Jeanette Vallandigham, et al., Petitioners, v. Clover Park School District No. 400, Respondent, 154 Wash. 2d 16

¶1 This case involves a difficult situation often faced by schools and teachers who serve students with severe disabilities. Two employees of the Clover Park School District have sued the district to recover for injuries caused by R.M., a severely disabled special education student. While Washington’s Industrial Insurance Act (IIA), Title 51 RCW, generally precludes employee recovery outside of the workers’ compensation scheme established by that statute, RCW 51.04.010, the legislature has created a limited exception, permitting tort recovery if an employer deliberately intended to injure its employee.

RCW 51.24.020. This court has interpreted this exception to mean that the employer acts with deliberate intention when it willfully disregards actual knowledge that employee injury is certain to occur. Birklid v. Boeing Co., 127 Wn.2d 853, 865, 904 P.2d 278. (1995). Bridge, J. ¶2 The trial court here granted

summary judgment in favor of the school district, concluding that neither actual knowledge of certain injury nor willful disregard existed in this case. Verbatim Report of Proceedings (VRP) at 40. The Court of Appeals disagreed in part, holding that the school district did have actual knowledge that injury was certain to occur. However, the Court of Appeals ultimately concluded that the school district did not willfully disregard that knowledge. Vallandigham v. Clover Park Sch. Dist. No. 400, 119 Wn.

App. 95, 107, 109, 79 P.3d 18 (2003). ¶3 We agree with the trial court and hold that the school district in this case did not have actual knowledge that employee injury was certain to occur. We reiterate that in order for an employer to act with deliberate intent, injury must be certain; substantial certainty is not enough.

Given the unpredictability that is so often inherent in the behavior of a student with severe disabilities, the school district here could not have been certain that its strategies for modifying R.M.’s behavior would fail such that R.M. would continue to injure school staff. ¶4 In addition, we recognize that a conflict exists between Divisions One and Two of the Court of Appeals as to how to evaluate whether an employer has acted with willful disregard. Compare Vallandigham, 119 Wn. App. at 108 with Stenger v. Stanwood Sch. Dist., 95 Wn. App. 802, 977 P.2d 660 (1999).

We note that this court has held that negligence, even gross negligence, cannot satisfy the deliberate intention exception to the IIA. Birklid, 127 Wn.2d at 860. An inquiry into the reasonableness or effectiveness of an employer’s remedial measures sounds in negligence, and we reject any notion that a reasonableness or negligence standard can be applied to determine whether an employer has acted with willful disregard. I Statement of Facts ¶5 R.M. was born in 1986. He has been diagnosed with epilepsy, mental retardation, and autism, which cause severe impairment of his verbal communication skills.

He has the cognitive ability of a two- to three-year-old. In 1997, R.M. began attending Woodbrook Middle School in the Clover Park School District where he was placed in the self-contained life skills classroom with approximately six other students. ¶6 During the 1997-98 and 1998-99 school years, R.M. was in the life skills program where Mark Brinkhaus was the teacher and Melinda Clarke was an aide. The record reflects 14 days in the 1998-99 school year on which R.M. injured staff or students, sometimes more than once. The majority of the injuries were scratches and slaps.

A chart provided by the plaintiffs reflects that R.M. caused a cluster of injuries in March 1999. Otherwise, R.M.’s injury causing outbursts were infrequent and irregular. See Clerk’s Papers (CP) at 380 (seven days of recorded injuries between March 1, 1999 and March 30, 1999; seven other days of recorded injuries interspersed throughout the school year). ¶7 In the fall of the 1999-2000 school year, Jeanette Vallandigham became R.M.’s teacher. Clarke, the classroom aide, reported that September went relatively smoothly for R.M.

Around late September and early October 1999, R.M.’s behavior became increasingly aggressive and he began to inflict injuries on staff and students. Vallandigham and other school personnel believed that this increased aggression was caused by the fact that R.M. was no longer taking one of his medications.

