to render final decisions in challenges to agency actions, a power that had previously been held by the agencies themselves. See 2011 N.C. Sess. Laws 1678, 1685-97, ch. 398, §§ 15-55. Prior to the enactment of the 2011 amendments, an ALJ hearing a contested case would issue a recommended decision to the agency, and the agency would then issue a final decision. In its final decision, the agency could adopt the ALJ’s recommended decision in toto, reject certain portions of the decision if it specifically set forth its reasons for doing so, or reject the ALJ’s recommended decision in full if it was clearly contrary to the preponderance of the evidence. See [N.C.G.S.] § 150B36, repealed by 2011 N.C. Sess. Laws 1678, 1687, ch. 398, § 20. As a result of the 2011 amendments, however, the ALJ’s decision is no longer a recommendation to the agency but is instead the final decision in the contested case. [N.C.G.S.] § 150B–34(a).

Under this new statutory framework, an ALJ must “make a final decision . . . that contains findings of fact and conclusions of law” and “decide the case based upon the preponderance of the evidence, giving due regard to the demonstrated knowledge and expertise of the agency with respect to facts and inferences within the specialized knowledge of the agency.” Id.

AH N.C. Owner LLC v. N.C. Dep’t of Health & Hum. Servs., 240 N.C. App. 92, 98-99

(2015). Thus, our review of substantive issues will be based on the ALJ’s final

decision.

Having established the proper scope of our review, we are entirely satisfied

that substantial evidence exists to support each of the arguments Duke raises on

appeal. While Duke argues that the ALJ’s decision was reversible insofar as it found

UNC’s application favorable on the basis of geographic access in a zip code with no

16 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

residents, the ALJ cited substantial evidence indicating that the immediately

adjacent zip codes are densely populated—to say nothing of the potential usage the

proposed location may receive from those who work, rather than reside, in the

proposed location of the UNC facility. As to UNC’s allegation that the Agency

deviated from its mode of analysis in previous reviews, rendering its decision

arbitrary and capricious, we cannot say a deviation without a more specific argument

as to why the analysis employed in this case was deficient that such an alleged

deviation constitutes reversible error, especially absent any directly binding law on

point to support such a proposition. The task before the Agency is multifaceted, and

the CON review process does not demand that it apply a fixed lens to every case,

especially where some considerations may be more salient in a given case than in

others. The ALJ’s findings and conclusions with respect to geographic access are

affirmed.

B. Relative Impact on Competition

Second, we address whether the ALJ properly affirmed the Agency’s

conclusions as to the Duke and UNC applications’ relative impact on competition.

Duke argues that the ALJ erroneously affirmed the Agency’s decision with respect to

this comparative factor because the Agency believed the comparative factor of

promoting market competition would always favor a new market entrant and because

the Agency failed to consider “quality, cost, and access” as part of the competition

factor. With these arguments, too, we disagree.

17 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

While the ALJ’s final decision does discuss this factor, we note that Duke’s

stance on this issue takes the form of a broad methodological critique rather than an

allegation that a specific analytical error occurred, making reproduction of this

portion of the record unnecessary. To the extent this argument constitutes an

allegation of legal error, we apply the de novo, rather than whole record, standard of

review. N.C.G.S. § 150B-51(c) (2023) (“With regard to asserted errors pursuant to

subdivisions (1) through (4) of subsection (b) of this section, [subsection (b)(4)

referring to “other error[s] of law[,]”] the court shall conduct its review of the final

decision using the de novo standard of review.”).

At the threshold, we note once again that Duke’s arguments principally

concern the determinations of the Agency and not the ALJ. However, as the ALJ’s

final decision is the proper object of our review, see AH, 240 N.C. App. at 98-99, we

base our analysis primarily on that decision. Bearing that in mind, very few of the

issues raised by Duke on appeal directly apply to the ALJ’s final decision. The

alleged defect that the Agency believed the competition factor would always favor a

new market entrant—a view found neither in the Agency’s written decision nor the

final decision of the ALJ, but sourced to testimony by Agency employees before the

Office of Administrative Hearings—was not present in the reasoning of the ALJ, who

indicated a typical preference for a new market competitor rather than a categorical

one.

However, even if the ALJ’s view had been as categorical as the view Duke

18 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

imputes to the Agency, this would hardly be a case where such reasoning would merit

reversal on appeal. Duke has not disputed the ALJ’s finding that, of the 1,388 acute

care beds in Durham County, only twenty are outside Duke’s control. Nor has Duke

otherwise presented us with any reason to believe UNC’s facility would present more

of a threat to competition for this service in Durham County than its own market

dominance.2 Rather, its arguments largely reduce to a contention that it could not

realistically “win” the competition factor. Barring radically extenuating

circumstances, we do not think an entity controlling more than 98% of a service

within a county should realistically expect to “win” when a neutral third party

considers whether a new market entrant would be the healthier choice for

competition. Cf. Craven Reg’l Med. Auth. v. N.C. Dep’t of Health & Hum. Servs., 176

N.C. App. 46, 57 (2006) (“[The petitioner]’s argument appears to be that if it operated

all three of the MRI scanners this would somehow foster competition rather than if a

competitor operated one of the MRI scanners. [The petitioner], in effect, argues that

giving it a monopoly in the service area would increase competition. We decline to

adopt this incongruous line of reasoning.”).

2 Duke points out that UNC, despite currently operating no acute care beds in Durham County,

is already a major medical provider in the greater triangle region, and it further contests the adequacy of the ALJ’s analysis as to competition on this basis. While we recognize Duke’s concern insofar as a regional oligopoly may be unhealthy for the state of market competition in the absolute sense, the ALJ’s assessment of competition was relative, not absolute. Thus, we cannot say the ALJ erred in its determination that, as between the two regionally dominant providers being considered in the competitive application process, the one not currently operating acute care beds within Durham County creates a more favorable impact on competition within the county than the one currently wielding a near-monopoly for that service.

19 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

Duke also argues that the failure to consider cost and quality of care within

the scope of the competition factor rendered its decision reversibly arbitrary. This

argument is meritless. Impact on the health of market competition is one of eleven

factors considered in the competitive CON review process, several others of which

account for cost and quality of care. We affirm the ALJ’s determinations as to relative

impact on competition.

C. UNC’s Compliance with Criterion 3

We next address whether the ALJ properly affirmed the Agency’s conclusions

as to UNC’s compliance with N.C.G.S. § 131E-183(a)(3). N.C.G.S. § 131E-183(a)(3),

or “Criterion 3,” provides that a certificate of need applicant

shall identify the population to be served by the proposed project, and shall demonstrate the need that this population has for the services proposed, and the extent to which all residents of the area, and, in particular, low income persons, racial and ethnic minorities, women, handicapped persons, the elderly, and other underserved groups are likely to have access to the services proposed.

N.C.G.S. § 131E-183(a)(3) (2023). With respect to Criterion 3, Duke argues that

UNC’s application was insufficient because it relied on unrealistically low projections

for the number of out-of-county patients the proposed facility could be expected to

attract and because UNC’s application allegedly failed to account for the absence of

20 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

high-acuity care at the proposed facility.3 As these arguments are derived from

factual disagreements with the Agency findings—which, in the ALJ review, were

supported by substantial evidence, see Surgical Care Affiliates, 235 N.C. App. at 622-

23—we affirm the ALJ.

In its final decision, the ALJ affirmed the Agency’s conclusion that UNC’s CON

application was in compliance with criterion 3, finding, in relevant part, as follows:

85. Criterion (3) requires the applicant to “identify the population to be served by the proposed project” and to “demonstrate the need that this population has for the services proposed, and the extent to which all residents of the area, and, in particular, low-income persons, racial and ethnic minorities, women, handicapped persons, the elderly, and other underserved groups are likely to have access to the services proposed.” ([N.C.G.S.] § 131E- 183(a)(3); Jt. Ex. 1, p. 1502).

86. To find an applicant conforming with this Criterion, the Agency engages in a four-part analysis: (1) the applicant must identify the population to be served, also referred to as the patient origin; (2) the applicant must demonstrate the need of the identified population for the services proposed; (3) the applicant must project the utilization of these services by the identified population in the first three operating years of the project; and (4) the applicant must project the extent to which the projected population, and particularly those in medically underserved groups, have access to the proposed services. (Jt. Ex. 1, p. 1502; Hale, Vol. 2, p. 224; see also Meyer, Vol. 5, p. 936). To be found conforming, the information provided by the applicant must be reasonable and adequately supported. (Hale, Vol.

3 Duke also argues that UNC’s alleged nonconformity with criterion 3 brings it out of conformity with criteria 1, 4, 5, 6, and 18(a). However, because we determine below that Duke’s arguments with respect to criterion 3 are without merit, we need not independently evaluate this argument.

21 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

2, pp. 223-24).

i. Patient Origin

87. The first element of Criterion (3) discusses patient origin, which is where the applicant projects patients will come from to utilize the proposed services. (Jt. Ex. 1, p. 1509; Hale, Vol. 2, p. 225). To analyze patient origin, the Agency reviews the information provided by the applicant and determines whether that information is reasonable and adequately supported. (Hale, Vol. 2, pp. 225-26).

88. The UNC Application provided that the patient origin for UNC Hospitals-RTP would include 90 percent Durham County residents, with some in-migration from Wake, Chatham, and Caswell Counties. (Jt. Ex. 4, p. 43; Carter, Vol. 10, pp. 1690-92).

89. To determine its projected patient origin, UNC considered the limited size of the facility and the overwhelming need in Durham County. While UNC could have used a higher percentage of in-migration in its projections, doing so would have been more aggressive, especially given that a small hospital would be less likely to attract patients from outside of the county. (Carter, Vol. 10, pp. 1692-93).

90. Ms. Sandlin acknowledged that her opinions regarding UNC’s projected patient origin, in-migration, and patient population were not based on any Duke facilities of similar size, since there are none. She also did not perform any analysis of the patient origin of a hospital of similar size developed by UNC in developing her opinions. (Sandlin, Vol. 7, pp. 1165-66).

91. Daniel Carter, one of UNC’s expert witnesses, opined that UNC’s 10 percent in-migration assumption was well- supported, reasonable, and conservative. (Carter, Vol. 10, pp. 1695-96). The UNC Application analyzed in-migration at all 116 acute care hospitals in North Carolina to reach its 10 percent in-migration assumption, and it also

22 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

accounted for UNC Hospitals-RTP’s smaller size and densely populated location. (Jt. Ex. 4, pp. 146-47; Carter, Vol. 10, pp. 1693, 1695).

92. Mr. Carter analogized UNC Hospitals-RTP to UNC Johnston Health in Clayton, a 50-bed community hospital which is approximately the same distance from Wake County as UNC Hospitals-RTP would be. At UNC Johnston Health, there is approximately 9 percent in- migration from Wake County despite its proximity. (Carter, Vol. 10, pp. 1693-94).

