Udesky et al Environmental Health 2019 18 99 Page 2 of 16. Background metabolized chemicals such as phthalates and bisphenol A. Chemical measurements play a critical role in the study BPA in blood or other non urine matrices despite the. of links between the environment and health yet many fact that urine is the preferred matrix for these chemicals. researchers in this field receive little if any training in Phthalates and BPA are present at higher levels in urine. analytical chemistry The growing interest in measuring and when the proper metabolites are measured there is. and evaluating health effects of co exposure to a multi less concern about contamination from external sources. tude of chemicals 1 2 makes this gap in training in including contamination from plastics during specimen. creasingly problematic as the task at hand becomes collection 9. ever more complicated i e analyzing for more and for More commonly however exposure studies simply do. new chemicals of concern If steps are not taken not adequately report on QA QC or describe how QC re. throughout sample collection and analysis to minimize sults informed reporting and interpretation of the data In. and characterize likely sources of measurement error the context of systematic review and weight of evidence. the impact on the interpretation of these valuable mea approaches not reporting on QA QC may result in a. surements can vary along the spectrum from false nega study being given less weight For example the risk of bias. tive to false positive as we will illustrate with real tool employed in case studies of the Navigation Guide for. examples from our own data Systematic Review includes reporting of certain QA QC. Some important considerations when measuring and results in its criteria for a low risk of bias rating e g ref. interpreting environmental chemical exposures have erence 10 When we applied the Navigation Guide QA. been discussed in other peer reviewed articles or official QC criterion to 30 studies of biological or environmental. guidance documents For example a recent document measurements that we included in a recent review of en. from the Environmental Protection Agency EPA pro vironmental exposures and breast cancer 11 we found. vides citizen scientists with guidance on how to develop that more than half either did not report QA QC details. a field measurement program including planning for the that were required for a low risk of bias assessment or if. collection of quality control QC samples 3 The Cen they did report QA QC did not interpret or use them ad. ters for Disease Control and Prevention CDC also gives equately to inform the analysis e g reported poor preci. guidance related to collection storage and shipment of sion but did not discuss how whether this could affect. biological samples for analysis of environmental chemi findings see Additional file 1 for details Similarly when. cals or nutritional factors 4 To assess the quality of LaKind et al applied their study quality assessment tool to. already collected data LaKind et al 2014 developed a epidemiologic literature on BPA and neurodevelopmental. tool to evaluate epidemiologic studies that use biomoni and respiratory health they found that QA QC issues re. toring data on short lived chemicals with a focus on lated to contamination and analyte stability were not well. critical elements of study design such as choice of ana reported 12 Of note several of the studies in our breast. lytical and sampling methods 5 The tool was recently cancer review that did not provide adequate QA QC in. incorporated into ExpoQual a framework for assessing formation had their samples analyzed at the CDC Envir. suitability of both measured and modeled exposure data onmental Health Laboratory It is helpful to include. for a given use fit for purpose 6 Other useful guid summaries of QA QC assessments in published work. ance has been published for example on automated even if researchers are using a well established lab be. quality assurance quality control QA QC processes for cause this provides a useful standard for comparing QA. sensors collecting continuous streams of environmental QC in other studies. data 7 and for establishing an overall data management Over many years of collecting and interpreting environ. plan including documentation of metadata and strat mental exposure data we have developed a standard ap. egies for data storage 8 proach for 1 using field and laboratory QA QC to. Despite these helpful documents there is still a lack of validate and qualify chemical measurement data for envir. readily accessible practical guidance on how to interpret onmental samples and 2 presenting our QC findings in. and use the results of both field and laboratory QC checks our research publications e g reference 13 These. to qualify exposure datasets i e flag results for certain methods are based on data validation procedures from the. compounds or certain samples that are imprecise esti EPA Army Corps of Engineers and U S Geological Sur. mated or potentially over or under reported and this vey 14 17 and the guidance of the many experienced. gap is reflected in the environmental health literature chemists with whom we have collaborated In this com. While the vast majority of environmental health studies mentary we compile our methods into a practical guide. report robust findings based on high quality measure focusing on how to use the information to make decisions. ments questions about measure validity have led to con about data usability and how to make the information. fusion and lack of confidence in some topic areas For transparent in publications Our guide is organized in. example a number of studies have measured rapidly three sections presenting questions to consider during. Udesky et al Environmental Health 2019 18 99 Page 3 of 16. study design implementation and data analysis We de Because the number of QC samples available is often. scribe key elements of QA QC including for assessing limited by budgetary constraints many of the methods. precision accuracy and sample contamination and we in we use rely on visualization and conservative action i e. clude suggested graphics Additional files 2 and 4 and removing chemicals from our dataset or qualifying their. table shells Additional file 2 that clearly present QC data interpretation unless there is evidence that the analytical. emphasizing how it may affect interpretation of study method was accurate and precise rather than on statis. measurements Minimizing and characterizing potential tical methods Whether statistical methods are incorpo. errors requires close collaboration between the re rated or not tabulating visualizing and communicating. searchers who may have designed the study and plan to about QA QC for environmental exposure measure. analyze the data and the chemists performing the analysis ments is important in order to reveal systematic error in. so our guidance also includes example correspondence the laboratory 20 or in the field and support future use. Additional file 2 to help establish this relationship at the of the data 6. start of a project, We present a detailed approach based on our own Study design. studies acknowledging that this is an example not a What can we measure and how. one size fits all approach Every study is unique and One of our first priorities when designing a new study is. some will require specialized quality assessment not cov to