6 ms·

Nearly half of drivers killed in (Ohio County) crashes had THC in their blood

87 points by pogue 1y ago

hackingonempty 1y ago
How many drivers killed in crashes had accessible phones in their vehicles? I bet it was more than half.
pogue 1y ago
As you can see, and will probably read in upcoming media headlines, the original headline without mentioning that this study was carried out in a single county in Ohio (Montgomery County) is extremely alarmist and even mentioning where they're talking about isn't listed till quite a way down the page. That isn't to say intoxicated driving isn't a problem. I'm sure that it is (and there may even be data on it I'm unaware of). How to address it without law enforcement having an easy to use immediate test like a breathalyzer for alcohol, I don't know. Driving high/drunk/buzzed on any intoxicating substance is reprehensible and should involve education, public messaging, law enforcement action, and further research into all the aspects of it. But, headlines like these, IMHO, do more of a disservice than a benefit.
- ratelimitsteve 1y ago
  they grow cherries in montgomery county? they certainly like to pick cherries there.
- notmyjob 1y ago
  Well, most users drive on it and we know it slows reaction time so it’s easy to believe the headline. Agree bad science is often worse than no science.
animitronix 1y ago
This literally means nothing. Do you know how long THC lasts in your blood after smoking?
- o11c 1y ago
  It means a lot give how small a percentage of people even touch THC with a 10-foot pole.
  - daeken 1y ago
    According to this, 16.5% of Ohio residents used a marijuana product in the last year in 2023: https://www.statista.com/statistics/723822/cannabis-use-within-one-year-us-adults/?srsltid=AfmBOopBG__ny6cU7xoRjqNfloIm00jmI_xCtFB7jmJekt5Ct7AaSp_A https://www.statista.com/statistics/723822/cannabis-use-with... Hard to argue that's a small percentage.
    - coloneltcb 1y ago
      without knowing the methods of that survey, if that 16.5% was self-reported you can expect that the actual number is even higher
  - trenchpilgrim 1y ago
    > Cannabis use in the past year and past month remained at historically high levels for both adult age groups in 2023. Among adults 19 to 30 years old, approximately 42% reported cannabis use in the past year, 29% in the past month, and 10% daily use (use on 20 or more occasions in the past 30 days). Among adults 35 to 50, reports of use reached 29%, 19%, and 8%, respectively. https://nida.nih.gov/news-events/news-releases/2024/08/cannabis-and-hallucinogen-use-among-adults-remained-at-historic-highs-in-2023 https://nida.nih.gov/news-events/news-releases/2024/08/canna...
  - sjsdaiuasgdia 1y ago
    Have you been to a weed store in a state with legal recreational usage? In my experience, the stores are always busy and you see a wide cross-section of society in line. Half of Americans have tried pot at least once: https://news.gallup.com/poll/509399/fully-half-americans-tried-marijuana.aspx https://news.gallup.com/poll/509399/fully-half-americans-tri...
  - bryanlarsen 1y ago
casey2 1y ago
Tell people not to drive unless it's vital and to take public transit whenever possible. The fact is that people are bad at estimating risks so they need to be told the facts.
- stronglikedan 1y ago
  > Tell people not to drive unless it's vital and to take public transit whenever possible. That sounds like a miserable way to get around. Driving is the bees knees.
  - recursive 1y ago
    Driving is a tragic waste of human attention.
paulv 1y ago
> No Effect from Legalization: The rate of drivers who tested positive for THC did not change significantly before or after legalization (42.1% vs. 45.2%), indicating that legal status did not influence the behavior of those who chose to drive after use. Ohio recently passed a marijuana decriminalization bill.
alphazard 1y ago
Testing for impairment from THC is not as simple as testing for impairment from alcohol. It's a molecule that is similar to normal endogenous chemicals, and the human body is good at building tolerance to it. Here is a video (timestamped to the relevant content) of an expert explaining to Huberman why it is so difficult to assess impairment from blood levels. https://www.youtube.com/watch?v=jouFvyRZntk&t=6072s https://www.youtube.com/watch?v=jouFvyRZntk&t=6072s The study in TFA may just be finding that nearly half of drivers in Ohio are habitual cannabis users, or used cannabis in the past few days.
- IAmBroom 1y ago
  I am always amused when I drive by a PSA billboard in my city that reads, "A little high is too high to drive." I mentally add, "... because we can't even test for it when we pull you over, so... please don't."
  - nekusar 1y ago