Between September 14, 1999 and October 25, 1999, he assaulted or injured students or staff approximately 18 times by scratching, hitting, pulling hair, biting, pinching, head butting, and grabbing glasses. ¶8 On October 21, Vallandigham sent an e-mail asking for an Individualized Education Program (IEP) team meeting about R.M. and requesting a one-on-one aide for him. As a result, on October 25 the school district implemented a plan: (1) the school nurse would contact R.M.’s doctor to discuss the medication situation;

(2) Terry Howard, Woodbrook’s special education department head, would conduct a functional behavioral analysis; and

(3) Vallandigham would continue to document R.M.’s behaviors. When these steps were completed, an official IEP meeting would be convened to discuss changes to R.M.’s program. In her deposition, Vallandigham testified that she believed that these actions were appropriate. ¶9 On the next day, October 26, 1999, when Vallandigham intervened to prevent R.M. from scratching another student, R.M. turned and head-butted Vallandigham. She fell backward, hit her head on a counter, and lost consciousness. Vallandigham filed a workers’ compensation claim for this injury.

Then, on October 27, 1999, R.M. bit Melinda Clarke on her breast, leaving a mark. She too filed a workers’ compensation claim for the injury. ¶10 Both Terry Howard and Mark Brinkhaus completed behavior evaluation reports on R.M. on October 29, 1999 and November 1, 1999. Brinkhaus reported that R.M. physically hurt other students or teachers more than once a week, while Howard reported that R.M. hurt other students or teachers “daily at various times.” CP at 221. In his deposition, Howard opined that if nothing were done to stop R.M.’s behavior, he believed that it would continue.

A separate functional behavioral assessment form that was completed on October 29, 1999 indicated that R.M. would frequently scratch other students. It also noted that when restrained, R.M. would “kick, pull hair, bite, and head butt to get away.” CP at 184. These problem behaviors occurred daily and began when R.M. went off his medications. ¶11 Between October 26 and November 12, 1999, R.M. caused a total of 20 more injuries, mostly scratches.

Virginia Alonzo, the school district special education officer, stated that at the end of October, the district was aware that it was a “probability,” but not “certain,” that R.M would have future outbursts that injured staff and students. CP at 596. ¶12 As planned, on November 12, 1999 the school district conducted an IEP meeting, which Vallandigham attended. Various placement options were considered, but the IEP team concluded that the self-contained special education classroom was still the most appropriate placement for R.M. at that time.

Vallandigham’s requests for a male one-on-one aide for R.M. and an isolation room, to be available at all times for the purpose of containing R.M. when needed, were incorporated into the program. This IEP also contained a detailed behavior plan for dealing with R.M.’s behaviors on a day-to-day basis. The behavior plan included a continuum of responses ranging from redirection, to time out in the isolation room, to suspension for the rest of the day. If these strategies were not successful, the team concluded R.M.’s

schedule would have to be reduced to half days. The IEP also contained an aversive therapy plan. Vallandigham signed the IEP and testified in her deposition that she had agreed with these strategies. However, she also noted that she was told by school officials that R.M.’s placement could not be changed until the school tried a one-on-one aide and a half day schedule. f 13 Around the time of the November IEP meeting, the district began its process for hiring a one-on-one aide to work exclusively with R.M.

In the meantime, Gabriele Williamson, a paraeducator from another classroom, was transferred to assist with R.M. beginning in early November. On November 29, an outside consultant was hired to perform another functional behavioral assessment on R.M. The consultant believed that changes could be made in Vallandigham’s classroom that would improve R.M.’s behavior. He suggested that R.M.’s day be more structured and that his activities be more directly targeted toward his IEP goals and objectives.

The consultant also recommended changes in the use of R.M.’s isolation room such that the room be used only for short periods of time and only as the result of clearly identified behaviors. ¶14 In December 1999 a full time one-on-one aide, Wanda Dalton, was hired to work with R.M. within the life skills classroom. Even so, R.M.’s behavior problems persisted. Between November 15 and December 31, 1999, approximately 38 injuries to students and staff were reported.

Again, most were scratches, but the injuries included bites, slaps, and muscle strains that occurred when staff attempted to restrain R.M. ¶ 15 In January 2000, R.M.’s IEP was again amended by Vallandigham and the rest of the IEP team, adding more specific structure to his daily routine. The IEP amendment also included a plan, at R.M.’s parents’ request, to call 911 and have R.M. transported to Madigan Hospital if the school was unable to calm him down after a significant period of aggression. Still, between January 1 and March 3, 2000, 15 injuries were reported.

