[MUSIC] Hi, I'm Mary Ellen Mortensen. I serve as the Chief Medical Officer in the Division of Laboratory Sciences, in the National Center for Environmental Health at the Centers for Disease Control and Prevention, which is in Atlanta, Georgia. I have served in this capacity for quite some time, long enough to be very familiar with bio-monitoring, and I'm happy to have been invited to present to you what I call bio-monitoring or biological monitoring in public health 101. So the outline of my talk shown here, these are the points I'm going to cover. First, I'll start with a general overview, and several important considerations about biomonitoring. Next, I'll talk about biomonitoring in the US population, which includes a high level overview of the National Health and Nutrition Examination Survey or NHANES, which is an ongoing survey conducted by CDC. Then I'll talk about some of the public health uses of biomonitoring data and close with some comments about the limitations of biomonitoring data. In everyday life, people everywhere are exposed to chemicals in their environment. Familiar exposures to those of you out in the audience might include chemicals in the air from such point sources as coal burning power plants, and other general ambient air exposures such as from vehicular exhaust emissions. In recent years, there's been an increasing awareness of chemical exposures in our daily lives. Such things as medications, chemicals that may be used in foods for preservation, or for enhancing the appearance or texture, and personal care products such as cosmetics that we use. Personal care products such as body washes, and a number of other chemicals we use around the house. These are just a few examples, and I'm sure you can come up with many more. Now, to evaluate or assess exposure to these chemicals, there are some different approaches that can be used. Four examples are shown are. First, interviews and questionnaires can be used to gather information from people who may have exposures. Then next, environmental monitoring, making measurements, for example, of chemicals in air, water, food, soil. These approaches often employ mathematical models to estimate the amount of chemicals that get from the environment and into people. Personal monitoring has also grown tremendously in recent years with many new technologies and personal tracking methods, such as smartphone apps that are available. And then the forth of these, is biomonitoring which is particularly valuable because it involves the actual measurements of the chemicals of interest in the body, fluids, and tissues of people. Of course, these different methods can be combined, and they can be used with calibrated and validated models to improve exposure assessment. At CDC, we define biomonitoring as the assessment of internal dose by measuring the parent chemical, or it's metabolite, or metabolites, or reaction product in human specimens. By reaction product, I mean such things as adducts. Often chemicals can react with proteins in the blood to form an adducts or with DNA to form an adducts. Now blood and urine are the most commonly used types of human specimens. And there are a number of validated methods available for many environmental chemicals that we may wish to measure. When we wish to measure, these biomonitoring measurements are typically made in human specimens, such as blood and urine. A couple of points I'd like to make about biomonitoring. These measurements that we make in blood or urine, typically integrate all sources and routes of exposure. For example, a blood lead measurement reveals the lead concentration in blood that results from all sources of exposure, such as air, dust, water and soil, as well as from all routes of exposure. That is, from breathing, from ingestion, which includes transfer of lead from hands to mouth, and then being swallowed, as well as from food and drinking water. So, the biomonitoring measurement is something of an integrating measure. Most biomonitoring measurements are made at very low concentrations. In fact, often at trace concentrations, very close to the analytical method of detection. This is one reason that measurements require trained, experienced analysts and validated methods and adherence to quality control and quality assurance procedures. It's also now worthy that biomonitoring measurements are concentrations. They are not, per say, exposures. Because a biomonitoring measurement can be a useful indicator of an exposure that has occurred, we sometimes refer to these as exposure biomarkers. Biomonitoring measurements have been employed in both large scale surveys, such as NHANES, and in more targeted studies. In large scale surveys of populations, the measurements can be used to generate reference values. NHANES provides US population biomonitoring results for older children and adults. I will talk a little bit more about NHANES a bit later. An epidemiologic studies of specific populations often use bio-monitoring measurements to classify the exposure status among those in the study. Usually, the biomonitoring results are used along with clinical and laboratory data, interview and medical information to try and determine disease risk, disease etiology or other risk factors. My remarks are going to focus on the large scale survey NHANES, because for many environmental chemicals, it provides basic information about exposures in the US population, who is exposed and to what extent, and allows us to develop reference ranges for environmental exposures. This basic information is crucial before we start trying to examine associations between exposures and health effects. Now, from a laboratory perspective, meaningful