CHRIS BARRETT: And it's especially a pleasure that one of our team members, Christine Olson, indeed one of the core team members who conceptualized the project, is going to kick us off. So Christine will talk today about this issue of hunger and food security in the United States and share with us some of the work that she and her colleagues and students have been doing on very fundamental questions about persistent poverty manifest in hunger and food security issues. So Christine, thank you.
CHRISTINE OLSON: Thank you, Chris. It's great to see so many of you here today. I'm going to be talking about professional work that's taken up about 25 years of my life. And I'm going to try to do that in an hour. And I'm going to restrain myself and not go off onto tangents, which probably every slide there is a potential tangent track. But I'm going to discipline myself and not go off on those tangents.
The talk's divided into four major sections. I'm going to talk first about the measurement of hunger and food insecurity in the US, then correlates-- we might think of those as antecedents and consequences of food insecurity-- the dynamics, and then finally, save a little time to talk about what can be done. And that will transition us into a discussion and question-and-answer period.
So we're going to first talk about measurement. In the early '80s, we actually had what I would call a real-world research problem dumped in our lap. You will recall that in the early '80s was the last time we had a major economic recession in our country. The mayors of our major cities were showing up on the steps of the White House, talking to Ronald Reagan, reporting that they were seeing in their cities a most serious emergency. They had more people showing up at soup kitchens and food banks than they could handle. And they wanted help.
The attorney general at the time, Ed Meese, said, I don't know of any authoritative figures that there are hungry children. I've heard a lot of anecdotal stuff. President Reagan went on to appoint a Food Assistance Task Force that looked into this issue. And what's interesting is the last part of this quote that basically says that at this time it's "impossible to estimate the extent of hunger. We cannot report on any indicator that will tell us where and by how much hunger has gone up in recent years."
So that was the beginning of some work here at Cornell University on hunger and food insecurity. Now, the problem with the term hunger at the time-- and some of you who are about my age or a little bit younger will remember that in the early '80s was one of the times that there were major famines in Africa. And the average American's picture of hunger was a starving child from Africa with a swollen belly, flies around the eyes. And they knew that wasn't what hunger looked like in the United States.
So from sort of public policy perspective, and from a public perception perspective, using the term hunger represented a problem because hunger, as it was being used by policymakers, didn't represent what the average American thought the experience of hunger looked like in the US. For them, hunger equaled starvation.
So I had a very bright doctoral student at the time named Kathy Radimer. And she said, I think we can take on this issue. And I think we can do research that will advance the definition and measurement of hunger and food insecurity in the US. And this is the title of her doctoral dissertation.
So she began with a study, an in-depth interviews study, guided by a naturalistic paradigm where what you're trying to do is to get an understanding of the problem of food deprivation from the perspective of people who've experienced it. So she did in-depth interviews with 32 women from central New York. Most of these women had children. We worked in both rural and urban areas of central New York.
And we started out looking for women who said they had gone hungry. And we weren't getting very many people. So we then said, ah. Have you been close to going hungry? And we got more people to volunteer to be in the study. And we went to places where you would expect to find families who'd been struggling to feed themselves.
And then based on that qualitative work, we then did a more quantitative survey with 189 women with children from the same geographical areas. We analyzed the in-depth interviews using the constant comparative method. And I'm going to give you a quick overview of some of the results of that qualitative phase, since it was probably the most important to the understanding of food insecurity and hunger.
This next slide is a quote from one of the women in the study that addressed the first problem we had, which was what is hunger? And what is this something else that we're not quite willing to label hunger, but it's viewed as problematic? And when you do this kind of work, you transcribe the interviews. The interviews are taped. You transcribe them.
And when this quote came up, Microsoft Word told us, you've got a double word there. Because this woman talked about hungry hungry, which was her way of saying this was really a problem. And that's when there was absolutely nothing to eat in the house. But you can see, she then goes on. And she only uses the word hungry once.
And she says, "but also going hungry is when you have to eat the same thing all week long, and you have no variation from it. You know sooner or later you're going to run out of that, too, because it's only going to go so far. So each day, you cut down the portions a little bit smaller and a little bit smaller. And you have a tendency to send your kid off to play with somebody else, so that they're there at mealtime so they do eat."
So this woman was clearly making a distinction between real going without eating-- there's nothing to eat-- and this other problem where you eat the same thing. You try to cope with what you see as impending depletion of your household food supply. So you eat a little less and a little less. And some of those constructs that she mentioned were key in Kathy Radimer's definition, the definition that she developed from the interviews with, obviously, other women besides this woman.
So from Kathy Radimer's work, we came up with this definition of this less-severe-than-hunger problem. And we called it food insecurity. Because that term had been around for a while in international circles to describe the food situation within countries. We were using it to describe the food situation within households and individuals. So food insecurity is "the inability to acquire or consume an adequate quality or sufficient quantity of food in socially acceptable ways, or the uncertainty that you'll be able to do so."
This table gives little another depiction of the results. Here, what we're looking at is a conceptual structure. And you can see that from the interviews, Kathy determined that there was a quantitative component that dealt with the amount of food available, a qualitative, the quality of the food, the dietary, nutritional quality of the food available. There was a psychological component and a social component to this construct of food insecurity.
Food insecurity was experienced differently and had different core elements at the household level-- and that's what's shown in the middle column-- than it did at the individual level. And at the individual level, adults in households-- and our case, it was the women, because that's who we interviewed-- and children experience these things differentially.