    > I mentally add, "... because we can't even test for it when we pull you over, so... please don't." Thats not entirely true. I knew 2 people who loved to drive high. Evidently it feels great (their words, not mine). Now, being incapacitated or inebriated with any substance is illegal. Even prescription drugs. And basic roadside sobriety tests like "walk in straight line", can be an easy fail if you cant stand up. Now weirdly, ive never heard of anyone ticketed or pulled for driving sick or very tired. Even though both of those conditions can also be just as dangerous.
    - mothballed 1y ago
      The roadside test is basically a test to see if you are a moron dumb enough to consent to it. If you perform an FST good luck doing it in a way that can't be used against you. In my state you're only required to submit to chemical testing, which the courts have ruled the little portable breathalizer machine isn't, and even if you refuse all they can do is suspend your license and then try to get a warrant.
lukeschlather 1y ago
This seems fundamentally intended to be misleading. Blood THC concentration has no correlation with current level of impairment the way it does with blood alcohol. "far exceeding most state impairment limits" specifically seems like it was written by someone trying to make a technically correct statement that is essentially wrong. The "legal limit" for THC is either 0 or there isn't one.
- cluckindan 1y ago
  Even if there was such a limit, people who use cannabis for medical reasons would have to be exempt.
  - kelipso 1y ago
    No..impairment is impairment whatever the reason for getting impaired is. Prescription meds get you in trouble in DUIs too.
Youden 1y ago
This is interesting but the causal relationship inferred by the article isn't supported by the evidence. Unlike alcohol for example, there's no clear dose of THC where it can be concluded beyond a reasonable doubt that the person is impaired. A dose that might give a regular user a gentle buzz could render a first-time user completely stoned. It's possible that these people were all incredibly stoned while driving but it's also possible that many drivers in Ohio are regular THC users and have such a high tolerance that their function is unimpaired. And like always, there's the definite possibility of confounding factors, like reckless drivers also enjoying recreational drug use. It's also important to note that the article's focus on legal limits is somewhat pointless. As there's no clear threshold above which one is impaired, the legal limits are somewhat arbitrary and are determined by other factors, like whether THC can be reliably tested at the given concentration. Ultimately, as far as I can tell, the current state of things is that we're fairly certain that THC is able to impair driving ability but we have no idea how much THC is needed to do it or how impaired drivers become. For a somewhat reputable source, the NHTSA did a report to Congress in 2017: https://www.nhtsa.gov/sites/nhtsa.gov/files/documents/812440-marijuana-impaired-driving-report-to-congress.pdf https://www.nhtsa.gov/sites/nhtsa.gov/files/documents/812440...
- xienze 1y ago
  > A dose that might give a regular user a gentle buzz could render a first-time user completely stoned. > It's possible that these people were all incredibly stoned while driving but it's also possible that many drivers in Ohio are regular THC users and have such a high tolerance that their function is unimpaired. Can't you make this same argument with alcohol? > Ultimately, as far as I can tell, the current state of things is that we're fairly certain that THC is able to impair driving ability but we have no idea how much THC is needed to do it or how impaired drivers become. So how exactly would you propose states move forward with allowing people to consume THC and operate vehicles (or not)? We apparently can't even tell if someone is impaired in a reliable way.
  - hamdingers 1y ago
    > Can't you make this same argument with alcohol? No because a once-daily user of THC will continue to test positive for 14+ days after quitting completely. For alcohol it's 6-12 hours. They are not comparable.
baby 1y ago
We need self driving cars and VR so people can get high with their friends without having to drive
- Perizors 1y ago
  There are Uber, taxi and other public transportation. You can get high with your friends and don't drive. Just plan ahead.
  - spookie 1y ago
    This. If you are drunk, high, tired af, or just sleepy, stop being a danger for others. A car is a hunk of metal going at high speed, stop being a child about it.
    - baby 1y ago
      Fix the system, because you cant fix the people
  - baby 1y ago
    In the city yeah, I believe most drunk driving accidents happen in places and times where uber would be too overpriced or public transport too inconvenient
stego-tech 1y ago
