A long it takes you to drive 3 hours. Oh, really? Joint, that one accident was oh, typically 2.5 That's true. Because I remember that for me I think was 2.15 minutes. That was hours fast. Well, no, I think it's like from like so I basically live here 10 minutes walk 12 minutes, walk to Lamont. So that's probably not living right at Lamont. The Pre. Yeah. Okay. I somehow thought because everyone was like when I was there, everyone was talking about they live in Manhattan. So I was like I used to. Uh huh. My mother in law was not happy with that. Yeah. Control compartment. Oh, I never would have. Yeah, three bedroom. It's in Manhattan. This sounds like a great a great I just came back from bears like during the Labor Day. Yeah. So this starts in a few minutes? Yes, I think so. I feel like because we just have coffee hour. What time is it now? Okay. Even 30? Yeah. Okay. I'm going to do one thing for my son. Okay. Yeah. And I can introduce you while you're doing these things. Launch. Luis. Just Dr. Dr. Jim Biddle. Are you alone? Where is your colleague conference? I currently alone and hoping someone comes to keep me company, just go out the door and search out as well. Just tell you, one of our classes is doing the Brandywine River today, so a lot of people had to drive early. Holly men to meet. You just want to me say hi because she has a whole day man today that she has to deal with. Yeah, no sugar head of the gun and there are lots of things. Yeah. Yeah. I don't think should we just start? Yeah. I just Go ahead. Okay. Yeah. Okay. Well, so it's my honor to introduce Dr. Steven Chert. Steve is Lamont Research Professor at Dot Earth Observatory of Columbia University. He is trained as a geochemist focused on analytical chemistry. I asked to come up with a title for the seminar and he basically give me five or six to choose from, which really illustrate his wide interest in environmental research. He's currently served as the Director of Exposure Assessment, Core facility of the Center for Environmental Health in northern Manhattan. And he got his Phd from Columbia University, and he never left. Instead of hearing something about learning science, usually what we study, I ask to share something he's most exciting about. Now sounds like we'll hear something about air pollution in Aa. Thank you. So I was born at Columbia University, so I really never left anyway. So we're going to talk about household air pollution today and specifically about how, what's normally considered meta data wearing compliance for a personal monitor. Did they really have it on their body or did they leave it somewhere for a little bit all the time and how that really helped analyze the data. And this is part of a very large collaboration. We work with the Ministry of Health at Ghana. They have three research divisions, one for the southern third of the country, one for the middle third. On the northern third, the Tampa Health Research Center is the middle third. It's a mainly very rural with a couple of cities. But that's not where we do our work, at least not this work. My outline for the talk today is I'm going to talk a little bit about how a sediment geochemists got into the world of working with public health scientists and specifically on air pollution. And then I'm going to give you a intro to what is household air pollution and why is it important? Then I'll talk about the study. I'll talk a little bit about future directions then if there's still time left. If I don't go too much into the weeds, I'll talk a little bit about the diversity equity inclusion issues that are central to working in low and middle income countries on science projects. What I did for my dissertation, what I did for my postdoc, what I thought I was going to do for my whole life was reconstruct sediment transport by taking advantage of contaminant tracers of fine grade sentiments to see where the Ss moved, recreate the contaminant. Do I need to talk somewhere for the speaker to pick me up? She be good? Yeah, you good to reconstruct the history of contamination and investigate policies and whether they were successful or not. That's what this is a summary slide of. It is one of the publications where as part of this large superfund project and the super fund was how I was first introduced to working with public health investigators. Because it's this great mechanism that requires them to collaborate with environmental science or environmental engineers. And we were invited to collaborate and here we were trying to understand PCBs and metals and how they moved around the city. As part of that, we went around collecting end member sediment cores that represented different types of sources. Here we went to Central Park Lake, Rowboat Lake. If you've ever been to the Central Park and been out on the rowboats, that's where we collected these. We went there to try to understand how much leaded gasoline fell into the into New York. And it turned out to be a lot more exciting because it turned out that lead and gasoline had very little to do with this record. Not that there wasn't a lot of cars going in and out of New York City, but it got dwarfed by this incineration signal that just dominated from the '30s to the '70s. And so there was a lot more lead in the garbage that was being burned than the lead in the cars going in and out of New York. Well, this was highly contentious, and even my public health colleagues didn't believe it. They didn't