So we then, when we went on to develop the items for the survey, used the words of the women who we had interviewed to construct items that tapped each of these components at each of the levels. And we tested those items and did reliability analysis. And then in a later study, we did some validity work, which I'll tell you about.
So the products of Radimer's research were a definition of food insecurity that was relevant to a food-rich country like the United States, a conceptual framework or an understanding of the key constructs, and then items to measure food insecurity. So that's really where we were. This was published in '92. But the work was-- she defended her dissertation in '90. And the work was done at the end of the '80s.
In 1990, our federal government passed an act to create a national nutrition monitoring and related research program. This was a time when in countries around the world, nutritional surveillance of the population was very popular. And many countries had surveillance systems to detect impending famines and that kind of thing. And in the US, we weren't looking for famines. But we were trying to create a system for monitoring the nutritional status of the population. And that's what this act was about.
And one of the tasks that was included in the legislation was developing a mechanism and an instrument for defining and obtaining data on the prevalence of food insecurity or food insufficiency in the US and then methods that could be used at the federal, the state, and the local level for doing it. USDA was the agency of the federal government that was given the responsibility for doing this.
At the same time, our professional nutrition organization working with the Life Sciences Research Office that's part of this Association of Experimental Biologists came up with their definition of food insecurity, which is actually quite close to the Radimer Cornell definition. And that's not so surprising, since there were a few people from Cornell on the committee. Jean Pierre Habicht was one of them. And also a definition of hunger.
So we had the items from the Radimer Cornell measure. And we'd done some reliability analysis. But we hadn't done the validity work that I think, really, you would want to have done if you were going to put these items on any kind of a national survey or even for research purposes. So we were very fortunate. New York state was interested in doing surveillance of the nutritional status of populations in New York state. They provided some funding for us to do a validation study of the Cornell Radimer measure.
And we selected a relatively typical rural county in upstate New York, in part because we could do a random sample of the population. And that was one of the key things that needed to be done in this work. The work to date had been done among people who were showing up at food pantries and soup kitchens. You know they're having a problem. But we wanted to do this in a general public population.
So we were able to select our stratified random sample of households with women who had children in the home. We over-sampled the low-income population, because we wanted have a big enough sample of low-income households. We conducted in-home interviews using a questionnaire. And we weighed all the women in the household. And we weighed some of the kids in a standardized way using a scale.
And we returned three weeks later and did a second 24-hour recall for getting dietary intake. And then we did a complete food inventory, because we thought that would be the gold standard. If there was no food in the house, the household was probably experiencing food insecurity.
So our sample was 193 households. And I'm not going to show you all the validation results, just a few selected ones. I've tried to put the papers at the bottom of each slide. And after this talk, we will put a list of the references up on the ISS website so you can go find the references if you want to. I was hoping to have a handout. But the time gods were not with me on that.
So here we have households that were food secure. There where 90 of them. And from then on, we have households that were food insecure with different degrees of severity. So for those who had household food insecurity, that's the least severe level. Individual level-- and this is adult food insecurity. And this is households in which the children in the household were going hungry. That's the most severe level.
Children in this country are protected within households from food insecurity. Parents know that your children can be taken away from you. It's deemed neglect if you do not feed them. Parents do everything, practically, to make sure their kids are fed. They go without eating themselves. And that's why this-- it's important at the individual level to make the distinction between adults and children.
Now, don't worry. These don't add up to this. I'm just showing you selected of the food groups that were in the inventory. And basically, higher numbers mean that there's more of this food in the house. So this is the sum of dairy products. And you can see that when it came to dairy, the households that were food secure had much more dairy than those in which the situation had deteriorated to the point that the kids were going hungry.
And that's true for all the food groups. These are all highly statistically significant. And when we add other groups to this and look at the total, you can see that there's a difference. So indeed, the Radimer Cornell instrument was able to distinguish between-- was related to the amount of food that was in the household. We also did 24-hour dietary recalls. The main foods that differed on the 24-hour recall were fruits and vegetables and the nutrients that you would expect to find in fruits and vegetables, vitamin C, vitamin A, fiber, folate, those kinds of things. So that's one of the pieces of validation work that we did.
The second piece was really looking at what we would call the sensitivity and the specificity. And I don't want to go into a lot of detail on how we developed this. But we developed a definitive criterion measure. And then we had two independent raters read through all the material we had on each of the households and say whether they thought this household was food secure or food insecure. And then we looked at how this household scored on the Radimer Cornell measure.
And in terms of specificity, these are the two things that are important. This is the Radimer Cornell measure with one of the uncertainty items deleted. So in 63% to 71% of the cases, if the Radimer measure scored the household as food secure, it was also scored that way by the definitive measure. I should've said it the other way around. I'll do it better the other way. The sensitivity, which is how good is this Radimer measure? How good is it at defining a household in the same way that these independent raters did on the criterion measure? And here, you can see the sensitivity is even higher, 89% to 84%. So this measure seemed to be pretty darn good. Yes.
SPEAKER 1: What are some of the measures in the Radimer [INAUDIBLE]?
CHRISTINE OLSON: I worry my household is going to run out of food before I'm able to buy more. I can't afford to eat a balanced meal, a balanced diet, are two examples. So it's those kinds of questions. And then people say how often is that true of your situation, the situation of your household? Yes.