93. Mr. Carter also noted that had UNC proposed higher in-migration, it would also have the effect of increasing UNC Hospitals-RTP’s utilization and the financial feasibility of the project, which would strengthen its application for both Criteria (3) and (5). (Id. at p. 1693). Furthermore, he noted that UNC could have supported an assumption of 20 percent or even 30 percent in-migration without going beyond its maximum utilization. (Id. at pp. 1694-95).

94. Based upon the information provided in the UNC Application, the Agency determined that UNC adequately identified the patient origin for the population it proposed to serve. (Jt. Ex. 1, p. 1511; Hale, Vol. 2, pp. 226-27).

ii. Demonstration of Need

95. The second element of Criterion (3) analyzes whether the applicant demonstrates that the population proposed to be served needs the proposed services. (Jt. Ex. 1, p. 1511; Hale, Vol. 2, p. 231-32). To conduct its analysis of need, the Agency reviews the information provided by the applicant and assesses whether that information is reasonable and adequately supported. (Hale, Vol. 2, pp. 231-32). This differs from the need determination of Criterion (1), which focuses on the need determination in the SMFP, rather than the needs of patients for the proposed services.

96. UNC provided several reasons why the patients it

23 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

proposed to serve at UNC Hospitals-RTP needed the proposed services. The Agency determined that UNC’s methodology and resulting projections were both reasonable and adequately supported. (Sandlin, Vol. 7, p. 1214).

97. The first reason provided by UNC is the population growth and aging in Durham County. (Jt. Ex. 4, pp. 48- 50). UNC noted that Durham County is the sixth most populous county and the third fastest growing county in North Carolina, with the growth rate expected to continue into the next decade. (Id. at 48-49). This growth, combined with the aging of the population, demonstrated that there will be more patients needing acute care services. (Id. at 49-50; Carter, Vol. 10, pp. 1700-01).

98. The second reason provided by UNC is the need for a new hospital in Durham County. As of the date the applications were submitted, there were no acute care beds in the southernmost zip codes in Durham County, where most of the population and growth exists within the county. (Jt. Ex. 4, pp. 51-55). The UNC Application contained the following map illustrating the location of existing hospitals in Durham County and the proposed UNC Hospitals-RTP location:

24 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

(Id. at 51; see also id. at 53; Carter, Vol. 10, pp. 1710-11).

99. Additionally, UNC demonstrated that its proposed services were needed because (1) there has not been a new hospital opened in Durham County in over 45 years and (2) Durham County lacks a full-service community hospital. (Jt. Ex. 4, pp. 51-52).

100. The UNC Application included a table which displayed UNC’s existing market share of certain zip codes within Durham County. This table showed that UNC already has a strong market presence in southern Durham County (including zip codes 27703, 27713, 27707, 27709) despite not having any facilities there. (Id. at 54; Carter, Vol. 10, pp. 1711-12).

101. The UNC Application also included a table which displayed the historical population growth by region and zip code within Durham County. This table showed that a

25 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

majority of the Durham County population lives in the southern zip codes. As of 2020, 165,824 out of 326,262 people live in the southern zip codes. In addition, those southern zip codes are the fastest growing zip codes with a compound annual growth rate (“CAGR”) of 2.4% between 2015 and 2020 and expected CAGR of 1.9% between 2020 and 2025. (Jt. Ex. 4, p. 55).

102. In further support of the need for a community hospital in southern Durham County, UNC described the development of roadways and businesses in southern Durham County to emphasize the “sustained growth and development” of southern Durham County that supports the need for UNC Hospitals-RTP. (Id. at pp. 56-58; Carter, Vol. 10, pp. 1713-14).

103. While the SMFP never states that there is a need for any hospital, the fact that there is a need for both beds and ORs in the same area offers the potential for a new hospital. Combined with the need for low acuity services in southern Durham County, there is a need for a community hospital in Durham County. (Carter, Vol. 10, pp. 1696-98).

104. UNC examined the entire Durham/Caswell service area when deciding where to locate its hospital. UNC determined that Caswell County was not an ideal location for a hospital due to its relative lack of population and determined that southern Durham County was ideal based on the need in those densely populated zip codes that lacked a hospital. (Id. at pp. 1699-702; Jt. Ex. 4, pp. 50-55).

105. A third reason provided by UNC is the need for UNC Hospitals hospital-based services in Durham County. A significant number of patients from Durham County use UNC Health facilities and developing a community hospital closer to them would meet their needs for higher frequency, lower acuity services. (Jt. Ex. 4, pp. 58-60; Carter, Vol. 10, pp. 1714-15).

106. UNC already has physicians in Durham County that

26 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

are part of UNC Health. UNC is focused on meeting the physician needs in the area and would recruit physicians to meet those needs. (Carter, Vol. 10, pp. 1715-16; see also Jt. Ex. 4, pp. 58-59, 382-511). Moreover, UNC Hospitals- RTP would have the same provider number as UNC Hospitals, so the same medical staff that performs surgery in Chapel Hill could do so at UNC Hospitals-RTP. (Carter, Vol. 10, pp. 1716-17; see also Jt. Ex. 4, p. 152; Hadar consistent testimony at Vol. 8, pp. 1464-65).

107. UNC already serves a large number of Durham County residents even without having a hospital in Durham County. Moreover, around one-half of patients in a hospital may not need surgery, and the hospitalists that would provide those services at UNC Hospitals could also provide those services at UNC Hospitals-RTP. (Carter, Vol. 10, pp. 1718-19).

108. The UNC Application further supported the need for UNC Hospitals services in Durham County by describing how UNC Hospitals-RTP “represents an exciting opportunity to develop a new hospital facility with innovation as a central design tenet.” (Jt. Ex. 4, p. 59). Mr. Carter explained that UNC felt that this opportunity to build a new hospital in Durham County, which had not presented itself for over 40 years, would allow UNC to provide care in a more modern, unique, and innovative way, as it described doing at its other facilities. (Carter, Vol. 10, p. 1720; Jt. Ex. 4, pp. 58-61).

109. The UNC Application provided examples of its “long history of embracing innovation to deliver the highest quality care with the best patient experience.” (Jt. Ex. 4, pp. 60-61). In developing this application, administrators of REX Holly Springs and Johnston Health Clayton provided input of lessons learned from the development of these relatively new hospitals that could be incorporated into the development of UNC Hospitals-RTP. (Carter, Vol. 10, pp. 1721-23; Jt. Ex. 4, pp. 60-61).

110. As a fourth supporting reason, UNC explained that

27 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

UNC Hospitals-RTP meets the need for acute care beds by providing lower acuity community hospital beds in particular, as it projected that convenient, local access to community hospital services was the primary driver of need for additional acute care beds in the service area. (Jt. Ex. 4, pp. 62-69; Carter, Vol. 10, pp. 1723-30).

111. UNC identified certain lower acuity, high volume services as “selected services,” and then analyzed Truven data to illustrate how, “despite the growth at existing tertiary and quaternary facilities in Durham, the basis of this growth was the need for lower acuity, community hospital services.” (Jt. Ex. 4, p. 65; Carter, Vol. 10, p. 1726).

112. UNC demonstrated that of the existing hospitals in Durham County, Duke Regional is the fastest growing. (Jt. Ex. 4, p. 64; Carter, Vol. 10, p. 1727). UNC then showed that the selected services were experiencing greater growth than other services in the existing Durham hospitals as a whole, and at DUH and Duke Regional in particular. (Jt. Ex. 4, p. 65; Carter, Vol. 10, pp. 1727-29).

113. UNC further demonstrated that south Durham County residents are seeking lower acuity services more than the central and north regions of Durham County, with over 94 patients daily seeking lower acuity services at existing hospitals. (Jt. Ex. 4, p. 66; Carter, Vol. 10, pp. 1731-33).

114. The UNC Application showed that UNC currently provides the most days of care and experiences the greatest growth for Durham County residents out of all other hospitals except for Duke facilities, and that out of those patients, the highest volume originates from the south region of Durham County. (Jt. Ex. 4, pp. 68-69; Carter, Vol. 10, pp. 1734-36).

115. The UNC Application further showed that UNC Hospitals-RTP meets the need for ORs by providing additional hospital-based ORs, which are well-utilized and provide flexibility and capacity not otherwise available

28 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

when those ORs are placed in an ambulatory surgical facility. (Jt. Ex. 4, pp. 69-71). Notably, UNC pointed out that while inpatient surgeries have grown at a slower rate than outpatient surgeries statewide, that trend is the opposite in Durham County. (Id. at pp. 69-70; Carter, Vol. 10, pp. 1736-37). UNC also indicated that there has been significant growth in outpatient ORs at ASCs, but that hospital-based ORs would provide the flexibility to meet the need for inpatient surgeries while still allowing for outpatient surgeries to be performed as well. (Jt. Ex. 4, pp. 70-71; Carter, Vol. 10, pp. 1737-38).

116. UNC also supported the need for other services at UNC Hospitals-RTP, including observation beds, procedure rooms, C-Section rooms, imaging, laboratory, and other services, which are needed to support the patients to be seen at UNC Hospitals-RTP. (Jt. Ex. 4, p. 71; Carter, Vol. 10, p. 1738).

117. Based on the information UNC provided, the Agency found UNC’s analysis of need to be reasonable and adequately supported. (Jt. Ex. 1, []p. 1512; Hale, Vol. 2, pp. 232-34).

iii. Projected Utilization

125. The third element of Criterion (3) evaluates the reasonableness and adequacy of the support for the applicant’s projected utilization. (Hale, Vol. 2, p. 235).

126. The Agency does not require applicants to use particular assumptions or methodologies to develop their utilization projections; instead, the assumptions and methodology used by each applicant must be reasonable and adequately supported. (Cummer, Vol. 4, p. 670; Sandlin, Vol. 6, pp. 1115-16).

127. Ms. Sandlin acknowledged that projected utilization at a facility may not necessarily line up with an applicant’s

29 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

actual experience for various reasons. (Sandlin, Vol. 7, pp. 1193-94).

128. The need methodology and projected utilization for the UNC Application were contained in Form C Utilization – Assumptions and Methodology in Section Q of the application. (Jt. Ex. 4, pp. 141-60). UNC projected utilization for the acute care services, surgical services, and ancillary and support services proposed in its application. (Jt. Ex. 1, pp. 1512-20; Hale, Vol. 2, pp. 236-39).

129. UNC used Truven data as the basis for its utilization projections, which both the Agency witness and expert witnesses agreed is frequently utilized by applicants and is a reliable source of data. (Hale, Tr. pp. 237-38; Meyer, Vol. 5, pp. 941-43; Carter, Vol. 11, pp. 1953-55).