believe that a sediment cores could tell you anything about what's going on in the environment. Certainly not what people are breathing in. This was probably big particles, not small particles. I went and found the Department of Energy had this great time series of filters that were archived in lower Manhattan from all over the world, including New York City. And I made a few measurements and I went to an Earth Institute talk. Earth Institute was this umbrella organization trying to get the university to work more together. I gave a flash talk and I showed about three slides of these air filters. If you notice on the sediments where this is that in the core time that we did some data, the core had to be multiplied by 20 to be on the same scale. Well, our T is in the air filters is also 20 and the time series overlapped. Well, and that was basically all I really said. And this guy walked up to me and said you can measure metals on air filters. Let's you're at a proposal to get I totally pivot and my research barely had. I only had one grant after this one that worked on sediments. I've had a lot of grants working on air conditions. I still do some groundwater work. Quite a bit of groundwater work. This was the next study we did. It was all about the subways. That first study I showed you in this study is what got me ten year. This is specifically the students always be open to the pivot, because the pivot might be where your future is. So household air pollution. Household air pollution is jargon like lots of jargon we use for the cooking and heating done with solid fuels. Typically in unvented stoves and indoors. Turns out we work in Ghana, it's typically outdoors, but it's solid fuel. It's wood and charcoal and crop residue. But the type of cooking they do is right over the fire. The cooking, this very viscous dish called, they're still heavily exposed at the time of our proposal. The global burden of disease estimates put household air pollution as the number four top for modifiable risk factor in Ghana at the time it was number one. Now these estimates change radically year to year as they get better estimates and more data. It's still in the top 20 and it's still suggesting 1.8 million deaths per year. With the estimates that were done in 2019, a large fraction of the people are exposed. In fact, again, the estimates keep changing. At the time of the proposal for the 3.5 billion people were supposedly exposed based on surveys. Now that's down to 2.8 billion. That's pre surveys that we are doing a lot to try to get people off of solid fuels and onto cleaner fuels, but it's a big deal. 1.8 million deaths per year. A lot of deaths less than age. 560 million daily adjusted life years or something like that. I always forget the daily being. Our study, we chose to work, we actually did a whole bunch of pilots in different African countries. We chose data because there was a direct flight from New York City, it was a stable government. We've been working in Bangladesh a lot. We were always worried of being kicked out. We chose data. We're really glad we have, because we also found a really good local partner with this Ministry of Health Research Center. And that's why we chose these trends and sell fuel use. These have been going, but notice that in Africa, it's a much smaller decrease. Actually, I just saw a headline yesterday saying that the total number of people on the African continent cooking on solid fuels is actually increasing, even though the percentage is going down. And so it's the rapid population growth that allows that to happen in Gaza. Specifically, this use of wood is a major issue for sustainability because they're deforesting their country, especially in the north and the desert is moving south because of that. But even in the area where we are, which is in this transition zone, wood is still over harvested, partly due to their own cooking, but mainly due to the fact that the supplement, their farming income, they make a lot of charcoal they ship to the city. If you've ever been in this part of the country, you just see charcoal trucks one after another going off to the city. It's a very inefficient, artisanal production of charcoal. What farmers can do on their own for a really long time, actually still going on, there's this big debate how clean is clean enough, improve health, and have an impact on sustainability. How do you get there? Do you make the available clean? Do you make biomass stoves better? More efficient combustion chambers? Locally, people can provide income as they have a local fabrication and will be acceptable. And all these things make the available clean or make the clean available. Here's a whole bunch of examples of these improved biomass. But almost every study has shown that, though they work great in the lab, they fail miserably in the field. It's really hard to beat the flexibility of a three stone fire. You can put any size piece of wood on it. You can put a massive pot to heat you. Washing water on, walk away, come back 3 hours later. Your washing water is ready. None of those improved biomass can do that. They all require constant feeding. That is the Achilles heel. The opposite is make the clean available for Ghana. That means LPG. They have fossil fuels that they're developing off the coast that produces LPG, which is a mix of butane and propane gases. They have a source of it where electricity. Even though there's been a huge electrification