Blackwell, Justice. Rochelle Frett was injured when she slipped and fell at her place of employment during a scheduled lunch break. She filed a claim for benefits under the Workers’ Compensation Act, OCGA § 34-9-1 et seq., but the State Board of Workers’ Compensation denied her claim. Frett sought judicial review, and the superior court upheld the denial of her claim. Frett then appealed the decision of the superior court, and in Frett v. State Farm Employee Workers’ Compensation, 348 Ga. App. 30 (821 SE2d 132) (2018), the Court of Appeals affirmed. Relying on Ocean Acc. & Guar. Corp. v. Farr, 180 Ga. 266 (178 SE 728) (1935), the Court of Appeals held that Frett suffered no injury compensable under the Act because she sustained her injury during a scheduled break, and her injury, therefore, did not arise out of her employment.[1] We issued a writ of certiorari to reconsider Farr and review the decision of the Court of Appeals in this case. For the reasons that follow, we overrule Farr, and we reverse the decision below. 1. The facts in this case are undisputed and are summarized accurately in the opinion of the Court of Appeals: At the time of the incident, Frett worked as an insurance claims associate for State Farm Insurance Companies (“State Farm”). Each workday, Frett had a mandatory, unpaid 45-minute lunch break. An automated system scheduled staggered lunch breaks to ensure enough associates were available to handle calls. After logging on for the day, Frett would see her schedule, including the time for her lunch break. At her scheduled lunch break time, Frett would log out of the phone system. All parties agree that Frett was free to do as she pleased on her break and could leave the office for lunch if she wished. Frett was not expected or asked to do work during her lunch breaks. Generally, Frett brought her lunch and would walk to the State Farm employee breakroom on her floor to prepare her food. During the spring and summer, she would eat her lunch on a bench outside of the office building or in her car in the parking lot. State Farm has a suite within the shared office building, but does not own the parking lot or the surrounding outdoor areas. On the day of the incident, Frett logged out of the phone system at her assigned time and walked to the breakroom where she microwaved her food. As Frett started to exit the breakroom to take her lunch outside the building, she slipped on water and fell[, suffering an injury]. It is undisputed that Frett was still inside the breakroom when she fell. Frett, 348 Ga. App. at 31. Frett filed a claim for workers’ compensation benefits. After a hearing, an administrative law judge awarded benefits, but the appellate division of the Board later reversed the award and denied benefits. The Board found that Frett had not sustained a compensable injury under the Act because, although her injury “arose in the course of her employment, it did not arise out of her employment but, instead, arose out of a purely personal matter.” Frett sought judicial review in the Superior Court of DeKalb County,[2] which ultimately affirmed the denial of benefits. Frett then filed an application for discretionary review,[3] and the Court of Appeals granted her application. See Frett, 348 Ga. App. at 30. Affirming the judgment of the superior court, the Court of Appeals perceived a conflict in the case law with respect to the compensability of an injury that is sustained in circumstances like those presented here. Broadly speaking, one line of cases held that an injury may be compensable if it occurs as an employee is entering the premises of the employer to begin her work or exiting the premises when her work is done. This principle is commonly known as the “ingress and egress rule.” See Frett, 348 Ga. App. at 33. Another line of cases, beginning with Farr, held that an injury is not compensable if it occurs during a “scheduled” break. These principles came into conflict, the Court of Appeals noted, in cases in which an employee was injured as she was leaving the premises of her employer at the beginning of a scheduled break or entering the premises at the end of a break. See id. at 34-35. Attempting to resolve this conflict, the Court of Appeals determined that its prior decisions applying the ingress and egress rule to scheduled lunch breaks were “an improper dilution” of Farr, and so, it “disapprove[d]” those decisions, noting that “ any decision to apply the ingress and egress rule to the scheduled break exception should be made by our Supreme Court, particularly because the Supreme Court has never expressed its view on the ingress and egress rule generally.” Id. at 36. The Court of Appeals then concluded that Frett’s injury was not compensable because it occurred during a scheduled lunch break, when she was “free to do as she pleased.” Id. 2. The Workers’ Compensation Act provides for compensation for injuries that occur “in the course of’ employment and “aris[e] out of’ employment. See OCGA § 34-9-1 (4). These two prerequisites to compensation, which have remained unchanged since the original adoption of the Act in 1920, are “independent and distinct,” and any claim for compensation under the Act must satisfy both prerequisites. See Mayor & Aldermen of Savannah v. Stevens, 278 Ga. 166, 166 (1) (598 SE2d 456) (2004). See also New Amsterdam Cas. Co. v. Sumrell, 30 Ga. App. 682, 687 (118 SE 786) (1923). The Court of Appeals below did not address the “in the course of’ prerequisite. It instead held only that the injury sustained by Frett is not compensable because it did not “arise out of’ her employment. See Frett, 348 Ga. App. at 36. For a proper analysis of this case, however, each prerequisite to compensation must be examined. (a) In the Course of Employment. An injury arises “in the course of’ employment when it “occurs within the period of the employment, at a place where the employee may be in performance of her duties and while she is fulfilling or doing something incidental to those duties.” Hennly v. Richardson, 264 Ga. 355, 356 (1) (444 SE2d 317) (1994). This statutory prerequisite “relates to the time, place and circumstances under which the injury takes place.” Potts v. UAP-Ga. Ag. Chem., Inc, 270 Ga. 14, 15 (506 SE2d 101) (1998). See also Stevens, 278 Ga. at 166 (1) (“An injury arises in the course of certain