The opinion was delivered by
Luckert, J.:The plaintiffs, Badr Idbeis, M.D., Gaiy S. Benton, M.D., Robert H. Fleming, M.D., and John D. Rumisek, M.D., are thoracic, cardiothoracic, and cardiovascular surgeons who filed this action seeking a restraining order and injunction prohibiting their employer, defendant Wichita Surgical Specialists, P.A. (WSS), from enforcing restrictive covenants contained in their employment agreements with WSS and seeking a declaration that those covenants are unenforceable.
Upon the filing of the petition, the trial court granted the plaintiffs a temporary restraining order (TRO). A hearing was conducted after WSS filed a motion to vacate that restraining order and requested its own TRO enforcing the restrictive covenants. The trial court granted a temporary injunction in favor of Drs. Benton, Fleming, and Rumisek and prohibited WSS from enforcing the restrictive covenants in their employment agreements. As to Dr. Idbeis, the court granted a temporary injunction in favor of WSS, requiring Dr. Idbeis to begin making payments of liquidated dam*757ages pursuant to the formula set forth in his employment agreement or to cease practicing medicine in Sedgwick County, Kansas, as required by the geographic restriction in his employment agreement.
Approximately 1 year later, after conducting discovery, the parties tried the case to the court after which the trial court verbally ruled that the restrictive covenants in the employment agreements of Drs. Benton, Fleming, and Rumisek were enforceable. However, the trial court granted these plaintiffs their alternative request that they be allowed the option of paying liquidated damages. As to Dr. Idbeis, the trial court ruled that the restrictive covenant in his employment agreement was enforceable and that he must pay a liquidated damage award or cease practicing medicine in Sedgwick County, Kansas.
WSS filed a timely motion to amend the district court’s judgment pursuant to K.S.A. 2004 Supp. 60-252(b) asking the court to change the liquidated damages formula. The court entered an order amending the final judgment by changing the liquidated damages formula. Plaintiffs filed a timely notice of appeal, and WSS filed a timely notice of cross-appeal. This court granted a motion by WSS to transfer the case to this court pursuant to K.S.A. 20-3017.
Factual Background
Wichita Surgical Group, P.A. (WSG) was originally founded in 1969. WSG did not initially require restrictive covenants in its employment agreements. However, as it grew, WSG began requiring restrictive covenants for the purposes of improving the quality of the group, protecting the strength and viability of the group, and protecting the group’s referral base.
The structure and ownership of WSG changed in 1994, resulting in the formation of WSS, a professional corporation organized and authorized to do business in Kansas. Although WSS formed a single entity for tax purposes, it organized five divisions each of which was managed by its own division council and through shareholder votes. This structure allowed each division to maintain a distinct existence and independently manage its own affairs, including employment decisions. WSS had a board of directors comprised of *758representatives of each of the divisions; those directors were members of their individual division councils.
The surgeons specializing in thoracic and cardiovascular surgery were members of the division known as the WSS division. The WSS division required restrictive covenants, although some other divisions did not. With some exceptions, if a surgeon from outside the Wichita area joined WSS, the employment agreement included a restrictive covenant with time and geographic restrictions. If a surgeon from an existing Wichita practice joined WSS, the employment agreement included a restrictive covenant with time and geographic restrictions but also a liquidated damages or “buyout” clause. If a surgeon brought a new specialty to WSS, no restrictive covenant was required. For new surgeons, WSS provided a guaranteed minimum salary for the first 2 years while the surgeons developed their referral base.
When Dr. Idbeis, who is board certified in general and thoracic surgery, joined WSG in December 1988, he signed an employment agreement which contained a covenant not to compete. In October 1994, he signed a new agreement with WSS. Because Dr. Idbeis had an existing practice in Wichita for 8 years before joining WSG, his employment agreement with WSS contained a 2-year geographic restrictive covenant and a liquidated damages provision:
“9. Restrictive Covenant. During the term of this Agreement and for a period of two (2) years after the termination of this Agreement for whatever reason, Doctor will not, without the prior consent of Employer, directly or indirectly engage in the practice of medicine and/or surgery, nor own, manage, operate, control, be employed by, invest in, participate in, advise, consult with, or be connected with the ownership, management, operation or control of any business engaged in the practice of medicine and/or surgery within 75 miles of the city of Wichita, Kansas.
“In the event that Doctor does engage in the practice of medicine in violation of this covenant, then Doctor shall pay to Employer, as liquidated damages, five annual payments in an amount equal to 20% of tire total taxable compensation received by Doctor from Employer for the twelve month period immediately preceding the termination of Doctor’s employment. Such annual payments will commence upon termination of this agreement and continue on each anniversary thereof until five annual payments have been made.”
