The Divisional Court has finally settled the question: is spilling coffee in a car an “accident” under the SABS? The answer is yes.

The definition of “accident” in section 3(1) of the SABS is broad and ever-expanding, sometimes in surprising ways. A reveler who broke his neck while “cavorting around [a] stripper pole” in a limo,1 a car surfer who fell of a bumper,2 and a runner who crashed into a parked vehicle3 have all been deemed “accident” victims. More recently, slipping and falling near vehicles joined the list.4 But coffee-related casualties? Until last week, the law couldn’t quite make up its mind.

In Miceli v TD Insurance, the Divisional Court finally spilled the beans: unless a barista deliberately throws a drink at a client, an injury caused by spilling hot coffee in a car qualifies as an “accident.”

The Facts

The case involved a rather prosaic misfortune: a couple pulled up to a McDonald’s drive-thru, he ordered an extra-large black coffee and handed it to her in the back seat. The lid was loose, coffee spilled on her hands, it burned, and in the reflexive chaos, the entire cup landed on her legs. Injuries followed, as did lawsuits—against both McDonald’s and the auto insurer.

In the end, it was the sheer banality of the event that proved fatal for the insurer.

The Mistake

At first glance, the car had little to do with what happened. How is spilling coffee inside a car any different from spilling it in a café? After all, if the lid had been properly sealed, there would have been no injury.

This was more or less the insurer’s reasoning when it denied the claimant benefits under the SABS—and the LAT agreed, dismissing the application at a preliminary issue hearing. A reconsideration at the LAT was denied, but the Divisional Court saw things differently on appeal.

The Correction

The Court found that the LAT misinterpreted the definition of “accident” in section 3(1) of the SABS.6 For over 20 years, this definition has been shaped by the legal tests set out by the ONCA in Greenhalgh v ING Halifax Insurance Co:7

  1. The “purpose test”: whether the incident arose out of the use or operation of an automobile. The later means some ordinary and well-known activity to which automobiles are put.
  2. The “causation test”: whether the use or operation of an automobile directly caused the impairments.
  • The “but for” consideration: but for the use or operation of the automobile, there would have been no injuries.
  • The “intervening act” consideration: there was no intervention of other events that cannot be said to be part of the ordinary course of use or operation of the automobile.”
  • The “dominant feature” consideration: whether the use and operation of the automobile was the dominant feature of the injuries.

Driving through a drive-thru is an ordinary use of a car. The extra-large black coffee wouldn’t have been purchased, poured into a cup, a lid secured (in this case, improperly), handed over, and ultimately spilled on the claimant had she not been a passenger in a vehicle engaged in this very ordinary activity. Therefore, the purpose test was met, and “but for” causation was in place.

The real issues before the Divisional Court were: (a) whether there was an intervening act, and (b) what was the dominant feature of the incident. The Court found that the loosely attached lid was neither. The real cause—and the dominant feature—of the incident was the use of the car itself, which naturally includes the common activity of having a hot drink while inside. In reaching this conclusion, the Court highlighted several key points.

First, an “intervening act” must be an unexpected one. For example, an attendant throwing hot coffee on a claimant is unexpected and thus an intervening act. This would also be the case if someone poisoned the coffee. In contrast, spilling coffee is entirely foreseeable, making it neither unexpected nor intervening. A loose lid changes nothing; it’s just as common.

Second, an “intervening act” must be beyond a reasonable risk created by the use or operation of an automobile. The risk of a barista launching a coffee attack or lacing a drink with poison is wholly unexpected, not a normal driving hazard. On the other hand, handling a hot coffee in a car is an everyday risk—easily confirmed by a quick online search for “spill-proof coffee mug.”

Third, an “intervening act” is different from a “triggering event.” A triggering event might set an accident in motion, but the direct cause and dominant feature of the incident must still be the use or operation of the vehicle. This distinction serves as a reminder to look past the immediate cause and consider the bigger picture. An inadvertent spill remains an inadvertent spill, no matter how secure—or insecure—the lid was.

Fourth, spilling coffee in a car is different from spilling it anywhere else because the occupant is physically restrained from reacting in time to avoid the spill. Whether seat-belted or not, the claimant couldn’t stand up, step aside, or dodge the coffee mid-air. The car itself limited her ability to prevent the injury, reinforcing its role as the dominant feature of the incident.

