Note: The I AM GPH podcast is produced by NYU GPH’s Office of Communications and Promotion. It is designed to be heard. If you are able, we encourage you to listen to the audio, which includes emphasis that may not be captured in text on the page. Transcripts are generated using a combination of software and human transcribers, and may contain errors. Please check the corresponding audio before quoting in print. Subscribe now on Apple Podcasts, Spotify, Google Play or Stitcher Radio.
EP04 Interning at the NYC Dept of Health and Mental Hygiene with Amila Samarabandu
Deborah Onakomaiya: Hey guys and welcome to another episode of I AM GPH. I am your host, Deborah Onakomaiya, on the show today we have Amila Samarabandu. He's a second year biostatistics student in the College of Global Public Health at NYU. He works at the Department of Health and Mental Hygiene in the Office of Informatics and Research, Bureau of System Strengthening and Access. He is also the creator and writer of a science blog called Surreal Science Stuff. You guys should really check that out. His public health interests are in mapping health and demographic data to New York City geographies, identifying those unstably housed within those areas. Fun fact about Amila is that back in his undergrad days, he would play the saxophone as a busker in downtown Toronto. Thanks so much for coming on our show today Amila.
Amila Samarabandu: Thank you for having me.
Deborah Onakomaiya: Yeah, I mean I've heard so much about your work and do you want to tell our listeners a little bit about what you do?
Amila Samarabandu: Well, right now I'm working at the Department of Health in what's called the Bureau of System Strengthening and Access within the Department of Mental Hygiene. So we do a lot of informatics work and look at the city as a whole and see where there are infrastructure holes. So, for example, if there are areas in the city where it takes too long to get to a facility, like a treatment facility, then we can identify those.
Deborah Onakomaiya: That's interesting. I understand that you're going to be ... A lot of what we're going to be talking about is very data driven. A lot of our listeners are not data gurus. Like you said, we're going to have to break down some terms, you know, as the interview goes along.
Amila Samarabandu: I mean, I'm not exactly a data guru either at the start of this.
Deborah Onakomaiya: Wow, but you work in the informatics here. But I mean, how did you get into this informatics world? I understand that you're a biostats student. How did that start out for you?
Amila Samarabandu: Well, I did a statistics minor in undergrad. I guess towards, I mean, when I was in third year maybe, I realized that I was taking two majors that were both biology driven and most of what we were doing was sitting in class analyzing papers. When we would discuss these papers in a group setting, you know, we would always get caught up on the statistics, right? Like, what does this mean that this alpha is 5% here, but it's 1% on this other table. So no one really had any idea, and to me that was sort of an alarm bell ringing that I should probably start learning this stuff.
Deborah Onakomaiya: What kind of inspired you to come into the public health space in terms of, you know, with your statistics background?
Amila Samarabandu: Well, most of what I was doing when I was an undergrad was theoretical statistics, which I don't think ... I didn't know half of what was going on, first of all. It was all proofs. It was all the linear algebra formulations and stuff. Not actual like, how do we solve problems using statistics kind of thing. And I read up on Google flu. Have you ever heard of that?
Deborah Onakomaiya: No. Can you tell us what that is?
Amila Samarabandu: So that's a project that Google has tried to track flus using search data and that sort of captured my attention, right? Because that's a very modern way to track the spread of disease. And it's a really cool way too. You don't have to have people going out into the field retrieving data. You can just do it using Google search data. So I wanted to learn a little bit more about what kind of went on behind that process.
Deborah Onakomaiya: That's really interesting. You currently work at the Department of Health and I mean that's like #careergoals for most people, you know? How were you able to leverage your past experience in statistics in getting into that particular position?
Amila Samarabandu: Well, it wasn't so much of leveraging my past experience at statistics as it was in taking up an offer that my epi TA made at the end of our tutorial. Well, she said that she normally takes her class out for drinks and me and one other girl were the only ones who ended up going and it was there that she was like, "I need an intern." And so I sort of grasped at that.
Deborah Onakomaiya: Ah.
Amila Samarabandu: Yeah, I know it's not a very inspiring story but ...
Deborah Onakomaiya: But I mean it's an opportunity that you were able to ... Wow. And how has that experience for you at the DOH been so far?
Amila Samarabandu: Oh, it's been incredible.
Deborah Onakomaiya: Wow, yeah.
