Gun violence and public health--concerning language creeping in. The Colorado Sun's scare story on gas stoves leaves out detail. And, because it's Friday, a Detroit Diesel's "dog bone" head.
Gun Violence and Public Health
Using public health concepts with guns and gun violence prevention is an interest of mine. I first got interested (and concerned) when I saw how public health was used to drive policy during COVID. Think about it: if concepts from public health were used to clamp down on our freedoms then, it's not a stretch to imagine them being used in other areas that people might want to tinker in.
I recently watched the webinar linked below and wanted to share it with you. It's long. It's dry as a bone. It's also important for you to watch and consider if you too share my concerns about public health, gun violence, and gun control.
Even if all you take away from it is some vocabulary terms and an overview of some of the state of public health research into gun violence, you'll be better off. Think of it this way. If you want to join the conversation, you need to know the language.
I won't go into exhaustive detail (I'm happy to answer questions or to go over things with you if you have questions or concerns), but I want to highlight some of the things that concern me or that stuck out here. These are some general thoughts in no particular order. Please feel free to add yours to the comments if you'd like.
I am hugely concerned at the wording I hear in webinars like the one below. You will hear terms like "being exposed to a gun in a household" and allusions to guns in a house being like having secondhand smoke in the house.
My concern is twofold. One, "exposure to guns" and the analogy to secondhand smoke is too juicy a soundbite for an advocate or politician to resist. We also all know that any subtleties and caveats in social science are thrown out quickly in cases like these. Less than forthright advocates and politicians toss them out and a less than thoughtful (and incredibly credulous press) swallow them whole.
Two, guns as disease or disease vectors is a false analogy. Gun violence** is not a disease. It is not an environmental harm waiting to be visited upon unsuspecting people. Violence is an interaction of human choice and the things we surround ourselves with.
Bringing a gun into a house is not like putting an open container of radioactive cesium on the coffee table. There is (and this is missing from a good deal of public health treatment of guns) human agency in a gun's use or misuse. That is, humans must act for a gun to be harmful.
This quite fundamental flaw in gun violence public health research carries forward into any conclusions reached and renders them suspect at best.
To give a specific example, pay attention when the second scholar is presenting his work. Look at how cavalierly he traipses back and forth across this line. His carelessness could easily leave the impression (intentionally?) that bringing a gun into a house CAUSES a suicide. While it may make it easier for a suicidal person to "succeed" in their suicidal attempt, the human thought that becomes action precedes the gun being used. There is more to the story than simply "bring a gun home and watch the bodies pile up".
Now, with all this being said, there is room for public health taking a role here. To return to the analogy of smoking, when people are aware of dangers, the thoughtful among them respond and change their behavior. How many fewer smoke (and how many more wear seat belts) now than before?
Getting specific again, consider how an awareness campaign about the higher first time "success rate" of gun suicides paired with information on preventing depressed relatives from using your firearm could reduce suicides. These are the things that public health can well teach us. One thing public health professionals have considerable experience with is in understanding current health attitudes and designing campaigns to help inform us of dangers we may not be aware of.
Thus, taking off from what I said about human agency earlier, the proper role for public health should be in tackling the human element of gun violence, not treating the gun itself as a disease carrier (and opening the door for advocates and politicians words and false justification to further erode the rights of law abiding citizens).
Lastly, I want to leave you with some familiarity of what a cohort study is because the guns and public health literature is chockablock with them. I'll take the first example in the webinar, the LongSHOT gun study (they discuss the method in the webinar, but if you want more detail you'll find it in the second link below).
In many social science studies, you are comparing one group to another and then deriving from this comparison a relative risk for one condition or another. For example, let's say that I assembled a group of 100 ice cream eaters and 100 non eaters. I find that 50 ice cream eaters had a heart attack, and that 25 non eaters had heart attacks. I would therefore say that ice cream eaters are twice as likely to suffer a heart attack as non eaters.
