Michael Gomez’s Death Wasn’t a Random Event

Pre-teen and teenager deaths are rare in New York City. Out of nearly a million city residents ages 10-19, just 226 died in 2011, the most recent data year. That’s barely more than two deaths per 10,000. Sliced a bit differently, only four to five New Yorkers age 10-19 die each week.

With this backdrop, the news this morning of the death of one of five young teens whom a driver plowed into last week in Maspeth, Queens, leads naturally to speculation that the crash was a contributing factor. Yet 13-year-old Michael Gomez apparently wasn’t among the grievously injured — he was said to have suffered only a “swollen arm” when Francis Aung Lu drove his Honda Pilot SUV across the curb and onto the sidewalk as the teens walked to school — and he was discharged from the hospital several hours after the crash. Michael’s death last Sunday, three days later, couldn’t have been related to the crash. Or could it?

I think that from a statistical standpoint, the conclusion is inescapable that the September 12 crash contributed directly to Michael’s death on September 15. It’s a matter of simple statistics, resting on the actuarial fact, hinted above, that a city teen chosen at random has less than a one in 200,000 chance of dying in a given week.

Some published accounts are attributing Michael’s death to asthma. The Post, citing “law-enforcement sources,” reported that Michael “died after suffering a severe asthma attack.” The paper added, “Michael suffered only minor injuries and was taken to Elmhurst Hospital before being released that same day. But on Sunday, he suffered an asthma attack and was rushed back to Elmhurst Hospital, where he later died, sources said.”

There’s no reason to question that account. Yet the fact that Michael’s asthma flared to fatal proportions just three days after he was vehicularly assaulted lies far beyond the realm of ordinary coincidence. “Deaths due to asthma are rare among children,” says the American Lung Association, noting that in 2009, only 157 children under 15 died from asthma in the United States. For comparison, the entire U.S. resident population under age 15 is 61 million, of whom 9.5 percent, or nearly 6 million, currently have asthma. A death rate due to asthma of 157 per 6 million equates to less than 1 death per 38,000 per year, or 1 in 2,000,000 per week. Translated: A child under 15 who has asthma but is otherwise chosen at random has a one in two million chance of dying from asthma in a given week.

Michael may have been “chosen at random” as a victim of the SUV, but once he was injured and experienced the further trauma of seeing two schoolmates pinned under the vehicle, he moved to a different statistical orbit, one in which asthma attacks would necessarily be both more likely and more severe.

I never met Michael or the Gomez family, and I don’t know if after the crash he woke with nightmares or slept like a baby. What I do know is that before Mr. Aung Lu got behind the wheel of his Honda Pilot last Thursday, a statistician would have given Michael a 99.9995 percent chance of living another week, based on the 1 in 200,000 proportion noted above, and an even higher probability of not experiencing a fatal asthma attack.

At some point, the NYC Medical Examiner will issue an official finding on Michael Gomez’s death. Perhaps it will offer post-traumatic stress as a trigger to the fatal asthma attack, perhaps not. Almost certainly, the cause will be given not as vehicular injury but as chronic lower respiratory disease, a category that includes chronic bronchitis and emphysema along with asthma. Of NYC’s 226 pre-teen and teenager deaths in 2011 noted above, ten were classified as CLRD. That’s one every five weeks among nearly a million kids and young people. For the one-tenth who have asthma, the weekly death frequency is one per 500,000.

The only clear statistical link, then, is between the Honda on the sidewalk and Michael in a casket. As I see the numbers, that link is nearly as strong as if the driver had killed the 13-year-old on the spot.

  • statsman

    Bringing up that “1 in 200,000” stat after the fact is called the prosecutor’s fallacy. You can’t use an a priori probability after an unlikely event in that way.

    As parodied by Richard Feynman: “You know, the most amazing thing happened to me tonight. I was coming here, on the way to the lecture, and I came in through the parking lot. And you won’t believe what happened. I saw a car with the license plate ARW 357. Can you imagine? Of all the millions of license plates in the state, what was the chance that I would see that particular one tonight? Amazing!”

    That said, yes, it is certainly possible, maybe even likely, that the crash injury had something to do with this death. But I rather wait for actual evidence from the medical examiner over fallacious armchair statistical reasoning.