The other plaintiffs’ contracts did not contain the liquidated damages provision and had different geographic limitations.
*759Dr. Rumisek was new to Wichita when he joined WSG, although he had extensive experience. He completed his general and vascular surgical residency training in 1979 and his cardiothoracic surgical residency training in 1983, after which he was' awarded a fellowship in pediatric surgeiy at Children’s Hospital of Philadelphia, University of Pennsylvania. Before coming to Wichita, Dr. Rumisek, who is board certified in general and thoracic surgery, served as Assistant Chief of Cardiothoracic Surgery Service at Walter Reed Army Medical Center, Washington, D.C., and worked as an Associate Professor of Surgeiy at the University of Florida. He joined WSG in 1989 at which time he signed an employment agreement which was later assigned to WSS. The employment agreement stated, in part:
“9. Restrictive Covenant. During the term of this Agreement and for a period of two (2) years after the termination of this Agreement, Doctor will not, within Sedgwick County, Kansas, without tire prior written consent of Employer, directly or indirectly, engage in the practice of medicine and surgery or own, manage, operate, control, be employed by, invest in, participate in, advise, consult with, or be connected with the ownership, management, operation, or control of any business engaged in the practice of medicine and surgery.”
Both Drs. Benton and Fleming started their private practice experience with WSG. Dr. Benton completed his residency training in 1989 and additional training through a cardiothoracic fellowship from 1989 to 1992, during which time he obtained certification in general surgery. He joined WSG in 1992 and secured his board certification in thoracic surgery in 1993. His employment agreement contained tire same covenant not compete as was included in Dr. Rumisek’s agreement, restricting competition for a 2-year period with the geographic restriction covering Sedgwick County, Kansas.
Dr. Fleming joined WSG 2 years after Dr. Benton. He had completed his residency training in 1992 and had additional training through a cardiothoracic fellowship from 1992 to 1994. He secured his board certification in general surgery in 1993 and his board certification in thoracic surgeiy in 1995. Because WSS’s market broadened geographically during the 2 years between when Drs. Benton and Fleming began their respective practices, the restric*760tive covenant in Dr. Fleming’s employment agreement contained a slightly different 2-year geographic restriction which prohibited him from practicing within 75 miles of Wichita. In all other respects the covenant was the same as those of Drs. Rumisek and Benton.
By August 2000, the WSS division council began discussing whether the restrictive covenants in its physicians’ employment contracts should be amended. Dr. Benton was appointed to a subcommittee to consider possible alternatives and recommend options to the shareholders.
In November 2000, the WSS division shareholders met and voted 14-2 against eliminating restrictive covenants altogether. All four plaintiffs in this case voted against eliminating restrictive covenants. The shareholders then voted 14-9 to approve a modification of the restrictive covenant. The minutes state: “[T]he restrictive covenants will be changed to provide financial restrictions after five years of geographic restrictions.” The departing physician would pay 80 percent of the previous year’s net income in payments made over a 4-year period. The intent was to allow each physician to choose whether to add this modification to his or her contract. However, there was conflicting testimony from the physicians about what they believed was supposed to happen next and whether they believed the modification was immediately available.
In January 2001, the WSS division council unanimously approved the modification to the restrictive covenants. The modification was sent to lawyers for legal review, and when it was returned, it contained material additions including an attorney fees provision that had not been reviewed by the shareholders.
In May and June 2001, the proposed modification was presented to the WSS division council and the WSS division shareholders. The shareholders were told that finalization of the amendment would proceed. The minutes of an October 2001 division council meeting stated that the amendment would soon be sent out to the physicians. The minutes never mentioned any plan to seek WSS board of directors approval. By the time plaintiffs left their employment with WSS, the amendment still had not been presented to the physicians.
*761During the fall of 2001, WSS faced another unrelated problem: revisions in federal legislation (the Stark law, 42 U.S.C. § 1395nn, and regulations) which would force the group to alter its structure and management. Dr. Ammar, the CEO, decided not to proceed with finalization of the restrictive covenant modification because of problems regarding restrictive covenants which would need to be addressed by all divisions as part of Stark compliance. The trial court also thought it likely Dr. Ammar knew that withholding finalization of the restrictive covenant modification would be a bargaining chip in negotiating the possible termination of the plaintiffs’ employment. By that time, such termination was contemplated because most of the shareholders in the WSS division found unacceptable Dr. Idbeis’ involvement with a proposed spine hospital.