The Divisional Court relied on the previous ONCA decision in Dittmann v Aviva.8 The insurer attempted to distinguish Dittmann to no avail, asserting that the improperly fastened lid took this incident outside of the definition of an “accident.” The Court disagreed, and again reinforced the idea that physical restraint inside a vehicle can make an otherwise mundane mishap an “accident” under the SABS. Once again, the Court emphasized that the SABS must be interpreted broadly and flexibly, always in line with their consumer protection purpose.

Implications

Miceli aligns with the recent trend of broadening the definition of an “accident.” A direct physical connection between an automobile and an injury is no longer required.9 Here, the claimant was burned by coffee, not injured in an impact involving the car itself. The example of an attendant deliberately throwing coffee at a customer alludes to previously decided assault cases—one of the few scenarios where a car is not the dominant feature of an injury.10

The Court’s discussion of the lid and seat belt underscores a key point: insurers won’t escape liability through microscopic scrutiny of minor details if the overall incident falls within the ordinary use and operation of a vehicle. That said, some borderline cases will persist, and a detailed analysis of the mechanics of an incident will still be necessary.

For instance, the Court cited with approval an older case where FSCO found that a heart attack while driving was not an intervening act (but rather a triggering event), making the resulting crash an “accident.”11 This is to be contrasted with a recent LAT decision where an adjudicator refused to classify a stroke-related fall while exiting a car as an accident.12

Conclusion

In coverage disputes, any event suspected of breaking the chain of causation should be assessed against the following factors:

  • Was it an ordinary and foreseeable occurrence, or something entirely unexpected?
  • Was it an incidental risk of using a car, or did it fall outside the realm of reasonable driving-related risks?
  • Was it merely a triggering event within a broader chain of events linked to the car’s use, or a truly independent incident?
  • What is the underlying practical reason for extending insurance coverage in this case?

Direct causation requires more than the mere presence of a car—it must be more than incidental to the injury—but the threshold is low. We can expect more cases where LAT and the courts classify as an “accident” something that common sense might not.

See Miceli v TD Insurance, 2025 ONSC 496, online: <https://canlii.ca/t/k92db>.

1 Economical Mutual Insurance Company v Whipple, 2012 ONSC 2612, online: <https://canlii.ca/t/fr84k>.|
2 I.C. v Intact Insurance Company, 2017 CanLII 69443 (ON LAT), online: <https://canlii.ca/t/h8rwk>.
3 D.S. v TD Insurance Meloche Monex, 2017 CanLII 43837 (ON LAT), online: <https://canlii.ca/t/h4rb0>.
4 Davis v Aviva General Insurance Co, 2024 ONSC 3054, online: <https://canlii.ca/t/k4z78>; Wall v Economical Insurance, 2025 CanLII 1852 (ON LAT), online: <https://canlii.ca/t/k8xv7>.
5 Miceli v TD Insurance, 2025 ONSC 496, online: <https://canlii.ca/t/k92db>.
6 An “accident” is “an incident in which the use or operation of an automobile directly causes an impairment.”
7 Greenhalgh v ING Halifax Insurance Co, [2004] OJ No 3485 (ON CA), online: <https://canlii.ca/t/1hq8c>.
8 Dittmann v. Aviva Insurance Company of Canada, 2017 ONCA 617, online: <https://canlii.ca/t/h50wg>.
9 Madore v Intact Insurance Company, 2023 ONSC 11, online: <https://canlii.ca/t/jtpmp>; Davis v Aviva General Insurance Co, 2024 ONSC 3054, online: <https://canlii.ca/t/k4z78>.
10 E.g., Kandaiya v TD Insurance, 2024 CanLII 83754 (ON LAT), online: <https://canlii.ca/t/k6nj3>; Ji v Aviva Insurance Company of Canada, 2024 CanLII 74769 (ON LAT), online: <https://canlii.ca/t/k67j1>.
11 Ignazio Salamone v Aviva Canada Inc, 2016 ONFSCDRS 173 (CanLII), online: <https://canlii.ca/t/jq95k>.
12 Dudoukh v Aviva General Insurance, 2024 CanLII 43460 (ON LAT), online <https://canlii.ca/t/k4mmp>.

Author

  • Dmitry Shniger

    Dmitry romanticism is notorious. After 13 years of in-house counselling in several industries, teaching law at a law school, and litigation practice in Russia, his search for the rule of law brought him to Canada.

    View all posts