Again, most were scratches, but Wanda Dalton, R.M.’s one-on-one aide, was “beat up badly” on March 2, 2000. CP at 412, 559-60. The last injury to one of the plaintiffs occurred when R.M. scratched Clarke on the arm and wrist on February 9, 2000. ¶16 In March 2000, the Department of Social and Health Services arranged for R.M. to be sent to a residential placement at the Francis Morgan Center for one month. While at Francis Morgan, R.M. did not have any serious behavior problems.

Vallandigham visited the classroom at Francis Morgan and a doctor from Francis Morgan visited Woodbrook to observe the classroom and meet with staff to discuss methods for directing R.M.’s behavior. Clover Park also consulted with and received training from a specialist from the Tacoma Child Study and Treatment Center. These experts believed that with appropriate interventions, R.M.’s behavior in Vallandigham’s class could be controlled. ¶17 In April 2000, R.M. returned from Francis Morgan and his IEP team met again to amend his program.

Alonzo, Howard, Dalton, and Clarke attended the meeting and Vallandigham provided written input. This time, R.M. attended half days at Woodbrook and was instructed in the hallway by Wanda Dalton instead of within the self-contained classroom. R.M. was schooled at home by Wanda Dalton in the afternoons. In the meantime, Alonzo explored alternative placement options for R.M. See CP at 50. The Tacoma School District and the Franklin Pierce School District declined to accept him because they did not have adequate space. Alonzo also explored options at four other alternative placements.

None of these potential placements would accept R.M. at that time. Vallandigham also reported that she was unable to locate an alternative placement for R.M. after contacting three additional programs. Full time home-based instruction was not an option because R.M.’s mother suffered from mental illness and his father was away from the home for significant periods of time. ¶18 The plaintiffs allege that over the course of the entire 1999-2000 school year, R.M. injured other staff and students about 96 times. Most of these injuries were scratches.

The record contains a total of seven workers’ compensation claims for staff injuries caused by R.M. during the 1999-2000 school year. These included the October 26 and 27 injuries to Vallandigham and Clarke as well as additional injuries to Clarke in late November and early December, injuries to Howard and Williamson in December, an injury to Clarke in February, and injuries to Dalton in early March. ¶19 Overall during the course of the 1999-2000 school year, the school district took the following steps to try to modify R.M.’s behavior: 1. The school nurse discussed medication with R.M.’s doctor; 2.

School personnel performed functional behavioral analyses and assessments to determine the cause of R.M.’s behavior; 3. At least three IEP meetings were conducted to adjust R.M.’s program. One contained a detailed behavior plan involving a continuum of planned responses to R.M.’s behavior. Another directed school staff to call 911 to transport R.M. to Madigan Hospital if necessary; 4. First a paraeducator and then a permanent one-on-one aide were assigned to work directly with R.M.; 5. Several outside experts were consulted; 6. A separate area was created for use as an isolation or time-out space; 7.

The district offered restraint training and issued walkietalkies to selected staff; 8. R.M. was sent home from school on 17 occasions; 9. The school reduced R.M.’s school time to half-days and removed him from Vallandigham’s classroom; 10. The school district considered alternative placements for R.M. but declined these alternatives because they were inappropriate or unwilling to take R.M. See also Vallandigham, 119 Wn.

App. at 108-09. ¶20 Vallandigham and Clarke sued the Clover Park School District, arguing that they should be allowed to recover in tort for their workplace injuries based on the “deliberate intention” exception to the IIA. The school district moved for

summary judgment. The trial court concluded that there was no genuine issue of material fact and that as a matter of law, the plaintiffs had failed to establish that their injuries were the result of the district’s deliberate intention to injure them because the school district did not willfully disregard actual knowledge that injury was certain to occur. CP at 675; VRP at 40; see also Birklid, 127 Wn.2d at 865. ¶21 On appeal, Division Two of the Court of Appeals affirmed the trial court’s award of

summary judgment. Vallandigham, 119 Wn. App. at 98. However, the Court of Appeals disagreed with the trial court’s reasoning in part and held instead that the plaintiffs had presented enough evidence for a jury to conclude that the district had actual knowledge that injury was certain to occur. Id. at 107. However, the Court of Appeals agreed that a reasonable jury could not conclude that the district willfully disregarded that knowledge. Id. at 109. In doing so, Division Two criticized Stenger, 95 Wn. App. 802, in which Division One refused to grant a school district’s motion for