employment if the employee is engaged in that employment at the time the injury occurs.”). Injuries occurring “in the course of’ employment certainly include injuries sustained when an employee is actually engaged in the performance of her assigned work, but they also include injuries sustained when the employee is engaged in activities “incidental” to her assigned work. See Hennly, 264 Ga. at 356 (1). Such incidental activities include, among other things, ingress and egress to the place of work while on the employer’s premises. See Federal Ins. Co. v. Coram, 95 Ga. App. 622, 624 (98 SE2d 214) (1957) (“[G]oing to and from the parking lot in order to reach and leave her immediate working area was a necessary incident to the [worker's] employment.”). See also Longuepee v. Georgia Institute of Technology, 269 Ga. App. 884, 885 (605 SE2d 455) (2004). Incidental activities also include the employee reasonably attending to routine personal needs, such as eating a meal or using the restroom. See, e.g., Thornton v. Hartford Acc. & Indem. Co., 198 Ga. 786, 788 (32 SE2d 816) (1945) (“Acts of ministration by a servant to himself, such as quenching his thirst [or] relieving his hunger[,] are incidents to his employment and acts of service therein within the workmen’s compensation acts though they are only indirectly conducive to the purpose of the employment.” (Citation and punctuation omitted.)); Harris v. Peach County Bd. of Commrs., 296 Ga. App. 225, 231 (674 SE2d 36) (2009) (same). See also 1 Douglas T. Lay, Kissiah and Lay’s Georgia Workers’ Compensation Law § 5.08 (4th ed. 2017) (“An employee who, during the time period of his employment and on employment premises, engages in an act to minister to his personal comfort (whether it is satisfying his hunger, quenching his thirst, relieving himself, or otherwise), ordinarily does not leave the course of his employment by doing so . . . .”).[4] The latter principle is consistent with Farr, in which we assumed that a worker’s injury that occurred while he was preparing to eat lunch at his job site occurred “in the course of his employment. See Farr, 180 Ga. at 270. And we have directly applied this principle in a case in which a salesman on a business trip suffered a fatal accident while walking to his hotel from a cafe where he had eaten dinner. Thornton, 198 Ga. at 787. There, we held that the salesman’s injury occurred “in the course of’ his employment, and we reasoned that a traveling employee’s lodging in a hotel “or preparing to eat, or while going to or returning from a meal” was conduct “incident to his employment, unless he steps aside from his employment for personal reasons.” Id. at 790.[5] Applying these principles to this case, we conclude that Frett sustained an injury “in the course of her employment. It is undisputed that she was injured on the premises of her employer, in the middle of her workday, while preparing to eat lunch. This activity, being reasonably necessary to sustain her comfort at work, was incidental to her employment and is not beyond the scope of compensability under the Act. See Thornton, 198 Ga. at 787. See also Harris, 296 Ga. App. at 228 (injury occurred in the course of employment when employee was injured at work while bending over to pick up her personal medication); Edwards v. Liberty Mut. Ins. Co., 130 Ga. App. 23, 24 (2) (202 SE2d 208) (1973) (injury was compensable when it occurred while the employee was using the restroom during the workday).[6] The fact that Frett was not paid during her lunch break, or that she was free to do other tasks during that time, is not dispositive of whether her preparation to eat lunch was “in the course of” her employment. When analyzing the “in the course of prerequisite, courts generally focus on the nature of the employee’s activity at the time of the injury, not whether she was paid for it or was free to do something else. See, e.g., U.S. Fidelity & Guar. Co. v. Skinner, 188 Ga. 823, 829 (5 SE2d 9) (1939) (traveling employee’s injury did not occur in the course of employment because he was injured while taking a detour for personal pleasure, even though the detour occurred with the permission of the employer and at the employer’s expense); American Mut. Liability Ins. Co. v. Lemming, 187 Ga. 378, 380 (200 SE 141) (1938) (employee was not injured in the course of his employment with the company, even though the injury occurred while he was performing an errand at the direction of the company’s president, because he was not “fulfilling any duties he owed” to the company at the time); Swanson v. Lockheed Aircraft Corp., 181 Ga. App. 876, 878 (1) (354 SE2d 204) (1987) (“Clearly, even if the employee is on a scheduled break and even if the employee is free to use the break time as he pleases, if the employee is in fact engaged in employment-related activities, the injury is compensable under the Act.”). That said, the lack of payment and freedom to act may be significant factors in close cases, where the nature and timing of the employee’s activity at the time of the injury are only tenuously connected to her usual work hours or work-related activities. See Thornton, 198 Ga. at 786 (noting that an employer “usually pays the expenses of [a traveling employee's] lodging and meals”). But this is not such a close case. It is clear that Frett was injured during an ordinary lunch break in the middle of her workday in a breakroom provided by her employer for the use of employees during such breaks. And Frett was indeed using her break time to prepare and eat her lunch, not to run some personal errand. Thus, we have no trouble concluding, as did the Board, that Frett was injured “in the course of’ her employment.