130. At the hearing, Mr. Carter explained in detail the assumptions and methodologies used in the UNC Application. The UNC Application began by describing the service area and emphasizing the focus on Durham County, which “sets the stage for” UNC’s focus on Durham County in the methodology. (Jt. Ex. 4, pp. 141-42; Carter, Vol. 10, pp. 1739-40).

a. Selected Services

131. The UNC Application next discussed acute care bed utilization, looking first to all days of care for Durham County residents statewide. (Jt. Ex. 4, p. 142; Carter, Vol. 10, p. 1740). Mr. Carter notes that while many methodologies look no further than this, the UNC Application took the extra step of identifying certain high acuity services that it would exclude from the potential days of care to be provided at UNC Hospitals-RTP, as UNC did not propose to provide high acuity, tertiary and quaternary services at UNC Hospitals-RTP. (Jt. Ex. 4, pp. 142-43; Carter, Vol. 10, pp. 1740-41).

132. The remaining services utilized by UNC were called the Selected Services. (See Jt. Ex. 4, p. 143).

30 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

133. The decision to exclude certain services was the product of discussions within UNC and the expertise of Mr. Carter. Certain services like cardiac catheterization were excluded because there was no need for a cardiac catheterization unit in the SMFP; other services like neurosurgery could have been included, but given that UNC Hospitals is located nearby, it made sense not to duplicate those services. Moreover, given that UNC Hospitals-RTP is proposed to be a community hospital, UNC prioritized lower-acuity, high-frequency, high- volume cases. (Carter, Vol. 10, pp. 1744-45).

134. UNC decided not to include ICU services at UNC Hospitals-RTP in part based on its recent experience developing community hospitals in Wake and Johnston Counties. Through those facilities, UNC learned that it did not make sense to develop ICU units due to the low volume of patients needing those services compared to the resource-intensive staffing that is required for those beds. (Id. at pp. 1763-65).

135. As explained in the UNC Application, the rooms at UNC Hospitals-RTP were designed to be flexible spaces that would be built to standards such that they could provide ICU-level care as needed. (Jt. Ex. 4, p. 38). If UNC Hospitals-RTP learns as it begins operating that more ICU beds are needed, it could decide to make those beds permanent ICU beds, which would not require any additional construction or renovation, or any CON approval. (Carter, Vol. 10, pp. 1761-62, 1765).

136. UNC accomplished the exclusion of high acuity services from its analysis by removing diagnosis related groups (“DRGs”) associated with the excluded high acuity services from the dataset. (Carter, Vol. 10, pp. 1741-42, Vol. 11, pp. 1897-98). The exclusion of these services resulted in a 31.1 percent reduction in 2019 days of care for Durham County residents. (Jt. Ex. 4, p. 143; Carter, Vol. 10, pp. 1742-44).

31 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

137. While the Agency does not require applicants to exclude services in its methodology, UNC chose to do so to underscore the conservativeness of its projections and to reiterate UNC’s intention not to develop a quaternary academic medical center in Durham County. (Carter, Vol. 10, pp. 1742-43).

138. Ms. Sandlin did not conduct any analysis utilizing DRG weights to determine the reasonableness of UNC’s projections. (Sandlin, Vol. 7, p. 1222; Carter, Vol. 10, pp. 1767-68). She also opined that there is no specific cutoff or threshold for DRG weights that are associated with ICU level of care. (Sandlin, Vol. 7, p. 1223).

139. Mr. Carter likewise opined that there is no bright-line rule for a DRG weight for ICU services. (Carter, Vol. 10, pp. 1756-58).

140. Mr. Carter even analyzed the data UNC relied upon in its analysis and discovered that had UNC applied a bright-line rule excluding DRG weights of over 3.5, only approximately ten percent of the patient days of care for UNC Hospitals-RTP were over that threshold. (Id. at pp. 1759-61).

141. Moreover, those patients without exception had a comorbid condition or major complication that led their condition to progress beyond a 3.5 DRG weight. In those cases, if UNC Hospitals-RTP could not provide the higher level of care needed, they could be transferred to an appropriate facility. (Id. at pp. 1760-61).

142. Ultimately, even if there were ICU patients that were not excluded from UNC Hospitals-RTP’s selected services patients, the projections in the UNC Application would not be impacted. (Id. at p. 1762).

143. Ms. Sandlin created and utilized a Venn diagram as a demonstrative exhibit to show the alleged overlap between UNC’s selected services, ICU, post-ICU, and pediatric patients. (Duke Ex. 227). On cross-examination, however,

32 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

Ms. Sandlin admitted that she did not know what percentage each of the “bubbles” or “circles” on her diagram represented for each service and that her exhibit was not drawn to scale. (Sandlin, Vol. 7, pp. 1218-20). Ms. Sandlin further acknowledged that she did not quantify the numbers or percentage of patients that the diagram was intended to represent. (Sandlin, Vol. 7, p. 1220; Carter, Vol. 10, pp. 1765-67).

144. Regardless of the exclusion of certain high acuity services, UNC Hospitals-RTP will be able to stabilize high acuity patients in an emergency in need of tertiary or quaternary care and transfer them to another hospital that can treat their condition, as it does at its other community hospitals in the greater Triangle area. (Carter, Vol. 10, pp. 1745-46; Hadar, Vol. 8, p. 1454).

b. Methodology

145. Next, UNC projected potential days of care for the selected services in Medicine, Surgery, and Obstetrics through 2029, which is the third project year, using a CAGR based on historical growth rate for those services. (Jt. Ex. 4, pp. 143-44; Carter, Vol. 10, pp. 1746-47). Duke, in its expert testimony, did not criticize UNC’s growth rates or methodology included on page 144 of the UNC Application. Mr. Carter opined the growth rates and methodology to be reasonable based on the historical growth rates for Durham County. (Carter, Vol. 10, p. 1747). UNC then showed the potential days of care for Durham County residents for the first three fiscal years of the project. (Jt. Ex. 4, p. 144; Carter, Vol. 10, p. 1747).

146. After that, UNC discussed its market share assumptions for UNC Hospitals-RTP, which is typically analyzed for any new healthcare facility that needs to project a volume of services to be provided. (Carter, Vol. 10, pp. 1747-48). Since UNC already treats many Durham County patients at its existing facilities outside of Durham County, UNC conservatively projected that UNC Hospitals-RTP would serve three-fourths of UNC’s existing

33 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

market share of Durham County residents. (Jt. Ex. 4, p. 145; Carter, Vol. 10, pp. 1748-50). In the third full project year, this results in a 7.7 percent market share of Durham County patient days for the selected services, leaving 92.3 percent of Durham County patient days to be treated at any other facility in the state. (Carter, Vol. 10, pp. 1750- 52).

147. After isolating Durham County and narrowing down days of care based on selected services and UNC’s market share of Durham County patient days, UNC was then able to project the patient days by service for Durham County residents, yielding an average daily census (“ADC”) of 26.5 patients in the third project year. (Jt. Ex. 4, p. 146; Carter, Vol. 10, pp. 1768-69).

148. The next part of the methodology in the UNC Application demonstrated why the 26.5 ADC was reasonable. UNC noted that its 2019 ADC for Durham County residents for selected services at its existing facilities was 24.4. This highlighted how reasonable and conservative it is to project that UNC Hospitals-RTP would serve only about two more patients per day than UNC currently serves, after UNC Hospitals-RTP is open and operational. (Jt. Ex. 4, p. 146; Carter, Vol. 10, p. 1769). UNC also provided more information about its in- migration assumptions. (Jt. Ex. 4, pp. 146-47; Carter, Vol. 10, pp. 1769-70).

149. UNC further highlighted the conservativeness of its methodology by noting that the amount of patients UNC Hospitals-RTP projects to serve is only part of the projected growth of Durham County residents over the next ten years. (Jt. Ex. 4, p. 148; Carter, Vol. 10, pp. 1770-71). In comparison, the Duke Beds Application proposed to increase patient days by roughly 40,000 in less than ten years. (Jt. Ex. 2, p. 95; Carter, Vol. 10, pp. 1771-72). Based on this observation, Mr. Carter opined that it was not unreasonable for the UNC Application to project to reach 10,700 patient days over a ten-year period of time, especially since UNC already had more patient days for

34 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

these lower acuity services at hospitals outside of Durham County. (Carter, Vol. 10, pp. 1772-73).

150. In its Comments, Duke claimed that UNC relied on a shift in volume to support its projections. (Jt. Ex. 1, pp. 176-78; Sandlin, Vol. 6, p. 990). UNC responded, however, that this claim was incorrect, because UNC was taking a portion of the new growth in patient days in Durham County. (Jt. Ex. 1, pp. 309-12; Carter, Vol. 10, pp. 1773- 75). Regardless, Ms. Sandlin acknowledged that it is reasonable in theory to assume that developing a facility in an area where patients live will cause the existing market share for that provider to increase. (Sandlin, Vol. 6, pp. 1115-16).[4]

151. Ms. Sandlin testified that UNC’s projections were unreasonable because the patients that UNC currently treats are going to UNC Hospitals for specialty services. (Id. at pp. 994-96). Mr. Carter refuted Ms. Sandlin’s testimony, opining that Ms. Sandlin ignored UNC’s exclusion of high acuity patients in its methodology. (Carter, Vol. 10, pp. 1775-76). Moreover, Ms. Sandlin acknowledged that she had not done any analysis of the acuity level of services provided to Durham County patients currently seeking care at UNC. (Sandlin, Vol. 7, pp. 1159-60).

152. UNC also projected emergency department (“ED”) utilization in its assumptions and methodologies. (Jt. Ex. 4, pp. 149-51; Carter, Vol. 10, pp. 1776-77). A hospital is required to have an emergency department in North Carolina, though there are no statutes or rules that apply to emergency department projections. (Sandlin, Vol. 7, p. 1215; Carter, Vol. 10, pp. 1778-79).

4 At several points in its final decision—most notably, findings 150 and 155—the ALJ used

language that signaled the existence of conflicts in the evidence without explicitly clarifying which testimony it deemed more credible. While these areas of the final decision were not specifically challenged on the basis of indecisive wording, we note that, in other areas of our caselaw, a gesture to conflicts in the evidence without an explicit resolution by the factfinder may support a challenge on appeal to the finding in question. We therefore note that the better practice for a factfinder is to explicitly, rather than implicitly, signal how it resolves conflicts in evidence.

35 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

153. UNC’s ED utilization projections were not based solely on ED admissions in Durham County; rather, it analyzed all ED admissions of Durham County residents receiving care throughout the state. (Jt. Ex. 4, p. 150; Carter, Vol. 10, pp. 1777-78). As Mr. Carter opined, even if the ED utilization projection methodology was wrong, as a hospital, UNC Hospitals-RTP is required to include an ED, and there is no standard the Agency applies to ED utilization that would cause the UNC Application to not be approvable. (Carter, Vol. 10, pp. 1778-79).