summary judgment under similar circumstances. Vallandigham, 119 Wn. App. at 108. ¶22 The plaintiffs filed a petition for review, arguing that this court should grant review to resolve the disagreement between Divisions One and Two of the Court of Appeals. We granted the petition at 151 Wn.2d 1031 (2004). The plaintiffs now contend that the school district willfully disregarded actual knowledge that they would certainly be injured by R.M and thus, this court should reverse

summary judgment in this case. The school district urges us to affirm the ultimate decision of the Court of Appeals but asks us to hold that the district did not have actual knowledge that injury was certain to occur. II Analysis ¶ 23 Standard of Review: On review of

summary judgment, we engage in the same inquiry as the trial court. Atherton Condo. Apartment-Owners Ass’n Bd. of Dirs. v. Blume Dev. Co., 115 Wn.2d 506, 515-16, 799 P.2d 250 (1990).

Summary judgment is affirmed if “the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law.” CR 56(c). All facts are considered in the light most favorable to the nonmoving party, Atherton, 115 Wn.2d at 516, and

summary judgment is granted only if, from all of the evidence, reasonable persons could reach but one conclusion. Wilson v. Steinbach, 98 Wn.2d 434, 437, 656 P.2d 1030 (1982). The burden is on the moving party to show that there is no genuine issue as to any material fact. Atherton, 115 Wn.2d at 516. “If the moving party satisfies its burden, the nonmoving party must present evidence that demonstrates that material facts are in dispute.” Id. If the nonmoving party fails to do so, then

summary judgment is proper. Id. ¶24 The IIA and its Deliberate Intention Exception: In 1911, as the result of a “grand compromise,” the IIA granted Washington employers immunity from lawsuits arising from workplace injuries. Birklid, 127 Wn.2d at 859. In exchange, the IIA created an exclusive workers’ compensation system that provided swift and certain recovery for injured employees, regardless of fault. Id.; RCW 51.04.010.

However, the legislature created a limited exception when an employer intentionally injures an employee: If injury results to a worker from the deliberate intention of his or her employer to produce such injury, the worker or beneficiary of the worker shall have the privilege to take under this title and also have cause of action against the employer as if this title had not been enacted, for any damages in excess of compensation and benefits paid or payable under this title. RCW 51.24.020 (emphasis added).

Washington courts have consistently interpreted RCW 51.24.020 narrowly, holding that mere negligence, even gross negligence, does not rise to the level of deliberate intention. Birklid, 127 Wn.2d at 860-61 (citing Delthony v. Standard Furniture Co., 119 Wash. 298, 205 P. 379 (1922) and listing subsequent cases). Even failure to observe safety laws or procedures does not constitute specific intent to injure, nor does

an act that had only substantial certainty of producing injury. Id. at 860. Before 1995, Washington courts interpreted the “deliberate intention” exception to apply only where an employer or its agent physically assaulted an employee. Birklid, 127 Wn.2d at 861-62 (citing Perry v. Beverage, 121 Wash. 652, 209 P. 1102, 214 P. 146 (1922); Mason v. Kenyon Zero Storage, 71 Wn. App. 5, 856 P.2d 410 (1993)). ¶25 In 1995, in Birklid, this court interpreted the deliberate intention exception to reach beyond intentional physical assaults. Id. at 863.

In that case, a Boeing general supervisor reported to management that fumes from phenol-formaldehyde resin were making employees sick. Id. at 856. He anticipated that the symptoms would increase as temperatures began to rise and as production increased. Id. The supervisor requested improved ventilation, but Boeing management denied the request, explaining that the problem did not warrant the expenditure of funds. Id. As a result, workers became ill on the job and Boeing knew the symptoms were the result of exposure to the chemical.

Id. ¶26 The Birklid court held that the phrase “deliberate intention” in RCW 51.24.020 means (1) “the employer had actual knowledge that an injury was certain to occur” and (2) the employer “willfully disregarded that knowledge.” Id. at 865. However, the court announced that the first prong of the Birklid test would be met in only very limited circumstances. Significantly, the Birklid court rejected the reasoning adopted in Michigan, North Carolina, and other states under which a cause of action would be permitted if the employer knew that injury was “substantially certain” to occur.