[7] We now turn to whether her injury also “arose out of’ her employment. (b) Arising out of Employment. The “arising out of’ prerequisite deals with causation. “An injury arises ‘out of the employment when a reasonable person, after considering the circumstances of the employment, would perceive a causal connection between the conditions under which the employee must work and the resulting injury.” Hennly, 264 Ga. at 356 (1). See also Ray Bell Constr. Co. v. King, 281 Ga. 853, 855 (642 SE2d 841) (2007) (“The words ‘arising out of the employment’ refer to the causal connection between the employment and the injury.” (Citation and punctuation omitted)). As we explained in Thornton, The words ‘arising out of mean that there must be some causal connection between the conditions under which the employee worked and the injury which he received. The causative danger must be incidental to the character of the employment, and not independent of the relation of master and servant. The accident must be one resulting from a risk reasonably incident to the employment. And a risk is incident to the employment when it belongs to, or is connected with, what a workman has to do in fulfilling his contract of service. Thornton, 198 Ga. at 792-793. See also SCI Liquidating Corp. v. Hartford Ins. Co., 272 Ga. 293, 294 (526 SE2d 555) (2000) (“In order to ‘arise out of employment, an injury must be a risk of employment that a reasonable person could have foreseen due to the nature of the work.”); Ladson Motor Co. v. Croft, 212 Ga. 275, 278 (92 SE2d 103) (1956) (to establish the “arising out of’ prong, the claimant must show that his employment was a “contributing proximate cause” of his injury); Fried v. U.S. Fidelity & Guar. Co., 192 Ga. 492, 495 (15 SE2d 704) (1941) (“[I]f the injury can be seen to have followed as a natural incident of the work, and to have been contemplated by a reasonable person familiar with the whole situation as a result of the exposure occasioned by the nature of the employment, then it arises ‘out of the employment. But it excludes an injury which cannot fairly be traced to the employment as a contributing proximate cause, and which comes from a hazard to which the workmen would have been equally exposed apart from the employment.” (Citation and punctuation omitted)). In this case, consideration of the “arising out of’ prerequisite should be straightforward. It is undisputed that Frett was injured when she slipped and fell on the wet floor of the breakroom on her employer’s premises. It logically follows that her injury was causally connected to the conditions under which she worked, and her injury, therefore, “arose out of’ her employment. See, e.g., Hennly, 264 Ga. at 356 (1) (claimant’s injuries arose out of her employment where she was injured due to second-hand smoke on her employer’s premises); Thornton, 198 Ga. at 792-793 (injury arose out of employment where traveling employee tripped and fell while crossing the street on the way to his hotel). Compare Stevens, 278 Ga. at 167 (2) (claimant’s car accident did not arise out of employment because it “was in no way related to her work as a police officer” and the “hazards she encountered were in no way occasioned by her job as a police officer”); Croft, 212 Ga. at 278 (where night watchman was found murdered during his shift but outside the employer’s premises, the evidence was insufficient to show that the watchman’s death arose out of his employment). Nevertheless, a determination that Frett’s injury—sustained during her scheduled lunch break—arose out of her employment would seem to conflict with Farr, 180 Ga. 266. In that case, the claimant—a steam-fitter and plumber—was injured when he tripped and fell while walking down stairs to a boiler room, where he intended to eat his lunch. The injury occurred on a job site—the claimant was assigned to work in the boiler room—during a 30- minute lunch break. In deciding whether the worker’s injury was compensable under the Act, we first assumed that it occurred “in the course of his employment. See id. at 270. We held, however, that the evidence authorized a finding by the Department of Industrial Relations (which at that time administered the Act) that the worker’s injury did not “arise out of his employment. We reasoned that the injury “arose after he had ‘knocked off from his work, and while he was preparing to eat his lunch. His preparation for lunch and his eating lunch was his individual affair. It was not a part of his employer’s work.” Id. at 270-271. So, we concluded, the department was “authorized to find that the ‘noon time’ was [the worker's] to employ as he should choose and eat his lunch where he pleased, and that the accident arose out of his individual pursuit and not out of his employment.” Id. at 271. Because the facts in Farr and this case are so similar, Farr strongly suggests that the injury sustained by Frett is not compensable, and if Farr remains good law, the decision of the Board in this case must be upheld. 3. Generally, we “adhere to the principle of stare decisis, which directs the courts to stand by their prior decisions.” Georgia Ports Auth. v. Lawyer, 304 Ga. 667, 677 (3) (821 SE2d 22) (2018) (citation and punctuation omitted)). At the same time, “we recognize that stare decisis is not an inexorable command, and sometimes, there are compelling reasons to reexamine an earlier decision.” Id. (citations and punctuation omitted). See also Woodard v. State, 296 Ga. 803, 812 (3) (b) (771 SE2d 362) (2015) (“[W]hen governing decisions are unworkable or are badly reasoned, this Court has never felt constrained to follow precedent. Stare decisis is . . . a principle of policy and not a mechanical formula of adherence to the latest decision.” (Citation and punctuation omitted)). When we reconsider our prior decisions, “we must balance the importance of having the question decided against the importance of having it decided right” Olevik v. State, 302 Ga. 228, 245 (2) (c) (iv) (806 SE2d 505) (2017) (emphasis in original; citation and punctuation omitted). To that end, we consider several factors in deciding whether to overrule a precedent, namely, the “soundness of its reasoning,” the “age of the precedent, the reliance interests involved, and the workability of the prior decision.” Lawyer, 304 Ga. at 678 (3). “The soundness of a precedent’s reasoning is the most important factor.” Olevik, 302 Ga. at 245 (2) (c) (iv). With these principles in mind, we now consider whether to stand by our decision in Farr. (a) Soundness of Reasoning The reasoning of Farr is unsound, and it is completely untethered from the analytical framework consistently employed by this Court in workers’ compensation cases for nearly a century. In Farr, we determined that the injury at issue did not “arise out of’ employment because it occurred at a time when the employee had left his work duties and was engaged in an “individual pursuit.” Farr, 180 Ga. at 270-271. In so reasoning, we