Without getting into the technical weeds of dosing, tolerance, and impairment, this is an interesting data set that definitely paints a picture of lax enforcement and ineffective public messaging around substance use and associated impairments. That being said, I also don’t see how we can do better in the USA given that cars are necessary evils for the majority of people outside a select few cities with public transport options, and that we’ve been doing messaging about impaired driving for nearly a century to no real avail. Perhaps it’s time to start treating motor vehicle operation as an actual privilege rather than a “right” gated by a far-too-easy licensing regime. Start by tasking safety officers with more citations for things like failing to signal or illegal turns through dash camera footage (to avoid dangerous traffic stops), start suspending licenses earlier for consistent or serious infractions, and raise licensing standards higher to shunt more folks onto smaller vehicles or public transit. This isn’t a cannabis vs alcohol debate despite the framing as such: it’s a debate over how many people should be allowed to drive at all, and to what degree we hold them accountable to ensure good behavior.
cluckindan 1y ago
Really? What else did they have in their blood?
Uhhrrr 1y ago
The article doesn't mention what percentage had alcohol, with and without THC. Why not?
ratelimitsteve 1y ago
while I'm glad they're no longer doing urinalysis and convicted people of DUI based on evidence that doesn't prove anything beyond that they've used marijuana in the past several months, the standard they use to test for impairment (5ng/ml seems to be the highest standard I could find among states that have a standard) may not actually indicate recent marijuana use and in some states the standard for intoxication is so low that frequent users will still test positive after days of abstinence (https://pubmed.ncbi.nlm.nih.gov/32841811/ https://pubmed.ncbi.nlm.nih.gov/32841811/)
notmyjob 1y ago
This would be interesting IFF those drivers did not have alcohol in their system too. I can’t tell if that’s the case from the attached article. I also can’t tell whether the drivers were at fault or e.g. if they froze when another vehicle came at them, which would fit with the stereotype of the slow overly cautious stoned driver vs fast reckless driving you get from alcohol or being a young male (which might be more common in Dayton because of the Air Force base).
josefritzishere 1y ago
Correlation is not causation. I think the writer fails to make the case.
fmajid 1y ago
Why would anyone expect legalization to cause drug usage to go down? It helps eliminate the violent criminals that proliferate around any criminalized activity. Addicts are not doing drugs for the lure of the forbidden, so making drugs easier and cheaper to get should only increase usage.
couchdive 1y ago
in other news, half of Ohio county not dead in auto crashes, also test positive of THC
theturtle 1y ago
What else is there to do in Dayton?
nzeid 1y ago
Some people have hinted at the following point but I wanted to ask directly: How many recently dead Ohio adults have THC in their blood? Forget driving.
- Balgair 1y ago
  I mean that's a pretty good control. Granted not a lot of the deceased are going to get these autopsys though. Still, a good control group
joshcsimmons 1y ago
Seems alarmist. As many others have pointed out, THC blood concentration doesn't have the same fast decay as BAC.
pRusya 1y ago
Personal experience after living in Columbus for a few months - it felt like everybody was driving drunk and/or distracted by default. Columbus, Dayton and Cincinnati were not very fun to drive around and taking unprotected left turns felt like a gamble. On the other hand, Amish communities had glass-smooth roads and people were careful when passing horse carriages.
m463 1y ago
I remember when someone was prosecuted for a drug offense because their money was contaminated by cocaine... but it turns out 3/4 of currency is already contaminated: https://en.wikipedia.org/wiki/Contaminated_currency https://en.wikipedia.org/wiki/Contaminated_currency I wonder because it might be many people have THC in their blood for days, and it might not indicate what the prosecution infers. (I think thc lasts longer than say alcohol)
elzbardico 1y ago
Well, unless you're completely brainwashed in luxury beliefs, it is pretty obvious that legalization would increase the prevalence and frequency of usage. This is trade-off that any rational person could predict.
- joshstrange 1y ago
  > it is pretty obvious that legalization would increase the prevalence and frequency of usage. Except that's not at all what happened, something mentioned multiple times in the article. > No Effect from Legalization: The rate of drivers who tested positive for THC did not change significantly before or after legalization (42.1% vs. 45.2%), indicating that legal status did not influence the behavior of those who chose to drive after use.
more_corn 1y ago
Doesn’t thc stay in your blood for weeks while the psychoactive effects last hours? They’re basically saying in a random sampling of dead people 50% had consumed THC at some point in the last few weeks. So this is more an analysis of THC use in the population and in no way can be causally linked to the deaths.