SPEAKER 2: Overseas culturally, you'd see that food doesn't necessarily flow to the children but to the wage-earner laborer. Do you see any of that culturally in the US, or is it dominantly, as you said, that parents take care of their children first or get them taken away from them?
CHRISTINE OLSON: Well, among adults in the household, there is some differentiation. Generally speaking, the woman in the household is a household food manager. That's a term we sometimes use. She is the one who eats last. It's that plain and simple.
The kids will eat first before both adults. Then often it's the husband or the partner, the male partner, eats next. And the woman says things like, well, you know, somebody will bring donuts to work. We'll have coffee and donuts. I don't need to eat. I'll eat then. Or I'll eat what's left on the kids' plates after the kids have eaten. That's typically what many of the women in these households did. And that's what they--
SPEAKER 2: Was there cultural diversity in the New York sample group?
CHRISTINE OLSON: We have African American and Caucasian. We do not have-- this does not represent the racial ethnic diversity of the US as a whole. But we have, as you would expect in central New York and some of our major cities, you can find a fairly large African American population.
OK. So that's where we were about 1993. We'd developed this measure. We'd done some validation work. We were feeling pretty good about it. We worked with USDA, working with contractors like Mathematica Policy Research and Abt Associates to help them develop what was an 18-item food security measurement tool.
13 of those items came from the Radimer Cornell measure. Those were the ones at the food insecurity end of the scale. The hunger items came mainly from the Community Childhood Hunger Identification Project, sometimes called CCHIP. That was the other group that was involved with USDA in the development of the measure.
USDA funds annually the assessment of the prevalence of hunger and food insecurity through the CPS. And this just shows the rates from the beginning, about 1996. This is the prevalence rate for food insecurity. You can see it follows the poverty line. It started high, almost 12%. The low was in about '99, 2000, where it got down to a little under 10%. It's crept up. And this is a household measure. So these are proportion of households.
And in this next slide-- well, in a couple more slides, I'll show you what the latest prevalence statistics are. The only other thing to say is you see that this is now called very low food security rather than hunger. And there's a reason for that, which I'm going to tell you about very shortly.
So fast forward to 2006. And the USDA ended up requesting that the National Academy of Sciences review the concepts, methods, and measures of hunger and food insecurity. This was probably a politically motivated act. But it doesn't hurt to check on things every so often. So ultimately, in this document that I have right here, Food Insecurity and Hunger Measurement, the panel of the National Academy concluded that the USDA should continue to measure and monitor food insecurity regularly in a household survey.
Given that hunger is a separate concept from food insecurity, USDA should undertake a program to measure hunger, which is an important potential consequence of food insecurity. That is the reason the current what used to be hunger is now called very low food security. I will be very honest and straightforward that when we were involved in this work developing the Radimer Cornell measure, we worked really, really hard on the food insecurity end of the scale. And the reason we did is we came at this from a public health perspective.
From a public health perspective, early intervention to prevent something before it becomes a problem that you then have to treat is the most efficient approach. And so we said, that's where we're going to put a lot of our energy. You don't wait till somebody's in intensive care before you define them as sick. You don't wait until a household's food situation has deteriorated to the point that the kids are going hungry to say that there's a problem with food in that household. And that was our attitude.
So most of the items at the least severe end came from Cornell. The items at the hunger end of this scale, the most severe end, mostly came from CCHIP. And we had difficulty in our validation studies. We did not get the same kind of validity results at the hunger end that we got it the food insecurity end. But I have to say, given the public health perspective that we were coming from, we didn't worry about that too much.
But you can see that this problem has come home to roost. And so they're fine with the food insecurity stuff, the National Academy panel. But they see some problems with the hunger end of the scale to the point that they won't allow anybody to talk about this as hunger. They have to talk about it as very low food security. So this is an area of work that needs to be taken further. I don't know that you can do it through questionnaire methods myself. But maybe it's possible.
So here's the latest statistics for the US from 2007. This survey is done in December. So in 2008, it was done in December, 2008. USDA generally releases the data in the November of the next year. So in November of 2009, we'll see what happened to these statistics during-- I think all of us know the recession had started.
So we're ticked back up now to 11.1% of households have food insecurity. They have trouble with ready, steady access to sufficient, nutritious food to feed themselves. I've said hunger. So 4.1% of households have very low food security. I think I'm willing to call that hunger. The situation has deteriorated to the point that kids and households are going without meals and going days without eating at home. That's what that means. Nick, you look like you wanted to ask a question.
NICK: Oh, maybe I can wait.
CHRISTINE OLSON: OK. So food insecurity has been ticking up since about 1990, 2000. And at the end, you'll see what our public health goal is and where we want to be and how far we are from it. So that really closes out the measurement section. We did a lot of work. Cornell was very involved in helping USDA develop the measures.
We then, since we had a valid measure, started doing research on what predicts whether a family is food insecure and what are some of the consequences of this food insecurity. And we've not done a lot of work nationally, although we've done some analysis of national data sets. A lot of our work has been in rural New York state, where we are.
So obviously, poverty is associated with food insecurity. So families with children are more likely to be food insecure than families without children. And families with income less than 185% of the poverty line are more likely be food insecure. This is national data.
This is the thing that just worries me no end. And you'll see why when I show you some later results. In 2007, 35%, more than one third, of households with children who make less than 185% of the federal poverty line-- and that 185% of the federal poverty line for a family of four is about $40,000-- were food insecure. So on average, 11% of households may not seem like so many. But then you take households with children, including the working poor, and you bump this number up to a third of households. And that's a pretty high prevalence.