impermissibly conflated the “in the course of’ and “arising out of’ prerequisites to compensability. As seen from our discussion in Division 2, supra, whether a worker is performing an employment-related activity at the time of the injury is a question that belongs squarely to the “in the course of’ prerequisite. See, e.g., Stevens, 278 Ga. at 166 (1) (“An injury arises in the course of certain employment if the employee is engaged in that employment at the time the injury occurs.”); Potts, 270 Ga. at 15 (the “in the course of prong “relates to the time, place and circumstances under which the injury takes place”). The “arising out of” prerequisite, on the other hand, deals with causation—whether there is a “causal connection” between the employment and the injury, see Hennly, 264 Ga. at 356 (1), or, in other words, whether the employment was a “contributing proximate cause” of the injury, see Croft, 212 Ga. at 278. This Court in Farr said nothing at all about causation when it analyzed the “arising out of” prong, focusing instead on the fact that the worker had “knocked off from work and was free to do as he pleased. See Farr, 180 Ga. at 270-271. If the Farr court had engaged in the proper analysis, we believe that its conclusion would have been different. The Farr court was prepared to accept that the employee’s injury occurred in the course of his employment. See Farr, at 180 Ga. at 270. This makes sense because the employee was getting ready to eat lunch—an activity that we subsequently indicated generally is incidental to employment. See Thornton, 198 Ga. at 788. Next, the Farr court should have asked whether there was “a causal connection between the conditions under which the work [was] required to be performed and the resulting injury.” Sumrell, 30 Ga. App. at 688. See also Georgia Cas. Co. v. Martin, 157 Ga. 909, 910 (122 SE 881) (1924) (holding that a worker’s injury did not arise out of his employment because his employment was not the proximate cause of the injury). It appears to us that, similar to the case at bar, the worker’s injury in Farr occurred precisely as a result of his working conditions—he tripped and fell on the steps at his work site, while engaged in an activity incidental to his employment. It cannot be said that this injury was unrelated to his work or that the “hazards [he] encountered were in no way occasioned by [his] job.” Stevens, 278 Ga. at 167 (2). Farr’s misguided analysis was an outlier, even in its day. By the time Farr was decided, the causation-based nature of the “arising out of’ prong had been firmly established. See Martin, 157 Ga. at 910; Sumrell, 30 Ga. App. at 689.[8] And in the years since, we have never relied on Farr in connection with the “arising out of inquiry, consistently adhering instead to the proper, causation- based approach. For instance, in Fried, 192 Ga. 492 (decided six years after Farr), we considered whether an injury arose out of employment where the employee was injured after a fight with a third party, while attempting to collect a work-related debt on his employer’s behalf. Id. at 494-495. We discussed at length the causation-based nature of the “arising out of” prong, but we never mentioned Farr in connection with this discussion (we did cite Farr in the opinion, but only for the deferential standard of review applied to administrative workers’ compensation decisions). See id. at 496. Likewise, in Thornton, we never mentioned Farr in our discussion of the “arising out of prong. See Thornton, 198 Ga. at 792-795. Rather, we focused on causation, holding that the injury arose out of employment because it occurred “by reason of a risk or hazard of the streets,” with the employee’s “duties being such as to carry him upon the streets.” Id. at 795. See also Croft, 212 Ga. at 276 (holding, without any reference to Farr, that the evidence was “wholly insufficient” to show that the employee’s death arose out of his employment—”that his employment was a contributing proximate cause of his death”). At the same time, we have consistently treated the question of whether the employee was engaged in purely personal tasks (that were not even incidental to her work duties) at the time of the injury as a part of the independent and distinct “in the course of’ prerequisite. Merely three years after Farr, we decided a case in which an employee was fatally injured while hitching a ride home from work on the company’s truck. American Mut. Liability Ins. Co. v. Curry, 187 Ga. 342, 352 (2) (200 SE 150) (1938). In holding that the employee’s injury was not compensable, we relied on the general principle that “a workman injured going to or coming from the place of work is not ‘in the course of his employment.’” Id. at 353 (2).[9] In that regard, we observed that the employee “had left the premises of his employer and was on his way to his home; that his time was his own, unrestricted and unaffected by his employment to go whence and how he pleased; [and] that he was doing nothing in the furtherance of the company’s business.” Id. at 359 (3). We did not cite Farr to support our analysis, given that Farr had little to say about the “in the course of’ prerequisite (we cited Farr only for the proposition that the case at bar was a workers’ compensation case). See id. at 352-353 (2). A year after Curry, we decided a case in which a traveling employee was fatally injured in a car accident when, during a business trip to Savannah, he took an 18-mile detour (permitted by his employer) to Tybee Island for the personal purpose of eating seafood and seeing the ocean. Skinner, 188 Ga. 823. We stated that the “pivotal question” in the case was whether the employee was “acting in the course of his employment” at the time of the accident. Id. at 829. We held that he was not, and so his injury was not compensable. See id. at 830. Again, we did not cite Farr to support our “in the course of determination.