consult with a chemist to establish an analyte list and. ered here Still we anticipate that many environmental method for analysis. health scientists will find this example to be a useful. framework for building their own processes Chemical identities Given the complexity of chemical. synonyms it is helpful to be as specific as possible when. Wrangling guide communicating about the chemicals to be analyzed One. Our guide is organized by a series of questions that we approach is to send the lab a list of the chemical names. ask when we start a new study and then again when we avoiding the use of trade names which can be imprecise. receive measurement data from the lab Key QA QC Chemical Abstracts Service CAS numbers and configu. concepts are introduced in the Study Design section and rations e g branched or linear if relevant of all desired. are most thoroughly addressed in sections about Study analytes see Additional file 1 for example correspond. Implementation and Data Interpretation ence For biomonitoring it is also important to determine. Not every question is relevant to every study for ex if the parent chemical or metabolites will be targeted. ample researchers working with a lab to develop a new. analytical method will need to focus more on method Matrix Another consideration in developing the analyte. validation and quality control than those using a well list is what type of samples are available if working with. established method and credentialed lab Still control stored samples or will be collected As discussed previ. ling for issues related to sample collection and transport ously certain biological matrices are preferred over others. remain important in the latter scenario as does variation for measurement depending on the chemicals e g refer. in method performance and or sources of contamination ence 9 Matrix type is also relevant for environmental. when samples are analyzed at the laboratory in multiple samples for example physical chemical properties like the. batches Our guidance is most relevant to targeted or octanol air partitioning coefficient inform whether an ana. ganic chemical analyses which use liquid or gas chroma lyte is more likely to be found in air or dust 21. tography often in combination with mass spectrometry. to determine whether a pre defined set of chemicals are Method The process of determining a final list of ana. present in samples QA QC approaches for non targeted lytes will differ depending on whether the lab has an. methods where tentative identities are established by established method or is developing a new method and. matching to a library of mass spectra such as the Na whether it is targeted to a few chemicals with similar struc. tional Institute of Standards and Technology NIST ture versus many chemicals with different properties differ. database 18 are addressed elsewhere 19 ent polarities solubilities etc Targeting a broad suite of. This guide is not a set of rules but rather establishes a chemicals may limit the degree of precision and accuracy. framework for evaluating and reporting QC data for that can be achieved for each individual chemical and the. chemical measurements in environmental or biological lab may need to invest substantial effort to develop a multi. samples While it may be most useful to environmental residue method that is a method that can analyze for. health scientists who have little or no experience in ana many chemicals at once and determine a final list of target. lytical chemistry we hope that researchers with a range chemicals with acceptable method performance In any case. of experience will find it helpful to consult our approach a new method should be validated to characterize perform. for evaluating and presenting QC data in publications ance measures precision accuracy expected quantitation. Udesky et al Environmental Health 2019 18 99 Page 4 of 16. and method detection limits and the range of concentra establish reporting limits to meet the needs of occupa. tions that can be quantitated with demonstrated precision tional or regulatory safety compliance testing these. and accuracy before analyzing study samples If the lab limits may be much higher than levels that are meaning. already has an established method for the chemicals of inter ful for research questions about general population ex. est the research team should review method performance posure and could result in most data being reported as. measures to ensure they are consistent with study non detect or qualified as estimated and imprecise On. objectives the other hand lower reporting limits generally translate. to more expensive testing so researchers have the op. Method quantification The method of quantification portunity to balance sensitivity and cost. affects the types of QC data that are expected from the. lab Three common approaches include external calibra How to minimize sample contamination. tion internal calibration and isotope dilution a form of There are ample opportunities for sample contamination. internal calibration External calibration where the re during collection storage shipment and analysis espe. sponse i e chromatogram peak from the sample is cially when targeting ubiquitous chemicals commonly. compared to the response from calibration standards encountered in consumer products and in home and of. containing known amounts of the analytes of interest is fice furnishings or laboratory equipment An important. a simple method that can be used for a variety of differ aspect of method validation is to check for contamin. ent analyses However results can be influenced by inter ation of samples during field activities from collection. ference from other chemicals present in the sample containers during transport and storage and during la. matrix and resulting fluctuations in the analytical instru boratory extraction and analysis see discussion of blanks. ment response 22 With internal calibration on the other in the Study Implementation section The CDC s guid. hand one or more labeled compounds either one of the ance on sample collection and management identifies. targeted analytes or a closely related compound are some possible sources of contamination when analyzing. added to each of the samples just before they are injected for common chemicals like plastics chemicals antimi. into the instrument for analysis and used to correct for crobials and preservatives in blood or urine Key consid. variation in the instrument response The internal standard erations depending on the particular chemicals being. must be similar to the target compounds in physical chem targeted include selecting appropriate collection con. ical properties e g a labeled polychlorinated biphenyl tainers e g glass containers if analyzing for plastics che. should not be used to represent a brominated diphenyl micals avoiding the use of urine preservatives e g. ether Finally for isotope dilution methods which are when analyzing for parabens BPA and providing ad. the most accurate labeled isotopes for each of the target equate instructions to participants collecting their own. compounds are added to samples prior to extraction Add samples e g avoid using antimicrobial soaps or wipes. itional internal standards are added to the samples just during collection 4 As noted previously contamin. prior to injection to monitor loss of the labeled isotopes ation can also be minimized in biomonitoring of some. and the analytical