154. UNC began projecting OR utilization by assuming that each surgical inpatient is one surgical inpatient case. (Jt. Ex. 4, pp. 155-56; Carter, Vol. 10, p. 1779). UNC then analyzed projected outpatient cases and concluded that there would be 1.5 outpatient surgeries for every inpatient surgery. (Jt. Ex. 4, p. 155; Carter, Vol. 10, pp. 1779-80).

155. Although Duke’s expert witness testified that UNC’s OR utilization projections were unreasonable because its acute care beds projections were unreasonable, both of UNC’s expert witnesses refuted this testimony. Mr. Carter opined that UNC’s OR utilization projections were conservative. The projections showed that some of the surgical cases would need to be performed in procedure rooms based on the relatively small capacity of 2 ORs in UNC’s proposal. (Carter, Vol. 10, p. 1781). Mr. Meyer opined that UNC’s projections were reasonable, and conservative based on his experience in healthcare planning. (Meyer, Vol. 5, pp. 943-44).

156. UNC similarly projected utilization for imaging and ancillary services, observation beds, procedure rooms, and LDR and C-Section rooms. (Jt. Ex. 4, pp. 151-55, 159-60).

157. Based on the information provided by UNC, the Agency found UNC’s projected utilization to be reasonable and adequately supported, because UNC:

(1) used publicly available data to determine Durham

36 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

County residents’ potential days of care for UNC Hospitals-RTP’s projected services,

(2) used an historical 2-yr compound annual growth rate (“CAGR”) to project days of care going forward, and

(3) based its projected surgical, obstetrics, emergency, imaging/ancillary, and observation bed services on historical Truven data for Durham County residents, relevant historical UNC Hillsborough experience, or UNC Health services for Durham County residents.

(Jt. Ex. 1, p. 1520; Hale, Vol. 2, pp. 239-40).

158. The Agency also found UNC’s projection that 90 percent of its patient population would come from Durham County to be reasonable because the southern part of Durham County was highly populated, and any nearby Wake County residents have a number of healthcare and hospital choices in Wake County. (Hale, Vol. 2, p. 317).

In light of these findings, the ALJ made the following conclusions of law:

45. To conform with Criterion (3), an applicant’s projected patient origin, demonstration of need, and projected utilization must be reasonable and adequately supported.

46. The Agency correctly determined that UNC’s projected patient origin for UNC Hospitals-RTP, including 90 percent Durham County residents and its conservative 10 percent in-migration assumption, was reasonable and adequately supported.

47. The Agency also correctly determined that UNC’s demonstration of need for UNC Hospitals-RTP based on the population growth and aging of the population in Durham County, the need for a new hospital in Durham County (particularly the southern area), the need for UNC- Hospitals’ hospital-based services in Durham County, and

37 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

the need for acute care beds (especially community hospital beds) and ORs in Durham County, was reasonable and adequately supported.

48. The Agency further correctly determined that UNC’s projected utilization for all service components at UNC Hospitals-RTP was reasonable and adequately supported.

49. Substantial evidence in the record of this case supports the Agency’s determination that the UNC Application was conforming with Criterion (3).

As reproduced above, these findings and conclusions demonstrate that the ALJ

extensively considered UNC’s proposal with respect to the service of in-county

patients. While we will not belabor the issue by reciting the support for each of the

more than eighty findings by the ALJ pertaining to Criterion 3 generally, we

specifically note that the alleged underprediction of patient days provided by UNC’s

proposed facility in light of the absence of high-acuity services—one of the primary

issues raised by Duke in this appeal—was considered and rejected at finding 151, et

seq. This finding was supported by testimony in the record indicating that, despite

Duke’s expert having opined that UNC overestimated its patient day projections at

the new facility, UNC’s projection methodology specifically accounted for the absence

of high-acuity services at the new facility—a projected patient reduction of 31 percent.

Similarly, Duke’s argument on appeal that the UNC application unrealistically

projected the number of patients originating from Durham County to be served was

also addressed and rejected by the ALJ on the basis that UNC statistically grounded

its claims about the relative need for the facilities in Durham County and in-

38 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

migration rates at comparable UNC facilities, with the ALJ consistently noting that

UNC conservatively projected its Durham-resident patient volume to account for

such considerations. These findings, too, were supported by testimony on the record.

Despite this evidentiary support in the ALJ’s final decision, Duke asks us to

overturn the result below on the basis of alleged failures in the reasoning of the

Agency. However, our task on appeal is not to evaluate the reasoning of the Agency,

but the reasoning of the ALJ. Compare N.C.G.S. § 150B-51 (2023) (governing appeals

from the Office of Administrative Hearings to the Court of Appeals) with N.C.G.S. §

150B-23 (2023) (governing appeals from the Agency to the Office of Administrative

Hearings); see also AH, 240 N.C. App. at 98. Where the reasoning of the ALJ is

supported by substantial evidence, we will not overturn the ALJ’s final decision

simply because the ALJ weighed the evidence in a manner unfavorable to the

appellant, Mills, 251 N.C. App. at 189; and, here, the ALJ’s decision was amply

supported. We will not, therefore, overturn its determination that UNC’s application

conformed with Criterion 3.

D. UNC’s Compliance with Criterion 12

Finally, we address whether the ALJ properly affirmed the Agency’s

conclusions as to UNC’s compliance with N.C.G.S. § 131E-183(a)(3). N.C.G.S. § 131E-

183(a)(12), or “Criterion 12,” provides that a certificate of need applicant

shall demonstrate that the cost, design, and means of construction proposed represent the most reasonable alternative, and that the construction project will not

39 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

unduly increase the costs of providing health services by the person proposing the construction project or the costs and charges to the public of providing health services by other persons, and that applicable energy saving features have been incorporated into the construction plans.

N.C.G.S. § 131E-183(a)(12) (2023). Duke argues that UNC’s proposal was

nonconforming with Criterion 12 in that the hospital’s primary proposed location in

RTP was subject to restrictive covenants not accounted for in the application, while

the alternate proposed site occupies a property that straddles proposed expansion of

a highway and is otherwise limited by power lines, a public greenway trail, and water

hazards.

In its final decision, the ALJ affirmed the Agency’s conclusion that UNC’s CON

application was in compliance with Criterion 12, making the following findings of

fact:

200. Analysis of this Criterion contains three elements: (1) whether the cost, design, and means of construction proposed represent the most reasonable alternative; (2) whether the construction project will not unduly increase the cost of providing health services by the person proposing the project; and (3) whether energy-saving features have been incorporated into the construction plans. (Id.; Meyer, Vol. 7, pp. 1271-72).

201. The UNC Application satisfied the first element by (1) providing drawings of its site plan and floor plan in Exhibit C.1 and (2) explaining that the proposed construction and layout for the hospital was based on a “configuration that provides the most efficient circulation and throughput for patients and caregivers,” based on “best practice methodologies,” as well as “relationships and adjacencies to support functions while also preventing unnecessary

40 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

costs.” (Jt. Ex. 4, pp. 112-13, 233-39; Meyer, Vol. 7, p. 1273).

202. UNC satisfied the second element of Criterion (12) by explaining that while the UNC Hospitals-RTP project would be capital intensive, UNC set aside excess revenues to fund the project, such that the project could be completed without increasing costs or charges to the public to help fund it. (Jt. Ex. 4, p. 113). UNC provided a letter from the Chief Financial Officer of UNC Hospitals certifying the availability of accumulated cash reserves to fund the project. (Id. at p. 292; Meyer, Vol. 7, pp. 1273-74).

203. Finally, UNC satisfied the third element of Criterion (12) by showing that its proposed hospital would be energy efficient and conserve water, and that UNC would develop and implement an Energy Efficiency and Sustainability Plan. (Jt. Ex. 4, p. 113; Meyer, Vol. 7, p. 1274).

i. Zoning of UNC’s Primary Site

204. Because a CON is “valid only for the . . . physical location . . . named in the application,” applicants also are required to identify a proposed site for a new facility. (N.C. Gen. Stat. § 131E-181(a); Jt. Ex. 4, p. 114; Meyer, Vol. 7, pp. 1272, 1282). The applicant should specify an address, a parcel number, or intersection of roads. (Meyer, Vol. 7, p. 1272).

205. The primary site for UNC Hospitals-RTP identified in the UNC Application is located in southern Durham County in the Research Triangle Park (“RTP”) at the convergence of North Carolina Highway 54 and North Carolina Highway 147, also known as the Triangle Expressway. (Jt. Ex. 4, p. 114). At the time of the filing of the UNC Application, the property, also known as the Highwoods Site, was owned by Highwoods Realty Limited Partnership (“Highwoods”). (Id. at 115). UNC provided a Letter of Intent for UNC Health to purchase the property from Highwoods along with its application. (Id.at 517-23).

206. The CON Law does not regulate or even mention

41 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

zoning. (Meyer, Vol. 7, p. 1281). Nonetheless, Section 4(c) of Criterion (12) in the Agency’s application form is entitled “Zoning and Special Use Permits.” (Hale, Vol. 2, p. 244). This Section requires an applicant to first describe the current zoning at the proposed site, and then, “[i]f the proposed site will require rezoning, describe how the applicant anticipates having it rezoned[.]” (Jt. Ex. 4, p. 115; Hale, Vol. 2, pp. 266-67).

207. The Agency contemplates that a proposed site for a project may not be properly zoned for the proposed project at the time the application is submitted, by asking applicants the questions posed in Section 4(c). (Hale, Vol. 2, pp. 246, 267).

208. The fact that a site identified in an application may need rezoning does not make an application nonconforming with Criterion (12) or non-approvable. (Id. at p. 267; Meyer, Vol. 7, pp. 1281-82, Vol. 8, p. 1398). The Agency frequently approves applications that propose projects to be developed on sites that require rezoning before they can be used to develop the proposed services. (Hale, Vol. 2, p. 246; Meyer, Vol. 7, pp. 1277-78). In Mr. Meyer’s 25 years of healthcare planning experience, he cannot recall a time when the Agency denied an application due to the fact that a site needed to be rezoned. (Meyer, Vol. 7, p. 1278).

209. Moreover, the Agency is tasked with applying the CON Law and related rules, not with considering an applicant’s compliance with other laws like zoning ordinances. Therefore, the Agency does not review applicable zoning laws or restrictive covenants when it reviews an application. (Hale, Vol. 2, p. 266; see also Craven Reg’l Med. Auth. [v. N.C. Dep’t of Health & Hum. Servs., 176 N.C. App. 46, 57-58 (2006)]).

210. Rezoning of sites identified in CON applications typically does not occur until after a CON has been awarded. (Meyer, Vol. 7, p. 1277).