[10] In short, the reasoning in Farr is at odds with the analytical framework that this Court has consistently applied in workers’ compensation cases since the adoption of the Act in 1920. If that analytical framework is generally sound—and we have no reason to doubt that it is—then the reasoning of Farr is unsound. We now turn to the other factors that inform whether we ought to adhere to this unsound decision 8.S 8. matter of stare decisis. (b) Workability Besides its misapplication of the proper framework for deciding questions of compensability, the more specific holding in Farr—that an injury sustained during a scheduled break (in which the employee is free to do as he pleases) does not “arise out of’ employment—has produced a number of apparent inconsistencies in our law, which demonstrate the unworkability of that holding. As we discussed earlier, in Thornton (decided 11 years after Farr), we held that a traveling salesman injured on his way from dinner was entitled to compensation, and we reasoned, in part, that eating a meal was an activity incidental to his employment. See Thornton, 198 Ga. at 790. This holding is not entirely consistent with the rule of Farr, as there is no obvious reason why eating a meal is incidental to the work of a traveling employee but not incidental to the work of an ordinary employee. We did attempt in Thornton to distinguish cases like Farr, stating: [W]e think, in a case involving an injury to a traveling salesman while going to or from, or while eating, a meal, such a strict interpretation of the phrase, “arising out of and in the course of the employment,” as is sometimes made in cases involving employees injured during a regular noon hour, or similar periods, is too limited. In the latter class of cases the injuries are received during a definite period set apart as belonging exclusively to the employees, during which they may go where they wish and do what they please, subject to no orders from their employers and freed from all duty or responsibility with reference to their employment; and their employers do not usually pay the expenses of their meals. Id. at 789-790. But this dicta is unpersuasive.[11] Nothing in Thornton suggests that traveling employees are not free to do as they please during some parts of the day, or that they are always considered to be “in the course of’ employment while away from home. To the contrary, we stated that continuous employment “does not mean that [a traveling employee] can not step aside from his employment for personal reasons, or reasons in no way connected with his employment, just as might an ordinary employee working on a schedule of hours at a fixed location.” Id. at 790. We explained that a traveling employee might “rob a bank” or “attend a dance” or “engage in other activities equally conceivable for his own pleasure and gratification, and ordinarily none of these acts would be beneficial or incidental to his employment and would constitute a stepping aside from the employment.” Id. (emphasis supplied). That is why we held earlier, in Skinner, that a traveling employee’s injury did not occur in the course of his employment where he was injured while taking a detour to the beach for personal pleasure, even though the expenses of this detour were covered by his employer. See Skinner, 188 Ga. at 829. It makes little sense to say that a traveling employee who chooses to eat a meal rather than go to the beach during his free time is covered under the Act, but an ordinary employee who makes the same choice during a break is not. As noted earlier, nothing in the Act suggests that the words “in the course of the employment” means something different for traveling employees as opposed to ordinary employees. See OCGA § 34-9-1 (4). Indeed, Thornton recognized that the interpretation of the Act in Farr was “strict” and “limited” when applied to non-traveling employees, see Thornton, 198 Ga. at 789; such interpretation is at odds with the long-held principle that the Act “is intended to have broad application so as to cover a wide variety of injuries and the pain and suffering incident to such injuries. It is a humanitarian measure which should be liberally construed to effectuate its purpose.” King, 281 Ga. at 854 (citations and punctuation omitted); Lumbermens Mut. Cas. Co. v. Griggs, 190 Ga. 277, 287 (9 SE2d 84) (1940) (“With reference to [the workers' compensation] statute a liberal construction must be given, to effectuate the humane purposes for which it was enacted.”); Sumrell, 30 Ga. App. at 689 (“The [workers'] compensation acts, though in derogation of the common law, being highly remedial in character, should be liberally and broadly construed to effect their beneficient purposes.”). Thornton is not the only decision of this Court that is in tension with Farr with respect to meal breaks. A year after Thornton, in a case dealing with general employer liability, we held that if an employer “operates a cafeteria on its premises, in the immediate vicinity of the work, at which its employees are, expressly or by fair implication, invited to eat, and they accept the invitation by using the facilities provided, the relation of master and servant is not temporarily suspended during the noon hour of such employees.” Holman v. American Automobile Ins. Co., 201 Ga. 454, 459 (1) (39 SE2d 850) (1946) (emphasis supplied). We acknowledged Farr and attempted to distinguish it on the ground that, unlike the case at bar, Farr involved the Workers’ Compensation Act. See id. at 459 (1). But we did not explain why the employer-employee relationship is suspended during lunchtime in one case but not the other. See id.; American Hardware Mut. Ins. Co. v. Burt, 103 Ga. App. 811, 814 (120 SE2d 797) (1961) (the scheduled break rule “is founded on the proposition that during the lunch hour the employee turns aside from his employment for his own purposes, and the master-servant relationship is suspended” (emphasis supplied)). There appears to be no logical or statutory basis for such a distinction. Indeed, we subsequently cited Farr in a general liability case, to support a holding about the scope of an employer’s liability for the actions of his employee. See Chattanooga Publishing Co. v. Fulton, 215 Ga. 880, 882 (3) (114 SE2d 138) (1960) (holding that an employer who owned a truck was not liable for injuries sustained in an accident involving the truck because the employee-driver of the truck was not acting within the scope of his employment when the accident occurred). See also SCI Liquidating Corp., 272 Ga. at 294 (“ [T]he same reasoning used in workers’ compensation cases has been held to be applicable to general liability cases.”). Aside from the cases discussed above, this Court has said virtually