Latest / Wellness, Actually with Emily Oster & Perry Wilson, MD / Researching the Importance of Paid Leave: A look into how studies are conducted

Transcript

Researching the Importance of Paid Leave: A look into how studies are conducted · Wellness, Actually with Emily Oster & Perry Wilson, MD

• 0:01
  Speaker 1: If past time America caught up with the rest of
• 0:03
  Speaker 1: the world on paid leave. Folks, all this helps families,
• 0:09
  Speaker 1: that it grows the economy.
• 0:12
  Speaker 2: I'm Emily Oster. This is parent data and today.
• 0:16
  Speaker 3: The flashpoint issue that impacts millions of Americans the heated
• 0:19
  Speaker 3: battle over paid family leave.
• 0:21
  Speaker 1: Oh that's right.
• 0:22
  Speaker 2: The good old US of A is the only developed
• 0:24
  Speaker 2: country in the world that doesn't provide new parents with
• 0:27
  Speaker 2: any financial assistance. Federal law grant's workers just twelve weeks
• 0:32
  Speaker 2: of unpaid leave, and there are some stark limits on that.
• 0:35
  Speaker 4: You have to work for a company with fifty or
• 0:37
  Speaker 4: more employees, you have to have been there for over
• 0:39
  Speaker 4: a year. You have to be a full time salaried employee.
• 0:43
  Speaker 4: So if you're a freelance or contract or something like that,
• 0:45
  Speaker 4: it doesn't apply to you.
• 0:47
  Speaker 3: We can afford to do it, but another we can't
• 0:48
  Speaker 3: afford not to do it.
• 0:52
  Speaker 5: The United States is one of the only countries in
• 0:55
  Speaker 5: the world that doesn't guarantee paid parental leave. The other
• 0:59
  Speaker 5: countries without this leave are the Marshall Islands, Micronesia, plau
• 1:04
  Speaker 5: Papa in New Guinea. It's a small number of small places.
• 1:09
  Speaker 5: The US is big and about eighty percent of new parents,
• 1:13
  Speaker 5: that's more than one hundred million people every year don't
• 1:17
  Speaker 5: have paid time off after.
• 1:19
  Speaker 2: The birth or adoption of a child. We say this
• 1:23
  Speaker 2: fact a lot. I say this fact a lot, but
• 1:27
  Speaker 2: it's useful to.
• 1:28
  Speaker 5: Think about what that means. What does it mean for
• 1:31
  Speaker 5: a family when parents don't have paid leave and don't
• 1:35
  Speaker 5: have the ability to take time off when a baby's born.
• 1:39
  Speaker 2: It means a lot of things.
• 1:40
  Speaker 5: It means moms going back to work, still recovering, really
• 1:45
  Speaker 5: recovering from childbirth. It means parents struggling to figure out
• 1:51
  Speaker 5: childcare for their baby, and it often means babies going
• 1:55
  Speaker 5: to group childcare settings, which may be wonderful but do
• 2:01
  Speaker 5: expose them to germs, germs which are more dangerous when
• 2:05
  Speaker 5: babies are small than when they're big. There are a
• 2:10
  Speaker 5: lot of reasons that we talk about the importance of
• 2:13
  Speaker 5: paid leave. We can talk about these reasons and we
• 2:17
  Speaker 5: can see why they might matter for kids outcomes, for
• 2:21
  Speaker 5: family's outcomes, But to figure out how much they matter
• 2:25
  Speaker 5: and in what ways, that's what research is for. And
• 2:29
  Speaker 5: this past summer my guest today doctor Catherine Aarons and
• 2:32
  Speaker 5: doctor Jennifer Hutchins, who are both medical doctors and professors.
• 2:36
  Speaker 5: Wrote a paper titled Paid Family Leave and Prevention of
• 2:40
  Speaker 5: Acute Respiratory Infections in Young Infants. The paper is an
• 2:46
  Speaker 5: analysis of paid leave in New York State, and the
• 2:50
  Speaker 5: authors are looking at the impacts of that paid leave
• 2:53
  Speaker 5: on hospitalizations for infants, mostly for RSV.
• 2:58
  Speaker 2: Because the paper uses.
• 3:00
  Speaker 5: A policy change, Kate and Jen are able to look
• 3:03
  Speaker 5: at the causal effects of a policy and draw conclusions
• 3:08
  Speaker 5: where we can say this is the impact of paid
• 3:11
  Speaker 5: leave on something, not just paid leave is correlated with
• 3:14
  Speaker 5: some better outcome. The upshot of the research is that
• 3:19
  Speaker 5: the observed illness rate in New York is lower after
• 3:22
  Speaker 5: the new policy. Jen and Kate calculated an eighteen percent
• 3:26
  Speaker 5: decline in hospitalizations and er visits and a twenty seven
• 3:30
  Speaker 5: percent decline in RSV. In particular, they find this change
• 3:35
  Speaker 5: only in babies under twelve months, not in babies twelve
• 3:38
  Speaker 5: to twenty four months, which really suggests that this is
• 3:42
  Speaker 5: isolated to the ages where the change in policy would
• 3:45
  Speaker 5: have changed behavior.
• 3:46
  Speaker 2: Bottom line from the paper paid family leave keeps babies healthier, keeps.
• 3:51
  Speaker 5: Them out of the hospital, and now we have the
• 3:54
  Speaker 5: data to prove it and to show that the effects.
• 3:56
  Speaker 2: Are big in terms of numbers. So that's it.
• 4:02
  Speaker 5: There's a peer reviewed academic paper with actual, meaningful, real
• 4:05
  Speaker 5: world implications, not just for struggling parents who always could
• 4:10
  Speaker 5: use this kind of validation, but also for policymakers who
• 4:14
  Speaker 5: are thinking about what kinds of policy changes might actually
• 4:17
  Speaker 5: matter for families throughout the country. In this episode, we'll
• 4:22
  Speaker 5: talk a little bit about the findings, but I'm actually
• 4:25
  Speaker 5: more interested here in talking to Kate and Jen about
• 4:28
  Speaker 5: how they actually found these findings. Putting together a research
• 4:33
  Speaker 5: paper like this is surprisingly tricky. You need to know
• 4:36
  Speaker 5: what questions you're asking. You need to think about how
• 4:39
  Speaker 5: you're going to get causality rather than just correlation. They
• 4:44
  Speaker 5: really want to ask, does paid family leave reduce infin illness?
• 4:48
  Speaker 5: But how can you do that when families are different
• 4:52
  Speaker 5: in all kinds of other ways. So we're going to
• 4:55
  Speaker 5: take their research from idea to final peer reviewed paper,
• 5:00
  Speaker 5: and we're going to talk about everything that you always
• 5:02
  Speaker 5: wanted to know about how research is conducted, how we
• 5:06
  Speaker 5: get ideas, how we make the ideas, how we turn
• 5:08
  Speaker 5: them into the paper, how we deal with referees.
• 5:12
  Speaker 1: These are all the.
• 5:13
  Speaker 5: Questions you want to know the answers to but never
• 5:15
  Speaker 5: thought to ask because maybe you're not an academic, why
• 5:19
  Speaker 5: would you need to know about referee too. But trust me,
• 5:22
  Speaker 5: it's really interesting and neat, and more than that, it's
• 5:25
  Speaker 5: going to apply to more things than you think, especially
• 5:28
  Speaker 5: if you live in New York want the rest of
• 5:30
  Speaker 5: the country to look like New York in terms of
• 5:32
  Speaker 5: paid leave, or if you just want your babies to
• 5:35
  Speaker 5: be as healthy as possible after the break. Kate Aaron's
• 5:39
  Speaker 5: and Jen Hutcheon. Kate Aaron's and Jen Hutchin, I am
• 5:51
  Speaker 5: delighted to welcome you both to Marrin Data. I would
• 5:55
  Speaker 5: love if we could start by just having you introduce
• 5:57
  Speaker 5: yourself before we get into.