I cannot establish any causal connection with this method, and, of course, the results depend greatly on how carefully I parse out the two groups. After all, maybe all the non ice cream eaters are more health conscious in general and have better health outcomes. maybe people that have a genetic predisposition to eat ice cream also have a genetic predisposition to have heart attacks. The likelihood of bald chance playing a role in your comparison also scales with the number of people you track. In general, the more people you follow, the better your data and the less likely that you'll see a false pattern.
What the LongSHOT researchers did here was to connect voter registration records with California's ridiculously-detailed gun ownership records, and then statistics on death to compile two different groups: those households where one member is a gun owner and the others are not and those households where no one is a gun owner. Using California's records enabled them to assemble a gigantic sample.
They then compared the suicide rates and homicide rates of these two groups finding a sizeable increased risk of both for houses where one member is a gun owner and the others are not. Pretty scary at first, but, as is always the case, you need to read and think carefully about what it is that you're presented with and where its boundaries lie.
In order to cut down on the length of this already bloated post, I will leave it to you to listen to the researchers' justifications of their conclusions and methods in the webinar below. I don't have the space to go point by point through the questions I would have for them if we were to sit down together. Suffice it to say, there is plenty of room for doubt and questions about the validity of the comparison and the methods used here.
**In both the below and other arenas, I am now hearing gun violence morph into a generalized term which includes people inflicting harm on others and also harm on themselves; that is, researchers and others are now lumping things like suicides in with homicides. Be aware of this, be aware of your own definitions (I do not personally count suicide as violence, a tragedy yes but not violence with its connotation of harm to another), and how it will affect numbers.
https://www.ncgvr.org/events/gun-violence-prevention-webinar-series/firearm-access-homicide-suicide.html?utm_source=AdaptiveMailer
https://pubmed.ncbi.nlm.nih.gov/31662345/
**Related
In order to assemble the cohort in the LongSHOT gun violence study I reference above, the researchers made cross references among gun ownership records, death records, and voter registration records.
I am not sure about you, but I have concerns about my government keeping track of how many and what kind of guns I may or may not own. The idea that a state could have data such as the LongSHOT group used, and that it's publicly available, is chilling.
It got me wondering what the State of Colorado does. I called a licensed gun dealer to inquire. Here is what I was told:
When you buy a gun, the dealer runs a background check. The information you use for this background check (name, etc.) is kept on file along with information about what gun you purchased.
These files are kept by the dealer until he or she closes up shop (at which point, I believe they may go to the ATF). They do not (unlike CA) automatically get filed with the state of Colorado. They may be subpoenaed by the government--state or federal--in the case of a gun showing up as being used in the commission of a crime, but they're not on file somewhere in Colorado.
If anyone reading knows better, please correct any mistakes I've made.
Ah, if not for the Colorado Sun, how could struggling environmental advocates get their news out there? Especially if what they're after is an article touting their research sans any serious scrutiny.
When you read articles that cite some new study, you should remember to keep your skeptic's hat on (whether or not the reporter who wrote the thing did or no).
There are multiple things to cover, but to keep this post manageable I want to focus on a couple things you should ask yourself when you read about things like what the Sun covers in the first link below.
What is actually measured? Is this realistic for how people live and interact with their world?
How do these numbers work in the context of other measurements or limits?
Cultivating the habit of asking things like this is critical because it is all too common to come across articles where the reporter who wrote it is either openly advocating for something or too lazy to ask the things above him or herself.
The thrust of the Sun article is that a group found that gas stoves emit "dangerous" (the quotes will become obvious later) levels of benzene. More damning evidence of a need to make everything electric right?
I'm not so sure. Digging into the report deeper than the Mr. Booth, the Sun reporter, chose to paints a little different picture.
In this study, the researchers measured both the concentrations and emission rate (how much is emitted per unit time) of benzene from gas stovetops and ovens in Southern California and Colorado.