while we've been there, it doesn't have the bandwidth to deal with everyone cooking at the same time. But hopefully it will in 2030, 40 years, but not now. The big problem with LBG, even though it's locally made gun, is it's really expensive. Much more expensive than telling your pregnant wife to go out to the forest, chop down a tree, and haul it back. It's the women that cut the firewood and haul it back no matter how pregnant they are. However, work that we've been doing, we've realized it's very similar cost to charcoal in the cities. From a sustainability point of view, there might be a real opportunity to get people away from charcoal in the city's LPG because the cost is similar enough. We'll talk about that later. The graph study was designed to provide evidence of the efficacy of introduced during pregnancy to improve the abstention outcomes, birth outcomes, and infant respiratory health in the first year of life, which is when a lot of infants of lower respiratory disease and pneumonia specifically efficacy means what another lot of jargon intention to treat and randomized control trials is. If you ignore exposure and you ignore a lot of things and you just randomize everything really well. Then you just look at the average health outcomes in your control arm versus the average health outcomes in your intervention arms. And you do this differences differences, average is your intervention arms better are 02:00 P.M. Study hypotheses, birth weight and a significant reduction in the great in position and severe pneumonia during the first year of life. There were a bunch of secondary hypotheses, but these were the first two. Our study design was we had our traditional three stone fire. We had two bild stoves we gave out. These were improved biomass stoves that had a thermoelectric fan that really improves the efficiency of combustion. As long as you check up the wood pieces and set it every 5 minutes. We did pilots, they said they loved them. They lied to us. It's an issue. They have done so many studies because of the Ministry of Health has got a big presence there that they've learned to tell us what we want to hear, that's a big problem and working on are still today, you really have to work to tease out what you really want to hear, which is what they believe, not what they think we want them to believe. We also had an LPG little bit less than the other two arms is more expensive because we thought it would be such a big reduction in exposure that we wouldn't read such a big arm and had two partners. And we gave them three gas during all throughout the pregnancy and for the whole first year of life. And then we cut off the fuel. The control arm could get the stoves at the end of the study, but no free gas. Our method was to randomize based on community. He could be an LBG community or you could be a bio community, or you could be a control community. But our level of intervention was on the individual. We might have 20 individuals within a community of 1,000 That was the LPG R, but that means those 20 people were surrounded by people still cooking on wood fuel, very compact communities, and sometimes you could have four people as close as this, one of you would be the pregnant woman cooking on the LPG, and the other three we cook on wood. I'm predicting some issues here with our study. I was actually brought into the study after it was launched. The very first time I worked in the community, I was like, this is not going to work just way too densely packed We did a lot of pilots trying to look at area monitoring which is often used in indoor, truly indoor studies where there's a kitchen. But even in those, the other work has shown such huge variation in vertically and horizontally in an indoor setting that if you really want to know what people are exposed to next to this big point storage, you really have to have personal exposure as well in an outdoor setting. That's even more true if you have an area sampler and the wind is blowing the other way your area sampler is picking up anything. We did personal sampling, we primarily relied upon the very in industrial hygiene monoxides made by last then. We had a fairly expensive S center that was very innovative at the time that had a mini Nehme, a side selective in let dual station factor that would take out the big stuff and really just let the stuff define particulate matter smaller than 2.5 micros through after it went through the sensing chamber of the Nehme collected on a filter. This was great because we could then calibrate that light scattering device to a real physical measurement. The weight on the filter for a certain amount of volume of the filter, which is the definition of what PNG points, a mass definition. It also had an accelerating it, which is opened up this window into the analysis we're talking about today. We could know when they're wearing, when they're not. This is the way we had them wear sensor. When it rains there it is a deluge. For about half an hour, hour, we needed to protect it from the grain. Such fue, a bunch of holes in the bottom and let the gas get in that way. This is the harness that would allow the micro pen to be worn to coming out of the breathing cell right next to the CO monitor. The hope was that we could just do a subset for the PM and we get this great correlation between the PM and the CO. And then we could fill in the PM data for the that also failed. That's been seen in several other studies