  • Eric McClure

    I think we can always rely on the expert medical diagnoses of law-enforcement officials, no?

  • dl

    Yes, I like Streetsblog, but this sort of wild speculation doesn’t really help anyone.

  • Rhubarbpie

    It probably be helpful to understand more about what the WHO means when it cotes emotional trauma as a precipitator of an asthma attack. However, this thesis shouldn’t be dismissed out of hand, as some commenters have.

  • Rhubarbpie

    That’s “cites,” not cotes.

  • Arthur Goldberg

    No, Charlie is right.

    The point is that Michael Gomez’s death is a highly significant event. So Feynman’s example of noting an extremely rare event that occurred is not relevant. Numerous rare events like that happen constantly. If I could tell you the serial numbers on 3 bills in your wallet that would be amazing because the chance that anyone would know them is essentially zero. But if you tell me the serial numbers that’s not interesting at all, because you can just read them.

    The question is what’s the proximate cause of Michael Gomez’s death?

    Charlie examines all the likely causes given the information known. What’s the chance that a kid his age would die? What the chance that a kid with asthma would die?

    There may be other likely causes about which we don’t know. Some would implicate Mr. Gomez’s injuries, physical and emotional, from the recklessly driven SUV, others wouldn’t.
    For the former, we might ask Did he have a form of asthma that tends to worsen severely with emotional stress? Did he have other injuries that were not detected in his first visit to the hospital? Did the doctors who treated him initially communicate his injuries and experiences to his pulmonologist?

    Questions in the latter category would include Did other things happen to Michael Gomez recently that could contribute to fatal asthma attack? Did he have near-fatal asthma attacks previously?

    But it is clear to me that unless the basic facts change, one can make a good statistical analysis (Bayesian, if you like) that provides strong support for the hypothesis that Michael Gomez was killed by being hit by an SUV.


  • Jake Stevens

    I think you are misusing the term “statistical link.” This article is hysterical – and not in a good way.

  • Mfs

    Agreed. Wait for the family to speak.

  • Joe R.

    Despite the skepticism in some of the comments, this article puts forth a very plausible theory. While we can never prove in court whether the trauma of this event bought on the asthma attack which killed Michael Gomez, it’s well known that stressful events negatively impact health. For example, people living under the glide path near airports have more health issues than those who don’t. And motor vehicles cause asthma and cancer even when operated legally. In fact, some statistics indicate motor vehicles indirectly kill ten times as many by polluting the air as they do in collisions. In any case, I agree that there’s a very good possibility Michael Gomez would be alive today had he not been involved in this event.

  • Glenn

    I would call this article more the start of a very interesting investigation rather than a conclusion. Any traumatic event or injury can have repercussions that manifest themselves many weeks, months or years later whether they are physical or psychological. I would continue the search for a link between MVA and the impact on all cause mortality on a statistical basis. A test vs. control match would be the best – similar to this actuarial analysis. Certainly fertile grounds for further analysis.

  • BornAgainBicyclist

    Stress is a pretty common asthma trigger — ask any doc who works with asthmatics (or anyone with asthma). This article is hardly the wild speculation a surprising number of commenters see it as.

  • The statistical comments notwithstanding, the paragraph is a bit jarring:

    “Pre-teen and teenager deaths are rare in New York City. Out of nearly a million city residents ages 10-19, just 226 died in 2011, the most recent data year. That’s barely more than two deaths per 10,000. Sliced a bit differently, only four to five New Yorkers age 10-19 die each week.”

    I get that the point is that Michael Gomez’s death was statistically unlikely, but LOOK at those numbers. It’s not “only” four to five kids each week–it’s FOUR TO FIVE KIDS EACH WEEK. Dead.

  • carma

    Charles Komanoff: ” Im not a doctor, but I play one on streetsblog.”

  • Anonymous

    Boy … “wild speculation,” “prosecutor’s fallacy,” “hysterical,” “jarring” and poseur. Even granting that my exposition could have been clearer, I’m stunned by the level of denial and/or innumeracy in the critical responses.

    Let’s take it from the top. From https://www.health.ny.gov/statistics/vital_statistics/2011/table01.htm” I calculate that there are 971,757 NYC residents ages 10-19. I don’t have asthma data for NYC children, but nationally 10% of children under 18 have asthma (http://www.cdc.gov/nchs/data/series/sr_10/sr10_254.pdf, p. 11 of 88). The city rate is probably higher, but using “just” 10% we have ~ 97,000 NYC youths with asthma.