In Januaiy 2002, the WSS division council, which is the governing body of the division, voted to notify Dr. Idbeis of this fact. A letter was sent to Dr. Idbeis giving him three options: withdraw from the spine hospital project, resign, or the division would vote on whether he should be terminated.
The WSS board of directors terminated Dr. Idbeis’ employment; his last day was March 17, 2002. As a result of Dr. Idbeis’ termination, Drs. Rumisek, Benton, and Fleming gave notice of termination of employment; their last day was March 22,2002. The four doctors made plans to form their own cardiovascular surgery group in Wichita, called Mid-America Surgical Associates (MASA). By establishing MASA, the plaintiffs violated their restrictive covenants, but, through the filing of this suit, they questioned the enforceability of the covenants.
In making findings of fact specific to the question of whether the restrictive covenants were enforceable, the trial court found that both the physicians and WSS benefitted from the provisions. Addressing the benefit to WSS, the trial court found that WSG and WSS would not have enjoyed the same degree of success without at least a reasonable restrictive covenant. Addressing the benefit to physicians, in particular new surgeons, the trial court found that they benefitted from WSS’s reputation and contacts with area doctors and hospitals, particularly WSS’s close relationship with St. *762Francis Medical Center and its successor, Via Christi. The trial court also noted that a cardiovascular surgeon’s business is almost totally dependent upon referrals from others, predominantly cardiologists, family practice doctors, and in some cases, the cardiovascular surgeon’s employer. Because referral patterns are well established in the Wichita area, it is difficult for a cardiovascular surgeon to enter the Wichita market without joining an established group.
Focusing more specifically upon the plaintiffs, the trial court concluded that all four benefited from their employment with WSS, especially Drs. Rumisek, Benton, and Fleming, who were able to get a foothold in the Wichita market and build a foundation for referrals which would have been more difficult, if not impossible, had they opened their own practices in Wichita. However, the trial court determined that, while WSS initially enabled Drs. Rumisek, Benton, and Fleming to establish a referral base, over time, WSS’s ability to influence referrals dwindled because of the surgeons use of what were referred to as the three A’s: “ability, availability, and affability.”
Trial Court’s Conclusions of Lato
The trial court’s conclusions of law were, in large part, framed by the considerations stated in Weber v. Tillman, 259 Kan. 457, 913 P.2d 84 (1996), in which the court upheld a restrictive covenant which prevented a dermatologist from practicing within a 30-mile radius from his former employer’s office for a period of 2 years. The Weber court stated the following general principles:
“A noncompetition covenant ancillary to an employment contract is valid and enforceable if the restraint is reasonable under the circumstances and not adverse to the public welfare. [Citations omitted.] The rationale for enforcing a noncom-petition covenant is based on the freedom of contract. [Citation omitted.] However, it is well settled that only a legitimate business interest may be protected by a noncompetition covenant. If the sole purpose is to avoid ordinary competition, it is unreasonable and unenforceable. [Citations omitted.] Additionally, noncom-petition covenants included in employment contracts are strictly construed against the employer. [Citations omitted.]” 259 Kan. at 462.
The Weber court also set forth the following factors to be considered in analyzing whether a noncompetition clause is reasonable:
*763“(1) Does the covenant protect a legitimate business interest of the employer? (2) Does the covenant create an undue burden on the employee? (3) Is the covenant injurious to the public welfare? (4) Are the time and territorial limitations contained in the covenant reasonable? The determination of reasonableness is made on the particular facts and circumstances of each case.” 259 Kan. at 464.
In applying the Weber factors, the trial court first considered the time and territorial hmitations and found those limitations to be reasonable. This finding is not disputed on appeal.
The court also found that enforcement of the restrictive covenants would impose a burden on Drs. Rumisek, Benton, and Fleming, but not an undue burden. The court noted that forcing the doctors into the national cardiovascular job market, which has a glut, would impose a financial hardship on them. The court considered the remedy of modifying the restrictive covenant so that it would prohibit only the practice of cardiovascular surgery rather than the practice of medicine. However, the court concluded this was not a realistic remedy because the doctors “would lose their CV [cardiovascular] skills, and as a practical matter they would also struggle as non-CV surgeons.” Even given these difficulties, the court found the restrictive covenants would not impose an undue burden on the plaintiffs. The court did not specifically mention Dr. Idbeis in making its findings about undue burden; however, it is clear that the court believed Dr. Idbeis was under less of a burden than the other plaintiffs since his employment contract gave him the option of paying liquidated damages and remaining in the Wichita area. Again, these findings are not specifically disputed by the parties.
