[MUSIC] So next we'll move from thinking about epidemiologic research methods to considering how we might develop health equity interventions. So when we think about population health and health disparities, it's important to consider where we are now and actually where we need to be. And in the figure that I'm sharing, here we can consider that for quite some time we have had a lot of knowledge about health and healthcare disparities. When you look on the left-hand side of the figure, you'll note that we have had epidemiologic data, as well as data on clinical efficacy and effectiveness of various interventions. And we've also known quite a bit about some of the basic biomedical science that leads to favorable health outcomes. What we have not had or what we've had less of is a bridge across this chasm between our knowledge and actually achieving health equity. And in order to achieve this, what we need more of is more implementation science or what is often referred to as delivery science. And here this is when we take what we've learned from our knowledge of different studies that have been done showing us what could work in ideal settings. And we actually translate that over into the delivery of equitable healthcare on a population level. And it's in these cases that we can have the best opportunity to achieve health equity. And so in order to appropriately apply these lessons learned, we certainly continue to need more work around research in this area, around implementation science, including training in this area. This type of work does require a significant stakeholder engagement, including healthcare delivery stakeholders, as well as patient and other stakeholders. It involves dissemination and translation of new initiatives. And very importantly, in order for there to be a continued effect of this type of work, sustainability approaches have to be considered on the front-end of the work in order to ensure that they will be long-lasting. So I was told early in my career that a researcher should study what annoys them. And so I'm sharing with you this figure, this line graph that illustrates something that has always annoyed me. So this is a study that was conducted back two decades ago, in 1994, using data from the US Renal Data System. And what these investigators looked at was the risk of new cases of end stage renal disease comparing black and white patients in the United States. And when they looked at the rate of new cases of end stage renal disease noted on the y-axis and they also looked at level of income, and this is household income that's noted on the x-axis. What they found was that at the lowest levels of income, blacks had more than a two times greater risk of developing end stage renal disease than did whites. And what was notable was that looking at that lower end of the income range was where they found the greatest disparity. I'm sure you'll note that when they looked in the study at the higher income levels, the rates of end stage renal disease were different, certainly, between blacks and whites. But not as profoundly different as when they examined the lower income level. And so in later years, our team actually examined a similar relationship, but here we were looking at earlier kidney disease. So in the context of the reasons for geographic and racial differences in stroke study. Which is a population-based cohort study that primarily includes individuals from the Southeastern as well as Southern United States and what is regarded as the stroke belt. Because the rates of stroke in this region of the country are considered quite high relative to the rest of the country. We actually looked at this question of the relationship between income and albuminuria, which is a loss of protein in the urine and is considered an early marker of kidney disease. And what we found was something quite similar to what was found in the study from 20 years ago that we discussed on a previous slide. We found that comparing blacks and whites across level of household income that we found this gradient, wherein both groups had higher rates of albuminuria if they fell into the lowest level of income. But this gradient was steeper for the black participants in the study. And this rang true even when we went on to adjust for some of the known confounders of the relationship between race, income, and kidney disease. And so on this bar graph, the taller bars are indicating the black participants in the study and the bars a little bit shorter than those are indicating the white participants. And the gradient across levels of household income, as noted on the x-axis, can be noted. And the y-axis indicates the percentage of the participants in the study who had albuminuria. And so next, we were interested in whether among regard study participants if it mattered whether the participants actually lived in communities that were poor, but were perhaps surrounded by wealthier communities. Would that make a difference in terms of the participants' risk of developing, in this case, end stage renal disease? With the idea being that poor communities surrounded by more affluent or wealthier communities might see resources from those affluent communities filter down into the poor communities. Leading those individuals to not be at maybe as great of a disadvantage than individuals living in poor communities surrounded by other poor communities. And so in order to do this type of analysis, we used this metric that actually takes into account the density of poverty for communities. And this was developed by the Centers for Disease Control and Prevention. And so in this figure what I'm sharing with you is the county poverty category. Where the group that is indicated on the far left-hand side or the high outliers were those individuals in the study who lived in communities with high degrees of county level poverty. So meaning that in many cases 20 or more percent of the people living in that county actually lived below the federal poverty guideline for household income. And so that's on the far left-hand side. On the far right-hand side, these are people who were living in communities that were considered low poverty outlier communities. Where there were few people who were living in poverty surrounded by other counties or other communities that also had few people living in poverty. And so what hopefully you can appreciate is that the risk of incident, end stage renal disease is indicated on the y-axis, was highest for those people living in the high poverty outlier counties. What was interesting about this study, though, is that when we went on then compare what we found when we looked at the density of poverty using this county poverty metric. To outcomes for in terms of risk of end stage renal disease using a metric of household income or household level poverty. What we found was that the household metric actually was more predictive of outcomes in this population. Meaning that it seemed that an individual's income seemed to be a greater predictor of how they fared in terms of their chances of getting kidney disease or kidney failure. It seemed to be a greater predictor than that county metric. And so when we think conceptually about how social determinants relate to social disadvantage and risk of chronic kidney disease. This model, that was developed by Dr. Nicholas and published a couple of years ago, reflects three different pathways for how an individual's risk might be increased related to their social disadvantage. So on the far left-hand side, there are factors related to residential segregation that are at play for many individuals. In the middle of the figure is demonstrated some