    As noted in my post, 10 of those children die (of asthma) each year, or roughly one every five weeks. Michael died, reportedly from an asthma attack, three days after the SUV incident. Let’s call that “within a week.” Within that week, we could have expected 0.2 youth asthma deaths in NYC. Let’s bump that up five-fold (allowing for Sept-Oct being high allergen season and males being more susceptible than females), to 1.0.

    What these numbers tell us is that of 97,000 NYC kids suffering, like Michael, from asthma, 96,999 would have been expected to live out the week from the Sept 12 assault to Sept 19, while one would not. And, in fact, at least one did not: Michael Gomez.

    Why Michael, then? He might have been extraordinarily sensitive to asthma triggers. But to reduce to the 5% “significant level” (the standard threshold for deeming an occurrence to have been non-random) the probability of his being selected at random for death by asthma within a week of the incident, he would have to have been nearly 5,000 times as susceptible to a fatal trigger as the average asthma-exhibiting youth. (The 5,000 factor comes from inflating a 1 in 97,000 chance to 1 in 20.) That seems unlikely. If Michael was sensitive to that extent, then how did he manage to live with his asthma for so many weeks — a few hundred, presumably?

    True, none of this would be germane if trauma and stress were not potential triggers of asthma attacks. But they are. Please follow the links in my post.

    From the above, I don’t see how one can look at Michael Gomez’s death in the wake of the Sept. 12 SUV assault without inferring that he would be alive today but not for the driver’s terrifying assault on Michael and his schoolmates.

  • Guest

    I like Streetsblog because it is overwhelmingly thoughtful and reasonable. This is absolute garbage. It might indeed be possible that the accident in some way triggered the asthma attack. But Komanoff cites not a single piece of medical evidence to support this claim. Instead, this entire post is based on astounding statistical fallacies.

  • Rabi Abonour

    If Komanoff wants to discuss the medical link between stress and catastrophic asthma attacks, I’m all ears. That is a perfectly reasonable line of inquiry. But the statistical argument, made the focus of this post, simply obscures the issue.

  • Anonymous

    Charles, I feel like you have a better understanding of statistics and probability than is expressed in this article, so maybe you’re just a having a bad day?

    Based on the logic here, you could just as easily argue that Michael Gomez got hit by a car (a rare event) because he had asthma. Or because his first name starts with “M”.

    You can’t make a valid statistical argument getting hit by a car is a co-factor for fatal asthma attacks based on one observation.

  • Anonymous

    Have to say, I was pretty stunned by these responses too. I’m glad you responded.

  • BornAgainBicyclist

    Same here.

  • Kevin Love

    He did. And provided sources. Follow this link from his article.


  • Kevin Love

    It is the opinion, not of Charles but of the medical experts at the Mayo Clinic that a serious asthma trigger is “Strong emotions and stress.” I quote from the Mayo source Charles linked to in his article.

    Are you seriously suggesting that being the victim of a vicious assault by a multi-tonne lethal weapon is not stressful? That seeing your friends being seriously injured is not stressful? That not knowing whether or not they are dead is not stressful?

  • Rabi Abonour

    I know, but that was a small part of the article. He should have focused on that – his statistical rant does nothing for his argument.

  • Anonymous

    There may be a causal link between “strong emotions and stress” and fatal asthma incidence. That’s a different question, and requires more medical knowledge than I have to know whether it is credible.

    The main assertion of Komanoff’s editorial is that there is

    statistical evidence to support the hypothesis that Michael Gomez’s death from an asthma attack was caused by his being hit by a car a few days earlier. There is no statistical evidence for this hypothesis.

    It’s bad enough that the kid got hit by a driver who floored his SUV onto the sidewalk. We don’t need to make up things in order to try to prove that it was bad.


Death of Cyclist Shocks Melbourne, Prompts Bus Ban

As I wrote in a post last week, the City of Melbourne, Australia, is working hard to make cycling easier and safer — but not quickly enough to save the life of one cyclist. The day after my post a 33-year-old Melbourne woman was killed when her wheels slipped on tram tracks on Melbourne’s main […]