software then corrects for loss during chemicals by measuring a metabolite rather than parent. sample extraction and for effects of the sample matrix e g chemical and possibly by measuring a conjugated rather. presence of other compounds in the sample that interfere than free form of the metabolite 9 In some cases the. with the analysis 22 Many laboratories that analyze lab may need to pre screen collection containers or. chemical levels in blood urine or tissues including the other sampling materials to see if they contain any target. CDC National Exposure Research Laboratory use iso chemicals For example when we used polyurethane. tope dilution quantification However isotopically labeled foam PUF sorbent to collect air samples for analysis of. standards are not available for every compound and may be flame retardants plastics chemicals and preservatives. cost prohibitive If quantification is by internal or external we asked the lab to pre screen the PUF matrix for target. calibration researchers will likely need to review and report analytes Another important precaution was to ship the. more extensive QC data from the lab compared to when samplers wrapped in aluminum foil that had been baked. using isotope dilution as discussed in Study Implementa in a muffle furnace to ensure it was clean and uncoated. tion What QA QC is needed,How will the lab report the data. Method sensitivity Another important factor in Three key elements of data typically reported by the lab. selecting a method is to make sure it is sensitive enough are the identity of the chemical the reporting limit for. to detect the anticipated concentrations in the field sam each chemical and sample and how much of each. ples samples submitted to the lab down to levels that chemical is present in each sample Sometimes an add. are relevant to the research question For example com itional measure is needed to normalize mass of chemical. mercial labs measuring environmental chemicals may per sample for example grams of urinary creatinine. Udesky et al Environmental Health 2019 18 99 Page 5 of 16. urine specific gravity grams of serum lipid or cubic me data analysis challenges and while a discussion of the. ters of air see reference 5 for discussion of issues re best available methods and the problems with common. lated to matrix adjustment and presentation of approaches such as substituting the RL RL 2 or zero for. measurements non detects is beyond the scope of this commentary it. is a critical issue and we refer the reader to several help. Chemical identities It is helpful to request in advance that ful resources 23 26 Reporting estimated values is not. the lab report CAS numbers and configurations if relevant standard practice for many laboratories so it is import. along with chemical names see Additional files 2 and 3 for ant to raise this issue early on see Additional file 2 for. example reporting requests example correspondence If the lab reports data quali. fier flags it may be necessary to clarify the interpretation. Reporting limits Common terms used by laboratories of those flags including but not limited to which flags. to discuss reporting limits include instrument detection distinguish non detects from detects above the MRL and. limit IDL method detection limit MDL and limit of estimated values It is best not to make assumptions. quantitation LOQ The IDL and MDL are both related, to the level of an analyte that can be detected with confi Study implementation. dence that it is truly present The IDL captures the smal What QA QC is needed. lest true signal change in instrument response when an QA QC occurs both inside and outside the analytical la. analyte is present that can be distinguished from back boratory see Table 1 Field QC samples namely blanks. ground noise variation in the instrument response to and duplicates capture the sum of contamination and. blank samples while the MDL takes into account add measurement error from collection storage transport. itional sources of error introduced during sample prepar and laboratory sources We base the number of QC. ation e g the extraction process possible concentration samples we collect in the field on budget and our sample. or dilution of samples and thus is higher than the IDL size generally aiming for at least 20 QC samples e g. The MDL is also often referred to as the limit of detection if collecting 80 field samples then collect 16 field QC. LOD or detection limit DL The LOQ on the other samples though a higher percentage is needed in small. hand describes the lowest mass or concentration that can studies Lab analysts should be blinded to the identity of. be detected with confidence in the amount detected The field QC samples whenever possible Maintaining blind. reporting limit RL or method reporting limit MRL ing can be challenging so it is worth putting some. which is either the lowest value that the lab will report or thought into sample names e g QC samples should not. the lowest value that the lab will report without flagging have obviously different IDs than other samples should. the data as estimated is often but not always the same as not be labeled with a D for duplicate or B for blank. the quantitation limit or LOQ Logs retained at the site must contain sufficient informa. Before submitting samples for analysis it is helpful to tion to allow the data analysts to identify field QC sam. find out 1 the methods and terminology that the la ples and sample types. boratory will use to describe reporting limits LOD QC samples prepared in the lab can include spiked. LOQ etc and 2 whether reporting limits will be con samples or certified reference materials CRMs for tar. sistent within a chemical or whether limits could vary get chemicals to evaluate the accuracy of the analytical. between samples or batches Equally critical is to clarify method surrogate compounds added to field samples to. how the lab will report non detects Several different estimate recovery during extraction and analysis and. values could appear in the amount or concentration blanks to assess contamination with target chemicals. fields for non detects including but not limited to zer from some source in the laboratory While laboratories. oes the detection limit the reporting limit or ND generally conduct rigorous review of their own QC data. considering lab and field QC together can help to iden. Amount Another important point to discuss in advance tify specific sources of contamination imprecision and. with the laboratory is how they will report values for systematic error so we typically request to review the. compounds with a confirmed identity but measured at lab s raw QC data in conjunction with the field QC data. levels below what can be accurately quantitated For ex. ample when measuring chemicals of emerging interest Spiked samples and certified reference material. we ask laboratories to report estimated values below the Spiked samples and CRMs establish the accuracy of. RL and we flag them during data analysis This practice the method by assessing the recoveries of known. has some limitations 23 but is preferable to falsely re amounts of each target chemical from a clean or. ducing variance in the dataset by treating estimated representative matrix A CRM is a matrix compar. values below the RL as equivalent to non detects below able to that used for sampling e g drinking water. the detection limit Non detects can present significant that has been