However, the trial court’s conclusions of law regarding the other factors are disputed.
Did the Trial Court Err in Ruling That the Restrictive Covenants Are Not Unenforceable as a Result of the Incorporation of the AMA’s Declarations Regarding Restrictive Covenants Into the Plaintiffs’ Employment AgreementsP
Before we reach the remaining factors of the Weber analysis, however, we must address the plaintiffs’ argument that the Weber factors are displaced by the American Medical Association’s (AMA) *764ethics guidelines, which the plaintiffs argue are incorporated into the contract between the parties.
All four doctors’ employment agreements contained the following provision regarding professional standards:
“3. Professional Standards. Doctor hereby promises to engage in the practice of medicine and surgery in Employer’s behalf to the best of his or her ability and in accordance with the generally accepted community standards, and to faithfully adhere to the principles of medical ethics of the American Medical Association or other associations or boards concerned with the conduct in question.” (Emphasis added.)
The AMA Council on Ethical and Judicial Affairs has published an opinion which states:
“Covenants-not-to-compete restrict competition, disrupt continuity of care, and potentially deprive the public of medical services. The Council on Ethical and Judicial Affairs discourages any agreement which restricts the right of a physician to practice medicine for a specified period of time or in a specified area upon termination of an employment, partnership or corporate agreement. Restrictive covenants are unethical if they are excessive in geographic scope and duration in the circumstances presented, or if they fail to make reasonable accommodation of patients’ choice of physician.” AMA Council on Ethical and Judicial Affairs, Op. E-9.02 (1998).
The trial court ruled that, because all of the plaintiffs’ employment agreements required them to faithfully adhere to the principles of medical ethics of the AMA, the AMA principles were made part of the employment agreements. Accordingly, the restrictive covenants in plaintiffs’ employment agreements were required to make reasonable accommodation of the patients’ choice of physician.
Both plaintiffs and defendant contend that the trial court erred in its consideration of the AMA provisions regarding restrictive covenants. Their arguments involve the interpretation of the plaintiffs’ written employment agreements, which raises a question of law over which this court’s review is unlimited. See Unrau v. Kidron Bethel Retirement Services, Inc., 271 Kan. 743, 763, 27 P.3d 1 (2001).
The plaintiffs ask us to reach the same conclusion as did the trial court and determine that their employment agreements with WSS *765incorporated the AMA’s principles relating to restrictive covenants and that restrictive covenants violate those principles. However, the plaintiffs also contend that the trial court erred in failing to conclude that, as a result of the applicability of the AMA provisions, the restrictive covenants are unenforceable.
Defendant WSS contends that the trial court erred in ruling that the AMA provisions were incorporated into the plaintiffs’ employment agreements, and in relying on the AMA provisions to require that the restrictive covenants make reasonable accommodation for the patients’ choice of physician.
Both parties’ arguments on this point fail. The AMA provisions do not state that all restrictive covenants are unethical; rather, such covenants are only unethical “if they are excessive in geographic scope and duration in the circumstances presented, or'if they fail to make reasonable accommodation of patients’ choice of physician.” These requirements are really no different than the common-law requirement in Kansas that restrictive covenants be reasonable and not adverse to the public welfare. See Weber, 259 Kan. at 462; see also Community Hosp. Group v. More, 183 N.J. 36, 56, 869 A.2d 884 (2005) (AMA standards do not make restrictive covenants per se unethical but adopt a reasonableness standard similar to that applied by courts); Loeser, The Legal, Ethical, and Practical Implications of Noncompetition Clauses: What Physicians Should Know Before They Sign, 31 J.L. Med. & Ethics 283, 287 (2003) (“While noteworthy in its opposition to restrictive covenants, CEJA’s position has limited legal impact, as the opinion merely parrots the reasonableness standard applied by most courts to non-competition clauses generally.”).
Thus, even if the plaintiffs’ contracts did incorporate the AMA provisions, as urged by the plaintiffs and found by the trial court, the trial court was not required to automatically invalidate the restrictive covenants. Rather, the court was still required to review the restrictive covenants under the four-part Weber analysis. Consequently, the plaintiffs’ argument that the trial court erred in failing to set aside the restrictive covenants altogether fails.
Defendant WSS’s argument fails for the same reason, i.e., the AMA’s statements about when a restrictive covenant is ethical are *766really no different than common-law reasonableness requirements. Kansas courts have noted that patient choice is an appropriate consideration in determining whether a restrictive covenant is injurious to the public welfare. See Weber, 259 Kan. at 470.