• 5:59
  Speaker 2: Paid leave and illness, So Kate, maybe we can start
• 6:02
  Speaker 2: with you.
• 6:03
  Speaker 3: Sure, great to meet you. I am an epidemiologist and
• 6:07
  Speaker 3: associate research professor at the University of Southern Maine which
• 6:11
  Speaker 3: is part of the University of Main system.
• 6:14
  Speaker 6: And Jen, thanks, nice to meet you as well. My
• 6:18
  Speaker 6: name is Jan Hutchin and I am an associate professor
• 6:20
  Speaker 6: in the Department of Obstetrics and Gynecology and a perinatal
• 6:23
  Speaker 6: epidemiologist as well at the University of British Columbia, which
• 6:26
  Speaker 6: is in Vancouver in Canada.
• 6:29
  Speaker 5: So our goal today is to talk about this very
• 6:31
  Speaker 5: nice paper, which I highlighted for everybody.
• 6:33
  Speaker 2: At the top of the episode.
• 6:35
  Speaker 5: So I'm not going to review the findings again. But
• 6:39
  Speaker 5: the primary argument in the paper, the primary finding is that.
• 6:43
  Speaker 2: Paid leave reduces childhood illness.
• 6:47
  Speaker 5: And so I guess what I would love to start
• 6:49
  Speaker 5: by asking you, guys, is about the very beginning here,
• 6:53
  Speaker 5: how did you get interested in this topic? And you
• 6:56
  Speaker 5: can answer that question either how did you get interested
• 6:58
  Speaker 5: in writing this paper or why do you care about
• 7:01
  Speaker 5: this general area of the world.
• 7:03
  Speaker 6: Either one, I can go ahead and take that. So
• 7:07
  Speaker 6: we actually kind of migrated into this. Kate and I
• 7:10
  Speaker 6: had originally done a lot of work looking at research
• 7:13
  Speaker 6: in birth spacing and how we can try to prevent
• 7:15
  Speaker 6: rapid repeat pregnancies. And then NICCHD put out and notice
• 7:20
  Speaker 6: a special interest sort of a call for research to
• 7:23
  Speaker 6: try to prevent rapid repeat pregnancies that specifically was interested
• 7:28
  Speaker 6: in how different policies or practices could help prevent rapid
• 7:32
  Speaker 6: repeat pregnancies. And they actually specifically paid family leave is
• 7:36
  Speaker 6: one of the things that they were interested in evaluating
• 7:39
  Speaker 6: so we actually put in a proposal to look at
• 7:41
  Speaker 6: paid family leave and prevention of rapid repeat pregnancy, which
• 7:44
  Speaker 6: is kind of what got us into this topic. That
• 7:48
  Speaker 6: grant did not get funded, and as we were kind
• 7:50
  Speaker 6: of rethinking what to do with it for resubmitting and
• 7:53
  Speaker 6: things like that, one of the considerations that we had
• 7:57
  Speaker 6: is that when you're evaluating these policies, because not everybody
• 8:01
  Speaker 6: in the population is affected by a policy, so some
• 8:04
  Speaker 6: people will have already had leave or will already be
• 8:07
  Speaker 6: taking leave, some people still won't take leave even if
• 8:10
  Speaker 6: it's available. You need to have a fairly strong signal
• 8:13
  Speaker 6: in the group of people who are affected by the
• 8:16
  Speaker 6: paid family lead policy to be able to detect it overall.
• 8:19
  Speaker 6: And I think we were a little bit worried that
• 8:21
  Speaker 6: because there's a couple steps involved in prevention of rapid
• 8:24
  Speaker 6: for preid pregnancy, it has to be that you are
• 8:26
  Speaker 6: more likely to go to your postpartum visit and get
• 8:29
  Speaker 6: contraception to be able to actually prevent a rapid for prepregnancy,
• 8:32
  Speaker 6: that maybe we couldn't detect signal overall. And then I
• 8:36
  Speaker 6: happen to be working on another project that was looking
• 8:39
  Speaker 6: at prevention of lower respiratory tract infection and invents and
• 8:42
  Speaker 6: going through all the literature on the risk factors, of
• 8:44
  Speaker 6: which the main one that stands out is daycare and
• 8:47
  Speaker 6: older siblings. And so that kind of was the connection
• 8:50
  Speaker 6: between saying, hey, if we want a strong direct effect,
• 8:53
  Speaker 6: why aren't we looking at the effect of paid family
• 8:55
  Speaker 6: leave on newborn respiratory tract infections? And that's kind of
• 8:59
  Speaker 6: how we got.
• 9:00
  Speaker 2: To where we were. Okay, I love this.
• 9:02
  Speaker 5: I have many follow up questions because I understood everything
• 9:05
  Speaker 5: you said, but I suspect that for many people on
• 9:08
  Speaker 5: this they didn't. So let's step back to the first
• 9:10
  Speaker 5: part of this, which is when you are thinking about
• 9:13
  Speaker 5: what to research.
• 9:14
  Speaker 2: The first thing you said.
• 9:15
  Speaker 5: Was we saw grant submission, we saw a way to
• 9:18
  Speaker 5: have financing. That's something I think people have a hard
• 9:22
  Speaker 5: time understanding, like how important financing is in.
• 9:28
  Speaker 2: What we study.
• 9:29
  Speaker 5: So can you say a little bit, just like for
• 9:31
  Speaker 5: someone who isn't in this space, like what does it
• 9:34
  Speaker 5: mean when you say we saw a grant and that
• 9:36
  Speaker 5: was important? Why would you care about there being money?
• 9:39
  Speaker 5: Like why don't you just do whatever you want?
• 9:42
  Speaker 6: Because research costs money, So research costs money. Do anything?
• 9:47
  Speaker 2: So you can.
• 9:49
  Speaker 6: To get the data to answer these questions, you need
• 9:51
  Speaker 6: to buy it. You need to hire analyst time, you
• 9:54
  Speaker 6: need to pay for faculty time to focus on the project.
• 9:57
  Speaker 6: So it's possible to do research with out having funding,
• 10:00
  Speaker 6: but certainly in healthcare, I think it's almost necessary that
• 10:05
  Speaker 6: you have some sort of grant to help to help
• 10:07
  Speaker 6: pay for it. The grant we got for this was
• 10:09
  Speaker 6: actually a fairly small grant that really just paid for
• 10:12
  Speaker 6: our data and Kate's salary and a little bit of
• 10:14
  Speaker 6: help towards students, but that was really it. Most of
• 10:17
  Speaker 6: the time people would want a much bigger grant, and so.
• 10:21
  Speaker 5: Kate, when you look out at the sort of world
• 10:24
  Speaker 5: of grants, how much of what you're going to study
• 10:26
  Speaker 5: is like, I want to study this, and I'm looking
• 10:29
  Speaker 5: for money to finance it, versus I'm looking for money
• 10:32
  Speaker 5: in my general space, and it's going to push a
• 10:34
  Speaker 5: little bit in one direction or the other.
• 10:36
  Speaker 2: I'm just curious about that balance.
• 10:40
  Speaker 3: Well, definitely money is important. So I'm an entirely soft
• 10:43
  Speaker 3: funded money position. So if I don't have grant money
• 10:47
  Speaker 3: or contract money, I won't actually have a job. I
• 10:50
  Speaker 3: can't really dip below one hundred percent funding, so it's
• 10:53
  Speaker 3: really important, but I'm not perseverating about funding too much.
• 10:57
  Speaker 3: I'm kind of an opportunist, honestly searcher, So I work
• 11:01
  Speaker 3: on topics I'm directly interested in, but I'm also happy
• 11:03
  Speaker 3: to contribute my data analysis skills and writing skills to
• 11:08
  Speaker 3: topics that are slightly tangential to what I'm interested in
• 11:11
  Speaker 3: as a co investigator. But yeah, definitely, and IICCHD National
• 11:16
  Speaker 3: Institute of Child Health and Human Development shares a lot
• 11:18
  Speaker 3: of interest with what Jen and I do because it's
• 11:22
  Speaker 3: focused on like families and maternal health and infant health
• 11:27
  Speaker 3: and child health. So that's a good institute for us
• 11:31
  Speaker 3: to apply for funding from.