Amila Samarabandu: Most of the people that I'm working with are, have a pretty solid background in mathematics and statistics, so I get to learn a lot about how to analyze different types of data and how to combine data from different sources.
Deborah Onakomaiya: Yeah, and I mean, from the way you speak about it, it seems like data is kind of like a passion for you.
Amila Samarabandu: Oh, of course. Yeah.
Deborah Onakomaiya: What drew you to data analysis? How did you first get involved? I understand that obviously you did statistics, but how did that passion come about?
Amila Samarabandu: More of a philosophical passion I guess. So have you ever heard of epistemology?
Deborah Onakomaiya: Epistemology? No.
Amila Samarabandu: So it's a field of philosophy which deals with how we know things, right? And so I took a philosophy course when I was in high school and epistemology fascinated me because it was basically the conclusions that we came to were that we don't really know anything. And so in order to know things you have to go through this entire process of induction, or go through the scientific method in order to find it out. And it really ... When you really get into it, it's about ... What is evidence? What constitutes strong evidence? What's the difference between strong evidence and weak evidence? And many brilliant people over the last several hundred years have been solely answering that question. And most of the answers they come up with are statistical methods.
Deborah Onakomaiya: So statistics is our way of finding how everything works or how to get ...
Amila Samarabandu: Yeah, like how do we know things at all, right? Like how do you know gravity is a thing or that the speed of light is a certain value, right? Statistics is involved in every single process of scientific inquiry.
Deborah Onakomaiya: You putting that into perspective kind of makes things clearer for me because I was like, yeah ... I mean we might not know that we're going through the scientific method when we have things happening, but when you break it down like that, it makes things much more clearer. So in layman terms, for those who are not gurus, you talked a little bit about informatics and you mapping out certain areas across New York City in terms of like, with your work at the DOH. So could you just go a little bit more into detail?
Amila Samarabandu: So what I was doing at first was mapping data from the American Community Survey, which is conducted by the Census Bureau. And it's basically a survey of basic demographics, you know, race, ethnicity ... But also has a lot of other interesting things. Like what type of health insurance you're on. Things like income, mode of transportation to work and travel time to work. So that's all really interesting information from a healthcare perspective. The only problem is that American Community Survey data is organized in a really confusing format.
Deborah Onakomaiya: Really?
Amila Samarabandu: Yeah. It's really hard to access.
Deborah Onakomaiya: So is it like, are they ... Like in what sense do you mean it's confusing?
Amila Samarabandu: Meaning that there is all sorts of summary tables and each summary table contains a different set of information and then there are different tables for different levels of aggregation, meaning ... So like, if you're looking at a state level, there are different tables for that. If you're looking at a County level, there are different tables for that. And knowing what all of these actually are and linking them together is a huge, time consuming process. So what I did was work with a colleague to do that and then to put it so that you can see New York City as a whole and see the neighborhoods in New York city and see the counts of whatever variables of interest that you're looking at are by neighborhood. So if you want to look at Astoria, for example, it would show you, okay, this is other number of estimated African Americans living in Astoria. These are the estimated number of people under the age of 18. These are the estimated number of people who don't have health insurance. Basically it was about putting all of this really useful data that's not directly related to health in a format that health researchers could use.
Deborah Onakomaiya: You mapping out things like transport, those are kind of around like social determinants of ...
Amila Samarabandu: Oh yeah, of course.
Deborah Onakomaiya: Yeah. And so, all this data, are they just archived?
Amila Samarabandu: Well, the American Community Survey is a public dataset. So the aggregate level data ... Basically, instead of giving you, all right, person X has these characteristics ... What it tells you is how many people within a certain area have these types of characteristics. Because there's so much information that there's a huge ethical problem when it comes to releasing data at the individual level. So they release aggregate measures. So like, county measures. So how many people in this county are under the age of 18? And so by taking data at all of those different levels and mapping it in a way such that you can see the different levels. So, for example, you want to switch between boroughs and neighborhoods, right? There should be an easy way, like just a button that you click, for example. And then now, instead of seeing data by borough, you see it by neighborhood. Click another button, you see it by census tract. Click another one, you see it by block. That was the project that I was working on.