211. According to the UNC Application, UNC’s primary

42 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

proposed site “will require rezoning.” UNC noted that it anticipated having the property rezoned:

The proposed site is located in Research Triangle Park across the street from the Research Triangle Foundations Frontier and HUB RTP developments that have an SRP-C zoning designation. UNC Hospitals currently is working with land use counsel, the property owner, and Research Triangle Foundation management to have the property rezoned to permit hospital use. With the guidance of land use counsel, UNC Hospitals will engage with Durham Planning staff, the Durham Planning Commission, and the Durham Board of County Commissioners to complete the rezoning process. Additionally, UNC Hospitals will, with the cooperation of the Research Triangle Foundation, work with the Research Triangle Park Owners and Tenants Association (O&T) to amend the Research Triangle Park Covenants, Restrictions, and Reservations by resolution to permit hospital use. . . ..

(Jt. Ex. 4, p. 115; Hale, Vol. 2, pp. 268-69).

212. Applicants are not required to submit letters of support with their CON application; however, it is common for CON applicants to do so. (Hale, Vol. 2, p. 260; Carter, Vol. 10, pp. 1790-91). The UNC Application included a letter of support from Scott Levitan, CEO of the Research Triangle Foundation (“RTF”). (Jt. Ex. 4, p. 512). Mr. Levitan’s letter indicated that the RTF supported the UNC Application; however, it did not make any reference to the property being rezoned or restrictive covenants being amended. (Id.; Hale, Vol. 2, pp. 280-82).

213. UNC was not required to submit the letter of support from Mr. Levitan or anyone else on behalf of RTF to be approvable. (Hale, Vol. 2, pp. 280-81; Carter, Vol. 10, p. 1791).

43 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

ii. UNC’s Primary Site in the Research Triangle Park

214. The RTP is an approximately 7000-acre university research park located in Durham and Wake Counties, with 5,600 acres, or 80 percent, located in Durham County. (Levitan, Vol. 5, pp. 774, 799-800). There are currently no people living in the RTP. (Id. at 897).

215. Scott Levitan is the President and CEO of the Research Triangle Foundation (“RTF”), a position he has held for approximately five years. (Id. at 769). In this position, Mr. Levitan reports to the RTF Board, which includes representatives of UNC, Duke, NC State University, and North Carolina Central University. (Id. at 773-74).

216. The RTF is a 501(c)(4) entity founded approximately 63 years ago for the purpose of facilitating coordination among UNC, Duke, and NC State University and to enhance the wellbeing of the residents of North Carolina. (Id. at 769-70). The RTF administers the activities of the RTP Owners and Tenants Association (“O&T”). (Id. at 770). The RTF also owns certain property within the RTP. (Id.).

217. There are two types of zoning within the RTP: Science Research Park (“SRP”) and Science Research Park – Commercial (“SRP-C”). (Id. at 777-78). SRP-C zoning is more lenient than SRP zoning but only covers 101 acres in RTP known as the RTP Hub, which is a mixed-use development intended to serve as a “town center” for RTP. (Id. at 780-81). The Hub includes Boxyard, a retail center containing food and retail vendors; Frontier, an innovation campus for startups and emerging companies; residential multi-family apartments; and other businesses not focused on scientific research. (Id. at 781, 829-31).

218. There are also restrictive covenants covering RTP that restrict the property to certain uses. (Jt. Ex. 1, pp. 191- 255). According to Mr. Levitan, these restrictive covenants do not currently permit the development of a hospital at

44 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

UNC’s primary site. (Levitan, Vol. 5, p. 785).

219. The primary site for UNC Hospitals-RTP is adjacent to the RTP Hub. (Id. at 783-84). In the recent past, the RTF allowed a parcel of property adjacent to the RTP Hub to be rezoned from SRP to SRP-C to allow the development of a fire station in Durham County. The RTP also allowed a text amendment to the RTP restrictive covenants to allow a school on a particular parcel in Wake County. (Id. at 782- 83, 895-96).

220. David Meyer is a 35-year resident of Durham County in addition to his healthcare planning expertise. Mr. Meyer opined that UNC’s location adjacent to the RTP Hub made sense from a health planning perspective. He likened UNC Hospitals-RTP to REX Hospital’s adjacency to Cameron Village in Raleigh, now known as the Village District, to support the notion that a hospital being adjacent to a multi-use district in the midst of a highly populated area is sensible. (Meyer, Vol. 7, pp. 1274-76, Vol. 8, pp. 1389-91).

221. Initially, UNC explored purchasing a site owned by Keith Corp. within the RTP, but not adjacent to the RTP Hub, and having the site rezoned to allow UNC to build a hospital there. When approached by Keith Corp. about this proposal, Mr. Levitan was not comfortable setting a precedent of SRP-C zoning in areas other than the Hub; however, Mr. Levitan eventually suggested that UNC approach Highwoods about purchasing its property adjacent to the Hub. (Levitan, Vol. 5, pp. 832, 839-42).

222. Mr. Levitan discussed UNC using the Highwoods Site for its proposed hospital at a [11 February] 2021, RTF Development Committee meeting. (Jt. Ex. 119; Levitan, Vol. 5, pp. 843-44). Following that meeting, Mr. Levitan emailed members of the RTF Development Committee who were not affiliated with either Duke or UNC and obtained their approval to continue cooperating with UNC’s proposal. (Jt. Ex. 117; Levitan, Vol. 5, pp. 844-49).

45 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

223. In particular, RTF Board member Smedes York stated: “I believe this could be positive as it ‘anchors’ the location without changing the ‘sizzle’ of the Hub area. We need the ‘personality’ of Boxyard and other parts of what we have planned. Rex Hospital’s previous location was adjacent to Cameron Village which was a positive.” (Jt. Ex. 117).

224. To change the zoning of the primary site, UNC would need to seek approval for rezoning from Durham County and would also need to seek approval from the RTP O&T to amend the restrictive covenants. (Levitan, Vol. 5, p. 785, 798). To Mr. Levitan’s knowledge, there has never been a healthcare facility like a hospital permitted in the RTP. (Id.).

225. Although the ultimate decision to allow the development of UNC Hospitals-RTP on the Highwoods Site is up to the RTP’s O&T, Mr. Levitan has already begun the process of running the proposal through the relevant committees for a recommendation to the RTP’s O&T. UNC’s proposal was first brought before the RTF Development Committee. Mr. Levitan believed he “had the imprimatur of the Development Committee to continue conversations in support of the hospital application on the part of the foundation . . . .” (Id. at 796-97). Based on this direction from the Development Committee, Mr. Levitan cooperated with UNC in its efforts to build a hospital within the RTP. (Jt. Exs. 15, 42; Levitan, Vol. 5, pp. 837- 38).

226. Mr. Levitan did not discuss his letter of support with the RTF Board or Development Committee before signing it, as he is frequently asked to sign letters of support and does not generally bring those to the RTF Board or other committees for review. (Levitan, Vol. 5, p. 799).

227. Mr. Levitan gave conflicting testimony about whether he was aware Duke might be applying for the same need determined assets in Durham County as UNC. (Compare Levitan, Vol. 5, pp. 786-87 with pp. 822-23). Despite Mr.

46 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

Levitan’s apparent confusion, this Tribunal finds that Mr. Levitan appears to have been aware that Duke may have a conflicting interest with UNC’s proposed hospital, based on his [11 February] 2021 email to certain members of the RTF Development Committee. In this email, Mr. Levitan noted he was “[k]eeping conflicted folks out of the conversation”—i.e., people who were affiliated with either Duke or UNC—and sought their approval to recommend the Highwoods site to UNC. (See Jt. Ex. 119).

228. Mr. Levitan’s Letter of Support indicated that the RTF supported UNC’s Application; however, it did not make any reference to the property being rezoned or restrictive covenants being amended. (Id.; Hale, Vol. 2, pp. 280-82). At the time the letter was submitted, Mr. Levitan understood the letter would be used “as support for UNC’s certificate of need application for a hospital in RTP.” (Levitan, Vol. 5, pp. 790-92).

229. UNC reasonably believed its statements regarding the zoning of the primary site were accurate at the time UNC submitted its Application. In an email to Scott Selig and Tallman Trask, Levitan stated, “I think Duke is going to need to pursue its interests in this matter, but based on the direction from the DevComm meeting, we have cooperated with this initiative.” (Jt. Ex. 42; Hale, Vol. 2, pp. 283-287). Similarly, in a [20 May] 2021 meeting of the RTF Development Committee, the meeting minutes reflected that at a prior meeting, that “committee suggested to UNC that they could pursue extending the SRP-C zoning across the street if Highwoods was interested in selling their land.” (Jt. Ex. 15; Hale, Vol. 2, pp. 287-88).

230. The Agency’s Team Leader Ms. Hale did not review any documents prior to the Agency decision that suggested UNC would not be able to have the primary site rezoned or the restrictive covenants amended. (Hale, Vol. 2, p. 291).

231. On or about [13 May] 2021, the Triangle Business Journal published an article discussing UNC’s proposed new hospital in the RTP. (Jt. Ex. 130; Levitan, Vol. 5, p.

47 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

808). Following the publication of this article, Mr. Levitan was asked by the RTF Executive Committee to clarify his letter of support. (Levitan, Vol. 5, pp. 804, 816). The Executive Committee gave Mr. Levitan the language to include in his second letter verbatim. (Levitan, Vol. 5, pp. 808, 813-14, 827-28).

232. At the hearing and at his deposition, Mr. Levitan used the terms “clarify,” “rescind,” and “withdraw” interchangeably to mean the same thing. (Levitan, Vol. 5, p. 816). Given the text of the [12 July] 2021 Letter and Mr. Levitan’s testimony, the [12 July] 2021 Letter was a clarification of the RTF’s position on the UNC Application, rather than a rescission or withdrawal of support.

233. After the RTF Executive Committee decided a clarifying letter should be sent to the Agency, Mr. Levitan sent an email to the Agency stating that his letter of support, which he described as “an outdated correspondence” was included in the UNC Application. In that email, Mr. Levitan asked to speak with either Ms. Inman or Lisa Pittman, the Agency’s Assistant Chief of Certificate of Need, regarding “the process and deadlines for submitting comment on UNC Health’s application.” (Duke Ex. 200; Hale, Vol. 3, pp. 332-33; Levitan, Vol. 5, pp. 810, 812-13).

234. Mr. Levitan subsequently spoke with Ms. Inman, who informed him that the deadline for submitting public comments to the CON Section had passed. Ms. Inman told Mr. Levitan he could still submit a letter and that she would “make every effort” to ensure it was seen by the CON Section. (Levitan, Vol. 5, p. 810).

235. After speaking with Ms. Inman, Mr. Levitan sent his second letter, dated [12 July] 2021 to the Agency. (Jt. Ex. 46). Mr. Levitan submitted his [12 July] 2021 letter to the Agency after the end of the public comment period in this Review. (Hale, Vol. 2, pp. 283, 308-09, 336). Mr. Levitan stated in the [12 July] 2021 Letter, in relevant part, that he was “writing to clarify [his] prior letter dated 13 April

48 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

2021,” and that “[u]ntil a certificate of need has been awarded and any appeals to the determination of the Healthcare Planning and Certificate of Need Section have been exhausted, RTF will not consider a zoning change for the proposed site in RTP.” (Jt. Ex. 46; Levitan, Vol. 5, pp. 818-19).