• 11:33
  Speaker 5: Often people ask why do some things get funded and
• 11:35
  Speaker 5: some things get studied and some things don't. I think
• 11:38
  Speaker 5: that a piece of that is that a lot of
• 11:40
  Speaker 5: our funding comes from either companies, pharmaceutical companies, or from
• 11:44
  Speaker 5: the government, and the government has the sort of set
• 11:46
  Speaker 5: of things that they've said we're intersted in studying this,
• 11:48
  Speaker 5: and then that pushes research a little bit in various directions.
• 11:53
  Speaker 5: And then there's sort of pharma funding, which I would
• 11:56
  Speaker 5: not be appropriate for this kind of project because there's
• 12:00
  Speaker 5: nothing to patent. If you learn that paid leave is
• 12:03
  Speaker 5: good for families, you can't then sell that in a shot.
• 12:07
  Speaker 5: We're working on it, but it's proved to be very challenging. Okay,
• 12:12
  Speaker 5: so you sort of get into this space, but it
• 12:17
  Speaker 5: is not initially the thing you're most interested in. So
• 12:20
  Speaker 5: can you just say a little bit about this idea
• 12:21
  Speaker 5: of rapid repeat pregnancies and actually what that is and
• 12:25
  Speaker 5: where you started, why what were you trying to understand there?
• 12:30
  Speaker 3: Well, Jeth and I have worked a lot in this topic.
• 12:32
  Speaker 3: We've probably published maybe ten or twenty papers on this topic.
• 12:36
  Speaker 3: We actually let a workshop at when I worked with
• 12:39
  Speaker 3: the Office of Population Affairs in like twenty seventeen. You
• 12:42
  Speaker 3: let a workshop where experts come in and talk about
• 12:44
  Speaker 3: the methodological challenges of looking at this issue of rapid
• 12:48
  Speaker 3: repeat pregnancy, which is also called studying interpregnancy intervals, and
• 12:52
  Speaker 3: a lot of literature for decades has found that there
• 12:56
  Speaker 3: are poor infant outcomes if an infant is born shortly
• 13:00
  Speaker 3: after their sibling is born, and so there actually have
• 13:03
  Speaker 3: been new types of studies that have come out recently
• 13:05
  Speaker 3: doing different types of methods with this. With these data
• 13:09
  Speaker 3: that you can get at a population level, and some
• 13:11
  Speaker 3: of those studies were challenging the findings of previous studies
• 13:15
  Speaker 3: and there was inconsistency. So we were looking a lot
• 13:17
  Speaker 3: on the methodology of how to study interpregnancy intervals on
• 13:21
  Speaker 3: infant health and maternal health.
• 13:23
  Speaker 2: Why is that hard?
• 13:24
  Speaker 3: Well, it's actually the concern is confounding. So the type
• 13:28
  Speaker 3: of people who have children close together have different characteristics
• 13:32
  Speaker 3: than the type of people who have children farther apart,
• 13:36
  Speaker 3: and so to control for that confounding has been challenging,
• 13:39
  Speaker 3: and there's always been a sense of doubt that it's
• 13:41
  Speaker 3: not actually the spacing that matters, it might be the
• 13:44
  Speaker 3: characteristics that we haven't properly controlled for. So they started
• 13:48
  Speaker 3: doing these within family analyzes, so like within a mother
• 13:52
  Speaker 3: looking at the spacing between her first and second baby
• 13:55
  Speaker 3: and her second and third baby. But those also have
• 13:58
  Speaker 3: challenges those types of studies, so we're kind of just
• 14:02
  Speaker 3: doing a deep dive into these study designs.
• 14:05
  Speaker 2: And then how do you come to paid leave?
• 14:07
  Speaker 5: Then?
• 14:08
  Speaker 3: I can't remember. I mean as part of this is
• 14:10
  Speaker 3: that Gen's from Canada and I'm from the US. I've
• 14:13
  Speaker 3: never had a job that has had paid family leave.
• 14:15
  Speaker 3: In fact, my current i've never had a job that
• 14:17
  Speaker 3: we don't work here, it's not like Canada. So Jen
• 14:21
  Speaker 3: always finds it like ridiculous that we don't have paid
• 14:23
  Speaker 3: family leave. She thinks this sort of preposterous, But so
• 14:26
  Speaker 3: we've I've never had one off, all three kids, never
• 14:29
  Speaker 3: no paid family leave. So part of it was part
• 14:32
  Speaker 3: of just our interest in this topic and why it's
• 14:34
  Speaker 3: so different between Canada and the US. And you know,
• 14:38
  Speaker 3: it does seem like there are so many things we
• 14:40
  Speaker 3: think about for maternal and child health, like what can
• 14:42
  Speaker 3: we do, especially in rural health and telehealth and stuff
• 14:46
  Speaker 3: like that, But it does seem like one of the
• 14:48
  Speaker 3: obvious things we might be able to do that has
• 14:51
  Speaker 3: been shown to have many health benefits is to offer
• 14:54
  Speaker 3: some type of paid family leave. And you know, we're
• 14:57
  Speaker 3: only they're only thirteen states have enacted paid FAMI, with
• 15:00
  Speaker 3: Maine being the thirteenth state. So it's not the majority
• 15:05
  Speaker 3: of states in the US that have paid family LIAF
• 15:07
  Speaker 3: which is an outlier among many high income countries. I
• 15:10
  Speaker 3: think we're.
• 15:11
  Speaker 5: Literally all all of yeah, high income and actually most
• 15:15
  Speaker 5: of the low income countries too. Yeah, we're holding our
• 15:18
  Speaker 5: holding our own. So so Jen, you find our lack
• 15:21
  Speaker 5: of pay family leave disturbing and you.
• 15:23
  Speaker 2: Wanted to study that, is that basically you came judgmentally
• 15:26
  Speaker 2: from the North to show us we're wrong.
• 15:29
  Speaker 6: I'm not going to say that. It's definitely a difference.
• 15:33
  Speaker 6: And my husband has been doing a lot of work
• 15:35
  Speaker 6: in Boston right around the time we were thinking about
• 15:38
  Speaker 6: this study, and and so, and I had just I
• 15:41
  Speaker 6: was on Matt lead for part of the time. It
• 15:42
  Speaker 6: had just come off then just come off Matt leave,
• 15:44
  Speaker 6: and so it was almost just kind of mind boggling
• 15:47
  Speaker 6: to be like, how do people make this work? How
• 15:50
  Speaker 6: do their kids not yet sick all the time if
• 15:51
  Speaker 6: you put these very young kids into daycare. So it's
• 15:55
  Speaker 6: just I hadn't really thought about it much, to be honest,
• 15:57
  Speaker 6: until spending time in the US, when you realize like
• 15:59
  Speaker 6: just how challenging it is and sort of all the
• 16:02
  Speaker 6: carry on effects of what happens if parents go back
• 16:06
  Speaker 6: to work right away, what happens to their kids? Do
• 16:08
  Speaker 6: those kids get sick? All these kinds of things.
• 16:12
  Speaker 5: Yeah, So I think what's interesting then, as a researcher
• 16:15
  Speaker 5: is you sort of have this sense of like, and
• 16:17
  Speaker 5: I think a lot of our ideas often come from
• 16:19
  Speaker 5: our own lives or the experiences we're having. So you're like, oh,
• 16:22
  Speaker 5: here it is paid family leave. It seems like it's important.
• 16:25
  Speaker 5: And then you're connecting it very directly as I would
• 16:28
  Speaker 5: too to Okay, well would you do with your kid? Well,
• 16:30
  Speaker 5: I remember, you know, when I send my kids to
• 16:33
  Speaker 5: d here for the first time, they got really sick. Okay,
• 16:36
  Speaker 5: well if that's happening earlier, you know, you think about
• 16:38
  Speaker 5: infant help, do you sort of come up with this
• 16:41
  Speaker 5: basically a theory of how these things might relate.