To their credit, and unlike in other studies about the emissions of gas stoves, they actually made their experimental setup closer to reality in terms of how people actually use their kitchens and live in their homes.
Let's focus on the benzene concentration measurements. For these, the researchers either used what they termed an "unsealed" and/or "open" kitchen and bedroom ("unsealed" meaning no plastic enclosure but doors and windows closed and "open" meaning all interior doors left open).
Screenshot #1, from the group's paper linked second below, gives you a common experimental setup. You'll note some artificiality here, but take the word of someone who has read more than one paper on this topic, the researcher's set up here is an improvement on others' work in terms of the experimental condition matching more closely to the way people actually live.
Let's look at their results.
In screenshot #2 (from the paper), you'll see a graph of benzene concentrations in the various kitchens that the group studied. The concentrations are measured in in parts per billion. That is how many particles out of one billion randomly chosen particles are the target chemical. E.g. if you have a benzene concentration of 4 ppb, 4 particles of every billion are benzene.
The green circles tell you a baseline for the various kitchens studied. That's how much benzene is there with no oven running. The black line (which I highlight with a blue arrow) is the amount of benzene that would be in a house's air if there was a smoker living there. The orange circles and black triangles show you the concentrations of benzene for homes where you have a gas or propane burner or oven on high. The biggest concentration is from the propane range/oven on the far right and it's about 7 ppb.
Now look at screenshot #3 (again, from the paper linked second below). This is graph of benzene concentrations measured in a room farthest from the kitchen. The researchers put the ovens on 475 degrees for an hour and half. They say this mimics baking bread, but as someone who has made lots of bread this edges on not being realistic. The closest thing to this I could think of is baking a lean crusty bread going 500 degrees for about 45 to 50 min. The researchers then turned the ovens off and let their sniffers go for 6 hours more.
Six different kitchens are shown here, picked because they had the highest emissions in earlier trials. Notable here are the two different US standards that the researchers highlight in the graph. The upper black bar is the CA standard for non cancer exposure to benzene and, below that in blue that I drew, is the US Agency for Toxic Substances and Disease Registry (ATSDR) standard for benzene.
Houses 1 and 2 have excursions above the stricter ATSDR standard of about 2.5 hours and 1 hour respectively. The other 4 houses do not go above the ATSDR standard. Going back to screenshot 2 for the kitchen concentration for various stove burners (or an oven at 350 degrees), you see 4 of the 16 houses going above the ATSDR standard.
Scary stuff if you're in those houses no?
Maybe. Again, for a full and fair understanding here, you should know what the standards are and where they come from. Looking up the 365 day airborne minimal risk level (what I believe the researchers were quoting--they didn't give too much detail) for benzene in the ATSDR paper on benzene (see the third link below), you will find what I attached as screenshot #4.
Since we might expect people to cook nearly every day, it's reasonable to use the 365 day exposure standard for gas stoves in a home. The thing is, and you'll see it highlighted in blue in the screenshot, the 3 ppb limit was established by using Chinese shoe workers whose job exposed them to benzene daily. To make this analogy work here, we'd have to have people exposed to 3 ppb for the equivalent of a full time job. That is not a reasonable analogy to using your stove a couple times a day for short periods to make a meal.
Further, not everyone agrees on safe limits. The workplace exposure limits for benzene (for full time work) put out by the CDC in the 4th link below and highlighted in screenshot #5 show levels that are 3 orders of magnitude higher than that of ATSDR. Said another way, the ATSDR exposure limit is 1/1000th of the limit for workers given by the CDC.
Including that line on the graph of gas stove emissions for this study would make it so you couldn't even read the dots for even the worse kitchen, they would be so far down at the bottom. 3
What is the point here?
Read carefully. Make sure you follow back and check the context.
Ask yourself simple questions such as "what was measured?" and "how do those results fit into the world?"
All too often, the reporters who cite work like this are unable or unwilling to go into the kind of detail needed to give you the complete context of the research. They are also sympathetic to the cause of the researchers and liable to let their confirmation bias guide them into credulity.