87 points by pogue 1y ago

How many drivers killed in crashes had accessible phones in their vehicles? I bet it was more than half.

As you can see, and will probably read in upcoming media headlines, the original headline without mentioning that this study was carried out in a single county in Ohio (Montgomery County) is extremely alarmist and even mentioning where they're talking about isn't listed till quite a way down the page. That isn't to say intoxicated driving isn't a problem. I'm sure that it is (and there may even be data on it I'm unaware of). How to address it without law enforcement having an easy to use immediate test like a breathalyzer for alcohol, I don't know. Driving high/drunk/buzzed on any intoxicating substance is reprehensible and should involve education, public messaging, law enforcement action, and further research into all the aspects of it. But, headlines like these, IMHO, do more of a disservice than a benefit.

ratelimitsteve 1y ago
they grow cherries in montgomery county? they certainly like to pick cherries there.
notmyjob 1y ago
Well, most users drive on it and we know it slows reaction time so it’s easy to believe the headline. Agree bad science is often worse than no science.

animitronix 1y ago

This literally means nothing. Do you know how long THC lasts in your blood after smoking?

o11c 1y ago
It means a lot give how small a percentage of people even touch THC with a 10-foot pole.
- daeken 1y ago
  According to this, 16.5% of Ohio residents used a marijuana product in the last year in 2023: https://www.statista.com/statistics/723822/cannabis-use-within-one-year-us-adults/?srsltid=AfmBOopBG__ny6cU7xoRjqNfloIm00jmI_xCtFB7jmJekt5Ct7AaSp_A https://www.statista.com/statistics/723822/cannabis-use-with... Hard to argue that's a small percentage.
  - coloneltcb 1y ago
    without knowing the methods of that survey, if that 16.5% was self-reported you can expect that the actual number is even higher
- trenchpilgrim 1y ago
  > Cannabis use in the past year and past month remained at historically high levels for both adult age groups in 2023. Among adults 19 to 30 years old, approximately 42% reported cannabis use in the past year, 29% in the past month, and 10% daily use (use on 20 or more occasions in the past 30 days). Among adults 35 to 50, reports of use reached 29%, 19%, and 8%, respectively. https://nida.nih.gov/news-events/news-releases/2024/08/cannabis-and-hallucinogen-use-among-adults-remained-at-historic-highs-in-2023 https://nida.nih.gov/news-events/news-releases/2024/08/canna...
- sjsdaiuasgdia 1y ago
  Have you been to a weed store in a state with legal recreational usage? In my experience, the stores are always busy and you see a wide cross-section of society in line. Half of Americans have tried pot at least once: https://news.gallup.com/poll/509399/fully-half-americans-tried-marijuana.aspx https://news.gallup.com/poll/509399/fully-half-americans-tri...
- bryanlarsen 1y ago

casey2 1y ago

Tell people not to drive unless it's vital and to take public transit whenever possible. The fact is that people are bad at estimating risks so they need to be told the facts.

stronglikedan 1y ago
> Tell people not to drive unless it's vital and to take public transit whenever possible. That sounds like a miserable way to get around. Driving is the bees knees.
- recursive 1y ago
  Driving is a tragic waste of human attention.

paulv 1y ago

> No Effect from Legalization: The rate of drivers who tested positive for THC did not change significantly before or after legalization (42.1% vs. 45.2%), indicating that legal status did not influence the behavior of those who chose to drive after use. Ohio recently passed a marijuana decriminalization bill.

alphazard 1y ago

Testing for impairment from THC is not as simple as testing for impairment from alcohol. It's a molecule that is similar to normal endogenous chemicals, and the human body is good at building tolerance to it. Here is a video (timestamped to the relevant content) of an expert explaining to Huberman why it is so difficult to assess impairment from blood levels. https://www.youtube.com/watch?v=jouFvyRZntk&t=6072s https://www.youtube.com/watch?v=jouFvyRZntk&t=6072s The study in TFA may just be finding that nearly half of drivers in Ohio are habitual cannabis users, or used cannabis in the past few days.