Additionally, in Graham v. Cirocco, 31 Kan. App. 2d 563, 69 P.3d 194, rev. denied 276 Kan. 968 (2003), the Court of Appeals held that a 150-mile noncompetition covenant restriction on solicitation was reasonable but a 25-mile limitation on office placement and the prohibition of practice in the entire Kansas City metropolitan area were unreasonable. The court concluded that the non-competition covenant which prevented predatory behavior through the restriction on solicitation by the departing physician “while permitting patients and referring doctors to continue to exercise their own choices struck an acceptable balance among the interests of the two parties, the patients, and the referring doctors.” 31 Kan. App. 2d at 569. Thus, the court recognized patient choice as one consideration.
Consequently, even if the trial court erred in ruling that the AMA provisions were incorporated into the employment agreements, the court did not err in considering patient choice. Thus, defendant WSS’s argument to the contrary fails.
Did the Trial Court Err in Ruling That the Defendant Has Legitimate Business Interests to Protect Through the Enforcement of the Restrictive Covenants P
Returning to the Weber factors, plaintiffs next argue that the trial court erred in ruling that WSS has legitimate business interests to be protected through a restrictive covenant. WSS responds that it had a legitimate interest in its referral relationships and in maintaining its size, which allowed it to further its educational mission and introduce new specialties.
The parties agree that whether a restrictive covenant serves to protect an employer’s legitimate interests is a question of law over which this court’s review is unlimited. See Weber, 259 Kan. at 461-62.
First, the trial court found that the restrictive covenants in the plaintiffs’ employment agreements protected a legitimate business *767interest of WSS by enabling it to attain and maintain a larger size or “critical mass,” which allowed it to further its educational mission, introduce new specialties, and potentially negotiate favorable reimbursement rates from insurance carriers. Plaintiffs contend this interest is nothing more than a desire to prevent ordinary competition and point out that there is no authority in Kansas or anywhere else that maintaining the size of a business operation is a legitimate business interest which can be protected by a restrictive covenant.
Traditionally, courts have recognized that “a medical employer has a legitimate interest warranting protection by a noncompetition agreement if a departing physician was provided with (a) patient contacts, (b) training, or (c) confidential business information.” Berg, Judicial Enforcement of Covenants Not to Compete Between Physicians: Protecting Doctors’ Interests at Patients’ Expense, 45 Rutgers L. Rev. 1, 15 (1992). In Weber, this court noted a variety of legitimate business interests which have been recognized by other courts: “customer contacts, . . . the special training of employees, trade secrets, confidential business information, loss of clients, good will, reputation, seeing that contracts with clients continue, and referral sources.” 259 Kan. at 467. Weber also cited the Berg article for its review of noncompetition agreements between physicians.
None of these interests are related to the employer’s size or the special contributions the employer might be able to make to its community because of its size, and WSS’s argument that it has a legitimate interest in preserving its size is not persuasive.
The other business interest cited by the trial court as a legitimate interest for protection through a restrictive covenant was preservation of a referral base. Plaintiffs urge this court to conclude that WSS did not have a legitimate interest in referral sources.
Both sides cite cases from other jurisdictions which they claim are persuasive. For example, defendant WSS cites Medical Specialists, Inc. v. Sleweon, 652 N.E.2d 517 (Ind. App. 1995), a case upholding a restrictive covenant on the ground that the continued success of an infectious disease practice, which was dependent *768upon patient referrals from other doctors, was a legitimate interest worthy of protection.
We also note a recent decision of the New Jersey Supreme Court, which has- decided many of the leading physician restrictive covenant cases, in which the court reaffirmed that legitimate business interests for a hospital to protect with a post-employment restrictive covenant with a doctor include: “(1) protecting confidential business information, including patient lists; (2) protecting patient and patient referral bases; and (3) protecting investment in the training of a physician.” More, 183 N.J. at 58.
Plaintiffs cite several cases from other jurisdictions which refused to recognize referrals as a legitimate business interest to be protected by a restrictive covenant. See, e.g., Cardiovascular Surgical v. Mammana, 61 P.3d 210, 214 (Okla. 2002) (noting that referrals to cardiovascular surgeons are made based on reputation and not employment with a particular corporation, and “[o]ne surgeon has no legitimate business interest in another surgeon’s referral base regardless of a past employer-employee relationship”); Hoddeson v. Conroe Ear, Etc. Assoc., 751 S.W.2d 289, 290 (Tex. App. 1988) (“referring physicians are governed by die skills and qualifications of the receiving physician, rather than his associations” and employer “did not impart trade secrets, specialized training or confidential information to [employee]”).