also. But at the time we proposed it, it looks very promising. Here's our repeat measures. We had a baseline before the intervention of just CO. After the intervention, three weeks go by, and then we did a new round of both the PM and the CO. Two more, so we had four repeat measurements during pregnancy, and then we did three repeat measurements of CO. After the child was born, we remove both the mother and the child. Then we did PM for one of those just on the mother. I tried to get them to do it on the child, but they wouldn't let me and it was a big site. 14, 14 women live births that we've followed. This is just an example of accelerometry data over a 72 hour period. You can see much less variability when they take it off. And the Ssn, a lot of variability during the day, they took it off. Here during the day, it looks like they took it off. This is a nice example of knowing when they wear it when. This is an example of a raw Natalie data corrected, maybe for a baseline shift, it has a lot of baseline shift. You can see it's very spiky. Going up to over 2000 nominal micrograms per cubic meter. And if we compare this to a filter weight, our corrections were anywhere from, typically from 0.5 to 2.5 but occasionally they were much different than that. So that individual correction accounted for units that might have a misaligned laser misaligned detector, as well as differences in the optical properties of the particulate map, because it's a light scatter, right? So if a particle is white, scatters really well. If a particle is black, doesn't scatter so much. So that's why that individual level calibration is so important. Since this is going to focus on compliance a lot, I want to show what we got for compliance here it is hours per day. And this was looking between 06:00 A.M. and 10:00 P.M. and up to 16 hours per day. And we have a huge berry building going from basically zero to basically almost 16 hours a day. But the medians were very similar for the control of the events was slightly higher for the LPG. There's an issue that might have to be corrected for, but overall you can see that there's just a lot of variation And that the vast majority of the people into Tyler range went from just under 6 hours per day to maybe other arm's going down as low as 5 hours per day up to about 11 hours per day, which is actually really good for these types of studies. In fact, amazingly good in US. Studies, we hope to get the 6 hours because people just don't like wearing this type of equipment. Here is what is the distribution of the real time data across the whole study, including all the arms and everything. And so you can clearly see the cooking periods. The morning cooking period is not everyone cooks in the morning, a lot more cook afternoon. And you see that in the data averaging across all these people, some of the people who are wearing it, so people who are not wearing it, this is not directly to that yet. You got 90 percentiles, up to 175 Dro grams. And, uh, in the morning, almost up to 100, but the medians, we were much lower than that. So this is the big fight. When the graduate student came to me with a spot, I was like, wow, we're going to make something of this. You know, the bio control arm looked down and the LPG R was very, this was really upset and it was exciting because I'm a simple geochemist, I like simple interpretations. Why intercept we can interpret. We got a trend that ever plateaus off. So it looks like really the people who wore the devices more really capturing more exposure, probably capturing more of the. And we have this huge difference between the LBG clean fuel and the control of the bio light. We also see that the biol, horrible. So we spent a lot of time thinking about this. We actually modeled it using a mixed defense model. You know, but, but the interpretation you can get just from intercepts, slopes, differences. So we're going to walk through that. And one of the things the model did was re, multiple measurements because they were slightly different. We hope to have about two thirds of the cohort to wear this device. And we hoped to use the same people for the first measurement. In the second measurement that didn't happen, life gets to, on average, we had about 1.5 measurements for a participant. So the effect of the intervention, the differences in difference approach says that we've got about 32% lower exposure in the LPG arm and about 11% lower biol arm. But not sign, I don't think so. But if we look at this and we say well, our best estimate of the overall exposure for an arm is at the highest, at the highest wearing compliance. Then we can take the difference at the high end and say that's the effect of the intervention. If you do that, we get, you know, no effect for the bio light. And we got about a 45% improvement on the LBJ. Was a range -12% to -77% if you look at the 95% confidence. So this is bigger, but I think it's probably closer to the truth to what we did in the main traditional outcome paper. Well, what's our interpretation when I didn't wear? Well, if you're not wearing the mops, monitoring whatever space you left it in here. We took the leap of faith to say because air exchange is so high in their homes, because typically they have the doors open, the windows open. They often have fat shrews. So there's others that have a metal roof, but typically with a gap to let some exchange. So because of the really high exchanges, which the