IAmBroom 1y ago
I am always amused when I drive by a PSA billboard in my city that reads, "A little high is too high to drive." I mentally add, "... because we can't even test for it when we pull you over, so... please don't."
- nekusar 1y ago
  > I mentally add, "... because we can't even test for it when we pull you over, so... please don't." Thats not entirely true. I knew 2 people who loved to drive high. Evidently it feels great (their words, not mine). Now, being incapacitated or inebriated with any substance is illegal. Even prescription drugs. And basic roadside sobriety tests like "walk in straight line", can be an easy fail if you cant stand up. Now weirdly, ive never heard of anyone ticketed or pulled for driving sick or very tired. Even though both of those conditions can also be just as dangerous.
  - mothballed 1y ago
    The roadside test is basically a test to see if you are a moron dumb enough to consent to it. If you perform an FST good luck doing it in a way that can't be used against you. In my state you're only required to submit to chemical testing, which the courts have ruled the little portable breathalizer machine isn't, and even if you refuse all they can do is suspend your license and then try to get a warrant.

lukeschlather 1y ago

This seems fundamentally intended to be misleading. Blood THC concentration has no correlation with current level of impairment the way it does with blood alcohol. "far exceeding most state impairment limits" specifically seems like it was written by someone trying to make a technically correct statement that is essentially wrong. The "legal limit" for THC is either 0 or there isn't one.

cluckindan 1y ago
Even if there was such a limit, people who use cannabis for medical reasons would have to be exempt.
- kelipso 1y ago
  No..impairment is impairment whatever the reason for getting impaired is. Prescription meds get you in trouble in DUIs too.

Youden 1y ago

This is interesting but the causal relationship inferred by the article isn't supported by the evidence. Unlike alcohol for example, there's no clear dose of THC where it can be concluded beyond a reasonable doubt that the person is impaired. A dose that might give a regular user a gentle buzz could render a first-time user completely stoned. It's possible that these people were all incredibly stoned while driving but it's also possible that many drivers in Ohio are regular THC users and have such a high tolerance that their function is unimpaired. And like always, there's the definite possibility of confounding factors, like reckless drivers also enjoying recreational drug use. It's also important to note that the article's focus on legal limits is somewhat pointless. As there's no clear threshold above which one is impaired, the legal limits are somewhat arbitrary and are determined by other factors, like whether THC can be reliably tested at the given concentration. Ultimately, as far as I can tell, the current state of things is that we're fairly certain that THC is able to impair driving ability but we have no idea how much THC is needed to do it or how impaired drivers become. For a somewhat reputable source, the NHTSA did a report to Congress in 2017: https://www.nhtsa.gov/sites/nhtsa.gov/files/documents/812440-marijuana-impaired-driving-report-to-congress.pdf https://www.nhtsa.gov/sites/nhtsa.gov/files/documents/812440...

xienze 1y ago
> A dose that might give a regular user a gentle buzz could render a first-time user completely stoned. > It's possible that these people were all incredibly stoned while driving but it's also possible that many drivers in Ohio are regular THC users and have such a high tolerance that their function is unimpaired. Can't you make this same argument with alcohol? > Ultimately, as far as I can tell, the current state of things is that we're fairly certain that THC is able to impair driving ability but we have no idea how much THC is needed to do it or how impaired drivers become. So how exactly would you propose states move forward with allowing people to consume THC and operate vehicles (or not)? We apparently can't even tell if someone is impaired in a reliable way.
- hamdingers 1y ago
  > Can't you make this same argument with alcohol? No because a once-daily user of THC will continue to test positive for 14+ days after quitting completely. For alcohol it's 6-12 hours. They are not comparable.

baby 1y ago

We need self driving cars and VR so people can get high with their friends without having to drive

Perizors 1y ago
There are Uber, taxi and other public transportation. You can get high with your friends and don't drive. Just plan ahead.
- spookie 1y ago
  This. If you are drunk, high, tired af, or just sleepy, stop being a danger for others. A car is a hunk of metal going at high speed, stop being a child about it.
  - baby 1y ago
    Fix the system, because you cant fix the people
- baby 1y ago
  In the city yeah, I believe most drunk driving accidents happen in places and times where uber would be too overpriced or public transport too inconvenient