I just put this in here quickly because I'm going to talk a lot about the federal poverty line. And so it's always useful, I think, for people to see what that is. So this is if you're at the federal poverty line and you're in a family or household unit the size of one, if you make less than this amount of money, you're defined as poor. Family of four, this is the poverty line, $22,050.
SPEAKER 3: That's gross income?
CHRISTINE OLSON: Yep. Yep. Yep. This I thought was interesting to see, because without a doubt, if you're at-- this would be 100% of the federal poverty line. The rates are very high. But what's interesting here, if you follow the blue line, which is the food insecurity, you have to get up to fairly high incomes before this number goes below 5%. And I think this speaks to the high level of other expenses that many relatively well-off households have. Their housing is expensive. Their transportation is expensive.
And so I will just tell you in one of Kathy Radimer's subjects that she interviewed in these 32 women, she interviewed the wife of a professor at Syracuse University, believe it or not. The guy had left, emptied the checking account, and this woman was left with nothing. And she had three kids that she had to figure out how to feed.
So in a survey, she would be counted as having a fair amount of income. She would probably be up here someplace. But she's not the norm. I mean, what's more normal is households who have very high expenses. They live in very expensive places so that they can send their kids to good schools. And they don't have the money to feed themselves, because their money goes to other expenses.
In our work in rural upstate New York, we found that many of these households who are food insecure have very little wealth. They have no savings. This was actually the most important characteristic distinguishing between food-secure and food-insecure households was whether the household had savings or not.
What assistance there is is often insufficient to meet their food needs. These family spent very little money on food. They skimped on food to meet their other expenses. And then when something unexpected comes along-- the car breaks down, the furnace breaks down, they need a new roof-- they have to use that little bit of padding that they have to meet that expense to maintain their livelihood and they don't eat.
OK. I seem to attract Kathys. So another large body of work on the consequences of food insufficiency for children was done by Katherine Alaimo for her doctoral dissertation, which she defended in 2000. I think she got promoted. She's an associate professor at Michigan State in nutrition.
So Katherine did a lot of work on the so-what of this for children. And this-- just as an example of one of her papers that was published in Pediatrics. So children in food-insecure households don't do as well in school. They have behavioral problems. They have to see a school psychologist. They have difficulty getting along with other children. She's also looked at some health measures, whether these children are in fair, poor, or good health, find similar kinds of things. So that's one of the potential downsides of the experience.
We have done some work with the elderly, although maternal and child nutrition is my area, so I've done more with women and children. We did some interview studies to try to see if the conceptual framework for the experience of food insecurity in the elderly was different than it was in households with women and children.
And one of the things that we discovered in these low-income elders is 90% of them had a chronic condition that was managed by nutrition. Nutrition was part of the therapy-- diabetes, high cholesterol, high blood pressure. And one of the things that happens when they start running out of food is they can't follow their medically prescribed therapeutic diet.
So this is a quote from a woman who had diabetes. She takes insulin. She knows darn well she's supposed to eat after that insulin shot or she's going to go into insulin shock. And we are paying for this ambulance ride. This woman couldn't eat. She took her insulin. She went into insulin shock, meaning her blood sugar went really, really low. She passed out. They called the ambulance. She goes to the emergency room. That is paid for with a public dollar.
And as a matter of fact, one of my colleagues, Cara Nelson, who's a physician, has done a lot of work on diabetics who experience food insecurity and shows that the average individual who has diabetes who also has food insecurity uses four times as many physician phone calls per year, uses an estimated 2.4 million physician encounters. I mean, it's a place where clearly food would save a tremendous number of health care dollars.
OK. So just a quick review of food insecurity and its consequences, so decreased levels of food and nutrient intakes, dietary intakes. But this shows up in blood levels. You should do blood level analysis. You can see that. Compromised health and cognitive achievement of children, that's Alaimo's work. Inability to follow medically prescribed diets in the elderly and others with chronic conditions.
And then probably the one that has given people fits. And I remember when we approached the New York State Department of Health for the funding, for the validation of the Radimer study, somebody from the health department looked at me and said, lady, you gotta be nuts. We do not see how these people can be having any trouble feeding themselves because they're all fat. It was that straightforward.
So that is why, in that validation study, we thought it was really important to weigh and measure the women in particular and the kids if we could. The reason we didn't do all the kids is kids were often off at school when we were in the home because that was the most convenient time.
So we actually did one of the first studies finding this relationship-- and I'll explain this slide-- between body mass index, our measure of overweight and obesity, and then obesity as a percent. We couldn't get it published. We could not. We were done with this work in 1994. Nobody would believe it. So but finally, the Journal of Nutrition came through. And it got published, initially in a supplement.
So here again, this slide is laid out like that other one. These are the households that were food secure by the Radimer Cornell measure, least severe level of food insecurity, and then increasingly severe levels of food insecurity. So on the far right, these households are really struggling. The kids are going hungry. And you can see the measure of obesity body mass index is 25.6. It's about two units higher, a little more than two units higher, in those with the least severe level of food insecurity. Then it goes back down. So the households where there is something that's more like hunger look like the food secure in terms of the proportion of women who are obese or have a body mass index that's high.
We are not the only ones to find this. There's been now six or seven more studies and only one that finds something in the opposite direction.