biomonitoring measurements require consideration of several factors. Otherwise, the results maybe uninterruptible, inaccurate or even misleading. Some of these important considerations are listed here in this slide. First, what is the best chemical to measure. And by that, I mean what is the analyte that we want to measure. Is it the parent chemical? Is it a metabolite? Or is it an adduct, that is, the chemical bound to a protein in the blood. Next, what is the best time to obtain the specimen? And by that, I'm referring to the specimen, blood or urine. We often refer to that as the matrix. And it is important that not just the most convenient time to get a sample is considered, but what is the best time? What is the optimal time? In other words, we need to know something about the chemical so that we collect the sample at the best time to capture the exposure, not just the most convenient time. And then, what is the best specimen or matrix? And by matrix we are typically referring to blood or urine. These are the most common biomonitoring matrices that we use. Then there are analytical considerations, some related to the specimen, and some related to the method. Relative to the specimen, the chemical needs to be stable, or to be made stable by an, use of an additive. For example, if you want to measure mercury in urine, you have to add a preservative, or the mercury will just dissipate into the environment and be gone. Sample contamination is also a problem, or a consideration when low concentrations are being measured, of a parent chemical or a metal. And then of course, there's also the possibility of interfering substances that may invalidate the measurement or render it inaccurate. The considerations related to the method include, is it a validated method? Has it been shown to be adequately sensitive and specific? In fact an example of the sensitivity and specificity issue comes into play recently with the change in the blood lead reference value, which was recently decreased from ten to five micrograms per deciliter in the US. Some of the lead measurement instruments can not reliably measure below ten micrograms per deciliter, so there's a problem of sensitivity. The laboratory making bio-monitoring measurements needs to have qualified trained personnel, needs to be using validated laboratory methods and document quality assurance and quality control procedures. In some cases, proficiency testing programs may be available and participation can be a valuable way to maintain the level of quality performance of a laboratory. Now deciding what to measure, that is the analyte, and in what sort of specimen, that is the matrix, depends on several factors. This cartoon shows a typical concentration time curve in the blood in red, and the urine in green for elimination of a non-persistent chemical. That is a chemical that does not stay in the body very long. Typically, blood concentrations increase fairly rapidly, as you see in that curve. And then decline rapidly after a single exposure. Then, as you see here the green urinary elimination curve lags behind, usually by hours, and sometimes by days. Note that the urine levels may be detected longer than the blood, if you're getting a sample between days one and ten in this case. And because urine can collect for a few hours in the bladder, there may be a concentrating effect with higher concentrations being detectable in the urine because of this. For the non-persistent chemicals, which include many pyrethroid and organophosphate pesticides and many of the preservatives in personal care products, urine is usually the prefered specimen. The chemical is rapidly metabolized in the body and then eliminated in the urine. Metabolite can be measured, not the parent compound. Now an advantage of measuring the metabolite, is that external contamination is not an issue. It provides evidence the person was exposed to the chemical as well. In contrast, there have been efforts to measure some non-persistent chemicals in blood. This can be problematic if the sample is collected after the, the chemical has been removed from the blood, as many of them are very rapidly metabolized. And further more, if a chemical is widely present in the environment, it's possible that the blood sample can become contaminated during or after collection. This was a problem in trying to measure bisphenol A in blood. It was virtually impossible to ensure that the blood sample did not get contaminated from the, this bisphenol A in the environment or from the materials that were used in specimen collection. Clearly, urine metabolites of this bisphenol A were the better sample for assessing exposure. As for the timing of the specimen collection, if the exposure occurs over a short time or is a single one-time exposure, it's important to collect the urine as closely to when the exposure occurred as possible. Or else the urine metabolite just may be not at detectable levels. However if the exposure occurs repeatedly and is likely to be continuing, there's a greater likelihood that a urine specimen collected at a random time can yield a measurable result. This is in fact the case with many of the personal care product used daily, and that contain antimicrobials, or preservatives. Examples of these chemicals are the phthalates and parabens. Again, this curve pertains to chemicals that are eliminated rapidly from the body, typically within hours to days and do not linger. Now in contrast, the cartoon here shows you the fate of a persistent chemical in blood and urine. In contrast, persistent chemicals are those often stored in the body in the fat and other organs, so they may be best measured in blood or in serum. As