• 16:44
  Speaker 2: And then there's this challenge, which is and you talked
• 16:49
  Speaker 2: about confounding.
• 16:49
  Speaker 5: But just to be clear on the challenge, you couldn't
• 16:51
  Speaker 5: really just write a paper where you compare you know,
• 16:56
  Speaker 5: people who have leave to people who don't, because those
• 16:59
  Speaker 5: groups are really really different. You know, maybe they're in
• 17:01
  Speaker 5: different countries, but even if you looked within the US,
• 17:04
  Speaker 5: pay leave is more likely from an employer for people
• 17:07
  Speaker 5: who are white collar who have more resources in other ways,
• 17:09
  Speaker 5: So you wouldn't really be able to do that.
• 17:11
  Speaker 2: So your challenge there is to say, Okay.
• 17:14
  Speaker 5: I'm interested in this question of how pay leave might
• 17:17
  Speaker 5: impact infant illness, but I can't just compare people. So
• 17:22
  Speaker 5: what is your next step in trying to figure out
• 17:25
  Speaker 5: how you can do a better job figuring out a
• 17:28
  Speaker 5: causal relationship?
• 17:31
  Speaker 6: I think here our next step was to try to
• 17:33
  Speaker 6: look for sort of the natural experiment type situation that
• 17:38
  Speaker 6: helps do a better job of controlling controlling for confounding,
• 17:42
  Speaker 6: so doing a better job of helping you isolate if
• 17:44
  Speaker 6: there actually is an effect of the policy, or if
• 17:47
  Speaker 6: it's just the people who have paid family leave also
• 17:50
  Speaker 6: have better health outcomes for many other reasons, but that
• 17:53
  Speaker 6: introducing paid family leave won't actually make any difference there.
• 17:56
  Speaker 6: And so we started sort of looking around for cases
• 17:59
  Speaker 6: where there could be this type of natural experiment and
• 18:02
  Speaker 6: when we were originally and.
• 18:03
  Speaker 2: What is the difference?
• 18:04
  Speaker 5: Can you just what is the difference between a natural
• 18:06
  Speaker 5: experiment and a regular experiment?
• 18:09
  Speaker 6: Oh, a regular experiment, I guess in our world would
• 18:12
  Speaker 6: be where the researchers are manipulating what's happening, so are
• 18:18
  Speaker 6: deciding who's going to get a treatment and who's not
• 18:20
  Speaker 6: going to get a treatment. So in this case, you
• 18:23
  Speaker 6: know that would be randomly assigning paid family leave to
• 18:26
  Speaker 6: some people and not to others, which would be a
• 18:29
  Speaker 6: very challenging thing to do. I'm sure it's possible with
• 18:33
  Speaker 6: a lot of money, but.
• 18:33
  Speaker 2: You would need a much, much larger grant exactly.
• 18:39
  Speaker 6: But instead, what we did was looked for states that
• 18:42
  Speaker 6: had recently introduced paid family leave, which we sort of
• 18:45
  Speaker 6: took as this what we would call a natural experiment
• 18:47
  Speaker 6: because it's not anything we did, it's just something that
• 18:50
  Speaker 6: happened and what can we learn from it.
• 18:54
  Speaker 5: So you find a state, your state is New York.
• 18:57
  Speaker 5: You say something about what happened, Kate, Do you want
• 19:00
  Speaker 5: to tell us, like, what happened in New York?
• 19:03
  Speaker 2: What is the policy change?
• 19:05
  Speaker 3: Yes, New York is a large state, which is also
• 19:08
  Speaker 3: why it's a nice state to look at. Has maybe
• 19:11
  Speaker 3: two hundred and fifty thousand births per year, I think,
• 19:16
  Speaker 3: And it was the fourth state in the US to
• 19:19
  Speaker 3: implement paid family leaf and it implemented it in January
• 19:22
  Speaker 3: twenty eighteen. And it started out with eight weeks of
• 19:26
  Speaker 3: paid family leaf and you could take it for various
• 19:29
  Speaker 3: medical circumstances, but one of them was if you had
• 19:31
  Speaker 3: a newborn babe, if you gave birth to a baby,
• 19:34
  Speaker 3: you could take eight weeks of paid family leave within
• 19:36
  Speaker 3: the first twelve months of the baby being born, So
• 19:40
  Speaker 3: that was a generous policy. It had generous eligibility, and
• 19:45
  Speaker 3: the benefits were good, like the percentage of your wage
• 19:48
  Speaker 3: that was replaced was good. And we had seen reports
• 19:52
  Speaker 3: about New York's paid family leave were roughly a third
• 19:56
  Speaker 3: of newborn's had parents who took family leave fits that
• 20:01
  Speaker 3: first year was implemented, so there was a really high uptake,
• 20:04
  Speaker 3: which is something you want to look at when you're
• 20:06
  Speaker 3: evaluating a policy. It's much easier if there's like a
• 20:09
  Speaker 3: robust uptake of the policy.
• 20:12
  Speaker 5: Yeah, because your idea is you're going to compare before
• 20:14
  Speaker 5: and after the policy, not like almost independent of whether
• 20:18
  Speaker 5: people take it up.
• 20:19
  Speaker 2: You're just going to look before and after.
• 20:21
  Speaker 5: And if you know nine people take it, then of
• 20:24
  Speaker 5: course you're not going to see in effect. But that's
• 20:26
  Speaker 5: just because nobody took it. So this issue of uptake
• 20:30
  Speaker 5: is really important because these policies don't have they definitely
• 20:32
  Speaker 5: don't have one hundred percent uptake.
• 20:34
  Speaker 2: They're hard, people don't know about them. Other problems occur.
• 20:38
  Speaker 3: Yes, Yeah, so that was great about New York. I
• 20:40
  Speaker 3: think there was a paper that compared New York to
• 20:42
  Speaker 3: the other states at the time, and it had the
• 20:44
  Speaker 3: most robust uptake among the states that had paid family
• 20:47
  Speaker 3: leave at the time.
• 20:49
  Speaker 5: Okay, so you have your policy. It changes over time,
• 20:51
  Speaker 5: this natural experiment. So you're confident, you're more confident about
• 20:54
  Speaker 5: your causality and your confounding. But now your problem is
• 21:00
  Speaker 5: that you need to have data. And because what you
• 21:04
  Speaker 5: want to study is illness, you actually need some data
• 21:08
  Speaker 5: on whether kids get sick. And so if you were
• 21:12
  Speaker 5: in Europe, it would be great because in Europe they
• 21:15
  Speaker 5: know everything about everybody beginning at birth and going until
• 21:19
  Speaker 5: they're dead, including all of the things that they.
• 21:21
  Speaker 2: Had for lunch on Tuesday and all kinds of other
• 21:23
  Speaker 2: grade stuff. But in the US we don't really have that.
• 21:26
  Speaker 2: So where do you go looking for data of that type?
• 21:31
  Speaker 3: So Jen and I looked for data that was population based,
• 21:35
  Speaker 3: so it covered the whole state of New York. So
• 21:38
  Speaker 3: there are these administrative databases that include all hospital discharges
• 21:43
  Speaker 3: for the state of New York and all emergency department
• 21:48
  Speaker 3: visits data. So these are emergency department visits that didn't
• 21:51
  Speaker 3: result in hospitalization, so they're mutually exclusive. So you have
• 21:55
  Speaker 3: these data sets for most of the states in the
• 21:57
  Speaker 3: United States.
• 21:58
  Speaker 2: And why who collects this?
• 22:00
  Speaker 3: Like who who collects this data the state agencies. Different
• 22:03
  Speaker 3: state agencies collect the data, like in New York goes
• 22:06
  Speaker 3: the Department of Health and Human Services in New York,
• 22:08
  Speaker 3: I think, And they're yeah, differently collected. Sometimes they're sent
• 22:13
  Speaker 3: on in a packaged form to h CUP, which I'm
• 22:18
  Speaker 3: forgetting as hr Q, I'm forgetting who collected the data,
• 22:21
  Speaker 3: But they.