Id. at 864-65 (quoting Beauchamp v. Dow Chem. Co., 427 Mich. 1, 21-22, 398 N.W.2d 882 (1986)). In addition, the Birklid court rejected Oregon’s test, which focused on “whether the employer had an opportunity consciously to weigh the consequences of its act and knew that someone, not necessarily the plaintiff specifically, would be injured.” Birklid, 127 Wn.2d at 865 (citing Lusk v. Monaco Motor Homes, Inc., 97 Or. App. 182, 775 P.2d 891 (1989)). Instead, the Birklid court emphasized that it was “mindful of the narrow

interpretation Washington courts [had] historically given to RCW 51.24.020, and of the appropriate deference four generations of Washington judges have shown to the legislative intent embodied in RCW 51.04.010.” Id. Disregard of a risk of injury is not sufficient to meet the first Birklid prong; certainty of actual harm must be known and ignored. ¶27 The Birklid court found that the company knew in advance that its workers would become ill but put the chemical into production anyway. Id. at 863. The facts were sufficient for a jury to find that Boeing had actual knowledge that injury was certain to occur.

See id. at 865-66. ¶28 With regard to the second prong of its new test, the Birklid court emphasized that the facts in that case went “beyond gross negligence.” Id. at 863. Notably, Boeing had refused to expend funds to protect its workers from toxic substances that were certain to cause injury if used without ventilation. Id. at 856. Because both prongs of the newly announced test were met, the Birklid court held that the plaintiffs had presented sufficient evidence to survive

summary judgment because a jury could find that Boeing deliberately intended to injure its employees under the IIA. See id. at 865-66. f 29 In this case, we apply the Birklid test in determining whether the Clover Park School District willfully disregarded actual knowledge of certain employee injury. We recognize that to some extent the two prongs of the Birklid test are not independent of each other.

Where an employer has taken remedial steps to try to alleviate the risk of further injury to its employees, those actions are relevant both to the question of willful disregard and to the question of whether the employer was certain that injury would continue, in spite of its efforts. We address the conflict between the trial court and the Court of Appeals in this case with regard to whether the school district had actual knowledge that employee injury was certain to occur at the hands of R.M.

We also take this opportunity to address the conflict between Divisions One and Two of the Court of Appeals as to what can constitute willful disregard. ¶30 Actual Knowledge that Injury Is Certain To Occur: Since Birklid, this court has twice concluded that employers did not meet the first prong of the Birklid test. In Minton v. Ralston Purina Co., 146 Wn.2d 385, 388, 391, 47 P.3d 556 (2002) we determined that the employee injury was accidental and found no intentional tort where a condensation tank exploded, injuring the plaintiff. In Folsom v.

Burger King, 135 Wn.2d 658, 661, 667, 958 P.2d 301 (1998), we held that the employer had no actual knowledge that injury was certain to occur when a former employee murdered two other employees during the course of a robbery. The Folsom court noted that “[e]ven

an act that has substantial certainty of producing injury does not rise to the level of specific intent to cause injury.” Id. at 665 (emphasis added). The court explained that even if the employer knew that its former employee had a criminal history, that the back door to the restaurant did not have a peephole and did not lock properly, that cash was kept in the restaurant, and that there was no active security system, the plaintiffs could not show that by operating the business in this manner, the employer knew its employees would be killed. Id. at 667. Thus, the employer maintained its immunity under the IIA.

Id. ¶31 Similarly, the Court of Appeals has found in six cases that the employer did not have actual knowledge that employees were certain to be injured. See Howland v. Grout, 123 Wn. App. 6, 11-12, 94 P.3d 332 (2004) (employee injured when she tripped and fell); Anica v. Wal-Mart Stores, Inc., 120 Wn. App. 481, 486, 494-95, 84 P.3d 1231 (2004) (employee suffered a strain injury when lifting); Schuchman v. Hoehn, 119 Wn. App. 61, 65, 72, 79 P.3d 6 (2003) (14-year-old employee injured by ice auger); Judy v. Hanford Envt’l Health Found., 106 Wn.