nothing about the merits of Farr or the scheduled break rule. Instead, the contours of this rule were developed almost entirely by the Court of Appeals. See Aetna Cas. & Sur. Co. v. Honea, 71 Ga. App. 569, 572-573 (31 SE2d 421) (1944) (relying on Farr to hold that an employee injured during a lunch break on her employer’s premises was not entitled to compensation under the Act); Austin v. Gen. Acc., Fire, and Life Assur. Corp., 56 Ga. App. 481, 483 (193 SE 86) (1937) (where a worker was injured while getting ice during a 15-minute rest break, “[i]t is plain, under the decision of [Farr], that . . . the injury to the claimant did not arise out of her employment”). See also Wilkie v. Travelers Ins. Co., 124 Ga. App. 714, 715 (185 SE2d 783) (1971) (“[I]n ‘lunch break’ and ‘rest break’ cases, both the Supreme Court [in Farr] and this court [in subsequent decisions] have laid down the rule that where the employee is free to use the time as he chooses so that it is personal to him, an injury occurring during this time arises out of his individual pursuit and not out of his employment.”). But even though it was bound by Farr, the Court of Appeals has expressed discomfort with the scheduled break rule and has struggled to apply it consistently. See Wilkie, 124 Ga. App. at 715 (“Were this an open question, we might be disposed to hold that time set aside by an employer as a ‘rest break’ and for the performance of functions necessary for the health and comfort of an employee on the job should be considered as incidental to the employment, and that injuries sustained while engaged in such performance should be deemed compensable as arising out of and in the course of employment.”); Burt, 103 Ga. App. at 814 (attempting to limit the scheduled break rule on the ground that the evidence in Farr “authorized,” but did not mandate, the Board’s finding that “where an employee was injured while on his way to get and eat his lunch during the lunch hour period, on his employer’s premises, that the accident arose in the course of the employment but not out of it”). Beginning in 1950, the Court of Appeals has developed the “ingress and egress rule,” pursuant to which a worker is covered under the Act if she is injured on her employer’s premises while entering or exiting her place of work. See Employers Ins. Co. v. Bass, 81 Ga. App. 306, 306-307 (3) (58 SE2d 516) (1950) (coverage applied to worker who died before commencing work, but while on the employer’s premises, on the way to the job site); Coram, 95 Ga. App. at 624 (coverage applied to employee who was injured while walking to the parking lot at the end of her workday, as “going to and from the parking lot in order to reach and leave her immediate working area was a necessary incident to the claimant’s employment”). See also Smith v. Camarena, 352 Ga. App. 797, 799 (2) (835 SE2d 712) (2019) (the “ingress/egress” rule provides that “the period of employment generally includes a reasonable time for ingress to and egress from the place of work, while on the employer’s premises” (citation and punctuation omitted)). It was not long until the Court of Appeals applied the ingress and egress rule not just to employees entering and exiting their employers’ premises at the beginning and end of their workdays, respectively, but also to workers coming back to work from lunch. See Travelers Ins. Co. v. Smith, 91 Ga. App. 305, 311 (85 SE2d 484) (1954) (coverage applied to employee who was returning to his place of work after a lunch break). See also Chandler v. Gen. Acc. Fire & Life Assur. Corp., 101 Ga. App. 597, 599 (114 SE2d 438) (1960) (“[T]he claimant, having reached the employer’s premises upon her return from supper, was entitled to a reasonable time for ingress to her place of work and . . . an accident occurring during such time shall be construed as arising out of and in the course of her employment.”). That rule was further extended, quite logically, to employees who were leaving their workplace to go to lunch. See Rockwell v. Lockheed Martin Corp., 248 Ga. App. 73, 73 (545 SE2d 121) (2001) (exclusive workers’ compensation remedy applied to employee who was injured on her way to the employer’s parking lot for her scheduled lunch break). As it was developing the ingress and egress rule, the Court of Appeals also expressly distinguished between injuries occurring during “scheduled” lunch or rest breaks, which it held were not compensable under Farr, and “unscheduled” lunch or rest breaks, which it held could be compensable. See Edwards, 130 Ga. App. at 24 (2) (the scheduled break exception “will not be extended to unscheduled breaks under the facts here when the employee finds it necessary to go to the rest room, even though he is permitted to do so without obtaining permission from his supervisor, for the reason that the time is not released to him as free time during which he may do as he will and it cannot be construed as an altogether personal pursuit, as is the case during scheduled breaks”). See also Miles v. Brown Transp. Corp., 163 Ga. App. 563, 563 (294 SE2d 734) (1982) (exclusive workers’ compensation remedy applied to employee who was injured during her lunch break while on her employer’s premises because the lunch break was unscheduled, and, although she was free from duties during this particular break, she often performed job-related duties during other lunch breaks).[12] The result was that, within a few decades of the Farr decision, the law concerning injuries occurring during rest breaks or lunch breaks became confusing at best, if not altogether incoherent. An employee preparing to eat lunch on her employer’s premises was in an employer-employee relationship for purposes of general liability, but not under the Act. See Holman, 201 Ga. at 459 (1); Burt, 103 Ga. App. at 814. Yet even under the Act, that employee might have been acting “in the course of’ her employment, see Farr, 180 Ga. at 270271, but at the same time, that employee was engaged wholly in her personal affairs, and so any injury suffered by the employee would not “arise out of’ employment, see id., unless, of course, that employee was on a business trip, in which case the injury probably would arise out of employment, see Thornton, 198 Ga. at 795. Further, if the employer had not specifically scheduled the lunch break (but allowed the employee to have lunch at any time during the day), then the employee probably would be covered under the Act for lunchtime injuries, but if the employer designated a particular hour for lunch, then she probably would not be covered. See Edwards, 130 Ga. App. at 24 (2). And, until the Frett decision below, if the employee slipped and fell while exiting the break room to go back to work, she probably would be covered, but if she slipped on the same spot while exiting the breakroom to use the restroom before going to work, coverage likely would not apply. See Smith, 91 Ga. App. at 311.