• 22:21
  Speaker 5: Collected data acronyms you like, very rapidly dive in the
• 22:25
  Speaker 5: acronym hole.
• 22:26
  Speaker 2: But yes, some other, some.
• 22:29
  Speaker 3: Other entity in the government collects the data and harmonizes
• 22:33
  Speaker 3: it and you can actually purchase it. But we couldn't
• 22:36
  Speaker 3: use those those packaged data sets because we wanted two
• 22:41
  Speaker 3: pieces of information that aren't always available in those package
• 22:44
  Speaker 3: data sets. We wanted the infants age in days or
• 22:48
  Speaker 3: weeks or at least months, so you could get the
• 22:51
  Speaker 3: zero in one month old because we wanted under eight
• 22:53
  Speaker 3: weeks of age that wasn't available in all these package
• 22:57
  Speaker 3: data sets. And then we also needed zip code or
• 23:00
  Speaker 3: residents for a secondary analysis that we're doing right now,
• 23:03
  Speaker 3: and those aren't always available, so we had to actually
• 23:05
  Speaker 3: go to each state and apply to use the data
• 23:08
  Speaker 3: directly from each state's Department of health and we had
• 23:11
  Speaker 3: to go write an application and go before their Data
• 23:16
  Speaker 3: Governance Board and kind of asked for the data and
• 23:19
  Speaker 3: say our hypotheses and methods and everything. So we did
• 23:23
  Speaker 3: that for New York, Maine, Vermont, New Hampshire, and Massachusetts.
• 23:27
  Speaker 3: New York and then the control states did not have
• 23:30
  Speaker 3: any paid family leave at the time of January twenty eighteen.
• 23:34
  Speaker 5: Yeah, so this is a very important part of a
• 23:36
  Speaker 5: study like this. It's not enough to just get the
• 23:38
  Speaker 5: data from New York, which would let you look before
• 23:40
  Speaker 5: and after. You actually want some comparison places so you
• 23:44
  Speaker 5: can rule out the idea that there was just a
• 23:46
  Speaker 5: change over time by comparing to other comparable states.
• 23:51
  Speaker 2: How how do you choose your comparable states.
• 23:54
  Speaker 6: Well, here, I mean our choice of control states was
• 23:56
  Speaker 6: actually influenced to some degree by data availability. We were
• 24:00
  Speaker 6: looking at hospital visits, which all states collect, but not
• 24:04
  Speaker 6: all states collect information on emergency department visits. So we
• 24:07
  Speaker 6: had hoped to include Pennsylvania, but they don't include information
• 24:10
  Speaker 6: on emergency department visits, so we couldn't really use them
• 24:13
  Speaker 6: as a state. And I think we had also looked
• 24:16
  Speaker 6: into using Maryland, but they don't collect information on I
• 24:20
  Speaker 6: think it was age in weeks or days. They only
• 24:24
  Speaker 6: had it in years, so you could only know if
• 24:26
  Speaker 6: someone was a year older younger. But we couldn't identify
• 24:30
  Speaker 6: the infants that were less than two months, and those
• 24:33
  Speaker 6: are the ones who would benefit from paid family leave.
• 24:35
  Speaker 6: So the control states that we picked was sort of
• 24:38
  Speaker 6: this balancing act of do they have data, do they
• 24:40
  Speaker 6: have the data we want? And are they a reasonable
• 24:43
  Speaker 6: control state? And can we afford them because there were
• 24:46
  Speaker 6: some states I think that were very, very expensive to
• 24:48
  Speaker 6: get the data that we couldn't.
• 24:51
  Speaker 5: I mean, it's you know, I think when people read that,
• 24:54
  Speaker 5: certainly if you read like media coverage or a paper
• 24:56
  Speaker 5: like this, but even if you read the paper, so
• 24:58
  Speaker 5: much of this stuff is hidden. The paper's just like
• 25:00
  Speaker 5: we have, you know, these five control states, and you
• 25:02
  Speaker 5: list them, You're like, okay, sure, But then to dive
• 25:05
  Speaker 5: into like, well every one of those was a decision
• 25:07
  Speaker 5: about you know, what does this match what I need?
• 25:10
  Speaker 2: How much does it cost? Can I show up in
• 25:12
  Speaker 2: front of.
• 25:12
  Speaker 5: The you know, these people and convince them that, like,
• 25:15
  Speaker 5: my project is good and then you get the data
• 25:17
  Speaker 5: and then you have to do well, it's not like
• 25:19
  Speaker 5: they just deliver a result for you, Like, you get
• 25:21
  Speaker 5: the data, somebody has to clean it up and you know,
• 25:24
  Speaker 5: and make it nice before you do anything. And that's like,
• 25:28
  Speaker 5: it's so frustrating because it's so frustrating, but it's it's such.
• 25:32
  Speaker 2: A huge part of the work of a project like this.
• 25:35
  Speaker 5: And then it gets like three lines in the method
• 25:36
  Speaker 5: section like we use these control states.
• 25:39
  Speaker 2: You're like, but let me.
• 25:40
  Speaker 5: Explain to you about that meeting I had with the
• 25:42
  Speaker 5: people in Massachusetts, and I can't believe what they asked
• 25:44
  Speaker 5: me to do.
• 25:45
  Speaker 2: It's like, No, nobody gets to hear about that problem.
• 25:48
  Speaker 1: No.
• 25:48
  Speaker 6: And the one line that I'd hoped we could add
• 25:51
  Speaker 6: into our methods was getting our data from New York
• 25:53
  Speaker 6: State because we had applied to New York State first
• 25:57
  Speaker 6: to try to get sort of some pilot data to
• 25:59
  Speaker 6: see if we could, you know, is there even a
• 26:01
  Speaker 6: signal there that makes this worth pursuing in a grant.
• 26:03
  Speaker 6: So we had applied to them first, and to get
• 26:05
  Speaker 6: the data, you needed an INK signature on the approvals
• 26:10
  Speaker 6: for the data, and our data request made it to
• 26:13
  Speaker 6: them in March twenty twenty, when COVID was at its
• 26:16
  Speaker 6: peak in New York, you know, Department of public Health
• 26:19
  Speaker 6: had many, many, many more important things going on. But
• 26:22
  Speaker 6: this person who worked at the Department of Public Health
• 26:26
  Speaker 6: made a trip from her quarantined home to the house
• 26:28
  Speaker 6: of whoever it was who needed to make this ink
• 26:30
  Speaker 6: signature so that we could get our data in a
• 26:33
  Speaker 6: timely manner. And I, you know, that blew me away
• 26:37
  Speaker 6: that they did that. So I was very impressed with that,
• 26:39
  Speaker 6: and I wish that could have gone into the method section.
• 26:42
  Speaker 5: I totally agree that definitely deserves at least an appendix
• 26:45
  Speaker 5: of some sort more parent data, including what it feels
• 26:55
  Speaker 5: like to actually be reviewed by your peers, what do
• 26:58
  Speaker 5: you do as a researcher when you're research doesn't match
• 27:01
  Speaker 5: your hypothesis, and when it does, and where this research
• 27:05
  Speaker 5: might take public policy?
• 27:06
  Speaker 1: After the break.
• 27:26
  Speaker 5: Okay, so you get all of your data and then
• 27:29
  Speaker 5: you run your analysis. And I'm curious if there is
• 27:33
  Speaker 5: a moment when you have this theory. The regressions you
• 27:37
  Speaker 5: run are in some ways relatively simple. I mean, once
• 27:40
  Speaker 5: you have all the data and it's cleaned up and
• 27:42
  Speaker 5: it's organized and you have your policy thing, it's kind
• 27:45
  Speaker 5: of like a couple of lines of code to get
• 27:48
  Speaker 5: the answer. Is there that moment when, like you first
• 27:52
  Speaker 5: see that the answer is a thing like was that fun?