stego-tech 1y ago

Without getting into the technical weeds of dosing, tolerance, and impairment, this is an interesting data set that definitely paints a picture of lax enforcement and ineffective public messaging around substance use and associated impairments. That being said, I also don’t see how we can do better in the USA given that cars are necessary evils for the majority of people outside a select few cities with public transport options, and that we’ve been doing messaging about impaired driving for nearly a century to no real avail. Perhaps it’s time to start treating motor vehicle operation as an actual privilege rather than a “right” gated by a far-too-easy licensing regime. Start by tasking safety officers with more citations for things like failing to signal or illegal turns through dash camera footage (to avoid dangerous traffic stops), start suspending licenses earlier for consistent or serious infractions, and raise licensing standards higher to shunt more folks onto smaller vehicles or public transit. This isn’t a cannabis vs alcohol debate despite the framing as such: it’s a debate over how many people should be allowed to drive at all, and to what degree we hold them accountable to ensure good behavior.

cluckindan 1y ago

Really? What else did they have in their blood?

Uhhrrr 1y ago

The article doesn't mention what percentage had alcohol, with and without THC. Why not?

ratelimitsteve 1y ago

while I'm glad they're no longer doing urinalysis and convicted people of DUI based on evidence that doesn't prove anything beyond that they've used marijuana in the past several months, the standard they use to test for impairment (5ng/ml seems to be the highest standard I could find among states that have a standard) may not actually indicate recent marijuana use and in some states the standard for intoxication is so low that frequent users will still test positive after days of abstinence (https://pubmed.ncbi.nlm.nih.gov/32841811/ https://pubmed.ncbi.nlm.nih.gov/32841811/)

notmyjob 1y ago

This would be interesting IFF those drivers did not have alcohol in their system too. I can’t tell if that’s the case from the attached article. I also can’t tell whether the drivers were at fault or e.g. if they froze when another vehicle came at them, which would fit with the stereotype of the slow overly cautious stoned driver vs fast reckless driving you get from alcohol or being a young male (which might be more common in Dayton because of the Air Force base).

josefritzishere 1y ago

Correlation is not causation. I think the writer fails to make the case.

fmajid 1y ago

Why would anyone expect legalization to cause drug usage to go down? It helps eliminate the violent criminals that proliferate around any criminalized activity. Addicts are not doing drugs for the lure of the forbidden, so making drugs easier and cheaper to get should only increase usage.

couchdive 1y ago

in other news, half of Ohio county not dead in auto crashes, also test positive of THC

theturtle 1y ago

What else is there to do in Dayton?

nzeid 1y ago

Some people have hinted at the following point but I wanted to ask directly: How many recently dead Ohio adults have THC in their blood? Forget driving.

Balgair 1y ago
I mean that's a pretty good control. Granted not a lot of the deceased are going to get these autopsys though. Still, a good control group

joshcsimmons 1y ago

Seems alarmist. As many others have pointed out, THC blood concentration doesn't have the same fast decay as BAC.

pRusya 1y ago

Personal experience after living in Columbus for a few months - it felt like everybody was driving drunk and/or distracted by default. Columbus, Dayton and Cincinnati were not very fun to drive around and taking unprotected left turns felt like a gamble. On the other hand, Amish communities had glass-smooth roads and people were careful when passing horse carriages.

m463 1y ago

I remember when someone was prosecuted for a drug offense because their money was contaminated by cocaine... but it turns out 3/4 of currency is already contaminated: https://en.wikipedia.org/wiki/Contaminated_currency https://en.wikipedia.org/wiki/Contaminated_currency I wonder because it might be many people have THC in their blood for days, and it might not indicate what the prosecution infers. (I think thc lasts longer than say alcohol)

elzbardico 1y ago

Well, unless you're completely brainwashed in luxury beliefs, it is pretty obvious that legalization would increase the prevalence and frequency of usage. This is trade-off that any rational person could predict.

joshstrange 1y ago
> it is pretty obvious that legalization would increase the prevalence and frequency of usage. Except that's not at all what happened, something mentioned multiple times in the article. > No Effect from Legalization: The rate of drivers who tested positive for THC did not change significantly before or after legalization (42.1% vs. 45.2%), indicating that legal status did not influence the behavior of those who chose to drive after use.

more_corn 1y ago

Doesn’t thc stay in your blood for weeks while the psychoactive effects last hours? They’re basically saying in a random sampling of dead people 50% had consumed THC at some point in the last few weeks. So this is more an analysis of THC use in the population and in no way can be causally linked to the deaths.