These cases demonstrate only that courts are split on the issue of whether referral sources are a legitimate interest. In Kansas, however, the law is clear that referral sources are a legitimate interest which can be protected by a restrictive covenant even in the context of a medical practice. See Weber, 259 Kan. at 466-67; Graham, 31 Kan. App. 2d at 570. The trial court’s ruling diat WSS had a legitimate business interest in protecting its referral relationships was consistent with Weber and Graham.
The plaintiffs also ask us to uphold die trial court’s finding that any interest WSS had in referral sources no longer existed by the time the plaintiffs left their employment. The trial court concluded the plaintiffs had developed relationships with referring physicians based upon their own abilities and were no longer relying on WSS’s referral base by die time they left WSS. Defendant WSS claims *769the trial court’s conclusion that it lost its interest in referral sources over time is contrary to Kansas cases holding that the relationships developed by an employee during the course of his or her employment belong to the employer, not the employee. See, e.g., Eastern Distributing Co., Inc. v. Flynn, 222 Kan. 666, 567 P.2d 1371 (1977) (interest in “customer contacts” is a legitimate interest subject to protection). Flynn also established that the ultimate determination as to whether a legitimate interest subject to protection is shown is an issue of law, although there will be underlying facts determined by the court. 222 Kan. at 673. Here, the dispute is not with the trial court’s factual finding but with the conclusion of law drawn from those facts. As such, our review of that question is unlimited. See Weber, 259 Kan. at 461.
The Court of Appeals considered and rejected a similar argument in Graham. Cirocco, the employee, argued he might have come into contact with those same patients and referral sources if he had started a practice by himself; the court stated: “[T]he fact is he did not. Instead, he came to an entirely new area of the country, became board-certified while working for Graham, and, for 6 years, took advantage of Graham’s established contacts in the community to make a name for himself.” 31 Kan. App. 2d at 570.
In Weber, the court noted that Dr. Weber had invested “years, education, and effort in estabhshing his practice and the value of goodwill developed over 17 years,” which benefitted Dr. Tillman who brought no patients with him but who took patients when he left. 259 Kan. at 468.
Similarly, even though the plaintiffs’ reputations may have become over time the paramount factor influencing referrals, they each benefitted from their association and from the investment of WSS and its contribution of goodwill. Furthermore, it is a situation such as present in this case where the association “ posits a substantial risk of loss of clientele to an employee because of the nature of his work’ ” which warrants a covenant. Flynn, 222 Kan. at 672 (quoting Blake, Employee Agreement Not to Compete, 73 Harv. L. Rev. 625, 657-59 [1960]).
Finally, Weber instructs us:
*770“It is the duty of courts to sustain the legality of contracts in whole or in part when fairly entered into, if reasonably possible to do so, rather than to seek loopholes and technical legal grounds for defeating their intended purpose. Although restrictive provisions in contracts of employment must be reasonable and not such as to contravene the public welfare, the paramount public policy is that freedom to contract is not to be interfered with lightly.” 259 Kan. at 474; see Foltz v. Struxness, 168 Kan. 714, 721-22, 215 P.2d 133 (1950).
Over time, there were discussions within WSS about whether restrictive covenants should be removed from contracts. A majority of shareholders, including the plaintiffs, voted to retain the covenants, implicitly recognizing that the agreements continued to protect a valuable interest of the group. Dr. Idbeis, who was the one plaintiff who brought a Wichita referral base with him when he joined the group in 1988, renewed the restrictive covenant in 1994 with the restructuring. No argument is made that there was not valid consideration for the new contract because the business interest which WSS sought to protect had evaporated with time. The parties again imposed the restriction.
Furthermore, there is no showing made and the trial court did not find that WSS merely sought to prevent ordinary competition by including restrictive covenants in the contracts. Thus, no showing has been made that the parties’ freedom to contract should be overridden.
We, therefore, conclude that WSS had a legitimate business interest in the referral sources which it could seek to protect through restrictive covenants.
Did the Trial Court Err in Ruling That Enforcement of the Restrictive Covenants Would be Injurious to the Public Welfare?
As its first issue on cross-appeal, WSS argues that the trial court erred in ruling that enforcement of the restrictive covenants would be injurious to the public welfare.