• 27:57
  Speaker 6: I would say it was more a big exil rather
• 28:00
  Speaker 6: than fun.
• 28:02
  Speaker 5: You know.
• 28:02
  Speaker 6: I think one of the challenges with doing research in
• 28:05
  Speaker 6: an area where you care quite a bit about what
• 28:08
  Speaker 6: the answer is is that it's very challenging if the
• 28:11
  Speaker 6: answer doesn't support your theory and you you know, you're
• 28:14
  Speaker 6: sort of torn between. Of course, you want to do
• 28:17
  Speaker 6: what's you know, most rigorous from a scientific perspective, but
• 28:21
  Speaker 6: when your emotions get in and it's not the answer
• 28:23
  Speaker 6: you want, it's tempting to say, well, maybe we should
• 28:26
  Speaker 6: just check to make sure we model seasonality right, maybe
• 28:29
  Speaker 6: we should just do some other checks. And so I
• 28:31
  Speaker 6: think that I found that a very stressful part about
• 28:34
  Speaker 6: working on this project. A lot of the other work
• 28:36
  Speaker 6: I'm working on right now is things related to like
• 28:39
  Speaker 6: drug safety, where if you find a signal of concern,
• 28:43
  Speaker 6: that's worth knowing and reporting on, but if you find
• 28:45
  Speaker 6: something safe that's also important and worth reporting on, Whereas
• 28:48
  Speaker 6: something like paid family leave is a bit harder because
• 28:51
  Speaker 6: of you know, the feelings invested in it. So I
• 28:53
  Speaker 6: think when we got our answer, I exhaled, and when
• 28:56
  Speaker 6: we did every single sensitivity analysis, I exhale a little
• 28:59
  Speaker 6: bit more. And then one of the last of our
• 29:01
  Speaker 6: sensitivity analyses that we did was the one where we
• 29:04
  Speaker 6: look to see if we saw a similar effect in
• 29:06
  Speaker 6: one year olds who were infants that we hypothesized would
• 29:10
  Speaker 6: not be affected by paid family LEA because their parents
• 29:12
  Speaker 6: weren't eligible. And we saw that there was basically no
• 29:15
  Speaker 6: change at the same time in those infants, and so
• 29:18
  Speaker 6: I think that's when I finally truly exhaled.
• 29:23
  Speaker 2: Kate, what about you? Are you an exhaler? Are you
• 29:25
  Speaker 2: like a celebrator?
• 29:26
  Speaker 3: Well, I think it is the best thing about research
• 29:29
  Speaker 3: actually is uncovering things and finding things and finding things
• 29:33
  Speaker 3: that are consistent with your hypothesis or inconsistent. Both of
• 29:37
  Speaker 3: those are very interesting about the job of a researcher.
• 29:42
  Speaker 3: So I do find I think we did it together.
• 29:44
  Speaker 3: I think so in this project, we didn't hire a
• 29:46
  Speaker 3: data analyst. I am the data analyst, and so once
• 29:49
  Speaker 3: the data were an aggregated format harmonized, Jen and I
• 29:52
  Speaker 3: did coding together on aggregated data, and so we looked
• 29:56
  Speaker 3: at the results and data together when it popped up,
• 29:59
  Speaker 3: and that was kind of exciting. We we didn't do
• 30:02
  Speaker 3: it on our own. We like waited for a regular
• 30:04
  Speaker 3: meaning to do it together. And then also with this
• 30:08
  Speaker 3: to point out this is something that you have maybe
• 30:10
  Speaker 3: some opinion about how the data should go. So we
• 30:14
  Speaker 3: had an elaborate analytic plan that we ran by our
• 30:17
  Speaker 3: co authors and we had like a meeting before and
• 30:20
  Speaker 3: so we were all on the same page because I
• 30:21
  Speaker 3: find that's the best way to do things, where you
• 30:23
  Speaker 3: don't get down into a well what if we look
• 30:26
  Speaker 3: at this way? What if we look at it that way?
• 30:27
  Speaker 3: We had a good pre analysis plan.
• 30:30
  Speaker 1: Yeah.
• 30:31
  Speaker 5: I always people ask like, what what is the best
• 30:33
  Speaker 5: thing about being an academic? And almost like as like
• 30:35
  Speaker 5: it's that moment where like you know something that no
• 30:37
  Speaker 5: one else does and it's like all this work has
• 30:40
  Speaker 5: gone into it's just like that's like so magical and
• 30:43
  Speaker 5: very difficult to I don't know, it's like difficult to describe.
• 30:48
  Speaker 5: There are other lessons parts which I want to get
• 30:49
  Speaker 5: in now. Okay, so then you write, you write the paper,
• 30:54
  Speaker 5: you have your results, you have your beautiful sensitivity analyses.
• 30:59
  Speaker 5: What is the pa between writing a paper like this
• 31:03
  Speaker 5: and having it published.
• 31:05
  Speaker 2: I'm not sure people have any sense of like what happens.
• 31:08
  Speaker 2: What do you do with your paper?
• 31:10
  Speaker 3: Well, first we Jen and I wrote the first draft,
• 31:13
  Speaker 3: and then we send it to co authors who are
• 31:15
  Speaker 3: excellent and gave a lot of feedback, and we revised
• 31:18
  Speaker 3: it based on all the co author comments, and then
• 31:21
  Speaker 3: we agreed upon the final draft as a team, and
• 31:24
  Speaker 3: then we submitted it to a journal. Tory, I would
• 31:28
  Speaker 3: like to say, Jen, the journal. Okay, So we submitted
• 31:31
  Speaker 3: to JAMMA, the Big DAMA, not DEMA Pediatrics, but the
• 31:35
  Speaker 3: over the big one, and it got reviewed, but it
• 31:39
  Speaker 3: got rejected.
• 31:40
  Speaker 5: Okay, So say what that say? Why that's actually meaningful?
• 31:43
  Speaker 5: So when you submit to a journal, you send it
• 31:45
  Speaker 5: in and sometimes they just tell you to go away,
• 31:48
  Speaker 5: like right away, and sometimes they send it to a
• 31:51
  Speaker 5: reviews yep.
• 31:51
  Speaker 3: So this they go away right away is called a
• 31:54
  Speaker 3: desk rejection. And this one they actually sent it out
• 31:57
  Speaker 3: to reviewers and we got comments back probably within a month.
• 32:01
  Speaker 3: They're pretty quick at JAMMA, and the reviewer comments were
• 32:06
  Speaker 3: actually not so bad. I don't think they were mostly concerned.
• 32:09
  Speaker 3: I think it was sent to health economists, and you know,
• 32:12
  Speaker 3: we're mostly epidemiologists. We have actually two health economists as
• 32:17
  Speaker 3: co authors. But Jen and I are more epidemiologists, and
• 32:20
  Speaker 3: so I think they were interested in that we had
• 32:23
  Speaker 3: a different method than is typically used in health economics
• 32:26
  Speaker 3: research and like a controlled different diff analysis or events study,
• 32:32
  Speaker 3: different differ analysis, and so we kind of felt like
• 32:35
  Speaker 3: these comments from these reviewers, though the paper was rejected,
• 32:38
  Speaker 3: they were actually pretty addressable. We had thought a lot
• 32:41
  Speaker 3: about our model and a lot about why not to
• 32:43
  Speaker 3: use this typical method that's used often by health economists.
• 32:48
  Speaker 3: So we actually asked Jamma if we could respond to
• 32:52
  Speaker 3: their reviewer comments and at the same time submit the
• 32:55
  Speaker 3: paper to GEMA Pediatrics, which is another jam of Family
• 32:59
  Speaker 3: journal but specializes in pediatrics, and they said okay, So
• 33:03
  Speaker 3: we did that. We responded to the comments of the
• 33:06
  Speaker 3: original reviewers at the same time we submitted to GEM
• 33:09
  Speaker 3: of Pediatrics, and that process went very smoothly, and it
• 33:13
  Speaker 3: wasn't actually sent out for a review again. They used
• 33:16
  Speaker 3: the original reviewers. It was a very efficient process.