236. In a [3 September] 2021, letter to Jud Bowman, Chairman of the RTF Board, Vincent Price, President of Duke University, characterized Duke’s position on the [12 July] 2021 Letter as follows:

[Mr. Levitan] then sent a follow up letter on July 12th to the State CON analyst stating that the Foundation would not consider a zoning change until after the CON determination and any appeals. This second letter is also deeply troubling. It did not withdraw the endorsement by RTF of UNC’s application. It continued to support placing a hospital within the RTP. It was also provided outside the prescribed public comment period, so cannot by law be considered by the State; thus, its purpose is unclear to me.

(Jt. Ex. 25).

237. Though the Agency received Mr. Levitan’s [12 July] 2021 Letter, the Agency did not consider Mr. Levitan’s second letter, and did not include the letter as part of the Agency File because the letter was submitted after the end of the public comment period. (Jt. Ex. 91; Hale, Vol. 1, pp. 177-78, 308-09, 336, 339). Mr. Levitan advised the RTF Executive Committee that he had submitted the clarifying letter and that it was submitted outside the public comment period. (Levitan, Vol. 5, pp. 814-15).

238. At the hearing, Mr. Levitan opined that UNC’s description on page 115 of the UNC Application regarding the zoning of the primary site was accurate. (Id. at pp. 833- 38).

49 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

iii. Issues Raised by Duke Regarding UNC’s Proposed Sites

239. Duke’s Comments raised issues regarding UNC’s primary site and pointed to UNC’s statement that rezoning was needed. Duke indicated that “the rezoning will require not only Durham County approval but also compliance with the applicable covenants and restrictions affecting Research Triangle Park to which the site is subject,” and attached the RTP restrictive covenants to its comments. (Jt. Ex. 1, pp. 185, 191-255).

240. Duke had no knowledge or factual basis to support its comments regarding the UNC Application’s primary site or conformity with Criterion (12).

241. Duke provided no expert testimony in support of its contention that the UNC Application was nonconforming with Criterion 12. (Sandlin, Vol. 6, p. 955).

242. Catharine Cummer was the only fact witness Duke called in its case. Ms. Cummer serves dual roles as regulatory counsel and in strategic planning for Duke and has primary responsibility for ensuring the preparation of all CON applications submitted by Duke. (Cummer, Vol. 3, pp. 410-11). Ms. Cummer was not tendered or accepted as an expert witness in this case. Ms. Cummer has never been qualified as an expert witness in any kind of case. She has no expertise in finance, is not a clinician and has never served as a healthcare or certificate of need consultant. Ms. Cummer has never been employed as a project analyst or in any other capacity by the Agency. She has never served on the SHCC or its subcommittees. (Cummer, Vol. 4, pp. 579-82). Ms. Cummer is not on the Real Estate Development Committee or any other committee of the RTF Board. She is not a member of the RTF Board of Directors. (Id. at p. 647).

243. Duke included multiple pages of comments regarding the primary and alternative sites proposed by UNC and its conformity with Criterion 12. Duke also included a copy of

50 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

the RTP Restrictive Covenants in its Comments against the UNC Application. (Id. at pp. 638-39; Jt. Ex. 1, pp. 191- 255). Ms. Cummer was sent a copy of the RTP Restrictive Covenants from Dr. Monte Brown. (Cummer, Vol. 4, p. 645).

244. Duke relied heavily upon its Comments filed against the UNC project as a purported basis for alleging Agency error in this matter and argued that the Agency failed to appropriately consider its Comments, in particular those comments regarding Criterion 12. In its Comments, Duke alleged:

Notably, the Board [Research Triangle Foundation Board] has historically denied all rezoning applications to allow for health care facilities. In fact, DUHS is informed and believes that UNC has previously asked for permission to put a healthcare facility on the RTP campus itself, which was denied.

(Jt. Ex. 1, p. 185).

245. Ms. Cummer was primarily responsible for the preparation of the Duke Comments regarding Criterion (12). On cross-examination, contrary to the above Comment, Ms. Cummer admitted she had no personal knowledge regarding any prior applications for rezoning related to healthcare facilities at the RTP and had no personal knowledge regarding what other applications, if any, had been submitted by UNC to the RTP. (Cummer, Vol. 4, pp. 646-49).

246. Instead, Ms. Cummer relied upon a discussion with Scott Selig, Vice President of Real Estate and Capital Assets for Duke University and a designated member of the Real Estate Development Committee of the RTF, for the factual basis of Duke’s contentions in its Comments to the Agency. (Cummer, Vol. 4, pp. 646-47).

247. On cross-examination, Ms. Cummer’s testimony was impeached by the following deposition testimony of Mr.

51 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

Selig:

Question: Okay. Well, regardless of who prepared it, there’s a statement in here, right here it says, ‘Notably, the board has historically denied all rezoning applications to allow for healthcare facilities.’ Is that accurate?

Answer: I have no idea.

Question: Okay. Can you recall a time when the RTF board has denied rezoning for a healthcare facility?

Answer: No.

Question: Okay. The following sentence says, ‘In fact, UNC has previously asked for permission to put a facility on the RTP campus itself, which was denied.’ Is that accurate?

Question: Do you know anything about UNC asking permission to put a facility on the RTP campus itself being denied?

(Jt. Ex. 157, p. 140; Cummer, Vol. 4, pp. 646-51). After such impeachment, Ms. Cummer agreed that she would defer to Mr. Selig’s personal knowledge of such questions regarding the history of the RTF and any submissions, approvals or denials made for zoning. (Cummer, Vol. 4, p. 652).

248. Ms. Cummer then testified that Dr. Monte Brown, Vice President of Administration for the Duke University Health System, had provided her with the factual basis for those representations made by Duke to the Agency. However, on cross-examination, Ms. Cummer’s testimony was impeached with the following deposition testimony of

52 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

Dr. Brown:

Question: And with respect to the primary site in the RTP, why do you say that was not a viable site?

Answer: Because we had always been told, the entire time I was here at Duke, that you can’t put healthcare in the RTP.

Question: Who had told you that?

Answer: I don’t know. It’s kind of folklore. Scott [Selig], Tallman [Trask], my predecessor, we had always stayed out of it.

(Jt. Ex. 147, p. 39; Cummer, Vol. 4, p. 654). Ms. Cummer acknowledged that she did not speak with any other persons regarding the content of this section of the Comments. (Cummer, Vol. 4, p. 655).

249. At hearing, Dr. Brown could not recall the factual basis supporting Duke’s contention in this regard. (Brown, Vol. 10, pp. 1630, 1634).

250. Despite Duke’s comments opposing the proposed site for UNC Hospitals-RTP, Dr. Brown sent an email communication to other Duke representatives calling the UNC primary location a “prime location.” (Jt. Ex. 12). Dr. Brown also sent an email stating that “DUHS honored the RTP rules and has purchased land at Page Road and Green Level Road to accomplish its goals outside the RTP. Had the RTP allowed for medical, we likely would have chosen differently.” (Jt. Ex. 17).

251. Dr. Brown acknowledged he made no investigation or inquiry whether the zoning for the primary site proposed by UNC could be modified by the Durham County zoning authorities. (Brown, Vol. 10, p. 1633).

252. The unrefuted factual testimony from UNC established that there was no factual basis supporting

53 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

Duke’s contention that UNC had previously sought permission to put a healthcare facility on the RTP campus and was denied. In its Response to Comments, UNC disputed Duke’s statements regarding UNC’s primary site as UNC was “not aware of the Research Triangle Foundation Board purportedly historically denying all rezoning applications to allow for healthcare facilities[,]” nor was UNC “aware of any situation in which it asked for permission to put a healthcare facility on campus.” (Jt. Ex. 1, p. 320). Ms. Hadar testified unequivocally, that UNC has not previously sought to put a facility on the RTP campus prior to the UNC Hospitals-RTP Application. (Hadar, Vol. 8, p. 1467).

253. Moreover, Ms. Hale’s testimony established that a project analyst may, but is not required to, research information outside of the application to understand what is contained in an application. (Hale, Vol. 1, p. 193). Ms. Hale was aware of the Agency doing such additional research in one other review—the 2016 Wake County MRI Review. (Hale, Vol. 1, pp. 194-97). While zoning ordinances, real estate deeds, and restrictive covenants may be public documents that the Agency could locate and review, the Agency was not required to do so and did not feel the need to do so with respect to UNC’s primary site. (Hale, Vol. 1, pp. 197-98, Vol. 2, pp. 300-01). Further, the Agency does not request additional information from applicants who are involved in a competitive review. (Hale, Vol. 2, pp. 277-78).

iv. The Alternate Site Identified in the UNC Application

254. UNC also identified an alternate site for its proposed new hospital. (Jt. Ex. 4, p. 114, n. 30). The alternate site is located along Highway 70 in Durham County and would not require any rezoning. (Id. at 515-16). The alternate site is also close to power, water, and sewer services. (Id. at 516).

255. Duke raised concerns about UNC’s alternate site in its Comments alleging the following: “However, that site has

54 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

even more fundamental obstacles to development than the primary site. . . . The bigger issue, however, is that the alternate site will be rendered unavailable for the proposed use by a NCDOT highway project in planning stages. . . .” (Jt. Ex. 1, p. 186). For that reason, Duke took the position in its Comments that UNC’s alternate site is not a viable possible location for UNC Hospitals-RTP. (Cummer, Vol. 4, p. 661).

256. By letter dated [3 September] 2021, during the Agency’s review of the UNC and Duke Applications, Dr. Vincent Price, President of Duke University, sent a four- page letter to the Chair of the Board of Directors for the Research Triangle Foundation, Jud Bowman (“Dr. Price Letter”). (Jt. Ex. 25). In his letter, Dr. Price aired several grievances regarding the UNC Hospitals-RTP project, its proposed primary site in the RTP, and the support letters from Mr. Levitan regarding the same. Dr. Price’s Letter represented to the RTF that:

It seems to me that the only cure for this highly concerning matter is for the Board to recuse itself going forward from any decision that relates to the CON application or eventual award, regardless of who is successful in the CON process. Note that UNC’s application does include an alternate site that does not require RTF action that does not require RTF rezoning.

(Id. at 3).

257. Thus, while the Comments filed by Duke represent that the alternate site is “not viable,” the Dr. Price letter to the RTF makes no reference to Duke’s public position on the alternate site and implies that the alternate site is viable.