literature does confirm, we interpreted if they weren't wearing it, it was then. Exposure representative of the community at large, not the ambient outside of the commit. So they're all pretty similar. Here are the levels from 37 up to 48 with 95% confidence intervals that all overlap. That gave us some confidence in that interpretation, but we had nothing to compare it to. We didn't have any ambient monitors. We couldn't afford them at the time. We were so overstretched. Adding the LPG, thank God we add the LPG. Otherwise, it would have been a lot of money. Just say once again on truth, biomass. But we had no ambient monitoring. In the paper, we did do some remote sensing work, but it was on a ten by ten grid. These communities are only 1 square kilometer out of that 100 square kilometers. Maybe it's 10 square kilometers and it was close, but they were lower. But you'd expect them to be lower because outside of the communities there's nothing, just forest or farmers fields. The other interpretation that we made was what I call the direct effect of cooking over the fire. Remember, these women are right over the fire stirring this viscous cassava to make the Baku, and the fumes are just going right into their space. And so if this is the unity, the difference between the y intercept and the value is the direct impact of cooking over the fire. On average they are that, that was our interpretation. And here you see a very, very small 55 feet at the high end, 46 y intercept. Only 9 micrograms/cubic meter, compared to 52 for the three stone five. Well that's an 83% reduction. That's awfully close to what we see in the lab. 90 to 95% reduction in the lab for wood cooked in the lab versus LPG in the lab. This is the first time anyone's ever come close to showing that LPG can have a really small anything close to what the emissions are in a lab setting. I was really excited by that. Only confirmation of the line, except at the time of this publication was saying, well, there is this dusty season in Ghana and most of this area after that, when the dry season, the winds come down from the Sahara desert, picks up a lot of dust, makes it much higher than a poll Yt did go up. But if we subdivide the data, just look at the Hermon versus the non Harmon. A lot less data during the Harmon because it's only a few months of the year versus the non Harmon. The non very si picture as before. There's a lot of data still and that was at the time. But since then we've put in low cost purple air monitors. How many of you know what a purple air monitor is? It's these little $250 light scattering devices can get. I think they're going up in price, but still very similar to the light scattering device that was in our personal air monitor, but these stay six. And so we have this data from 31 of the 35 communities we were in. And here it just shows one year of data basically. And you can clearly see the Hermon on average theory 31 line plots is much different than Heaton period and we have another year of this data but the student didn't send me the two year. If we compare the Hermaone averages from the purple air to the non hemat, we get very similar values. Also, Kevin calibrated these purple as, but we used an algorithm generated in a very different environment. So it might not be the right calibration, but it's, it's better than not doing anything to this raw data and it makes it look similar. So this is the next piece of evidence suggesting maybe the Y intercept is interpreted as an ambient measurement is right. So our basic conclusions focusing on the wearing compliance method is that personal monitors really need accelerometers. Especially when you're focused on major point sources where behavior is what drives the exposure. It might not be so important if everything is ambient. When the wildfires come through. But it's really important when you're close to these point sources. As in the slope of wildfire is far away. Where you're just plan, not when you're close to a wildfire then that's just like another point source. And this slope in acceptance by modeling approach allowed for three effects to be estimated. Two of which you couldn't get without the wearing. This is the only one that done without wearing appliances. The effect of the intervention on exposure. Now this was a health study, but I'll tell you a little bit about health results. The intention to treat results were not just like all the other randomized control trials on household. However, the exposure response, again very similar to other studies, showed really strong effects on physician assessed pneumonia and severe pneumonia and for a whole host of other health effects. Still showing that this pregnancy in the first year of life is a critical window of exposure. Now we've been following these children for longer, and the LPG stove did improve growth on average as kind of a intention to treat it also improved lung function. These are our latest results. This is super exciting. This is the first intention to treat related to a cook stove intervention that's ever been seen. And it was so exciting that they gave us another five years of money to continue following the kids for lung function. We also had some cardiovascular results that were exciting. So we got funded to look at cardiovascular health. So what were the policy implications? Well, the biomass improved stove was not effective at reducing exposure, confirming the need to move to clean fuel