SPEAKER 5: [INAUDIBLE] stuff part of how you're defining these things, though. Because you've got hunger. And you have malnutrition. And those are two different components of expression of food insecurity. And depending upon which one weighs more heavily than the other, you'll have these kinds of outcomes that we've seen.
CHRISTINE OLSON: What is happening here is one of the consequences of deprivation. The psychological experience of food deprivation is overeating when food becomes available. That's why diets don't work. That's why, in the dieting literature for weight loss, diets don't work.
In these households, the women were putting-- it wasn't that there wasn't anything to eat. But there were times in the month when the household food supply was really tight. And remember, I said to you the woman is the buffer. She's the one who doesn't eat when food supply gets tight so other people can eat.
So what's happening here, I think, is it's like what happens when people go on diets. They're deprived. Food becomes available. So the food stamps come in. The paycheck comes in. Somebody goes shopping. There's an influx of food into the household. There's the opportunity to eat because there's food there. And women overeat. It's a well-documented psychological response or a behavioral response to a psychological situation that happens with food deprivation.
And just think about it yourself sometime when you know you're not going to be able to get something to eat, what you do. I mean, you may overeat. Or you may, if you have the opportunity, you may not eat, but then when you have the opportunity to eat, you may go hog wild. And that's probably what's happening here.
Here, biology is taking over again. The food situation has deteriorated to the point that there isn't enough food in the household for everybody to eat. So people just aren't eating. And it's a chronic, long-term, low-food-supply situation that's going on there. And we actually have done behavioral measures of eating patterns. And they're consistent with the story that I told you.
DAN: But isn't it they're also buying cheap, high-calorie food? I just finished reading The Grapes of Wrath. And they were eating flour and lard.
CHRISTINE OLSON: This is a very, very important and interesting question. Dietary studies are the most difficult studies there are to do in nutritional sciences, in my opinion. Getting accurate measures of food intake is extraordinarily difficult. So that's the preamble. Interestingly, except for lower levels of fruit and vegetable consumption in food insecure household and the nutrients that go along with that, it is very hard-- and USDA has looked at it up one side and down the other-- to find higher proportion of calories from fat, higher proportion of calories from sugar, in food in food-insecure households. But we don't have the evidence to support the statement that you made.
I think it's real. But given the difficulties of measuring diet, it might be just that the error and the variability is so great we can't detect it. But we do not find the tremendous stacking up and eating of junk in these households.
DAN: --argument though, about people who are overweight are less productive in the workforce. They're more likely to be unemployed, so the cause and effect is exactly what would be [? reverse? ?]
CHRISTINE OLSON: So let me show you-- that leads nicely to this. So what Dan is saying is what's cause and what's effect? That's mostly what I'm going to talk about for the next part of this. And it starts with, actually, our work on obesity. So you asked a good question. Did you want to--
SPEAKER 6: Does the measure measure sort of access to foods like grocery desert or where groceries--
CHRISTINE OLSON: No. We don't. We don't in this measure. But that's a whole area that a lot of work could be done on. So we were really interested in trying to determine this direction of causality between food insecurity and obesity. And because of my other work on gestational weight gain, postpartum weight retention, we had a sample of women who had had children from the catchment area of Bassett Heathcare. And we had asked food insecurity initially. And we'd asked food insecurity at two years postpartum.
And what we found here is a strong interaction between food insecurity and body weight status and weight gain across time. So what you can see here is it was only among the women who were initially obese that food insecurity was related to major weight gain. So that made us think, well, there could be a lot of things going on. But one of the things is that both obesity and food insecurity have a common root. There's something that's related to both those things.
And that got us on to the whole issue of growing up poor. One of the things that's well documented in the literature by many investigators from other countries of the world who have access to longitudinal data sets-- kids who've grown up in poor households are more likely to be overweight or obese as adults than kids who've grown up in higher-income households. So what I was really interested in is what is the mechanism?
So using another data set that we had with a small sample size-- but we had really in-depth work-- we looked at women in households who, when they were growing up, their household is defined as low SES, meaning the household got welfare and the adults, the parents, did not graduate from high school. The higher SES households weren't on welfare and the parents had graduated from high school.
And you can see that there's about a two-fold difference in the proportion of the women. This is a small sample. And this wasn't designed to be a quantitative study. The one we were showing on the other side is really the bottom line. There was a two-fold difference in the proportion that were overweight and obese.
So what was the mechanism? And this comes back to this eating pattern. So you grow up in a poor household where you experience food insecurity, you develop attitudes and patterns of eating. Look at the difference here. It's a small sample. It's not-- but a huge difference in-- zero is good here. And very few of the low SES households got a score of zero. But 40% of the high SES households had a score of zero. So the experience of deprivation in early life leads to a patterning of eating, a use of food, a more emotion-based use of food, and attitudes about food that predispose to obesity and weight gain.
And so it turns out these interests of mine ended up being linked, obesity and food insecurity. And generally, people who are obese do have a more feast-famine pattern of eating. Other people have found that.
So that's kind of where we're at. This story is not over, how this food deprivation in early life is related to health in later life. But recall again, 35% of children in the US in households with incomes less 185% of the federal poverty line experience food insecurity. So I don't think obesity is going to go away for a while, if these findings that clearly need to be replicated and supported are anything close to the truth.
Now, let's see. I wonder what priority I should put on the dynamics. I think maybe I should march through, if that's OK. So now we're onto the dynamics section. And what we're really looking for is what is it that holds families back from escaping food insecurity? And I'll try to be quick.