The trial court considered whether enforcement of the covenants would result in a shortage of cardiovascular surgeons in the Wichita market. The trial court noted that Via Christi wanted two or three more cardiovascular surgeons to work in its facility, but the WSS division of cardiovascular surgeons opposed hiring more surgeons because some of them wanted more work. Dr. Idbeis *771opined that the Wichita market needs as many as 17.9 cardiovascular surgeons. A defense witness commissioned by Wesley Medical Center said there is demand for only 7.29 cardiovascular surgeons. Another defense witness believed Wichita requires 11 cardiovascular surgeons. Evidence was presented that advancing technology will decrease the future need for cardiovascular surgeons; however, evidence was also presented that, as baby boomers age, the future need for cardiovascular surgeons will increase. Conflicting testimony was presented about whether some of the cardiovascular surgeons in Wichita planned to retire in the near future and on whether patient care would suffer if three or four of the plaintiffs were removed from the Wichita market.
Ultimately, the trial court found it could not make a determination as to the need for cardiovascular surgeons based solely on numbers; rather, the quality of the surgeons being lost was also a factor. The trial court noted that, while there were 16 cardiovascular surgeons in Wichita, the four plaintiffs were some of the busiest. The court also noted that if Drs. Rumisek, Benton, and Fleming were required to end their cardiovascular practice, Wichita would lose one of the few surgeons who does pediatric neonatal cardiovascular surgery, one of the few surgeons who does beating heart surgery, and one of the few surgeons who does the Ross procedure. The court found all of these procedures were wanted and needed in the critical area of heart surgery.
First, WSS contends that the trial court improperly admitted Dr. Idbeis’ report and testimony regarding his assessment of the local need for physicians because Dr. Idbeis was not qualified to give an expert opinion. However, we need not address whether the trial court erred in admitting Dr. Idbeis’ report because, even if the admission was erroneous, the error was harmless because the trial court did not rely upon the report. The report addressed a purely numeric analysis of the demand for cardiovascular surgeons in the Wichita market. As WSS recognizes in its brief, the trial court’s findings were based upon what WSS labels a “subjective” analysis by the trial court, i.e., factors such as the specialized procedures performed by the plaintiffs and their good reputations for high quality care.
*772Second, WSS argues this subjective analysis was inappropriate. The problem with WSS’s argument is that it is not at all clear from the record that the trial court made any finding that enforcement of the restrictive covenants would be injurious to the public welfare because of the special qualities of the plaintiffs. Clearly, the court did refuse to make a finding that enforcement of the restrictive covenants would result in a shortage of cardiovascular surgeons, stating, “[T]his case is not just about numbers.” However, when the court reviewed the Weber factors it based its ruling that enforcement of the restrictive covenants would be injurious to the public welfare on the fact that, as to Drs. Rumisek, Benton, and Fleming, the restrictive covenants failed to make accommodation of patient choice, stating, “Having found the enforcement of the restrictive covenant in Drs. Rumisek, Benton, and Fleming’s contract — contracts injurious to the public welfare by virtue of the lack of a liquidated damage buyout . . . .” No mention was made of a shortage of cardiovascular surgeons, whether measured on a subjective or objective basis. Thus, the only clear finding of injuiy to public welfare did not deal with shortage of doctors.
Any restrictive covenant agreed to by a physician is going to malee some Hmitation on patient choice. Perhaps because of this, the AMA guideline condemns only those covenants which fail to make “reasonable accommodation” for patient choice. In each case, the varying circumstances must be considered in the effort to evaluate that impact. One valid consideration in this case is the nature of the typical relationship between a patient and a cardiovascular surgeon: it is usually short-term, lasting long enough to accommodate the surgical care and follow-up.
However, the trial court did not discuss such considerations, instead basing its ruling solely upon the lack of a liquidated damages provision in the contracts of Drs. Rumisek, Fleming, and Benton. In reaching this conclusion, the trial court appears to rely upon the court’s analysis in Weber in which, after noting that Dr. Tillman could continue to practice dermatology upon paying liquidated damages as provided by his employment contract, the court stated: “Thus, as a practical matter, the people of Hays and the surrounding area may not lose Dr. Tillman’s services as a dermatologist. *773Further, their welfare is not injured if they have to travel further to obtain dermatology services should Dr. Tillman elect not to pay liquidated damages . ; . .” 259 Kan. at 475.
However, the Weber court did not hold that the liquidated damages provision in Dr. Tillman’s contract prevented the enforcement of the restrictive covenant from being injurious to the public welfare. Rather, the court held that the public welfare was not injured regardless of whether Dr. Tillman paid liquidated damages and stayed in Hays or complied with the restrictive covenant and left the area. It is an overly broad reading of Weber to conclude that it is necessary that all restrictive covenants in the employment contracts of physicians have liquidated damages provisions in order to be enforceable, and the lack of one does not make the contract unenforceable.