certified to contain a specific amount. Udesky et al Environmental Health 2019 18 99 Page 6 of 16. Table 1 Summary of QC sample types interpretation and possible actions. QA QC Measure Interpretation Possible Actions, Accuracy Lab control sample recoveries Measure of whether the analytical Drop compounds with inaccurate. and or matrix spike recoveries method produces accurate quantification from the data analysis. Certified reference material quantification for each compound discuss with lab whether improvements. Isotope dilution quantification Matrix spike recovery evaluates can be made for future analyses. matrix effects on accuracy such If problems are modest and batch specific. as interferences include batch as a covariate in regression. Isotope dilution is the most rigorous model,approach to generating accurate. measurements in biomonitoring, Extraction Surrogate spike recovery in Measure for each field sample Consider dropping samples with poor. efficiency each sample of whether the chemical is extracted surrogate recovery from data analysis. completely from the sample matrix Consider applying a surrogate correction. e g blood dust factor 1 fraction recovery if the recovery. Isotope dilution approaches capture is consistent 15 20 in standard. and correct for differences in extraction deviation. efficiency, Detection limit Level above which the lab can detect See Method Reporting Limit. with confidence that the analyte is,present in the sample. Common terms include,Instrument detection limit IDL. Detection limit DL Method detection,limit MDL Limit of detection LOD. Quantitation Level above which the lab can quantify See Method Reporting Limit. limit with confidence the amount of chemical,in the sample. Common terms include,Practical quantitation limit PQL Limit of. quantitation LOQ Laboratory quantitation,level LQL Contract required quantitation. limit CRQL, Method Levels detected in blanks Level above which the researcher is Determine MRL by comparing. Reporting Limit lab blind field blanks solvent confident that the reported chemical the lab limit quantitation limit. MRL blanks matrix blanks storage measurement reflects a signal from unless not reported in which. blanks other types the media sampled considering all case detection limit to the levels. sources of measurement error in the blanks for each compound. especially potential contamination Qualify reported values below the. during sample collection and handling MRL as estimated. as well as in the laboratory, Potential Levels detected in blanks Measure of confidence in accuracy If evidence of contamination consider. contamination lab blind field blanks solvent of values reported above the MRL dropping a compound or dropping. Analytical bias blanks matrix blanks storage results for a compound in a particular. blanks other types batch,Identify source of contamination. e g lab vs field equipment to,inform future work,For compounds with consistent. contamination in blanks researchers,may correct field sample quantity by. subtracting the amount attributed to,contamination This is most important. when contamination is significant,relative to sample values e g 10. and for comparisons with external data, Precision Relative percent difference RPD for A measure of reproducibility of field Flag compounds with 30 RPD. side by side duplicate samples lab blind measurements including analytical Consider precision in combination with. or split samples lab blind if possible variability and sampling variability other QA QC when deciding to qualify. Udesky et al Environmental Health 2019 18 99 Page 7 of 16. of analyte with a well characterized uncertainty If performance of the method in a clean or representative. CRMs aren t available the laboratory can prepare labora matrix surrogate compounds are used to evaluate recov. tory control samples LCSs by spiking known amounts of eries from individual samples Recoveries of surrogate. target chemicals into a clean sample of the matrix of inter compounds can help identify any individual samples that. est such as a dust wipe air sampler purified water or syn may have inaccurate quantification for example due to. thetic urine or blood that has been analyzed and shown to extraction errors or chemical interferences Surrogates. be free of the analytes of interest or to contain a consistent like internal standards are spiked into each sample. amount of analytes of interest that can be subtracted from however surrogates are added prior to sample extraction. the amounts measured in the spiked sample to calculate a to assess the efficiency of this process Internal stan. percent recovery The LCS or CRM at least 1 per analyt dards on the other hand are added after extraction just. ical batch is run through the same sample preparation prior to injection into the chromatographic system to. extraction and analysis as the field samples to capture the account for matrix effects and other variation in the in. accuracy of the complete method calculating the percent strument response during analysis The ideal surrogate is. of the known spiked amount recovered for each analyte a chemical that is not typically present in the environ. tells us whether the method is accurate in the matrix ment but that is representative of the physical and. Another type of spiked sample called a matrix spike chemical properties of target analytes 16 It is best to. can be used to check the extraction efficiency for a com have a representative surrogate for each individual. plex sampling matrix that may interfere with the ana chemical though when analyzing for numerous chemi. lysis These samples are typically included if there is cals at once with multi residue methods cost and time. concern about interference from the sampling matrix restraints may result in one or a few surrogates being se. for example with house dust soil or sediment samples lected to represent a class of compounds In this case it. consumer products or biological samples like blood In is critical that the lab selects an appropriate surrogate. stead of recovery from a clean matrix these QC checks For analyses using external or internal calibration we. capture recovery from a representative field sample ask the lab to provide us with the recovery results for. Here the matrix refers to all elements of the sample each surrogate in each sample so that we can flag any. other than the targeted analytes this includes the sam samples or compounds that might have had extraction. pling medium e g dust PUF foam itself as well as any problems However if the lab uses isotope dilution quan. other chemicals present in the sample that might inter tification we are less concerned about obtaining this. fere with measurement of target chemicals A matrix raw data from the laboratory given that the reported re. spike can be created for example by splitting a repre sults are already automatically corrected for extraction. sentative sample collected in the field and spiking the and matrix effects. target analytes into one half prior to extraction and ana. lysis The recovery of spiked analyte is determined as the Blanks Collecting and preparing several types of blank. amount measured in the spiked sample minus the samples helps us to distinguish sources of contamin. amount measured in the non spiked sample divided by ation Laboratory blanks alert us to possible contamin. the spike amount A limitation of this approach is that ation originating in the lab