So here we've done a small study in a rural sample in upstate New York, 29 families we followed for three years. All these families have incomes less than 200% of the federal poverty line. We held on to all but one of them. We started with 30. Ended up with 29. And we're interested again in the mechanisms that allow families to leave or get stuck in food insecurity.
So at Wave 1 in our study, 17 of those 29 families-- so that's about 59%-- were food insecure. By Wave 2, about 50%-- so this is a year later-- had moved out of food insecurity. But by Wave 3 three years later, 65% of this original group was food insecure. So some people moved out and went back. And 35%, over that time period, left food insecurity.
As we had rich data-- and here I'm just showing you the results of a screening tool that we use for measuring depressive symptoms. And you can see that of those initially food insecure, those who had depression at Wave 2, the midpoint, 100% of them remain food insecure. The people who moved out, very few had depressive symptoms. They had no symptoms. And 33% of those were food insecure. It was highly significant. That was just published in the journal I showed at the bottom.
Now, these 30 families were part of a national data set that covered 10 states. And here, we were able to do some more quantitative things. These are the states that were involved. Those in orange lost 50% of their sample, so we excluded them from the analysis.
And again, we see this. In those households where the woman was not at risk for depression-- she didn't have depressive symptoms-- were much more likely to leave food insecurity. Again, this growing up poor was also-- those who grew up in better socio-economic circumstances were more likely to leave food insecurity. And those who weren't hungry were more likely to leave food insecurity.
So food insecurity's fairly persistent in these low-income families. This is not the same picture one gets when one looks at the whole population. When you look at the whole population, food insecurity looks pretty transient. But you look at poor populations, I think it's chronic. Human capital factors seem to be particularly important.
So programs and policies-- what can we do about this? It's good Cornell has given me an extension outreach responsibility for my 35 years that I've been here because it's what I really like to do. So we actually have a public health objective for the United States related to food insecurity. By 2010, only 6% of households are supposed to be food insecure. 94% are fine. And this goal came out of the 1996 World Food Summit.
We are going the opposite direction. We're at 11.1%, up from a little under 10% since 2000. And were supposed to be at half that. So we have not made progress. And we're not going to hit the target. I don't need to say anything about that.
In doing something about this, though, this is one of these cases where the national picture may mislead you, where you really do need to know about the target audience that you're working with. So if you looked around here in rural upstate New York, in our validation study, 71% of the households that were food-insecure were two-parent families. 64% had at least one household member who was employed. 81%, the adults had graduated from high school. But 62% made less than $20,000 a year.
SPEAKER 7: Can I just amplify that a little bit or puzzle it a little bit? Because all those characteristics that you have listed are probably within a few percentage points of what you would list if you were describing poverty, rural poverty. There are two-parent families. They have one that's got employment. They have high school-- they don't make a lot of money. All right. So that list correlates very well with rural poverty also. And so what's the poverty-- you know what I'm saying? So are you just talking about characteristics of rural households or rural poor?
CHRISTINE OLSON: Yeah. I think this is what-- national poverty, national food insecurity, doesn't look like what we see in the more rural, and I would even say, rural, quasi-suburban areas around small cities. It looks different. But I think the reason why we have to distinguish between food insecurity and poverty is that slide that I showed you where food insecurity doesn't go away for households that are making three and a half, four, five times the federal poverty line.
And then a second reason is because our major way-- one of our major types of assistance that we give to poor families now is food. So food insecurity, nutrition is tied up in poverty in a whole lot of convoluted and interesting ways.
So what are we doing? That's where I want to start. This figure, I think, is just stunning. And some of you have probably seen that. In July of this year, one in nine Americans participated in-- it's now called SNAP, but it's the food stamp program.
Our other big way of providing service, supposedly addressing hunger and food insecurity, is charity-based food assistance. Feeding America, which is now America's Second Harvest, provides food to 25 million Americans a year in any given week. And they've been careful not to double count. They serve 4.5 million different people.
So what we are doing now-- you all know the story of the babies coming down the river. This is my "babies coming down the river" slide, where we're throwing food at those cradles as they're going by. And I think we have to go upstream and look why the babies are getting in the river and address some of the underlying causes of food insecurity.
And it's the things that we-- low wages, under or unemployment, lack of education, and a skill level that prepares you for a job-- obviously, there's got to be jobs for people to have for this to make a difference-- and then the work on depression, I think, access to affordable, quality physical and mental health care.
When we've looked at the reasons that the people who are depressed don't leave food insecurity, it's that the depression is so debilitating that the members of the household can't work. And it might be that it's the woman who's depressed, so she can't hold a job. It might be that because she doesn't have proper care, her husband has to stay home to take care of her.
In some of our families, it's the kids who are depressed. They can't find child care for their kids. They can't work, because nobody will take care of their kids because their kids are so challenging. So some of these health conditions become debilitating and keep people out of the workforce. And we need to do something about them.
And then they always say, if you're a hammer, every problem's a nail. And you all know, I'm involved in nutrition education. So it's not so surprising that I would say this. This is part of that analogy. But we've done work that shows that people who know how to manage their bills, make a family budget, stretch their groceries, they know how to make food last, are less likely to be food insecure.