you can see in this cartoon, they really don't appear in the urine. So, these are not chemicals you would be anticipating measurement in the urine. They have minimal metabolism, and essentially, you're measuring the parent chemical. Exemplified in this cartoon, are chemicals such as dioxins, the polychlorinated biphenol compounds or PCBs, and several of the organal chlorine pesticides, such as chlordane and Myrex, the so-called persistent organic pollutants. Often these chemicals are either not metabolized very much, or are so slowly metabolized that they persist in the body, and can be detected in serum or blood for a long time. Because they are eliminated so slowly, and usually not in the urine to any extent, these chemicals are usually best measured in blood or serum. So, the timing of the specimen collection is not critical. As you can see, the chemical or it's metabolites may be detected in blood for a long time after the exposure ends. This is the case with such chemicals as PCBs, dioxins and the persistent pesticides. Now I'm going to talk a little bit about NHANES. The website is provided for you at the bottom of the page if you'd like to find out more about this survey. NHANES has been going for approximately 40 years. It started as an effort to collect health information from a representative sample of the US population. In recent years, CDC's National Biomonitoring Program has developed measurements for more than 300 environmental chemicals and been measuring them over this time. NHANES is an ongoing survey, that is representative of the US population. For this reason, measurements of the various environmental chemicals made in blood and urine, provides us with US national reference values. NHANES was devised and mandated to obtain information about the health status of US residents starting in the 1950s. This was in response to evidence of malnutrition within certain segments of the US population, and there was a lot of research using NHANES data linking diet and health. Various health status and dietary surveys were conducted beginning in 1969, and in 1999, NHANES became a continuous survey. The data that is collected on health status, behavior, diet, and the findings of physical and laboratory examination of the participants, is publicly available. It is released into your cycles or survey periods, and it is representative of the US population. The biomonitoring measurements that are made by CDC's National Center for Environmental Health are also available on the NHANES website. I'd like to focus now a little bit just on the biomonitoring results that we provided in NHANES. Most of the biomonitoring measurements are made in older persons. For blood and serum, it's mainly children ages 12 and older who are sampled, exceptions are blood lead, cadmium and mercury, which are measured in ages one year and older, and serum codeine which is measured in ages three years and older. Most of the urinary chemical measurements are made in a one third sub sample of the participants, and these participants who give urine samples are ages six and older, although NHANES is considering lowering the age of collecting urine to age three, which would be wonderful because there's not a lot of information in young children with regard to biomonitoring measurements. The subsamples used for each of these chemical measurements are determined so that they are representative of the US population. Now the biomonitoring results in addition to being published on the NHANES website, are also aggregated and presented in the National Reports on Human Exposure to Environmental Chemicals, which I'd like to talk to you a little bit about. If you want to actually review the National Reports on Human Exposure to Environmental Chemicals, the website is provided in the left hand corner of the slide. The cover is shown from the report that was published in 2009. Now, the data in the National Report are representative of the US population, and the sample data tables shown here is for total blood mercury. As you can see, we have broken down the results and calculated geometric means and provide selected percentiles according to certain age groups. In the case of mercury, we actually have data for children as young as one. Usually that's not the case. Most serum is in children 12 and up. We also provide by sex, that is male and female, and results are also shown for the major racial ethnic groups that NHANES collects information for. The geometric means are presented in the yellow column, and then the percentiles are shown over in the green columns. The data are broken down by each of the survey cycles. Each survey cycle is two years, and that's in the column labeled survey years. So these data are representative of the US population by age group, by sex and by racial ethnicity. In some cases, there is no geometric mean shown. The reason for this is that the detection frequency was too low, that is less than 60% for us to feel confident in calculating a geometric means. So, instead of a value, or an estimate, you will simply see an asterisk, which means that there were not enough detectable results. The National Reports have several goals. And I refer to them in the plural because there was a National Report published in 2009, and then to keep updating the information we have been providing updated tables. So there are two sets of documents available. The original National Report of 2009 and then the updated tables, which are periodically updated. So the purpose of these are many-fold and several of the important goals are listed here. First of all, they allow us to assess the exposure of the US population to a number of environmental chemicals. Which