interventions. Now this is an important policy because there's been tons of money, and there's still tons of money being put into improve biomass stoves. In lab settings, improved biomass stoves work great in field settings. They've been shown some of them to reduce exposure or reduce emissions a bit because they're more efficient. But they don't change exposure enough to health. If you look at a, at a relative risk versus exposure, it's not a straight line. It's actually, there's a see part of the curve and then it bends over. At the time of these studies, it wasn't all that well filled in. This far end was from active smoking. The next data point was from passive smoke. And then there was this huge gap. And they just assumed it was being bend over. All the cities and lower and middle income countries have gotten really dirty, really filled that in and it really does bend over. And so you need to reduce exposure to affect health. You might be able to affect sustainability in the amount of wood that's being burnt by a little bit. But even there, typically the people don't like the stoves. They don't work as well as a three stone fire, so they continue to use their three stone fires. Right now, there's still a lot of carbon credits going into people getting out the stoves, and yet everyone really knows they don't do it, that is a policy. So we saw our direct exposure was very low, but that our overall exposure is pretty high. And that we interpreted that that was due to just the integration of all the other, suggesting that individual level interventions based on this one study are not effective. But it was also true in a lot of other studies because they've all been no suggesting that we need community level interventions necessary to make big changes of health. And this led to privately funded Columbia World Project, where we were trying to get ready to do a large community scale intervention in a rural setting. But all our work that led up to it really convinced us that it's hopeless that you'd have the demand curve for price is that you have to give it away for free and that's just too expensive. And that's when we did the demand curve in the cities. That's where we got really excited that we did maybe have a big impact on the charcoal heaps, which is still the major consumption of wood that's leading to their deforestation and sustainability issues. We're now pivoting to work in peryurban areas and having an app. That will allow the big difference between LPG and charcoal. You can go buy charcoal and at whatever you want, it's done by the weight. They'll give you just what you need for me, LPG, you buy this big canister for the month. That's a lot of money for a family to outlive. We've worked with a local company in Ghana to build a mini banking system where they can put in money little by little bit electronically. All use electronic money now anyway. And then if they're short when they need a tank, we'll give them micro loan that they'll have to pay back. But again, a little bit by, little bit so that they can keep on using LPG. That intervention is going to launch sometime in January. Maybe we're excited by that now. At the same time we just got these results from age four, just Smith and health improvements are done by intervening on the individual at these critical time windows which then opens up. Maybe all these null results were just because they haven't followed the kids long enough. So that's why we're going to follow the kids longer. That's why the other big study there was a multi site study. Love our results also, they're going to try to follow their cohorts longer reinvigorated the field. The program officer, when we called, called her up to pitch a renewal was like, I'm done with household air pollution, everything is no, it's time for you guys to move on. And then we showed these results and the growth results and her eyes lit up and we got funded again. So where are we going from here? We're going to take advantage of, we haven't taken advantage at all of the continuous data on the CO, well, we have one paper on the CEO, a very small pilot on blood pressure, but we haven't taken advantage of the continuous minute by minute or hour by hour data at all. We're looking at health effects, so we're trying to each students to work on that. We have a huge series of biobank samples. Urine, placental tissues, children's baby teeth. Think we have hair, we have a blood spots, we have all this stuff that should allow us to ask more questions that we can answer. Looking at that, we're about to start analyzing the filters metals. I've already done a lot of the black carbon, but we're not quite done. Saves a lot of filters. But those are non destructive analysis. We can still do additional analysis on the filters. I already told you about the Pm World Project, and we have two new pants in addition to the purple air, low cost sensors, we just got funded by the Air Fund to put in community. The schools are on the edge of towns in these communities. We get funded to put air monitors a little bit higher quality in the centroids of the communities related to a new pregnancy cohort that Gates is funding that had no exposure assessment at all. We still have pending or plan proposals on neurological development, climate change impacts, and a host of other things. With that, I'd like to thank our funders, like to thank the study participants and the field workers who work incredibly long hours, six