App. 26, 30, 33, 22 P.3d 810 (2001) (employee injured after employer’s physician failed to inform employee that the demands of her job exceeded her grip strength and lifting capacity); Henson v. Crisp, 88 Wn. App. 957, 959, 961, 946 P.2d 1252 (1997) (employee suffered emotional trauma when an assistant manager pointed and fired a toy gun at her); Goad v. Hambridge, 85 Wn. App. 98, 100, 104, 931 P.2d 200 (1997) (employee injured when he reached his hand inside a planer whose safety mechanism had been disabled). ¶32 The Court of Appeals has found that the first prong of the Birklid test was met only three times.

Two of those cases involved facts that were similar to Birklid in that the employee injuries were caused by exposure to chemicals. In 1996, in Baker v. Schatz, 80 Wn. App. 775, 912 P.2d 501 (1996), employees sued when their employer repeatedly exposed them to toxic chemicals. Id. at 778. Like the employees in Birklid, the Baker employees repeatedly complained of illness and injury and the employer knew that workplace chemicals were causing the illnesses but refused to take steps to protect the workers. Id. at 778-79.

Significantly, the employer frequently reassured the workers that the chemicals were safe, despite the employer’s knowledge to the contrary. Id. at 784. Therefore, the Court of Appeals held that the employer had actual knowledge that the employees were certain to be injured. Id. ¶33 Similarly, in 2001, in Hope v. Larry’s Markets, 108 Wn. App. 185, 29 P.3d 1268 (2001), the Court of Appeals found that an employer had actual knowledge that injury was certain to occur when an employee repeatedly complained that cleaning chemicals were causing bums and a severe rash. Id. at 189-90,193-94.

The Hope court noted the unique role that chemicals play in causing workplace injuries and concluded that in cases involving chemical exposure, repeated, continuous injury and the observation of the injury by the employer can satisfy the first prong of the Birklid test. Id. at 193-94. Because the plaintiff presented evidence that the employer knew its employees were being continuously injured by chemicals, she had produced evidence sufficient to establish that the employer had actual knowledge that injury was certain to occur.

Id. at 194. f 34 Stenger is the only case in which a Washington court has found the first prong of the Birklid test was met absent injuries caused by chemical exposure. In that case, two school employees sued their school district to recover for injuries they suffered at the hands of a special education student, Jason Springstead. 95 Wn. App. at 803. Like R.M., Jason was diagnosed as “multihandicapped” with severe behavioral disabilities. Id. at 804-05.

The plaintiffs produced testimony that Jason had caused between 1,316 and 1,347 injuries to school staff over the course of more than four school years, at times inflicting injury on almost a daily basis. See id. at 806, 811, 812. The injuries included “scratches, gouges, bites, upper body strain; scalp, breast, neck, back, shoulder, leg, arm, wrist, hand, and finger injuries; and bruising.” Id. at 812. Several accident reports and Department of Labor and Industries claims were filed in the years before the plaintiffs’ injuries. Id. at 812-13.

School district administrators testified that they were aware of the staff injuries and believed that despite their precautions, school staff would continue to suffer injuries when working with Jason. Id. at 813. However, the Stenger court did note periods of improvement. For example, district employees reported that Jason’s communication skills, behavior, and ability to socialize with others improved during the 1991-92 school year. See id at 807. Again in November 1994, Jason’s IEP noted that a strict behavior modification system was working and the frequency of his outbursts had decreased dramatically.

Id. at 811. ¶35 Even so, “[g]iven the frequency of Jason’s outbursts, the number of injuries he inflicted, and the claims filed with the District,” the Stenger court held that “a jury could reasonably conclude that the District had actual knowledge that the staff would continue to be injured by Jason in the future.” Id. at 813. The Stenger court reversed the trial court’s grant of

summary judgment, id. at 817, and the Stanwood School District did not file a petition for review at this court. ¶36 Similarly, in this case, the Court of Appeals held that there was enough evidence for a jury to conclude that the school district had actual knowledge of certain employee injury. Vallandigham, 119 Wn. App. at 107. The Court of Appeals asked “whether R.M. had a known propensity to injure, and was this propensity so extreme that, from it, Clover Park certainly knew that another injury was forthcoming.” Id. at 101.