chemicals, who's exposed, and to what extend? In addition, we have been able to establish population reference ranges in the US, four of these chemicals by the various categories that I already mentioned to you, age, sex and race, ethnicity. And then third, as we measure these chemicals over time, we have an opportunity to see changes in the concentrations, and in the reference ranges. A good example of that that you may be familiar with, is blood lead, which over time has declined dramatically in the US population. And the fourth important goal of the National Reports, is that we can use the data to help set priorities in research and in public health policy. And they can be useful in studies that try to link exposures to various health outcomes. The data tables currently available at the website, which is given on the bottom of this slide, are available for more than 290 chemicals, with results that begin in 1999 and go through 2012. Now, there are some caveats about the reports, I want to point out. First of these, is that for many of the chemicals measured, concentrations are very low, and may be clustered near the limit of detection. In many cases, this makes it difficult to know the clinical significance of the measurements. More research is needed to know whether the measured concentrations are hazardous or might contribute to adverse health effects. And it is important to remember that the reports are exposure reports, that provide results based on all sources of exposure, via all routes of exposures, that is total exposure. Now the National Reports provide aggregate levels, that is statistical point estimates, for geometric means and selected percentiles for the entire US population. And as such, there are some limitations. The data are limited in children, particularly those younger than six years. It's also important to beware of comparing individual concentrations, that is results from a single individual, to results in the reports. Individual results may be quite different from any reasons, including the timing of when the specimens were collected, the differences between people such as the pharmacokinetic, that is how the chemical is handled in the body, a person's body size, the nature of the exposure, and many other differences. And then it's also important to be aware that the data in the reports are not representative of specific locations. It is not possible for example, to pull out results by county or by region. It is also not possible in many cases to examine particular groups of people. Seasons may be difficult to examine, and exposure, of course, to specific products, there just may not be information on that. Furthermore, the design of the enhanced survey does not consider exposure status. So often there may be limited or, if any, information about specific chemical exposures. One exception to this situation is tobacco smoke and smoking. There's quite a bit of information about tobacco smoke. I would also be remiss if I didn't talk a bit about the general limitations of biomonitoring results. So biomonitoring results don't tell us specific sources or routes of exposure. The measurements of a metabolite in urine or a chemical in blood reflects all sources and all routes of exposure to a chemical. Second, with regard to chemicals that are not persistent, that is they are fairly quickly metabolized and eliminated from the body, usually in the urine, it's important to consider collecting the urine relatively close to the time of the exposure if possible. Otherwise the urine may not have a detectable concentration, so it may be incorrectly assumed that there was no exposure. On the other hand, if the urine is collected very soon after the exposure, the urine concentration may be quite high, and lead to an incorrect assumption about an extreme exposure. This makes it difficult to interpret individual biomonitoring results. Particularly if there are only single measurements from an individual. Now results obtained from a population can overcome this limitation. Because once you get a large number of specimens being collected, the results are typically easier to interpret because these samples have been collected at different times, and after varying exposure levels. And third, biomonitoring measurements typically are made at very low concentrations, at the part per billion, or micrograms per liter range. The specimen then has to be properly collected, and the measurements need to be made by experienced, qualified laboratorians so that the results are accurate. Common problems in biomonitoring include inadvertent contamination of the specimen, and laboratory errors in making the measurements, often with inadequate Quality Assurance and Quality Control procedures, or in an inexperienced laboratory. In summary, what I hope I have shown you is that biomonitoring is a tool for assessing human exposure to environmental chemicals. A valuable tool, and it can be used with other methods as well. The National Report on Human Exposure to Environmental Chemicals and the Updated Tables that supplement that report, provide biomonitoring reference ranges for the US population. Third, the chemical properties, that is persistence in the body, the metabolism, the stability of the metabolites, and analytical considerations, are important factors in biomarker selection, and in specimen collection. And, last but not least, the detection of a chemical, does not mean that it is causing disease. Thank you all very much for your attention, I appreciate it, and I hope you will look a little bit further at the National Report and at NHANES, I think you'll find it very interesting to see the amount of data that are available. [MUSIC]