[13] The mishmash of arbitrary rules and exceptions described above shows that the scheduled break rule Farr created is not particularly workable. It is not based on sound reasoning and defies reasonable expectations about ordinary working conditions. It has led to inconsistencies in this Court’s case law and has caused the Court of Appeals to try to circumvent the rule through arbitrary and logically suspect distinctions. To be sure, the ingress and egress rule appears to be a sound principle, and the Court of Appeals acted quite logically in extending it to lunch breaks. Yet, as the Court of Appeals observed in this case, the “current case law regarding the intersection of the ingress and egress rule with the scheduled break rule creates anomalous and arbitrary results.” Frett, 348 Ga. App at 35. Because Farr has proven to work poorly, if at all, the workability factor weighs in favor of overruling it. See State v. Jackson, 287 Ga. 646, 658 (4) (697 SE2d 757) (2010) (workability factor did not weigh against overruling precedent where this Court and the Court of Appeals have been “unable or unwilling” to apply the precedent consistently). (c) Age of the Precedent and Reliance upon It Farr is more than 85 years old. Notwithstanding its age, however, it would be hard to assert with a straight face that Farr has become deeply entrenched in our jurisprudence. As we explained earlier, this Court has never cited Farr for its implicit holding that the “arising out of’ employment prerequisite is about something other than causation, we have never cited Farr for its specific holding that injuries sustained during a break do not “aris[e] out of’ employment, and we have not cited Farr for any proposition at all in the past 60 years. Although the Court of Appeals occasionally has relied on Farr to reject a claim for compensation under the Act, it more frequently and more recently has devised rules and exceptions that have effectively limited the precedential effect of Farr narrowly to its facts. Until the decision of the Court of Appeals in Frett below, jurisprudential reliance on Farr was limited at best. Moreover, we cannot say that Farr has created any other reliance interests that would be significantly impaired were we to overrule it. After all, at least until the Frett decision below, no employer or employee could tell for certain whether an injury occurring in the employer’s break room or during lunch hour would be covered under the Act—it all depended on the particular facts of each individual case. And we have no reason to believe that our overruling of Farr will detrimentally affect anyone’s course of conduct. Farr does not, for instance, create an incentive for employers to relax safety standards on their premises. See, e.g., CSX Transp., Inc. v. Williams, 278 Ga. 888, 889 (608 SE2d 208) (2005) (“Under Georgia statutory and common law, an employer owes a duty to his employee to furnish a reasonably safe place to work and to exercise ordinary care and diligence to keep it safe.” (Citation and punctuation omitted)). In this light, we can identify no contractual, property, or other substantive right that would be impaired by our overruling of Farr. Reliance is no reason to preserve Farr. See Lejeune v. McLaughlin, 296 Ga. 291, 298 (2) (766 SE2d 803) (2014) (“[W]hen the courts speak of reliance interests in the context of stare decisis, they refer to contract interests, property rights, and other substantive rights.”). See also Olevik, 302 Ga. at 245 (2) (c) (iv) (“Substantial reliance interests are an important consideration for precedents involving contract and property rights, where parties may have acted in conformance with existing legal rules in order to conduct transactions.” (Citation and punctuation omitted)). Consequently, only the age of Farr cuts in favor of its retention. But when balanced against the other factors, its mature age is not reason enough to preserve it. See State v. Hudson, 293 Ga. 656, 661 (748 SE2d 910) (2013) (“While [prior precedent] has been on the books for some time, we find this alone to be an insufficient rationale for adhering to it, particularly when weighed against the countervailing factors.”).[14] In sum, the stare decisis factors weigh in favor of overruling Farr,[15] and we do so today.[16] There is thus no impediment to our determination, as discussed above, that Frett’s injury occurred “in the course of her employment and arose “out of her employment under the Act. We thus reverse the decision of the Court of Appeals and remand this case for further proceedings consistent with this opinion. Judgment reversed. All the Justices concur, except Peterson, J., who dissents. Bethel, J., not participating, and Ellington and McMillian, JJ., disqualified. Peterson, J., dissenting. I agree with much of what is said in the majority opinion. I agree that Ocean Accident & Guarantee Corp. v. Farr, 180 Ga. 266 (178 SE 728) (1935), was probably wrongly decided. I agree that unless overruled, Farr controls the outcome in this case. And I agree that the result of the majority’s overruling of Farr is probably an improvement in the law. But right or wrong, Farr has been the law for 85 years. In my view, stare decisis counsels against overruling such a venerable statutory precedent without an unusually compelling reason. And I have grown increasingly troubled by the ease with which our Court overcomes stare decisis in statutory interpretation cases, a category of cases in which we claim that stare decisis applies with special force. Because I see no unusually compelling reason to overrule the 85-year-old statutory precedent in question here, I respectfully dissent. It is well-settled that stare decisis applies with the least force to constitutional