These blanks can capture. the analytes are spiked in an already dissolved state so it contamination during sample extraction solvent blanks. is possible that the analytes in the environmental matrix from reagents and other materials used in the analytical. would not be extracted as readily from the matrix as the method solvent method blanks or from typical back. spiked chemicals Thus the true extraction efficiency ground levels of target analytes present in the sampling. may be lower than represented by the matrix spike matrix matrix blanks Field blanks on the other hand. For newly developed methods where performance is not capture all possible contamination during sample collec. characterized we request results for all recoveries of spiked tion and analysis Field blanks are clean samples e g. samples and or CRMs so that we can perform visual checks distilled water air sampling cartridge detached from. that have at times revealed systematic problems with the pump immediately following calibration that are trans. analytical method that were not noted by the lab see Data ported to the sampling location and exposed to all of the. Interpretation Is the method accurate for discussion For same conditions as the real samples e g the sampler is. well established methods and particularly when isotope di opened if applicable except the actual collection. lution quantification is used it is sufficient to request a process We aim for at least 10 of our samples to be. table summarizing the spike recovery or CRM recovery re field blanks with an absolute minimum of 3 field blanks. sults by batch if relevant for reporting in publications Unfortunately in some cases there aren t good options. for representative field blanks For example field blanks. Surrogate recovery standards Whereas recoveries from can be created for biomonitoring programs by taking. LCSs matrix spikes and or CRMs tell us about the empty collection containers into the field and using. Udesky et al Environmental Health 2019 18 99 Page 8 of 16. purified water or synthetic urine or blood to create a will specify whether or not samples were analyzed in. blank 4 However important short comings of this ap batches it is a good idea to request that a variable for batch. proach are that 1 it is difficult to capture contamin be included in the results report. ation that can be introduced by sample collection In Additional files 2 and 3 we provide example corres. materials such as needles and plastic tubing used to col pondence for requesting QC data and consistent format. lect blood 2 water may not perform the same as urine ting from the lab. or blood in the extraction and analysis and 3 the lab. will likely be able to identify the field blanks Similarly it Data interpretation. is difficult to maintain lab blinding when using a clean What was measured. matrix like vacuumed quartz sand as a field blank for. vacuumed house dust Chemical identities No amount of QA QC can save a. dataset from basic misunderstandings about what is be. Duplicates Collecting side by side duplicate samples in ing reported After receiving data it is helpful to ask the. the field helps assess the precision of both the sample chemists to double check the analyte list chemical. collection and analytical methods Duplicate samples can name CAS isomer details against the list of standards. also be created by collecting a single sample and splitting used in the analysis particularly if this information was. it prior to analysis which is the only option for biological not included in the report from the lab It is worthwhile. samples however this method only captures the precision to make this verification even when chemical identities. of the analysis process 14 17 and could lead to un were specified in advance of the analysis as it is possible. blinding of the lab analyst if for example the split samples that the standard used for analysis was slightly different. are noticeably smaller than others When planning for du than planned Only through this process for example. plicate collection the best practice is to label these sam did we discover that a lab had accidentally purchased a. ples so that the lab analyst is blinded to duplicate pairs standard for 2 2 4 trimethyl 1 3 pentanediol isobutyrate. i e use different Sample IDs for the two samples Ideally rather than 2 2 4 trimethyl 1 3 pentanediol diisobutyrate. researchers should plan to collect or create that is split two different chemicals. one duplicate pair per every 10 20 samples collected and Table 2 summarizes some steps for getting acquainted. spread duplicate pairs across analytical batches with a new dataset received from the lab We have also. published sample R code on GitHub that may be helpful. Analytical batches Analytical performance can shift for getting acquainted with a new dataset including exam. over time and even between multiple extractions or in ining trends in QC and field samples over time 28. strument runs within a short time window Laboratories. often analyze samples in multiple batches that is sets of Were there trends over time. field samples and associated laboratory QC samples that. are analyzed together in one analytical run The time be Analytical batches Examining results by batch or even. tween batches can vary from days to months or even by sample run order can reveal trends in QC samples. years though ideally this time span is minimized in over time identifying systematic laboratory errors that. order to maintain consistent equipment and procedures. throughout the study Table 2 Get acquainted with your data. Two approaches help address batch to batch variability 1 Verify chemical identities. 1 randomizing participant samples between batches by. Check CAS number chemical name isomer type of reported analytes. specifying the order and grouping of samples and blind vs analytical standards purchased by lab see Additional file 2 for. field QC samples when submitting samples to the lab this example correspondence. may require corresponding with the lab to determine the 2 Count overall and by batch the number of. batch size in advance and 2 running CRMs such as Real samples compare these to the chain of custody that lists. standard reference material SRM from NIST 27 in the samples submitted for analysis to make sure all submitted. each batch of samples in order to characterize drift When samples were analyzed. CRMs are not available another option is for the researcher Lab control and or matrix spike recoveries. to prepare identical split reference samples We have done Reference samples e g CRMs. this for example by pooling together several urine speci Surrogate spike recoveries. mens and making many aliquots of the pool then including. Blanks solvent method field matrix other, 1 2 blinded samples from this pool with each set of sam. Duplicates, ples we send to the lab If the laboratory analysis is per. formed in multiple batches all QC elements should be 3 Examine any data qualifier flags reported by the lab and make sure. interpretation is clear, examined on a batch specific basis Not every laboratory. Udesky et al Environmental Health 2019 18 99 Page 9 of 16. may be missed by summary statistics or visualizations Table 3 Spiked samples and certified reference material. 