Considering the evidence in the light most favorable to the plaintiffs, the court concluded, based on R.M.’s history of inflicting multiple injuries on students and staff, that a reasonable jury could conclude that the school district had actual knowledge that injury was certain to occur.

Id. at 105-07. ¶37 The school district argues that even if it could have been substantially certain that injuries would continue to occur, given the unpredictability of R.M.’s behavior and the district’s consistent efforts to alter that behavior, a reasonable jury could not conclude that the district had actual knowledge that injury was certain to continue. To the extent that the Stenger case is inconsistent with this argument, the school district encourages us to disapprove the holding in that case.

We agree with the school district and reverse the Court of Appeals with regard to the first prong of the Birklid test. In doing so, we recognize that the first prong of the Birklid test can be met in only very limited circumstances where continued injury is not only substantially certain but certain to occur. ¶38 We cannot overemphasize that the Birklid court considered and rejected both a “substantial certainty” and a “conscious weighing” test. Birklid, 127 Wn.2d at 865.

Specifically, the Birklid court rejected the Michigan Supreme Court’s conclusion that “ ‘[i]f the injury is substantially certain to occur as a consequence of actions the employer intended, the employer is deemed to have intended the injuries as well.’ ” Id. at 864 (quoting Beauchamp, 427 Mich, at 21-22)). Instead, the Birklid court, mindful of Washington’s historically narrow

interpretation of RCW 51.24.020, made it abundantly clear that foreseeability, or even substantial certainty, is not enough to establish deliberate intent to injure an employee. Birklid, 127 Wn.2d at 865. Even an admission that the district recognized that injury would probably occur is not enough to establish knowledge of certain injury. Only actual knowledge that injury is certain to occur will meet this first prong of the Birklid test. Birklid, 127 Wn.2d at 865. ¶39 It is easy to see how the first prong of the Birklid test was met in Birklid, but not here; the facts of this case are easily distinguishable.

In Birklid, impact of the phenol-formaldehyde was predictable, Birklid, 127 Wn.2d at 863, Hope, 108 Wn. App. at 193-94; the behavior of a child with special needs is far from predictable. Countless variables can impact a special education student’s behavior from day to day, including whether or not the student has taken a prescribed medication. See CP at 17, 19. Therefore, the employer in the Birklid case was in a vastly different position than the employer in this case.

While Boeing knew that the phenolformaldehyde fumes would continue to make employees sick absent increased ventilation, the Clover Park School District could not know what R.M.’s behavior would be from day to day. No one could be sure that R.M.’s violent behavior would not cease as quickly as it began. Unlike Boeing, the school district in this case attempted a series of increasingly restrictive strategies for bringing R.M.’s behavior under control. See supra p. 24. Each of these strategies was clearly intended to contain or halt R.M.’s aggressive behavior.

The plaintiffs have not shown that the district could ever have been certain that any of these strategies would fail. ¶40 The plaintiffs seem to narrow their argument to address only the October 26 and October 27 injuries. Pet’r’s Resp. to Amicus Wash. Schools Risk Management Pool at 6. Even if we consider only the events occurring before October 26, the school district still could not have been certain at that point that staff injury would continue. On October 21, Vallandigham called for a meeting. As a result, R.M.’s teachers theorized that a change in his medication was causing his violent behavior.

CP at 17, 19. On October 25, the school nurse was ordered to contact R.M.’s doctor, Vallandigham was ordered to continue to document R.M.’s outbursts, and a functional behavior assessment was requested, all with the goal of bringing the results to an IEP meeting. CP at 17,19. Thus, at the time that the October 26 and 27 injuries took place, the school district was in the process of taking steps to alleviate the risk of injury to its employees.

It did not know at that time that these measures would be ineffective; therefore, the school district could not have been certain that staff injuries would continue to occur. ¶41 In sum, substantial certainty that employee injury would continue is not enough, and here a jury could not conclude that continued injury was certain. Thus, we hold that the first prong of the Birklid test was not met in this case.