precedents. See Ga. Dep’t of Nat. Res. v. Ctr. for a Sustainable Coast, Inc., 294 Ga. 593, 601 (2) (755 SE2d 184) (2014) (quoting Smith v. Baptiste, 287 Ga. 23, 30 (1) (694 SE2d 83) (2010) (Nahmias, J., concurring specially)) (“The doctrine of stare decisis is always important, but it is less compelling when, as in this case, the issue is the meaning of a constitutional provision.”) We have explained our reason for this approach: “it is much harder for the democratic process to correct or alter our interpretation of the Constitution than our interpretation of a statute or regulation.” Id. But even in constitutional cases, “[t]his doesn’t mean that we disregard stare decisis altogether, though; what it actually means is that the first stare decisis factor (soundness of reasoning) becomes even more critical. The more wrong a prior precedent got the Constitution, the less room there is for the other factors to preserve it.” Olevik v. State, 302 Ga. 228, 245 (2) (c) (iv) (806 SE2d 505) (2017). If stare decisis applies with little force to constitutional precedents because it is very difficult for the “People and their elected representatives [to] overrule [those] precedents, if they think them incorrect[,]“ Lejeune v. McLaughlin, 296 Ga. 291, 298 (2) (766 SE2d 803) (2014), it necessarily follows that stare decisis must apply with greater force to statutory precedents, which the People’s elected representatives can overrule more easily.[17] And, indeed, we have said so quite clearly. See, e.g., Radioshack Corp. v. Cascade Crossing II, 282 Ga. 841, 843 (653 SE2d 680) (2007) (“Even those who regard ‘stare decisis’ with something less than enthusiasm recognize that the principle has even greater weight where the precedent relates to interpretation of a statute.”) (citation and punctuation omitted). So has the United States Supreme Court. See, e.g., Ill. Brick Co. v. Illinois, 431 U.S. 720, 736 (97 SCt 2061, 52 LE2d 707) (1977) (“[W]e must bear in mind that considerations of stare decisis weigh heavily in the area of statutory construction, where Congress is free to change this Court’s interpretation of its legislation.”). But we are better at reciting this principle than following it. Ten years ago, we drew the four factors of our now-familiar stare decisis test from the United States Supreme Court’s decision in Montejo v. Louisiana, 556 U.S. 778 (129 SCt 2079, 173 LE2d 955) (2009): the soundness of the precedent’s reasoning, the age of the precedent, whether the precedent gave rise to reliance interests, and the workability of the precedent. See State v. Jackson, 287 Ga. 646, 658-60 (5) (697 SE2d 757) (2010). Since Jackson, this Court has overruled precedent every single time that a majority opinion has explicitly considered whether stare decisis warrants retaining a statutory precedent that the majority doubted was correct. See SRM Grp., Inc. v. Travelers Prop. Cas. Co. of Am., ___ Ga. ___ , ___ (3) (841 SE2d 729, 734-35) (2020) (overruling 11-year-old statutory precedent); State v. Burns, 306 Ga. 117, 123-24 (2) (829 SE2d 367) (2019) (overruling 30-year-old statutory precedent); Nalls v. State, 304 Ga. 168, 179-80 (3) (b) (815 SE2d 38) (2018) (overruling 18-year- old statutory precedent); City of Cumming v. Flowers, 300 Ga. 820, 825-34 (4)-(6) (797 SE2d 846) (2017) (overruling 21-year-old statutory precedent); Southall v. State, 300 Ga. 462, 467-68 (1) (796 SE2d 261) (2017) (overruling 45-year-old statutory precedent); State v. Springer, 297 Ga. 376, 382-83 (2) (774 SE2d 106) (2015) (overruling 12-year-old statutory precedent); id. at 383-85 (Benham, J., dissenting) (dissenting on stare decisis grounds and noting that “[t]he doctrine of stare decisis seems to be less viable year by year”) (quoting Grissom v. Gleason, 262 Ga. 374, 378 (418 SE2d 27) (1992) (Benham, J., concurring specially); Woodard v. State, 296 Ga. 803, 810-14 (3) (b) (771 SE2d 362) (2015) (overruling 24-year-old statutory precedent); Shekhawat v. Jones, 293 Ga. 468, 469-74 (1) (746 SE2d 89) (2013) (overruling 16-year-old statutory precedent); Jackson, 287 Ga. at 658-60 (5) (overruling 29-year-old statutory precedent).[18] Even more starkly, since Jackson, opinions in 45 cases decided by this Court have used the phrase “stare decisis.” of those 45, many have been constitutional cases (to which stare decisis applies with less force), and many have been concurrences and dissents. But in none of those 45 cases — regardless of subject matter — has a majority invoked stare decisis to retain a precedent of any kind while expressing doubts as to whether it had been correctly decided.[19] We overruled many of these statutory precedents with little reason more than those precedents were wrongly decided. But if stare decisis is to mean something, we need more than that to overrule a statutory precedent. Indeed, stare decisis does not even begin to apply until we doubt the correctness of a previous precedent. See Elliott v. State, 305 Ga. 179, 209 (III) (C) (ii) n.21 (824 SE2d 265) (2019) (“Because we conclude that Olevik was correctly decided, it is unnecessary to consider whether we should retain that decision under the doctrine of stare decisis. See Kimble v. Marvel Entertainment, LLC, 576 U. S. 446, 455 (135 SCt 2401, 192 LE2d 463) (2015) (“stare decisis has consequence only to the extent it sustains incorrect decisions; correct judgments have no need for that principle to prop them up”)). A precedent’s age is also an important consideration, and in my view, especially so when statutory precedents are considered. The longer our interpretation of a statute persists without legislative amendment, the more embedded our precedent becomes in the legal landscape, and the greater the force that stare decisis should have.[20] Farr is both a statutory precedent and an old one. If either of those factors were different, I could more easily join the majority. But they are not, and none of the other reasons the majority offers for overruling Farr seem to me sufficient justification for changing our mind after nearly nine decades. In short, “[i]f there is any inconsistency or illogic in [Farr], it is an inconsistency and illogic of long standing that is to be remedied by the [General Assembly] and not by this Court.” Flood v. Kuhn, 407 U.S. 258, 284 (92 SCt 2099, 32 LE2d 728) (1972). I respectfully dissent.