different factors that may come about as a response to experiencing discrimination. And then on the far right-hand side are healthcare related factors, including uninsured or underinsured status, which can lead to difficulties with healthcare access. So each of these factors can ultimately lead to biological mediators of risk for going on and developing chronic kidney disease and chronic kidney disease progression. And also can play a role in leading to a person's increased risk of end stage renal disease, as well as premature mortality. And so I'd like for us to focus our attention a bit on an area that I've spent some time working on that relates to this framework, and that is poor nutrition. And so individuals living in these residentially segregated areas who are at social disadvantage may experience difficulties in terms of nutrition. So first, I think it would be important to just cover some terms that are often used when referring to healthy food availability. So many of you may have heard of food deserts, and these are areas that have limited access to affordable and nutritious foods. Fewer of you may have heard of a term called a food swamp. And as it turns out, these are areas where there actually is a lot of food, but these foods are not very healthy foods. So these are areas where there's an overabundance of high-energy foods that inundate healthy food options. In these types of communities what you might see is a predominance of small corner stores and carry-out outlets, but very few healthy food sources. And then food insecurity is an important term to consider. And this is the limited or uncertain ability to acquire nutritionally adequate and safe foods in socially acceptable ways. And so the way that this plays out is that these are individuals who have very limited incomes and are unable to purchase foods. And so in some cases, they may have to skip meals or try to make their meals stretch. In some extreme cases, they may actually have to steal food in order to be able to sustain themselves. They may have to eat from places like dumpsters in order to, again, to sustain themselves. And so Baltimore city and many of its areas are actually, there are food deserts as well as food swamps. And so in the community actually that surrounds Johns Hopkins Hospital and the Johns Hopkins School of Public Health, we are actually situated in what is both a food desert and a food swamp. And so in this figure, which is from a study by Dr. Manuel Franco, who is both a faculty member at our School of Public Health and one of the guest lecturers in our course. What I'm sharing with you is a map of Baltimore city that is surrounded by Baltimore County, with the central portion of the map focused on Baltimore city. And so in the gray shaded areas, these are communities that are predominantly white. And in the white shaded areas, these are communities where the inhabitants are predominantly black. And then the green circles indicate the amount of healthy food that was noted to be available in the stores in these communities, with the larger green circles indicating greater amounts of healthy food availability. And so what you can hopefully appreciate is that the larger green circles mostly overlie the darker gray shaded communities. And in fact, what Dr. Franco and colleagues found was indeed that the prevalence of healthier food was greatest in predominantly white and in higher income communities than in predominantly black and lower income communities. And so how might these barriers relate to chronic kidney disease? Which, as I mentioned, is my area of study. And so what we have been working from in our research group is this model. Where we think about how limited availability of healthy food, as might be the case for people with food insecurity, how that might lead to poor dietary patterns, including patterns that include limited fruits and vegetables. And how those poor dietary patterns can ultimately go on to lead to new cases or incident chronic kidney disease as well as progression of chronic kidney disease. Through both indirect pathways, such as through its influence on some of the established chronic kidney disease risk factors, like obesity, diabetes, and hypertension. But also through some direct pathways, where these types of dietary patterns actually directly influence the kidney and its chances of disease developing in the kidneys. And so what I'd like for us to do is to focus first on this issue of limited availability of healthy food. And so our group was interested in the relationship of food insecurity and chronic kidney disease. And we used data from the National Health and Nutrition Examination Survey to look at this question in the context of work conducted in the Centers for Disease Control and Prevention Chronic Kidney Disease Surveillance Team. And so what we found was that for those individuals who had either marginal or high food insecurity, there were certainly some differences between them and those without food insecurity. And I will highlight some of those for you, which included, and here in this study, we also did find that 26% of US adults did have either marginal or high food insecurity. And there were some differences in that the people who had food insecurity were a bit younger than those who did not have food insecurity. There were some racial and ethnic differences, including there being a greater proportion of non-Hispanic blacks comprising the individuals with food insecurity than in the non-food insecurity group. And then certainly, individuals who were food insecure were more likely to live in poverty, they were less likely to have health insurance, they had lower levels of educational attainment. And they were also more likely to use tobacco than were people who were not food insecure. And so when we went on to examine the relationship between food insecurity and chronic kidney disease, and in this case, we were looking at prevalent chronic kidney disease or just a chronic kidney disease in one snapshot in time. What we were somewhat surprised to find is that we only observed this relationship among individuals who had some of the established risk factors for chronic kidney disease. And so in this figure what I'm sharing with you is that when we stratified the participant population by whether or not the individuals had diabetes or did not have diabetes. What we found was this association between food insecurity and chronic kidney disease that was only present among individuals who had diabetes. And so the bars in this figure are indicating the odds ratios for chronic kidney disease where there was more than a 60% greater odds of chronic kidney disease among the high food insecurity group as compared to those who did not have food insecurity. Which was what was our comparator in this study. While there was no difference when we looked at levels of food insecurity among people without diabetes, such that we did not find an association between food insecurity and chronic kidney disease among that population. And so when we tested for a statistical evidence of effect modification, as was covered earlier in our lecture, what we found was indeed that there was evidence of effect modification by diabetes status in this analysis. And then similarly, when we looked at the same population stratified by hypertension status, we found something quite similar. And in this figure, I'm sharing again that there was around a 40% greater odds of chronic kidney disease among individuals with hypertension who had high food insecurity. And there was no relationship between food insecurity and chronic kidney disease among persons without hypertension. [MUSIC]