20 Shifts in method performance over time may require Approach see Additional file 4 for example of this approach with real. batch specific corrections or dropping or flagging data data. from certain batches Notably a trend in QC sample re 1 Summarize percent recoveries for each chemical across analytical. sults over time can be problematic even if they remain batches and flag those chemicals with average recoveries outside of a. pre established acceptable range, within the acceptable limits established by the lab In our. We typically apply an acceptable range of 50 150 recovery for. own work for example examining our data by analytical most environmental samples particularly when we are analyzing for. batch revealed an upward trend in sample specific detec new chemicals or combinations of chemicals for which methods are. tion limits for some analytes such that detection limits in not well established For well established methods a more. conservative range 80 120 recovery is appropriate, later batches were within the range of sample results from. earlier batches Fig 1 The detection limits in the later 2 Visualize percent recoveries for each chemical across analytical. batches to assess consistency, batches still met the specifications of our contract with. the lab but it was clear that we would not be able to com If recoveries for a particular chemical or chemicals are consistently. out of range 150 or 50 across multiple batches this should. pare results in the latter two batches to those in the first be discussed with the laboratory analyst. three We showed the plot in Fig 1 to the lab and they If the laboratory analyst agrees that the method was not. agreed to re analyze the samples in the later batches successful we drop the chemical s from our dataset We do not. which resulted in more consistent detection limits report values or include such chemicals in any data analyses. If the laboratory analyst can explain the reason for consistent. Is the method accurate high or low recoveries and has confidence in the ranking and. relative values of the reported sample data the reported values can. be used for many data analyses but it will be difficult to compare. Spiked samples and certified reference material with levels from another study. Table 3 outlines our approach for analyzing LCS or If recoveries from one or a few batches are out of range we are. matrix spike recovery or CRM data The approach is similar concerned that results in those batches might be over under. for all of these samples However one distinction is that if estimated compared to the rest One way to investigate this concern. is to look for corresponding systematic differences in sample data. see Additional file 4 Figure S3, If field samples have been randomized into batches we check if. the variation in sample results correlates with spiked sample or. CRM recoveries by batch Note we still go through this step even if. we were not able to randomize field samples but in this case it. can be very challenging to distinguish systematic analytical. variation from other possible sources of variation in sample results. between batches e g if samples in different batches were also. collected during different seasons, If there are systematic differences e g the sample results for a. chemical are higher in the batch where the spike or CRM recovery. was high or if only one batch the sample results for a chemical. with high spike or CRM recovery are much higher than previously. reported levels we consider dropping the chemical results from. the affected batches from the dataset If an identical split reference. sample was analyzed in each batch these results can also be. helpful to resolve questions about whether and how to use the. data in this case, If there are no obvious systematic differences we keep the. chemical in our dataset but flag the results for that chemical in the. batch with the out of range spiked sample or CRM recovery. We note in summary statistics when the average spiked sample or. CRM recovery for a particular chemical was out of range. We note whether levels in our study might be systematically over or. under reported i e because of consistent high or low spiked sample or. CRM recoveries We especially note this if comparing to levels from. another study, For chemicals with low high recoveries in certain batches we may. Fig 1 Visualizing urine sample results by analytical batch data not perform sensitivity analyses for example by including lab batch as a. yet published revealed that sample specific detection limits in later covariate in regression analyses though this can be challenging for. small datasets, batches were higher and in the range of sample results in previous. batches After discussing with the laboratory samples in later. batches were re analyzed to achieve lower detection limits. Udesky et al Environmental Health 2019 18 99 Page 10 of 16. LCS recovery and other QC measures such as lab blanks employed we review the surrogate recovery standard. matrix solvent method or other are acceptable a poor data for each individual sample generally considering. matrix spike recovery higher or lower than acceptable 50 150 recovery to be acceptable Interpretation of an. bounds can alert chemists to interferences from matrix ef out of range surrogate recovery depends both on its direc. fects and suggest steps to address this such as matrix tion and on the levels of the associated analytes i e those. matched calibration 17 We typically only use data for ana represented by the surrogate compound measured in the. lytes that have average LCS matrix spike and or CRM recov sample In samples with low surrogate recoveries the. eries between 50 and 150 though this decision criterion concern is that if similar target analytes are present in the. can be adjusted based on the needs of the project If we do sample the measurements will be underestimated biased. retain data for chemicals with spike or CRM recoveries out low For samples with high surrogate recoveries on the other. side of this acceptable range we note in publications that hand we can be confident that similar target compounds. concentrations in our data may be under or over reported should be detected if present but the amount may be. Figure 2 illustrates a case from our own data where the overestimated or biased high If surrogate recoveries are out. laboratory reported that 1 2 5 6 9 10 Hexabromocyclodode of range in all samples and particularly if they are also out. cane HBCD a brominated flame retardant was mostly of range in blank samples this is likely indicative of a. not detected but the LCS recoveries which ranged from broader problem with the analytical method 16 29 Table 4. 