• 33:19
  Speaker 5: One of the reasons I wanted to talk to you
• 33:21
  Speaker 5: guys about this paper as opposed to one of the
• 33:23
  Speaker 5: many other papers I've talked about in the newsletter is
• 33:25
  Speaker 5: I think your paper is really good.
• 33:27
  Speaker 3: Uh.
• 33:28
  Speaker 2: And sometimes I'll write about.
• 33:30
  Speaker 5: Things where I think that the reviewers should have made
• 33:33
  Speaker 5: the paper better.
• 33:35
  Speaker 2: Do you think it is the job.
• 33:37
  Speaker 5: It sounds like you thought that your reviewers at least
• 33:40
  Speaker 5: maybe they didn't make the paper better, but maybe they were.
• 33:43
  Speaker 2: They were not problematic. Do you generally think this process.
• 33:50
  Speaker 5: Works at both sorting out what's good research and improving
• 33:55
  Speaker 5: the research or is this just like a weird waste
• 33:58
  Speaker 5: of everyone's time.
• 34:00
  Speaker 3: I'm curious with John thinks, but I actually do think
• 34:03
  Speaker 3: that most of the time it does improve your paper.
• 34:07
  Speaker 3: There are some times where you send a paper to
• 34:10
  Speaker 3: a journal and you get like a two or three
• 34:12
  Speaker 3: sentence review, and that's just so not helpful because someone's
• 34:16
  Speaker 3: clearly busy, and it's like, fine, they don't really think
• 34:19
  Speaker 3: it's that interesting, and they don't really provide many comments.
• 34:23
  Speaker 3: But I think for the most part, either between the
• 34:26
  Speaker 3: editor's comments or at least one of the reviewer's comments,
• 34:30
  Speaker 3: it does really point out things that could be wrong
• 34:33
  Speaker 3: or confusing or inconsistent with other papers or the literature.
• 34:38
  Speaker 3: I find it's a very helpful process for me. But
• 34:43
  Speaker 3: I'm also a reviewer all the time. I'm always reviewing
• 34:45
  Speaker 3: the paper, so I'm always i'm also part of this process,
• 34:48
  Speaker 3: so I have a stake in the game. What do
• 34:51
  Speaker 3: you think, Jen, how do you feel about your review?
• 34:54
  Speaker 2: I think it's really hit and miss.
• 34:56
  Speaker 6: I mean, I think there's many many papers that get
• 34:59
  Speaker 6: published that you just read and are like, how did
• 35:01
  Speaker 6: this make it through the peer review process? There are
• 35:03
  Speaker 6: so many problems with this paper? Where did things go wrong?
• 35:06
  Speaker 6: And that happens often, I would say, On the other hand,
• 35:10
  Speaker 6: I think the peer review process probably does a reasonable
• 35:12
  Speaker 6: job of sort of making sure that studies that have major,
• 35:17
  Speaker 6: major flaws in them don't make it into a top journal.
• 35:19
  Speaker 6: They're still going to get published somewhere, because that's the
• 35:22
  Speaker 6: nature of publication that if you pay pay money, you
• 35:24
  Speaker 6: can get it published anywhere. But I think, yeah, the
• 35:29
  Speaker 6: top half or the top half is a big mix
• 35:31
  Speaker 6: of ones that should be there and ones that definitely
• 35:35
  Speaker 6: shouldn't be there. So there's definitely a lot of room
• 35:38
  Speaker 6: for improvement in peer review. It's hard to think of
• 35:41
  Speaker 6: a time where I've gotten comments from reviewers where my
• 35:44
  Speaker 6: thought was, oh, I had never thought of that before.
• 35:46
  Speaker 6: That's such a great point. Most of the time it's
• 35:49
  Speaker 6: comments that are worth discussing a little bit more. But
• 35:52
  Speaker 6: because you have a limit on the number of words
• 35:54
  Speaker 6: you can put in your manuscript. You couldn't put it
• 35:56
  Speaker 6: in the first time, and so maybe you get to
• 35:58
  Speaker 6: discuss something in a little bit more detail later, but
• 36:01
  Speaker 6: it's hard to think of times where I've really felt
• 36:03
  Speaker 6: like they made a major improvement to my work.
• 36:07
  Speaker 5: Yeah, I mean, there's an interesting philosophical question about how
• 36:10
  Speaker 5: we like because a lot of different sub like fields
• 36:12
  Speaker 5: of academic field to do this differently. So you know,
• 36:16
  Speaker 5: in economics, there the review process is actually way more involved.
• 36:20
  Speaker 5: Though typically like for a top outlet, it would be
• 36:23
  Speaker 5: like five or you know, could be five people writing
• 36:27
  Speaker 5: you know, fifteen five pages each of detailed comments. It
• 36:31
  Speaker 5: takes a year to revise, Like it's very very long.
• 36:35
  Speaker 5: And then you have fields like like physics, where the
• 36:39
  Speaker 5: kind of review process is like you upload things to
• 36:41
  Speaker 5: the Internet and other people look at them, and then
• 36:44
  Speaker 5: if enough people like look at your it's basically like
• 36:46
  Speaker 5: some kind of like you know, Reddit voting thing where
• 36:49
  Speaker 5: like if enough people up vote your thing, you get
• 36:51
  Speaker 5: to like put it in shot quite like that, but
• 36:53
  Speaker 5: like it's basically a totally crowdsourced structure in like physics
• 36:58
  Speaker 5: or math and then I don't know. So it's it's
• 37:00
  Speaker 5: like it's suggested me there's the process can't be perfect.
• 37:04
  Speaker 2: Because every field is doing it differently.
• 37:06
  Speaker 5: So if we had a really great process, we probably
• 37:08
  Speaker 5: all have the same process, and it must be that
• 37:11
  Speaker 5: it's kind of in some way problematic. But your paper
• 37:15
  Speaker 5: got published, and then when it gets published, then what happens?
• 37:18
  Speaker 5: What's the like, what what do you do? Paper comes out?
• 37:22
  Speaker 5: Does anything happen?
• 37:25
  Speaker 3: Well, this is I think where we I do not
• 37:28
  Speaker 3: do well. So once a paper is accepted, then it
• 37:31
  Speaker 3: goes through this whole like you know, copy editing process
• 37:34
  Speaker 3: and stuff, and then it finally gets published maybe a
• 37:37
  Speaker 3: couple months later. And I think I have colleagues who
• 37:41
  Speaker 3: really promote their work, like on LinkedIn or research gate
• 37:45
  Speaker 3: or I guess Twitter or x or just do a
• 37:48
  Speaker 3: lot of podcasts.
• 37:49
  Speaker 2: Did they come on podcasts?
• 37:51
  Speaker 3: They come on podcasts? Or you know their university issues
• 37:54
  Speaker 3: of press release which actually are my university? Did are
• 37:59
  Speaker 3: they kind of have these?
• 38:00
  Speaker 1: Oh?
• 38:00
  Speaker 3: We had a one pager jen for our previous paper
• 38:03
  Speaker 3: that we issue that was kind of like in lay
• 38:05
  Speaker 3: Layman's language or something. But I kind of feel like
• 38:08
  Speaker 3: at that stage that I've like birtht this thing and
• 38:11
  Speaker 3: it's out there to be consumed. And I don't do
• 38:13
  Speaker 3: a lot of work promoting my papers at all, so
• 38:18
  Speaker 3: this is very unusual that I'm doing this podcast. But yeah, Jen,
• 38:21
  Speaker 3: what do you think? I feel like I get the
• 38:23
  Speaker 3: impression I should be doing more, but I.
• 38:25
  Speaker 6: Don't no my feelings exactly. I'm an introvert by nature,
• 38:31
  Speaker 6: and so going out there and sort of promoting my research,
• 38:35
  Speaker 6: promoting findings is not something that comes naturally to me.