And we've done work with the Expanded Food Nutrition Education Program, EFNEP, Cornell's program for nutrition education for low-income families, and show that graduates of EFNEP, those who've done the six lessons, are less likely be food insecure than similar people who've enrolled but didn't complete the program. If you're interested in the statistics, some of the national statistics, this is the website. And then also the feedingamerica.org, the new Second Harvest, also has a lot of information.
So with that I want to open it up for questions and discussion. And thank you for being so patient. So the people who wanted to say something before--
So Chris, you wanted to--
CHRIS BARRETT: You've put a lot of emphasis on behavioral adaptation. To what extent is physiological adaptation part of the story?
CHRISTINE OLSON: Yeah. On the obesity question, that's very, very important to keep in mind. When we are deprived of food, physiologically, we make a metabolic adaptation. And we become more efficient. That's another reason dieting is so hard. It's hard to lose weight when you diet. That same kind of thing may be going on with the obesity and food insecurity.
So the woman may actually not overeat in relation to what her needs were when she started. But because she got more efficient when she was going without food, metabolically, that same amount of food is more than she needs. So that could be going on in this obesity food insecurity connection. So it may be both a behavioral adaptation and a physiological adaptation that's operating against maintaining weight if you experience food insecurity.
Now, there was something said about malnutrition. One of the things that's really, really hard-- because the severity of deprivation of food isn't at the level that you would see in third-world countries, it's very hard, using traditional methods for assessing nutritional status, to see some of these differences.
You do see more anemia from iron deficiency. You do see some lower blood levels of nutrients that are associated with fruits and vegetables. But other kinds of blood indicators, you don't see much difference. So it's really hard, using physiological methods, to detect food insecurity. And it's why we went down the behavior experience route to measure food insecurity and not down the biological, physiological route to measure food insecurity.
For hunger, possibly that physiological root-- you know, analytic methods have improved dramatically. Maybe there's some payoff there. I don't want to rule it out. But I don't know. It'll be tricky. [INAUDIBLE]
SPEAKER 8: They are interesting, given that most of the people who have food insecurity, most of the households earn less than $20,000 a year. It's hard for me to see how you could not be food insecure at that level, given that some households are still food insecure when they earn five times that much. Why are you stressing a program that would change the behavior of the individuals or the households, giving them nutrition education instead of stressing doubling the minimum wage for a living wage or some other things to change the environment within which they operate? Is it really their fault? So if they're just doing what you tell them they'll be OK?
CHRISTINE OLSON: So you see though, these things. I do think low wages are a problem. But I have to say after working on this for 25 years, it is stunning, absolutely stunning, how none of this has resonated in public policy circles. I mean, maybe they'll ratchet up the food stamp allotment a little bit. But that's about it.
And that lack of the public sector stepping in-- the private sector, the charity-based food networks-- when we started in this business, if you went to a soup kitchen or a food pantry, by definition, you were food insecure. Now 25 million Americans a year are using those services. Now, part of the reason it expanded is industry gets tax breaks for donating. And all of us seem very willing to bring a couple of cans of food to the movie theater to put in the box or put in a box outside church or whatever. Nobody wants to pay taxes for what some of these public policy initiatives will cost.
But I think, as I said, related to the diabetes, in some of these cases, addressing the hunger would save money. I mean, we've all heard the figure for the WIC program, the Supplemental Nutrition Program for Women, Infants, and Children. Feeding a pregnant woman pays off in $3 of savings of health care cost because the likelihood of that baby having to go to a newborn intensive care unit is decreased substantially.
So I think some of these things, feeding, making sure that elderly people with chronic conditions and other people with chronic conditions that are managed by nutrition that they get access to the food they need would save a lot of health care dollars. So those are some places to start, I think.
DAVID: I have two questions. One is vis-a-vis WIC and some of these other programs. I know there's a substantial amount of evidence that you're referencing in terms of saving, improving health outcomes, but I see nothing that suggests that has anything to do with the food as opposed to the related types of care that's provided to WIC mothers, for example. I'm wondering if there is any that has compared the types of programs that provide the services associated with WIC, with or without food.
My second question, which motivates that first one is that there are two or three very interesting studies from developing countries now that are showing that programs like Oportunidades and other conditional cash transfers are contributing to increased rates of obesity in a rather substantial way, including a recent study by the [INAUDIBLE] colleagues down in Mexico, which is really quite alarming. So I am wondering, if part of the goal or much of the goal here is to reduce health care costs over the long term, how much do you worry about expansion? I mean, the one in nine people in the United States who are on food stamps, I could see as quite alarming from a different perspective, not that they're poor, but they're getting fatter and unhealthier. How much should I be concerned about that?
And actually, I can throw in one final question related to kind of the [INAUDIBLE] Barker hypothesis issue. My understanding is that all the evidence, in terms of long-term implications of in-utero stresses that contribute to chronic diseases in older age groups, is from studies that have been done on children or women who are having babies that are substantially underweight and substantially nutritionally deprived, basically, not what I would normally expect to be-- women in the United States basically don't fall in that category. Is there any evidence, again, from the US that substantiates that Barker hypothesis at all, relevant to a US-type population?
CHRISTINE OLSON: So David's given me a lot of questions. So the first one I want to start with is sort of welfare approaches, improving access to food, improving nutrition marginally, whether it's cash transfers or however you do it. Now, from this slide, you would predict if families are down here, and you give them enough assistance to move them here, so what you want to do is you want to go from here over to here.