2 to 1670 and averaged about 750 indicated that the outlines our approach for analyzing surrogate recovery data. method was not able to accurately quantify this chemical Figure 3 shows an example where our examination of. We removed this compound from our dataset and did not surrogate recoveries on a batch specific basis indicated. report on it Examining spike recoveries thus prevents us trends in the recoveries over time even though most. from reporting a chemical as not detected or from remained within the generally acceptable range 50. reporting an unreliable detect if the analytical method is 150 This plot led to a discussion with the lab analyst. not performing accurately for that compound who suggested that stock solutions for surrogate com. A summary of the recovery information should be in pounds may have concentrated over time as solvent. cluded in the peer reviewed manuscript to demonstrate evaporated until a new stock solution was prepared for. accuracy See Additional file 2 Tables S1 S2 and Figure the last batch On the advice of the lab analyst we. S1 for an example of how to present this information looked at trends in the spike check solvent that is. spiked with target analytes but not extracted or concen. Were there problems with certain samples trated sample recoveries Spike check recoveries indi. cated good reproducibility giving us confidence that the. Surrogate recovery standards When isotope dilution drift in surrogate recoveries did not reflect changes in. quantitation with automatic recovery correction is not instrument calibration over time. Fig 2 a Results for flame retardant HBCD measured in air samples collected in 105 homes All but three samples were non detects open circles. Samples were analyzed in six different analytical batches b Summary of laboratory control spike recovery data for HBCD across the six analytical. batches shows very poor accuracy and indicates no confidence for this analyte in the indoor air samples. Udesky et al Environmental Health 2019 18 99 Page 11 of 16. Table 4 Surrogates Table 4 Surrogates Continued, Approach see Additional file 4 for example of this approach with real Similarly if 100 of samples are detects we also flag the minimum. data value if it is from a sample associated with a low 50 recovery. and note that in this case the minimum might be underestimated. 1 Count high and low recoveries for each surrogate chemical across. analytical batches For any statistical analyses if possible i e if large enough dataset. we run sensitivity analyses, We typically apply an acceptable range of 50 150 recovery for most. environmental samples particularly when we are analyzing for new Excluding samples with out of range surrogate recoveries. chemicals or combinations of chemicals for which methods are not. Controlling for lab batch if surrogates were problematic for a. well established For well established methods a more conservative. range 80 120 recovery would be appropriate particular batch. 2 Identify any sample where all surrogate recoveries were low e g. 50 This suggests a potential problem with the extraction for that. Is there evidence of contamination or analytical bias. Discuss with lab analyst Consider dropping sample, 3 Visualize surrogate recoveries for QC samples lab blanks lab control or Blanks Once we have determined that we can accurately. matrix spikes across analytical batches See Additional file 4 Figure S4. for an example, measure the target analytes in our sampling matrix the. next step is to ensure that we are confident about whether. If these recoveries are out of range this suggests a larger problem. with the analytical method rather than with particular samples those target analytes came from the study site or participant. Summarize information about the surrogate recoveries in the QC or from somewhere else Table 5 outlines our approach. samples as well as lab control or matrix spike recoveries for the to reviewing data from blank samples When it is not. associated chemicals and discuss with lab analyst, straight forward to collect field blanks e g for blood sam. 4 Visualize percent recoveries across all samples for each surrogate by ples any assessment of contamination introduced from. analytical batch See Additional file 4 Figure S6 for an example. Note any trends upward or downward in the distribution of. surrogate recoveries across batches Such trends should be discussed. with the laboratory analyst even if all recoveries are in the 50 150. acceptable range see Fig 3 for an example, If the surrogate is a deuterated version of one of the target. chemicals it can be helpful to compare a plot of the surrogate. recoveries by batch to the sample data for the corresponding un. deuterated target chemical by batch We would be concerned and. would seek guidance from the lab analyst if we saw a trend for the. target chemical results that matched the trend in the surrogate. recoveries, Note if many surrogate recoveries e g more than half are out of. range in a particular lab batch If yes flag the results in that batch for. the chemical s represented by that surrogate, 5 Visualize sample results flagged by surrogate recoveries For each. individual sample with an out of range recovery for a surrogate flag the. results for the chemical s associated with that surrogate Plot all sample. data with indicators e g different colors for whether the representative. surrogate for each sample was out of range See Additional file 4 Figure. S7A D for an example, Note whether samples with high surrogate recoveries consistently. have the highest results for the associated chemical s. If yes we would be concerned that samples with high recoveries. are all overestimated Discuss with lab analyst Consider applying a. surrogate correction factor to sample results multiplying by 1. fraction recovery, Note whether samples with low surrogate recoveries were. consistently non detects or very low level detects for the associated. chemical s, If yes we would be concerned that samples with low recoveries Fig 3 In this example from our data recoveries of surrogate d4 di. are all underestimated Discuss with lab analyst Note in. n butyl phthalate from air samples showed notable upward and. publications that levels and detection frequencies for associated. chemicals might be underestimated downward trends over time despite largely staying within the 50. 150 acceptable bounds Here we were examining surrogate. Reporting recoveries in batches of samples from different studies analyzed at. In summary statistics we note whether any maximum value is from the same laboratory The last two batches Sept 2014 and May 2015. a sample associated with a high 150 surrogate recovery and were from the same study but collected approximately a year apart. note that in this case the maximum might be overestimated per the study design. Udesky et al Environmental Health 2019 18 99 Page 12 of 16. Table 5 Get acquainted with blanks, Approach see Additional file 4 for example of this approach with real. 1 Summarize results across all chemicals by blank type e g field blank. solvent method blank matrix blank etc with non detects set to zero. For chemicals with no detects in blanks the MRL will equal the lab. reporting limit and none of the subsequent steps in Tables 6 or 7 are. For chemicals detected in blanks, 2 Visualize levels in blanks by blank type Set non detects to lab. reporting limit and plot by analytical batch, Consider whether blank detects are consistent across batches Note. whether detects seem to occur mostly in one type of blank which. could indicate a source of contamination in the lab or field. If a particular source is suspected we investigate talk to lab look. at field logs etc, 3 Visualize levels in blanks by blank type along with field samples by. analytical batch Set non detects to lab reporting limit. Note whether blanks are in range of the samples, If field samples have been randomized into batches check if. variation in sample results correlates with blank results by batch. Note we still go through this step even when we were not able to. randomize field samples but it is more challenging to distinguish. whether contamination is driving differences in sample results in a. particular batch or whether other explanations are more likely e g all. samples in one batch were collected in a different season or from a. particular study site Fig 4 