to seven days a week, 14 to 16 hours a day. These workers really work hard. If you want more details, you can go to the published manuscript and check out the supplement because it has nice cartoons of how we interpreted the data. We can stop here and take questions, or I can make some comments about diversity, inclusion, equity, and accessibility, and working in low and middle income countries. What were the incentives for? The incentives for the participants was a bar of soap, so you have to pay money to get, so there's a big, so they have their own IRB and their own IRB is very sensitive to overcompensated. Don't, don't want to because the people are so poor, you don't want them to be part of a study just because they're giving something. So they have really low incentives. But they do want them to be meaningful. At age 78, we were running into the issue where they stopped wearing. The Amity said, well, in the US city setting, we pay incentives. We pay a large incentives for people to wear a monitors because it's a pain. We were going to give a kilo of price, or half a kilo of price, and they were like, whoa, whoa, whoa, whoa, whoa. That's way too much anyone would do anything for a kilo of rice, especially depending on the type of year. There is a type of time of year when the farmers don't necessarily have enough food to a kilo of rice at that time of year. Way too much from the ethical point of view, according to the local ethics sport. We went back to a bar of soap for the mother and to a classic reusable water bottle for the child. That plus re, sensitization in the communities about how important was to wear the air monitors. Encourage the people you see that are wearing it. Oh, you forgot your air monitor. Go back and get it. That changed our low war and compliance up to even better than what you saw there. So it doesn't take much of an incentive. Was it clear to me what carbon monoxide was for? I assume PM 2.5 was the culprit. Are you using COs a surrogate? So, even PM 2.5 is a surrogate, right? Pm two point. So, emissions from a fire are very complex. They've got all kinds of things in H what exactly is doing that. And so that's an area where we could try to take advantage of the filters that look into that or the bio samples that look into exposures. There are no good biological markers of exposure. There are some that have been proposed, but they always seem to fall apart. And so yes, both CO and P 2.5 are proxies for exposures of the smoke. The CEO was the primary exposure metric that was done seven times because it was affordable. The PM is thought to be the best proxy of the complex mixture, but it's a lot more expensive to do so, so we did a much smaller size. So when we, when we've continued to study, we shrink the covert down a lot and we're just doing, sorry. Where you burn wood. Big or small particles. What's coming out in terms of particular matter, Chemically, Very the composition wise, it's very different from what's coming out of you say propane and butane? Yes. Right. Are we comparing apples and oranges? Yeah. And I mean people, well like the godfather of household air pollution, it was Kirk Smith from Berkeley. And this was a hot topic of what should we be measuring, should we be measuring all these chemicals, such a complex mixture. Pm 2.5 itself is a complex mixture, right? It's just what you weigh in the filter. And his advice was, don't go down that graphical look at what happened in the cigarette smoke. They spent hundreds of millions of dollars trying to find out what exactly it was. And they have some indications, but they don't really know. He goes, you know, the money is so critical, these settings in countries, let's not go throw that much money into what's the key for that being said, we have some hypotheses that are testing love function and it might be some sugars that are, Yeah, we do have a mechanism and the no sugar but not sugar. Sugar was converted into something. Yeah, sugar in the well would gets converted into sugar that you would have that is active. There are, there are sugar the results on an exposure to 2.5 are really convincing. But then the connection to health, I mean, obviously we know that connection exists. But you're, you're providing the mothers with a whole bunch of gas, you're receiving a bunch of money, you're completely changing the way that they're tending their they're heating. Would the right comparison be to give both control people? Helper who brings the free charcoal and puts into their sofa for them. That's a really good point. There is one thing that we saw was very different than the LPG women than any other breastfed children. Why? Because they weren't going down to the forest with a machete down the tree. So that could be. We all know that breastfeeding is healthy. So that could be the mechanism and someone needs to look into that. The study design would be great save save them the same amount of time so that they can have the same benefits of prosecuting kids longer than it's just the exposure right on the money related to James question, I'm sure you look at it. We wouldn't to participating LPG studies. These pregnant women, are they do you see like any socio economic difference or like education difference? No, the grammarizationd. Good job at that. You know, it was the Ization that gave the LG. At least that's my memory. It's not the thing that I look at, trust my colleagues, but my memory is that now there's all