To the extent that the Stenger case is inconsistent with this holding, we disapprove that opinion. ¶ 42 Willful Disregard of Certain Injury: Because we hold that the school district did not have actual knowledge that employee injury was certain to occur, the school district could not have willfully disregarded such knowledge. However, we note that when evaluating Birklid’s second prong in this case, the Court of Appeals rejected the reasoning set forth by Division One in the Stenger opinion, creating a disagreement between Divisions One and Two as to what constitutes willful disregard. See Vallandigham, 119 Wn.

App. at 108. ¶43 In Stenger, Division One focused on “whether a jury could conclude that [the district’s] efforts to accommodate Jason in the classroom were inadequate and thus constitute willful disregard under the Birklid rule.” Stenger, 95 Wn. App. at 813 (emphasis added). Following Steiger, the Hope court noted that it could find willful disregard if attempted remedial measures were ineffective. Hope, 108 Wn. App. at 195.

The Court of Appeals in this case took issue with the Stenger court’s reasoning and opined that “[b]y focusing on the efficacy or adequacy of the remedial measures, Stenger impermissibly erodes the requirement of ‘deliberate intent.’ ” Vallandigham, 119 Wn. App. at 108.

Because evaluation of the effectiveness of a remedial measure is merely another way of evaluating its reasonableness, we agree that the Stenger and Hope language, at least to some extent, adopted a negligence standard. ¶44 We note that this court has been abundantly clear that negligence, even gross negligence, cannot satisfy the deliberate intention exception to the IIA. Birklid, 127 Wn.2d at 860. Therefore, we reject any notion that a reasonableness or negligence standard should be applied to determine whether an employer has acted with willful disregard.

We disapprove of the holdings in the Stenger and Hope cases to the extent that they suggest that a finding of willful disregard can be based upon the simple fact that an employer’s remedial efforts were ineffective. Stenger, 95 Wn. App. at 813; Hope, 108 Wn. App. at 195. ¶45 In sum, the school district has met its burden to show that there is no dispute of material facts in this case. The burden shifts to the plaintiffs to raise a genuine issue of material fact.

We conclude that given the inherently unpredictable nature of special education students like R.M., at no point could the school district have been certain that injury to staff would continue. The school district was thus entitled to judgment as a matter of law. III Conclusion ¶46 RCW 51.24.020 provides only a very limited exception to the IIA’s workers’ compensation scheme in circumstances where an employer deliberately intends to injure an employee. Even substantial certainty that employee injury will occur by virtue of an employer’s action (or inaction) is insufficient.

In this case, the plaintiffs have not presented sufficient facts from which a reasonable jury could conclude that the district knew, despite its increasingly restrictive remedial measures, that R.M. would continue to injure its staff members. Therefore,

summary judgment was properly granted, and we affirm the decision of the Court of Appeals. Alexander, C.J.; C. Johnson, Madsen, Owens, and Fairhurst, JJ.; and Ireland, J. Pro Tem., concur. The federal Individuals with Disabilities Education Act (IDEA), 20 U.S.C. §§ 1400-1487, requires that each student with a disability have an IEP, a written program drafted and reviewed according to IDEA requirements. A student’s IEP is reviewed as needed, but at least annually by an IEP team consisting of teachers, other school personnel, and parents. 20 U.S.C. § 1414(d).

The IEP team is charged with ensuring that the student receives a free and appropriate public education in the least restrictive environment. 20 U.S.C. § 1412(a)(1)(A), (a)(5)(A). To the maximum extent appropriate, at home placement and residential programs should be avoided unless education in the regular school cannot be achieved satisfactorily with the use of supplementary aids and services. 20 U.S.C. § 141'’(a)(5)(A). Presumably, the IEP team would have considered the IDEA’S least restrictive environment requirement, preferring placement in the regular school whenever possible. See 20 U.S.C. § 1412(a)(5)(A).

While the record does not specifically define the difference between a paraedncator and a classroom aide, it is clear that a paraeducator’s duties involve helping the classroom teacher with both instructional and behavioral matters. The IDEA requires that an IEP be revised to address lack of expected progress toward stated goals. 20 U.S.C. § 1414(d)(4)(A). Where a child’s behavior impedes his or her learning or that of others, the IEP team shall consider, when appropriate, various strategies, including positive behavioral interventions and supports. 20 U.S.C. § 1414(d)(3)(B).

In the 2000-01 school year, R.M. began attending Lakes High School. Apparently, R.M. has done very well there.