Furthermore, the provision for liquidated damages is not a provision dictating the scope of the covenant not to compete. Rather, it represents the parties’ agreement on an available remedy if the covenant is breached. This is an important distinction, especially in the context of the trial court’s findings. It is the scope of the restriction, not the presence of a remedy, which makes the covenant enforceable. The remedy for breach of the parties’ contract is either an injunction to enforce the contract or a computation of monetary damages. The trial court made no finding that the liquidated damages provision which it grafted into the contract had any relationship to actual damages.
Thus, the trial court erred in finding that the restrictive covenants of Drs. Rumisek, Benton, and Fleming were injurious to the public welfare because of the lack of a liquidated damages provision.
Did the Trial Court Err in Allowing Three of the Plaintiffs to Pay Liquidated Damages Where Their Employment Agreements Did Not Contain Such a ProvisionP
The trial court further bolstered its ruling to add the liquidated damages provision by stating it was “merely finalizing that which was not finalized.” The trial court stated that this was the “controlling ruling.” The trial judge then added, “Now, which is prob*774ably a little bit unusual for a Court to do, in case I was wrong in my previous decision, I am going to set forth and address the issue regarding the amendment, a possible amendment to the contracts.”
The trial court found that, given the manner in which the divisions operated independently of WSS and the history of making modifications to employment contracts without the WSS board of directors’ approval, the November 2000 vote of the division shareholders to approve the modification to the restrictive covenants did not have to be approved by the WSS board and did not have to be in writing, even though the contract required modifications to be written to be binding on WSS.
Modification of a contract, like the creation of an original contract, requires a mutual assent or a meeting of the minds. Fast v. Kahan, 206 Kan. 682, 481 P.2d 958 (1971). Despite 53 pages of transcript, there is no finding which specifies what constituted the subsequent contract under the facts of this case. Was it the November 2000 version which was approved by the shareholders and ambiguously recorded in the minutes as “[T]he restrictive covenants will he changed to provide financial restrictions after five years of geographic restrictions”? Was it the June 2001 version which included the language added by attorneys? Or, was it some other agreement? Second, the trial court also found:
“Since it was not presented for signature, this Court cannot determine the identity of physicians who would have made the prepared written restrictive covenant amendment a part of their employment terms, who would have declined any change to the restrictive covenant, and who would have required a restrictive covenant that did not add material provisions to the shareholders’ approved modification.”
In light of this finding of fact, there cannot be a determination that there was a meeting of minds as to an agreement. The trial court did not make a finding specific to Drs. Benton, Fleming, and Rumisek regarding their acceptance. Rather, the trial court’s reasoning seems to be that the possibility of acceptance still existed at the time these plaintiffs resigned.
The trial court found the modification did not apply to Dr. Idbeis’ contract because, upon his termination, Dr. Idbeis
*775“did not accept the modification to the restrictive covenant as part of his contract and instead chose to negotiate the buyout that was part of his 1995 contract. The Court’s of the opinion that he is bound by his election not to have his contract modified with the new liquidated damages provision.”
As WSS argues, this must mean that the trial court determined that the doctors’ acceptance or rejection of the amendment occurred during • settlement negotiations which did not take place until well over a year after the shareholder vote. Again, this finding of fact is inconsistent with the conclusion of law that the doctors had agreed to a modification. Furthermore, WSS notes that in the same letter the other doctors proposed paying an amount different than that proposed in the draft amendment and, therefore, the conclusion must be that all of the plaintiffs rejected the modification.
As the trial court stated in its “controlling ruling,” any modification would be “finalizing that which was not final.” The trial court’s alternative findings do not support the conclusion that there was a modification which the plaintiffs accepted.
Plaintiffs also make arguments based on ratification, promissory estoppel, equitable estoppel, unclean hands, and breach of fiduciary duty. None of the plaintiffs’ arguments were mentioned by the trial court as a basis for its ruling, and the trial court did not make the findings of fact necessary to support these claims or allow our review of the issues. There cannot be ratification of that which is not final, and the trial court did not make findings to support reliance, fraudulent conduct, unclean hands, or breach of a fiduciary duty.
In conclusion, the restrictive covenant in each of the plaintiff s employment contracts is enforceable. Because only Dr. Idbeis’ contract included a liquidated damages provision, only he is entitled to pay liquidated damages in lieu of ceasing the practice of medicine as provided in his restrictive covenant.
*776Affirmed in part, reversed in part, and remanded for entry of judgment consistent with this opinion.
GERNON, J., not participating. Larson, S.J., assigned.