• 38:37
  Speaker 6: I appreciate the importance of making sure the research doesn't
• 38:40
  Speaker 6: just sit on a shelf and collect dust, but I
• 38:43
  Speaker 6: think in general, the people who are drawn to doing
• 38:45
  Speaker 6: science might not necessarily also be people who are drawn
• 38:48
  Speaker 6: to public speaking and kind of engaging with the public
• 38:52
  Speaker 6: and about their findings and that kind of thing. So
• 38:54
  Speaker 6: I think I agree, I could also should also do
• 38:57
  Speaker 6: more to make people hear about the findings.
• 39:00
  Speaker 2: I guess yeah, And I think it's a pretty different skill.
• 39:03
  Speaker 5: But it does also strike me that the relationship between
• 39:07
  Speaker 5: the quality of the work and how much like attention
• 39:10
  Speaker 5: it gets in the media is not I would have
• 39:12
  Speaker 5: said that correlation was about zero, maybe negative, and so
• 39:16
  Speaker 5: it's it's sort of it almost feels like there should
• 39:20
  Speaker 5: be another layer, like it's not really your job too.
• 39:24
  Speaker 5: I'm not sure that it should be the job of
• 39:25
  Speaker 5: the researchers to push their research on the public, but
• 39:30
  Speaker 5: it would be good, especially in a case like this.
• 39:33
  Speaker 5: I think people should know about this. I mean, this
• 39:34
  Speaker 5: is yet another reason we should have more pay leave.
• 39:37
  Speaker 5: It really matters for people's lives, and somebody should I
• 39:42
  Speaker 5: guess I feel like somebody should be helping you do that.
• 39:45
  Speaker 2: But you know, once you, as Kate.
• 39:47
  Speaker 5: Eloquently says, once you've given birth, like can't somebody else,
• 39:51
  Speaker 5: why do you else have to be a tiger mom?
• 39:52
  Speaker 2: You know, like, can't someone else take over?
• 39:54
  Speaker 6: Now?
• 39:54
  Speaker 2: It's in all the hard work.
• 39:56
  Speaker 3: And you have so many other projects in the pipeline
• 39:58
  Speaker 3: to work on, you know, So it's like it's kind
• 40:01
  Speaker 3: of like you don't have a lot of I don't
• 40:03
  Speaker 3: have a lot of time to devote to promoting previously
• 40:05
  Speaker 3: published papers.
• 40:07
  Speaker 1: Yeah.
• 40:08
  Speaker 3: Yeah.
• 40:08
  Speaker 5: And the reward, I mean, guess the other thing to
• 40:10
  Speaker 5: say is the reward inside academia for doing that is
• 40:12
  Speaker 5: quite limited. Like from the standpoint of the things that
• 40:16
  Speaker 5: you know your job cares about, like they will care
• 40:18
  Speaker 5: that you have a jabb at pediatrics paper, for sure,
• 40:21
  Speaker 5: But if that paper is covered in your Times or
• 40:23
  Speaker 5: is covered like that matters less from a public from
• 40:27
  Speaker 5: a promotion prospect.
• 40:29
  Speaker 2: Just it's just not a It's not.
• 40:30
  Speaker 5: A core part of how we're evaluated as academics.
• 40:34
  Speaker 3: When we do reports to NIH on our grant every year,
• 40:37
  Speaker 3: they're a ripper. I forget what it stands for, some
• 40:40
  Speaker 3: kind of report that you do each year. I think
• 40:42
  Speaker 3: they do like to actually hear about in addition to
• 40:45
  Speaker 3: academic publications. They like to hear about presentations or like
• 40:49
  Speaker 3: podcasts or some kind of like dissemination on YouTube or something.
• 40:55
  Speaker 3: They like to hear about alternative ways of disseminating or
• 40:57
  Speaker 3: study findings. So it's not just academic plications.
• 41:01
  Speaker 2: All right, So TikTok, you guys should be on.
• 41:03
  Speaker 3: Ticketak exactly, is what I'm saying. That would be perfect, Jen.
• 41:08
  Speaker 5: I think Jen, you would be perfect on a little dance,
• 41:11
  Speaker 5: you know, simply it seems like we really could we
• 41:14
  Speaker 5: could get something good here.
• 41:16
  Speaker 3: All right?
• 41:16
  Speaker 2: Before I let you guys go, tell me what's next?
• 41:18
  Speaker 5: What are you Is there a new project that we
• 41:20
  Speaker 5: should be covering on this in a few months when
• 41:21
  Speaker 5: it comes out.
• 41:23
  Speaker 6: We are working right now at looking some paid family
• 41:26
  Speaker 6: leave and to see how it affects different disadvantage groups
• 41:29
  Speaker 6: in the population. So looking to see if there's different
• 41:32
  Speaker 6: effects of the policy in different Yeah.
• 41:35
  Speaker 3: So looking at disparities. So one of the biggest concerns
• 41:38
  Speaker 3: about paid family leave or any kind of policy effect
• 41:40
  Speaker 3: is it can have unintended consequences. And we actually wrote
• 41:43
  Speaker 3: a piece about this for another paper that came out
• 41:46
  Speaker 3: in Gymipediatrics recently. It can exacerbate or reduce disparities. So
• 41:51
  Speaker 3: what we're doing now is we're using the same data,
• 41:54
  Speaker 3: but we're looking to see if the effect of the
• 41:56
  Speaker 3: policy was the same across different race groups, across Medica
• 42:00
  Speaker 3: versus commercial payer, and then by Childhood Opportunity Index, which
• 42:05
  Speaker 3: is kind of like a socioeconomic index that's been developed
• 42:08
  Speaker 3: for kids, to see if there was an equal effect
• 42:11
  Speaker 3: of the policy across these groups. So that's that's the
• 42:14
  Speaker 3: paper right now.
• 42:15
  Speaker 2: That's great. Thank you guys. This was absolutely a delight.
• 42:21
  Speaker 5: I really appreciate you being here and promoting your promoting
• 42:24
  Speaker 5: your work on this podcast, which I hope will go
• 42:27
  Speaker 5: in your Ripper report for that.
• 42:30
  Speaker 3: It definitely, it definitely will thank you for having us.
• 42:33
  Speaker 3: Thank you.
• 42:41
  Speaker 5: Parent Data is produced by Tamar Avishai with support from
• 42:44
  Speaker 5: the parent Data team and pix. If you have thoughts
• 42:47
  Speaker 5: on this episode, please join the conversation on my Instagram
• 42:50
  Speaker 5: at Prof Emily Ostar, and if you want to support
• 42:54
  Speaker 5: the show, become a subscriber to the parent Data newsletter
• 42:57
  Speaker 5: at parentdata dot org, where I write weekly posts on
• 43:00
  Speaker 5: everything to do with parents and data to help you
• 43:03
  Speaker 5: make better, more informed parenting decisions. For example, earlier this fall,
• 43:08
  Speaker 5: I wrote an article that went deeply into genicates papers
• 43:11
  Speaker 5: findings and explain where this paper fits in the world
• 43:14
  Speaker 5: of public policy, which I think we can all agree
• 43:17
  Speaker 5: is in desperate need of data like this.
• 43:20
  Speaker 2: Read it at parentdata dot org.
• 43:23
  Speaker 5: There are a lot of ways you can help people
• 43:24
  Speaker 5: find out about us. Leave a rating or a review
• 43:27
  Speaker 5: on Apple Podcasts. Text your friend about something you learned
• 43:30
  Speaker 5: from this episode. Debate your mother in law about the
• 43:33
  Speaker 5: merits of something parents do now that is totally different
• 43:35
  Speaker 5: from what she did. Plose the story to your Instagram,
• 43:38
  Speaker 5: demunking a panic headline of your own. Just remember to
• 43:41
  Speaker 5: mention the podcast too. Write Penelope right well, We'll see
• 43:46
  Speaker 5: you next time.