Sadly, despite it being their job, they won't do this work for you.
https://coloradosun.com/2023/06/23/colorado-gas-stoves-benzene-danger-stanford/
https://pubs.acs.org/doi/10.1021/acs.est.2c09289
https://www.atsdr.cdc.gov/ToxProfiles/tp3.pdf
https://www.cdc.gov/niosh/npg/npgd0049.html
Chew on this “dog bone” head from a Detroit 8V-71
Last one of the day and you know what that means, something for fun and not related to politics!
Earlier this summer we were working on rebuilding some of the cylinders and the head in my dad's old semi (a 1965 GMC 6500). The engine is an 8v-71 Detroit. Translated to English this means it's a V8 two-cycle diesel engine, and each cylinder has a displacement of 71 cubic inches.
The head (the part of the engine that bolts down on top of each cylinder and has all the valves and etc. to make the engine work) needed to be rebuilt and machined. Also, 3 of the 4 cylinders on that side needed new piston rings and liners. When pistons are loose in the cylinder the engine can't develop as much power.
As we were reassembling the head after getting it back from the machine shop, I saw something that I thought you might find interesting. Look at picture #1 attached.
Do you see the part circled in red? If you do, it will be small wonder as to why these heads are called "dog bone" heads.
Here's the question before us: what is the connection between these dog bone shapes and the metal fingers in picture #2 attached?
Got an idea?
They are both examples of devices to provide relief for thermal expansion.
All materials expand (and contract) with the addition (or removal) of heat energy. I forget if I've mentioned it in the past, but you can think of temperature as being a rough measurement of how quickly the atoms and molecules that make up a substance are moving: increase the temperature and the little constituents of matter start to move faster on average.
It makes sense, then, that when you add heat to a body, the temperature rises signaling that the little particles are moving faster, and particles moving faster means they'll need more space. The analogy I always use in my physics classes is that you can safely fit more people into a dance floor when they aimlessly shuffle than when they are really zipping around.
What does thermal expansion look like? What are the controlling factors?
If you look at screenshot #3, you'll see the equation that describes this (at least in a linear sense, i.e. along one dimension). The things that control how much something will expand are:
--It's current size. If it's bigger it, will expand more.
--The change in temperature. If you change the temperature more, it will expand more.
--The characteristics of the material. This information is encoded in the expansion constant. Some things just expand more than others due to the shape of their crystals, the type of atoms, etc. If this number is bigger, the material will expand more all other things being equal.
So, for example a 100 foot reinforced concrete bridge girder will differently than the same 100 foot steel bridge girder. And a 50 foot steel girder will expand differently than a 100 foot girder.
The expansion in the girders is what is underlying the steel fingers from screencap #2. They actually covering a large gap in the road that is there to allow the bridge girders to expand and contract with heat and cold. Without them there would be quite a bump in the road, especially on cold days because a big gap would open up.
And since girders are different material and operate under different temperature ranges than engine parts, the gap for expansion looks different when you bring it to the level of the head in a diesel.
The 8V-71 in my dad's truck doesn't have as many passages for water to circulate, and thus it gets a lot hotter than later models. When engineers made this engine originally, they didn't have these dog bone relief grooves and they noticed a big problem. The engines were failing quicker and developing problems in the head. As the engine hit and stayed at a high operating temperature, the heads would crack because they wanted to expand and the only way they could was by warping and cracking.
After considering the problem and what to do about it, the solution became clear. They milled this shape out of the metal to allow it room to expand as it got hot. It's an expansion joint, similar to that in a bridge.
One last bit of trivia. If you didn't before, look again at picture #1 and note that the edges on that bone are rounded. Why? The rounding is there because squared corners concentrate stresses in a material. When you have a relief cut out, you want the corners in the shape to be rounded so stress isn't concentrated there. Another way to minimize cracks.
That's it. Have a good Friday!