And that takes a little more of something. Or it might take a slightly different approach. Actually, there's some evidence that food stamps in the United States, they don't get families out of food insecurity. But they do move some families from hunger to food insecurity. So some of this assistance, we do have to think pretty carefully about. How much do we do? And how much do we do for certain outcomes? And how much do we do for certain other outcomes? Because you might prevent hunger. But you might increase the risk of obesity. You might not do enough to get people to a healthy weight. So that's one way to think about that.
Food stamps, of course-- around 2006, when they were reevaluating the food insecurity measure, and there was a lot of brouhaha in Washington, DC about food-assistance programs making people fat, USDA undertook a major review of this. And there's sort of one study in the literature that might point in that direction. But there is no evidence that the level of assistance that you get on food stamps in the US is making people fat.
Now, on the Barker hypothesis and this sort of in-utero under-nutrition, or let's just say nutrition in utero, and its relationship to risk of obesity in childhood or adulthood, so the offspring's risk of obesity. I don't know of a Barker study that's been done in the US. But there have been studies in England and other kind of Western-type countries.
And the mechanism that seems to be operating here is under-nutrition in utero increases the likelihood of a central fat pattern. So later in life the individual has a higher body mass index. But that fat is disproportionately around the middle. And in terms of chronic-disease risk, that's a bad place to have your fat is around the middle. You're better off having it on your butt or someplace like that.
We have done some work-- and this may be what's a little more relevant in the US, over-nutrition in utero. In other words, exposure to high levels of blood glucose, excess calories in utero, have shown that women in our Bassett Mothers Health Projects who gain too much weight in pregnancy, their kids are at increased risk for being overweight at age three. So over-rich, particularly in the form of fat and glucose, environment in utero may predispose to increase body mass index. And that seems to be being fat all over, not the central pattern.
So under-nutrition in utero may lead to a centralized patterning of fat. Over-nutrition in utero may be related to increased risk of obesity all over, fat all over. Have I answered your questions? Ken.
KEN: You mentioned the role of unexpected expenses for many people as part of food insecurity. And I was just curious to link it up to some other big policy debates in the US today. To what extent do health care expenses in individuals, families who don't have health insurance or are under-insured-- do you find that intersecting extensively with these problems? Is it something that you all looked at, kind of put a number on, estimate the role that that played?
CHRISTINE OLSON: I can't put a number on it. And our work was done long enough ago that I don't know that the numbers would be exactly relevant to the health care debate that's going on now. But I can tell you when we were doing that criterion validity kind of study-- so we were looking at all the information-- there were some households that popped up as food insecure that you couldn't quite figure out. The annual income looked fine. These people had college educations. You're going, I can't figure out what's going on here.
And invariably, it would be that there was a health crisis. It was that or somebody lost their job. But it was more often a health crisis that sometimes put somebody out of work. And one of the things that we sometimes don't think about, related to health crises in lower-income working families that get paid wages, they actually may have some insurance. But if you don't go to work, you don't get paid. It's not like us. We have sick days.
If you check out groceries at P&C-- or maybe I shouldn't pick on P&C, but that type of a job, a service kind of a job-- if you don't go to work-- or you paint houses. You drive a cab. You don't go to work, you don't get paid. And so there's increased expenditures of money out of the house. And there's decreased inflow of income into the house. So these health crises work in two ways in these low-wage working families. Yes.
SPEAKER 9: I just wanted to-- I mean, we're so wedded to the poverty measure as a way to benchmark a lot of things. And one of the things that I have learned is that you have to look at disposable income, that poverty is not the-- the official poverty measure is not useful. It's what the disposable income is, whether there's any wealth defined for a buffer. Can you borrow against your car? Do you have a car? It's sort of wealth and taking care of other necessities first.
There's a really nice article on whether you eat or you heat your home. And that trade-off is very present, very, very out there strongly. The other one is the unexpected health expenditures. And the other one is the car to get to work. So you've got health expenditure. You have high feeding expenses, like in the Northeast. And you have car repairs.
CHRISTINE OLSON: You won't get an argument from me about the current poverty measure. There's been National Academy of Science report after National Academy of Science report saying we need to change it. And I think we probably do. There doesn't seem to be the political will to do much about it.
What I will say-- and I think this will be the closing. So if you're really interested in this, you know the poverty measure is based on the Thrifty Food Plan, a market basket of foods that do not meet nutrient needs over the long term. It's only meant for short-term use.
One of the visiting scholars is Parke Wilde, who's a faculty member at Tufts University. On October 5, he's going to give the community nutrition seminar over in Savage Hall on the Thrifty Food Plan. And then on the sixth, he's going to give this seminar on the dynamics of food insecurity. So for those of you who are still here, if you were interested in what you heard today, come back lots of times. But be sure to come back in October, October 6, and hear Parke. So I think we should close because we're getting to the witching hour here.
We've received your request
You will be notified by email when the transcript and captions are available. The process may take up to 5 business days. Please contact firstname.lastname@example.org if you have any questions about this request.
How many hungry people are there in the US? In the mid-1980s this question set off a flurry of activity that ultimately resulted in annual measurements of the proportion of US households and individuals who are food insecure and hungry.
In this seminar Christine Olson, Cornell professor of nutritional science, addresses the history of food insecurity measurement, and the antecedents and consequences of food insecurity and hunger in the US. She also describes current research addressing the dynamics of food insecurity.
Olson is a member of the Institute for the Social Sciences theme project on persistent poverty and upward mobility.