258. Duke attempted to distinguish its position in these two documents by claiming that it was merely pointing out that UNC had represented the alternate location to be viable and that the “alternate site has nothing to do with

55 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

the Research Triangle Park or Research Triangle Foundation, so there would be nothing for the board to do as to the viability or not of an alternate site.” (Cummer, Vol. 4, p. 668). Dr. Brown confirmed in his testimony that he did not discuss whether this representation by Dr. Price was inconsistent with the representations in Duke’s Comments. (Brown, Vol. 10, p. 1645). Though it could cite no factual support for the same, Duke continued to stand by its Comments in Opposition. (Id. at 1652). Nonetheless, this answer did not explain why Dr. Price addressed UNC’s alternate site at all if its existence was not relevant to the RTF.

259. Ms. Cummer, the author of the Comments, also reviewed and provided comments on a draft of Dr. Price’s Letter prior to it being sent to the RTF (Cummer, Vol. 4, p. 666), and was therefore aware of the inconsistent representations made by Duke to the Agency regarding the alternate site and those made to the RTF regarding the same.

260. At hearing, Dr. Brown acknowledged that he provided the information in Duke’s Comments about the proposed NCDOT highway project on UNC’s alternate site. Yet, he also conceded that he did not investigate whether (1) the proposed alternate site had actually been acquired for the highway project or (2) whether there were any restrictions on what UNC could do with the alternate site property if it had not been acquired by NC DOT or if UNC had acquired the property. (Brown, Vol. 10, pp. 1635-36). Dr. Brown also testified that UNC admitted, in its application, that a highway project was planned for its alternate site. (Id. at p. 1635).

261. However, Mr. Carter clarified that the UNC Application provided information about the alternate site but did not speculate “as to the future of that parcel of land or how it may be used other than for a proposed hospital.” (Carter, Vol. 10, p. 1792).

v. UNC Can Make a Material Compliance Request if it

56 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

Ultimately Cannot Develop a Hospital at its Primary Site

262. A material compliance request is a letter to the Agency stating why the applicant cannot proceed with the project exactly as described in its application. (Hale, Vol. 2, pp. 247, 276-77; Meyer, Vol. 7, p. 1283). The applicant would include in its request the reasons why they could not develop the project at the site and identify an alternate site for the Agency to consider as a location for the assets awarded in the CON. (Hale, Vol. 2, pp. 247-48; Meyer, Vol. 7, p. 1283). Through this process, a modification in plans can be deemed by the Agency to be in “material compliance” with the representations in the approved application.

263. The Agency routinely approves material compliance requests and has approved material compliance requests to develop projects at alternate sites. (Hale, Vol. 2, p. 248; Cummer, Vol. 4, pp. 680-81; Meyer, Vol. 7, p. 1283). For example, in 2018, Mr. Meyer assisted an ASC in making a material compliance request to the Agency seeking to develop its ASC in a location within Brunswick County at a different site. The Agency approved this request. (Jt. Ex. 100; Meyer, Vol. 7, pp. 1284-85).

264. Regardless of whether UNC develops UNC Hospitals- RTP at the primary site, UNC would be able to submit a material compliance request to the Agency to approve a new location for the facility. UNC could make a similar request if it ultimately was unable to have the primary site rezoned appropriately. (Meyer, Vol. 7, pp. 1285-86).

265. Notably, Duke itself experienced issues with a site identified in a 2018 CON application for ORs in Orange County. (Id. at p. 1286). The 2018 Orange County OR Review was a competitive review in which Duke and UNC both applied for 2 ORs in Orange County. (Cummer, Vol. 4, p. 681). The Agency ultimately awarded the CON to Duke, and UNC challenged this award in a contested case. (Id. at p. 681-82). Duke engaged Keystone Planning, Mr. Meyer’s company, to develop Duke’s application, and later serve as an expert witness, in that review. (Meyer, Vol. 7,

57 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

pp. 1286-87).

266. In that review, Duke had leased a location on Sage Road, which location was approved by the Agency. However, during the course of the Agency’s review of the application, Duke identified certain remediation and code issues that it believed made it financially more favorable for the project to be developed at a different location. In response, Duke determined that it could make a successful request for a material compliance determination to change the location. (Cummer, Vol. 4, pp. 685-88; Meyer, Vol. 7, pp. 1286-87).

267. Duke did not inform the Agency during the course of the review that it had identified potential issues with its proposed site. (Cummer, Vol. 4, p. 691). Because the original site was still available to Duke during the course of the review, the “information in the application that the site was available was correct.” (Id. at p. 693). According to Ms. Cummer, “[s]o unless an[d] until we were interested in seeking a different site or doing anything else, there was nothing to inform the agency of.” (Id.)

268. In both his expert report and deposition testimony in the 2018 Orange County OR Review, Mr. Meyer emphasized that the issues with Duke’s ASC site in its CON application were immaterial, as Duke could submit a material compliance request, which the Agency routinely approves. (Jt. Exs. 101, 102; Meyer, Vol. 7, pp. 1287-89).

269. Ms. Cummer also cited to an occasion when Duke previously withdrew a CON application after learning it had relied upon incorrect and overstated data. She explained that the data error was so significant that it made the application infeasible as presented. (Id. at pp. 697-98).

270. Mr. Meyer’s opinion concerning UNC’s conformity with Criterion (12) and the ability of an approved applicant to submit a material compliance request in the event of site issues is consistent between this Review on behalf of UNC

58 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

and the 2018 Orange County OR Review on behalf of Duke. (Id.).

271. Mr. Carter agreed with the Agency’s conclusion that the UNC Application was conforming with Criterion (12), as UNC provided all information requested by the Agency for this Criterion. (Carter, Vol. 10, p. 1790). Mr. Carter opined that the Agency’s analysis of this Criterion was consistent with the way the Agency has analyzed Criterion (12) in previous reviews. (Id. at 1792). Mr. Carter also opined that the specific location of UNC Hospitals-RTP was not material to UNC’s demonstration of need for this project, but rather the location of the facility within the southern region of Durham. (Carter, Vol. 11, pp. 1982-83).

272. Ms. Sandlin offered no opinions with respect to UNC’s conformity with Criterion (12). (Sandlin, Vol. 6, p. 955; see also Jt. Exs. 54, 146).

273. The Agency considered Duke’s Comments in its analysis of UNC’s conformity with Criterion (12). In its analysis of Criterion (12), the Agency noted “there is some question as to whether or not the first site can be rezoned for a hospital” and indicated it had reviewed Duke’s Comments. (Jt. Ex. 1, pp. 1575-76; Meyer, Vol. 7, pp. 1280- 81, Vol. 8, pp. 1393-94). The Agency was aware that the site has not yet been rezoned and that Duke questioned the possibility of rezoning the site. (Id.).

274. Ultimately, the Agency found that UNC had adequately explained its proposed project and its plans for accomplishing the required rezoning, such that it was conforming with Criterion (12). (Jt. Ex. 1, pp. 1575-76; Hale, Vol. 2, pp. 274-75).

In light of these findings, the ALJ made the following conclusions of law:

73. The Agency correctly determined that the UNC Application identified a proposed site and adequately demonstrated that the cost, design, and means of construction of UNC Hospitals-RTP represent the most

59 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

reasonable alternative, will not unduly increase the cost of service to the public, and incorporates energy saving features.

74. UNC provided adequate information requested by the Agency in the application related to Criterion (12), including describing how it anticipated having the property rezoned.

75. The Agency reasonably assessed potential zoning and restrictive covenant issues with the primary site for UNC Hospitals-RTP and correctly determined that the UNC Application was conforming with Criterion (12) nonetheless. Moreover, the Agency did not err in not seeking additional information regarding the zoning and restrictive covenants at the primary site. “There is no provision in [N.C.G.S.] § 131E-183, nor Chapter 131E, which permits the Agency to independently assess whether the applicant is conforming to other statutes.” (Hale, Vol. 2, p. 266; see also Craven Reg’l Med. Auth., 176 N.C. App. at 58[] . . .). Therefore, the Agency did not err in not engaging in further analysis of the zoning or restrictive covenants beyond what was contained in the Agency findings.

76. The letter of support from Mr. Levitan was not necessary to the approval of the UNC Application; nonetheless, Mr. Levitan’s support letter was consistent with UNC’s representations in the UNC Application and its Responses to Comments.

77. The Agency was correct to exclude Mr. Levitan’s clarifying letter of [12 July] 2021 from the Agency File because it was submitted after the end of the public comment period. Had the Agency considered that letter and used it as a basis to deny the UNC Application, it would have been reversible error.

78. Mr. Levitan’s clarifying [12 July] 2021 Letter did not state that the RTF would deny any efforts to rezone the primary site; instead, it simply noted that the RTF would

60 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

not take action until a CON has been awarded and any appeals exhausted. (Jt. Ex. 46; see also Jt. Ex. 25). Thus, had the Agency considered the [12 July] 2021 Letter, the Agency would have been incorrect to use it as a basis for UNC’s nonconformity with Criterion (12).

79. While Duke raised questions about UNC’s alternate site, Duke presented no competent evidence as to the unavailability of that site. Neither Ms. Cummer nor Dr. Brown are qualified as an expert in real estate, condemnation, or highway construction. Their testimony suggesting UNC could not develop a hospital at the alternate site is unreliable, and the undersigned gives it no weight.

80. If UNC is ultimately unable to develop a hospital at the UNC Hospitals-RTP primary site due to zoning or restrictive covenant issues, UNC may submit a material compliance request for another suitable site, consistent with prior Agency decisions approving alternate sites following issuance of a CON. (See [N.C.G.S.] § 131E-181; Hale, Vol. 2, p. 248; Meyer, Vol. 7, pp. 1283-89; Jt. Exs. 100- 102). The Agency has the discretion to evaluate any request to develop the proposed hospital at a different location and determine whether such project would be in material compliance with UNC’s representations in the UNC Application. [N.C.G.S.] § 131E-189(b).

81. Substantial evidence in the record supports the Agency’s determination that the UNC Application was conforming with Criterion (12).