Deborah Onakomaiya: Yeah, I think I needed that when I was doing my social behavioral assignment. But yeah, that's an inside joke for the listeners. But yeah, I mean that, I think that kind of makes things easier to view data. So we're going to switch gears a little bit. You mentioned that you're a biostats student. How ... I mean, obviously as students we all face challenges. What NYU resources have you found most useful with your time being here?
Amila Samarabandu: Ooh, what specifically, what NYU resources? I would say NYU libraries as a whole because there's, first of all, there's their journal database, which is amazing. I feel like people don't understand how difficult it is to get scientific journals. Any scientific journal that's been published in the last ... As long as it wasn't published yesterday, they can probably get it to you, which is kind of amazing. But they also have tons of different tutorials on how to use different types of software. So like you can get GIS training, you can get R training, you can get Python training all from the library for free.
Deborah Onakomaiya: What exactly the GIS, the R ... For our listeners.
Amila Samarabandu: So ArcGIS is a program that allows you to map things. So map boundaries, map data ... R is basically a statistical programming language. It allows you to do statistical analysis, but it's very flexible and it's open source. A lot of people have developed different packages for it. Python is a programming language.
Deborah Onakomaiya: Okay, okay. And have you made use of all of these? Have these resources helped you?
Amila Samarabandu: Oh yeah, absolutely. For my work I had to use all of those different things, rather than try and learn it from scratch yourself, why not use what the library has available for you? Which are these beautifully made tutorials that anyone can just sort of pick up.
Deborah Onakomaiya: Yeah, and they are for free guys. Right? So, a quick Google search about you ... The first thing that comes up is Surreal Science Stuff. Seriously, you're famous on the internet Amila.
Amila Samarabandu: That only because no one else has my last name.
Deborah Onakomaiya: But I Googled you and I was like, oh my gosh, this is so cool. Can you just tell us a little bit about your blog, the Surreal Science Stuff, how did that start? What's it's about and so on?
Amila Samarabandu: Well, I love science. I think it's really cool.
Deborah Onakomaiya: What? You love science? Are you kidding me?
Amila Samarabandu: Well like, I don't know, I feel like when you really get down into modern scientific research, the reason why people don't engage with it as much anymore is because it's just too complicated. For example, there was a paper published last week called how long non-coding RNAs regulate transcription. That's actually a pretty simple concept compared to some of the stuff ... That most of the stuff that's being published on the air. But in order to understand that you have to understand different types of RNA. What is a long non-coding RNA? What is transcription? What is a transcription regulation mechanism?
Deborah Onakomaiya: Biology 101.
Amila Samarabandu: Yeah. So all of those things, just those basic concepts require biology 101. And so immediately anyone who hasn't taken that is automatically shut out. And then in order to actually get into the nitty gritty and be able to see how do long non-coding RNA regulate transcription? To answer those questions, you need to have a little bit higher knowledge of biology. So if you're in working through that process, you learn a lot about yourself, about the cells of your body and like how exactly it works. And it lends you a sense of wonder because you realize just how complicated your own body is and the universe around you and how like ... First of all, it's humbling. Second of all, it gives you a lot of hope almost, right? Because you realize that like a lot of the problems that we face in society are solvable by science and by technology. It's just a matter of understanding them. Basically, when I was looking at the research that people were doing and looking at what was being published in news articles, there was a huge disconnect. The kind of stuff that people publish in the New York Times is so different from what scientists are actually doing. Did you hear about the gravitational waves discovery a couple ... Maybe a year ago?
Deborah Onakomaiya: Sure ...
Amila Samarabandu: Okay, so the idea was that at the beginning of the universe ... Oh, this is a bad example. At the beginning of the universe, the universe started very, very small and it expanded really quickly. Like, faster than the speed of light and that expansion had to result in something. Some trace of it. And so there's a physicist that theorized that it was gravitational waves would be sort of the signal. And so I think it was a year ago, maybe it might've been a little bit longer than that … The headlines were blaring that they had discovered gravitational waves and it got ridiculous. There was a video of one of the grad students with a bottle of champagne at the physicist's door being like, "We found it." And there was this teary reunion and later they found that it was instrument error.
Deborah Onakomaiya: Instrument error?
Amila Samarabandu: Instrument error.
Deborah Onakomaiya: Wow.