kinds of interventions where the government, you know, once the government takes on a policy, we've done interventions in Bangladesh, that once the government takes it over, the intervention is horribly skewed to the wealthy. The deep wells in Bangladesh that were done by the government all went in the back yards of wealthy individuals that then did not share the deep well with the rest of the community. That was a massive trucks. Yeah, yeah. Okay, cool. First of all, a great talk. Thank you. So I was wondering about these stoves. So they get the LPG stove and they get the free gas for the first year of their child's life. They can't afford the gas after that fact? Generally, yes. So that's something I should have said, especially when we're talking about the four. As soon as we stop providing the fuel, 98% of people stop buying fuel. It's just so expensive. Yeah. A farming family to buy the LP and you know, they're used to telling their pregnant wives to go chop down the tree and call I still on the I just have this vision of this very pregnant woman machete. Just whacked down the tree as like seed to my mind. So as soon as intervention ended the exposure and we had data that came back to all being the same across all three. And so that was a good question, which is why as we continue it, the real exposure differences as well as the lactation differences all from that pregnancy in the first year of life. After that, everything went back to the same. And so it's very powerful for looking at this window. Just well question when you look at the slope and use the PM 2.5 as the exposure index is because there's a intercept. And do you think like, I guess, how does the indoor exposure, PM 2.5 from the stove cooking compare with the ambient light? Do you think that's right? So that's the question. Since we've added these ambient monitors and we're continuing, we can begin to really look at that question. And we really haven't taken advantage of the real time data disabled, but even the ambient monitors are lower at night. Just saying that PM is lower at night is not telling you whether it's different from ambient. I will say another thing that is really different in Ghana, even in the cities then in the West, where do we spend most of our time indoors or outdoors? Indoors, 90, 95% of our time is indoors. Exactly. Slip the other way, the only time they're indoors is sleeping even in the city. If you go into an apartment of someone in the city, there is nothing in the apartment other than what they cook, whatever they sleep on, they have nothing in their apartment. They're always out working, or socializing or whatever they do. They spend so little time at home. That's also very different. In fact, I was just reading a paper yesterday that estimated ozone exposure and health risk and lower middle income countries. They show, as you would expect, that Subsaharan Africa has really high ozone, super hot. But then they predict the health outcomes for ozone is much lower than in other parts of the country that have similar or even lower ozone. Because they predict that the kids are only exposed to ozone 16% of the time I was like, that is just dead. These kids are outside all the time and so something needs to be fixed there Or I didn't understand there's something in the paper. I didn't understand sides. The dry season, the homicide caused an increase in particulate matter, but that might be different than like the particulate matter being produced over a stoke. Very different. He doesn't translate to any health outcome. Different health outcomes depending on when a baby is born. In the excellent question. Something we want to look into in that new study with prisma study. But in general, when they've looked at health outcomes, there's hints of differences that the sources combustion is thought to be more toxic dust. At the same time, I'll tell you if you're there for more than a week in the dusty season, you have a headache, you do not like it, it is miserable, and everyone is so pleased when the dusty seats, there are some hypotheses that the dust is carrying more than just dust that's carrying meningitis infections with it. And they don't really know if it's carrying it or if it's just making so much weaker then the infections can penetrate better. But yeah, that's, that's something that this new cohort with all this ambient data could try to put. Because it's a much bigger cohort. So it's not just 14, 6,000, it's actually a multi multi site study. But we're only focused on that. I don't, we don't know if any of the other sites are also a exposure system. Looks like we are looking forward to the ambient air pollution data. Well, great, you've read my points. That's all I really had to say at the end. I didn't have anything much for it, thank you very much. Was there anyone in the room? It was people there but at a lot base.
Steven Chillrud - SMSP Fall Seminar Series 2023
From Taylor Link September 08, 2023
3 plays
3
0 comments
0
You unliked the media.
"Household Air Pollution in Rural Ghana
Novel Approaches for Integrating Wearing Compliance Data for Assessing Personal Exposure to Airborne Particulate Matter."
Steven Chillrud
Lamont Research Professor
Earth Observatory of Columbia Univ.
Zoom Recording ID: 94428117948 UUID: dDLd/L7hTtONYLJE30qv6Q== Meeting Time: 2023-09-08 02:53:44pmGMT
- Tags
- Department Name
- School of marine science and policy
- Appears In
Link to Media Page
Loading
Add a comment