Here, while the ALJ’s decision critiques at length Duke’s failure to ground its

contentions concerning medical providers’ historical inability to create facilities in

RTP in fact, it does admit that the primary location is currently subject to zoning

requirements and restrictive covenants that would, as they stand currently, prevent

the construction of the proposed facility. Moreover, under N.C.G.S. § 131E-181(a),

61 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

“[a] certificate of need shall be valid only for the defined scope, physical location, and

person named in the application.” N.C.G.S. § 131E-181(a) (2023). The application in

this case concerned only the RTP location and not the proposed alternative location

discussed by the ALJ, so the scope of the consideration should have been limited to

the primary proposed location.5 Thus, much of the ALJ’s reasoning was unsound

insofar as it treated the presence of the zoning requirements and covenants as

unproblematic and considered the alternative site in the determination of whether

the CON should issue.6

As we review the determination as to Criterion 12 only for substantial evidence

on the record and do not interfere with the credibility and weighting determinations

of the ALJ, Surgical Care Affiliates, 235 N.C. App. at 622-23, we note that the

reasoning of the ALJ concerning UNC’s compliance with Criterion 12 may have been

5 In so holding, we express no opinion on whether the ALJ could have permissibly considered

an alternate site for the proposed facility if that alternate site had been included in UNC’s application. 6 Moreover, to the extent the ALJ used the subsequent possibility of UNC filing a material

compliance request to justify its reliance on the availability of the alternate site, we have treated the material compliance request process arising under N.C.G.S. § 131E-181(b) as analytically independent of, and distinct from, the grant or denial of a CON ab initio. See Craven, 176 N.C. App. at 59 (“The CON Section granted [the] request for a material compliance determination after the CON was issued. [The petitioner] is asking this Court to review events which occurred after the issuance of the final agency decision.”); see also N.C.G.S. § 131E-181(b) (2023). We understand the possibility of rectifying issues with a proposed facility as a remedial mechanism, not an invitation to lower the threshold at which an initial proposal is deemed satisfactory under our statutory criteria, and the absence of any caselaw in the course of our research in which the future possibility of a material compliance request has constituted substantial evidence to grant a CON appears to confirm this view. While the ordinary rule is that the ALJ is “authorized to establish its own standards in assessing whether an applicant” conforms with the criteria in N.C.G.S. § 131E-183(a), this rule only applies where review requirements have not been specified by our General Assembly. AH, 240 N.C. App. at 100; see also N.C.G.S. § 131E- 177(1) (2023). In this case, our General Assembly clarified in N.C.G.S. § 131E-181(a) that an application’s consideration is limited to the physical location described. N.C.G.S. § 131E-181(a) (2023).

62 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

independently supported, but not definitively so. Namely, even setting aside the

ALJ’s reasoning concerning the alternate location and its qualms with the support

proffered by Duke for its challenge to UNC’s CON application, the ALJ’s invocation

of prior cases where certificates of need have been awarded prior to zoning

amendments and finding that RTP has recently altered its zoning restrictions to

accommodate a fire station and its covenants to accommodate a school suggests it

found the proposal at the location listed in UNC’s application satisfactory under

Criterion 12. However, given the possibility that the ALJ would not have awarded

UNC the CON without the additional consideration of the proposed alternative site

and a future material compliance request, we have no way of knowing whether the

ALJ’s conclusion would have followed from only the allowable considerations.

Under N.C.G.S. § 150B-51(b), “[t]he court reviewing a final decision may affirm

the decision or remand the case for further proceedings. It may also reverse or modify

the decision if the substantial rights of the petitioners may have been prejudiced

because the findings, inferences, conclusions, or decisions are[,]” inter alia,

“[u]nsupported by substantial evidence . . . .” N.C.G.S. § 150B-51(b) (2023). For the

reasons explained above, the ALJ’s decisions as to Criterion 12 were, for purposes of

our review, supported by substantial evidence. However, the use of considerations

outside the scope of the ALJ’s review casts doubt on whether the ALJ herself would

have reached the same conclusions as to Criterion 12 when taking only the proposed

location in the application into account. Accordingly, we remand to the ALJ for

63 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

consideration of whether UNC’s application, taking into account only the site

proposed in its application and setting aside the possibility of a future material

compliance request, satisfied Criterion 12.

In particular, the ALJ should give due consideration to the possibility that a

potential inability to change RTP’s applicable covenants could result in substantial

cost being passed to patients. While the ALJ appears to have been satisfied with the

likelihood that both the zoning restrictions and applicable covenants could be

amended as necessary to accommodate the proposed UNC facility given a recent

history of amendments to permit the construction of a fire station and a school, the

final decision makes no meaningful reference to the financial ramifications of a

failure to amend either. This is especially troubling with respect to the restrictive

covenants, the termination of which requires the consent of the owners of 90% of the

subject property and the amendment of which is subject to judicial scrutiny to ensure

any changes are “reasonable in light of the contracting parties’ original intent” in the

event one of the affected property owners is dissatisfied with the amendment.

Armstrong v. Ledges Homeowners Ass’n, Inc., 360 N.C. 547, 559 (2006); but see Kerik

v. Davidson Cnty., 145 N.C. App. 222, 228 (2001) (emphasis added) (“[A]doption,

amendment, or repeal of a zoning ordinance is a legislative decision that must be

made by the elected governing board[.]”). When considering the potential for property

owners with an interest in maintaining these covenants to disallow the construction

64 DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

of the new facility7 in isolation of UNC’s ability to pivot to a location not listed in its

application, the ALJ may make a new determination in accordance with whether it

is satisfied that UNC has demonstrated that the project “will not unduly increase the

costs of providing health services” at the site proposed in the application. N.C.G.S. §

131E-183(a)(12) (2023).

CONCLUSION

We affirm the ALJ with respect to geographic access, competition, and

Criterion 3; however, because we cannot determine whether the ALJ would have

found UNC’s application in conformity with Criterion 12 without considering matters

outside the scope of its CON application, we remand to the Office of Administrative

Hearings for further findings.

AFFIRMED IN PART; REMANDED IN PART.

Judge STADING concurs.

Judge GRIFFIN concurring in part and dissenting in part by separate opinion.

7 Or, perhaps more concerningly, consent only for an exorbitant price.

65 No. COA23-351 – DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

GRIFFIN, Judge, concurring in part and dissenting in part.

I concur with Parts A, B, and C of the majority opinion. However, I dissent

from Part D because there was substantial evidence that UNC’s application

conformed with Criterion 12 and I would therefore affirm the ALJ’s decision.

Criterion 12 provides that

[a]pplications involving construction shall demonstrate that the cost, design, and means of construction proposed represent the most reasonable alternative, and that the construction project will not unduly increase the costs of providing health services by the person proposing the construction project or the costs and charges to the public of providing health services by other persons, and that applicable energy saving features have been incorporated into the construction plans.

N.C. Gen. Stat. § 131E-183(a)(12) (2023); see N.C. Gen. Stat. § 131E-183(a)(1), (3).

The majority holds the ALJ erred by considering evidence regarding a secondary

location that was not included on UNC’s CON application when determining whether

the application for the RTP location conformed to Criterion 12.

The standard of review is set forth by section 150B-51 of the North Carolina

General Statutes. “With regard to asserted errors pursuant to subdivisions (5) and

(6) of subsection (b) of [N.C.G.S. § 150B-51], the court shall conduct its review of the

final decision using the whole record standard of review.” N.C. Gen. Stat. § 150B-

51(c) (2023). The whole-record test requires this Court to determine whether the

Agency’s decision is supported by substantial evidence. Craven Reg’l Med. Auth. v. DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

GRIFFIN, J., concurring in part and dissenting in part

N.C. Dep’t of Health & Hum. Servs., 176 N.C. App 46, 52, 625 S.E.2d 837, 841 (2006)

(internal citations omitted). Substantial evidence is relevant evidence that a

reasonable mind could conclude supports a decision. Parkway Urology, P.A. v. N.C.

Dep’t. of Health & Hum. Servs., 205 N.C. App. 529, 535, 696 S.E.2d 187, 192 (2010)

(internal marks and citations omitted).

This Court may not “replace the agency’s judgment as between two reasonably

conflicting views” even if it may be possible to reach a different result if the matter

were reviewed de novo. Id. “Rather, a court must examine all the record evidence –

that which detracts from the agency’s findings and conclusions as well as that which

tends to support them – to determine whether there is substantial evidence to justify

the agency’s decision.” N.C. Dep’t. of Env’t. & Nat. Res. v. Carroll, 358 N.C. 649, 660,

599 S.E.2d 888, 895 (2004) (internal marks and citations omitted). Substantial

evidence is “relevant evidence a reasonable mind might accept as adequate to support

a conclusion.” Dialysis Care of N.C., LLC v N.C. Dep’t of Health & Hum. Servs., 137

N.C. App. 638, 646, 529 S.E.2d 257, 261 (2000) (internal marks and citations omitted).

The majority correctly points out that a CON is specific to what is listed on the

application. N.C. Gen. Stat. § 131E-181(a) (2023) (“A certificate of need shall be valid

only for the defined scope, physical location, and person named in the application.”).

While an ALJ may generally “establish standards and criteria or plans required to

carry out the provisions and purposes of [a CON]”, N.C. Gen. Stat. § 131E-177(1)

(2023), the ALJ may not utilize requirements that conflict with what has been

-2- DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

specified by our General Assembly, AH N.C. Owner LLC v. N.C. Dept. of Health &

Hum. Servs., 240 N.C. App. 92, 100, 771 S.E.2d 537, 542 (2015) (internal citations

omitted).

Here, the ALJ considered a secondary location not included on the application.

These considerations were error. However, as the majority states, the ALJ’s decisions

concerning Criterion 12 were supported by other allowable substantial evidence.

UNC provided drawings of its site plan and floor plan and explained how the

construction was designed to be efficient for the provision of services based on “best

practice methodologies” while preventing unnecessary costs. UNC also explained

that even though the project would be capital intensive, there was funding set aside

to ensure the project could be completed without increasing costs. A letter from the

Chief Financial Officer of UNC Hospitals was included to certify the availability of

funds to be used on this project. Additionally, UNC showed that it would design and

implement an Energy Efficiency and Sustainability Plan to demonstrate that the

proposed hospital would be energy efficient and conserve water. Although UNC’s

proposed site required rezoning, UNC anticipated having the property rezoned and

indicated that it would work with Durham County and the Research Triangle

Foundation to achieve the rezoning required. UNC also supplied a letter of support

from the CEO of the Research Triangle Foundation. There was also testimony at the

hearing indicating CON applications are almost never denied due to the fact that a

site needs to be rezoned.

-3- DUKE UNIV. HEALTH SYS., INC. V. N.C. DEP’T OF HEALTH & HUM. SERVS.

All of this evidence is permissible as it relates only to the primary site that is

included on the application. See Living Centers-Southeast, Inc. v. N.C. Dep’t. of

Health & Hum. Servs., 138 N.C. App. 572, 580, 532 S.E.2d 192, 197 (2000) (“Our

review of the individual statutes within the CON Statute . . . indicates that this

article grants applicants a full contested case hearing at which they are allowed to

present testimony and evidence contained in their applications.” (emphasis added)).

I would hold that this is substantial evidence as a reasonable mind may accept this

evidence as adequate in support of the conclusion that UNC’s application conforms

with Criterion 12.

Our standard of review demands we stop here. N.C. Gen. Stat. § 150B-51(b)

(2023) (“The court reviewing a final decision may affirm the decision or remand the

case for further proceedings. It may also reverse or modify the decision if the

substantial rights of the petitioners may have been prejudiced because the findings,

inferences, conclusions, or decisions are . . . [u]nsupported by substantial evidence.”

(emphasis added)). As UNC’s application provided substantial evidence supporting

the ALJ’s decisions regarding Criterion 12, I would affirm that part of the ALJ’s

decision, as well.

-4-