Amila Samarabandu: That didn't exactly make headlines. Right? So it became slowly, this process of like, "Okay, wait a second. How are we going to verify that this is actually taking place?" That whole process was a really interesting process. They double check the results. They went to a different instrument and then a different facility and then perform parallel experiments to see whether they could actually match it up. It was honestly, it was gorgeous to watch, but none of that made the news. It's just this whole process that we're continually engaging in and trying to discover things about the world. Has all these little quality control checks built into it.
Deborah Onakomaiya: So that gap, that identified gap, was what inspired you to start this for?
Amila Samarabandu: Well, the gap between what scientists are actually doing and what we learn about as a public. Like, can you tell me anything that happened in chemistry in the last couple of years? And that's ... Chemistry is a huge field, right? Why aren't we learning about that? Why are we instead learning about all ... Like, what certain celebrities are doing on a daily basis and stuff? Why aren't we learning about the world and the universe and what we're discovering on a daily basis about it?
Deborah Onakomaiya: Wow, that's really, really, really interesting. Everything that's on that blog, is it basic science stuff like chemistry, physics, biology, or like, what exactly are you talking about on it?
Amila Samarabandu: So usually what I do is I take a paper that's been published recently, usually in the last week and try to explain it in terms that someone who doesn't have a scientific background can understand. To be honest, it's mostly for my own benefit in that it's really good practice reading papers and communicating and writing. Engaging with science writing. So it's mostly practice in that regard. Occasionally I'll write articles about some really cool concept that will probably change the world in the future. For example, graphene: Thin sheet of carbon that's really conductive, really strong, and really flexible, which is amazing. Imagine building a building out of this stuff. It would be a solar panel and a building at once. That kind of stuff really gets me. But that's the kind of stuff that I would write about that isn't necessarily directly related to a paper that was published.
Deborah Onakomaiya: Just for research purposes so I can steal your research skills ... How do you find papers that have just been published like a week ago? How do you know what to look for?
Amila Samarabandu: Usually lists of journals. You start with a field. So, say chemistry ... Or physics. And then you have the annual physics review, right? So that'll give you a good source. When that comes out they'll have a bunch of really good papers. That'll be reviewing the literature on physics. That's a good source of broad, more broader research topics.
Deborah Onakomaiya: And then your selection is based off of your interests?
Amila Samarabandu: Yeah.
Deborah Onakomaiya: In terms of after you've viewed, like okay these are all the papers that have been. Wow, that's interesting.
Amila Samarabandu: Yeah, I guess it just kind of forces me to read up on what's being done in science recently. It sort of forces me to go through these different journals and be like, "Oh that's the kind of stuff that they're doing now. And this is how it's different from what they were doing the last time I checked."
Deborah Onakomaiya: And is this like a weekly activity like, between getting an MPH in biostats and working at the DOH mapping stuff? How do you have time for ... ?
Amila Samarabandu: It used to be a weekly thing. It's not nearly so much anymore. And this conversation is making me want to reengage with it more.
Deborah Onakomaiya: You should though. Because it was really cool stuff. I'm like, "What am I doing with my life?" You know? But I mean, it's really, really interesting stuff. It's really engaging stuff and I think at this time, rather than, you know, reading about what one celebrity's doing, especially now that we are all in grad school ... It's important to also look at, what are the new things happening in science? And I mean, the news tries as much as it can to do that. I mean, I try to listen to NPR, but I mean with your type of blog, it kind of goes into detail and breaks it down for someone like me or our listeners that might not necessarily understand. So it's, I mean, it's really good. You seem to be a science lover and I mean, you said that. Can you just explain your love for science?
Amila Samarabandu: I'm going to have to think about that. I think when I was very, very ... Okay. Yeah, I do remember when it was. It was ... Do you remember those little book orders that you get? Did you have that in elementary school where they would send you a news pamphlet and they would have a lot of different books on it? One of the books I got from there, was this really detailed astronomy book, and to this day I haven't come across a book this good because it explained solar system and all the different planets and like all the different gases in the planets and stuff. At the time, I was in grade one and I could actually follow what was going on because it was so clear. I was like six, seven maybe. But it wasn't like ... It's not me. Other people who read this book also had similar results. It was the book. It was just an incredibly well designed book, I guess. And then from then on, I guess I've always been really interested in science.
Deborah Onakomaiya: Wow. That's really, really inspiring. It was so wonderful having you on the show today, I learned so much. I feel like I need to go brush up on my biostats 101, but thank you so much for being here.
Amila Samarabandu: Thank you for having me.