Phthalate DEHP measured in air in college dorm rooms. before and after occupancy data not yet published Levels in our. samples purple dots were higher post compared to pre. occupancy but this plot revealed that levels in field blanks blue. sampling e g pre screening of collection materials should dots were also higher post compared to pre occupancy and within. be thoroughly described and limitations acknowledged the range of field samples We also saw a matrix blank green dot. Figure 4 illustrates an example from our study com well within the range of the field samples in the pre occupancy. paring levels of chemicals in air in college dorm rooms batch These data suggest DEHP contamination in both batches for. before and after students moved in data not yet pub the post occupancy batch we hypothesized this might have come. from the plastic bags in which the samplers were shipped We will. lished where field blanks proved particularly crucial not report results for this chemical from this study given the. Our first look at the sample data suggested that bis 2 evidence of contamination LLOQ Lower Limit of Quantitation. ethylhexyl phthalate DEHP a chemical commonly, used in plastics was present at notably higher levels after. students moved in However upon further review we detects above the MRL In the example of the potentially. found that DEHP levels in the field blanks were also DEHP contaminated plastic bags used to transport sam. higher and in the range of the sample data at the post ples however we decided not to report DEHP levels for. compared to pre occupancy time point At the same the post occupancy samples given the evidence that con. time levels of DEHP in the laboratory blanks matrix tamination might have significantly biased the results in. and solvent method were not elevated A conversation that batch Unexpected findings such as a chemical or che. with the lab revealed that different plastic bags may have micals detected at much higher levels in a lab blank. been used to transport samples during the later round of matrix solvent method or other than in the field blanks. sampling i e the post occupancy sampling These bags warrant further investigation In this case we might suspect. may have contained higher levels of DEHP that the lab blank was contaminated by another sample. Typically we use blanks to qualify values rather than re examining the sample run order which must be requested. move measurements from our data Specifically we use de from the lab see example correspondence in Additional file. tected values in field blanks and sometimes other blanks 2 could shed light on whether a very high sample was run. see Table 6 as a basis to qualify data by raising the method directly before the lab blank. reporting limit MRL flagging low values as estimated After we establish the MRL for chemicals that are de. until we feel confident in the levels we re reporting Values tected in blanks we are confident that levels in samples. reported by the lab but below the MRL are considered above that value are true detects and that they are cor. estimated see Fig 5 for example of graphical presentation rectly ranked but there may still be concern about con. distinguishing estimated detects below the MRL from true sistent bias in the actual numeric values being reported. Udesky et al Environmental Health 2019 18 99 Page 13 of 16. Table 6 Consider Raising Method Reporting Limits MRLs Table 6 Consider Raising Method Reporting Limits MRLs. Approach see Additional file 4 for example of this approach with real Continued. data Graphical presentations should distinguish estimated from true. 1A For chemicals not detected in blanks the MRL is equal to the detects e g by plotting as different shapes see Fig 5. laboratory reporting limit For reporting in tables we use median sample volume across. 1B For each chemical detected in blanks if there are detects in blanks in samples to convert mass based MRL to a single concentration based. all batches establish the MRL as follows otherwise proceed to 1C MRL for each chemical if applicable. Compare the lab s reporting limit to the 90th percentile of field There are different approaches for incorporating estimated or 0 5. blanks computed with non detects set to lab s reporting limit The flagged values in statistical analyses including performing analyses. higher value is the new MRL weighted by estimates of the measurement precision below the MRL. or using censored regression methods 23 However any approach. However if we observe many detects in other types of blanks that incorporates estimated values is preferable to procedures that. e g matrix solvent we consider determining the MRL by substitute with the DL DL zero or remove these values a practice. comparing the lab s reporting limit to the 90th percentile of ALL which can introduce bias 25. blanks computed with non detects set to lab s reporting limit. The higher value is the new MRL, It can be helpful here to plot sample data with different possible. MRLs to gain understanding of precisely what is being achieved by. both from contamination in the field or lab or from bias. raising the MRL i e are we successfully flagging data that we are in the analytical method Consistent bias in levels would. not confident in and at the same time leaving data in which we not be a major concern for ranking individual exposure or. have confidence unqualified See Additional file 4 Figure S9 for. an example of this type of plot, comparing groups within a study but is misleading when. comparing to levels reported in other studies For each. Note we use the 90th percentile of the blanks rather than using. the maximum value or the mean because the 90th percentile is less chemical we check for evidence of consistent bias across. sensitive to extreme values and can be estimated for data that are many blanks and correct concentrations reported in sum. not normally distributed However if the overall study is small e g in mary tables in our papers to reduce this bias see Table 7. our practice when we have 5 blanks we set the MRL equal to the. maximum blank mass,How precise are these measurements. 1C For each chemical detected in blanks if detects in blanks are. clustered in one or a few batches, Duplicates Duplicate samples indicate whether variation. If just one extremely problematic batch consider dropping the. sample data from that batch in our data is explained by imprecision If duplicate. If multiple field blanks were run in each batch can consider. determining MRL as above but on a batch specific basis. In this case the way to proceed will very much be a judgment. call Spend time with the data considering various approaches. Data from reference material and duplicate samples can be helpful. in deciding which data points should be qualified because they are. in the noise, 2 After determining the MRL we flag each sample result as follows. 0 flag measurement reported by the lab as non detect. 0 5 flag measurement falls below the MRL These are considered. estimated detects, 1 flag measurement falls above the MRL These are considered. true detects, Note that our data qualifier flags may differ from those used by others. For example NHANES flags non detects with a 1 and detects with a. 3 Normalize MRL, If the MRL is determined on a mass basis but sample results are. normalized by some factor such as sample volume we compute a. sample specific concentration based MRL by dividing the mass based. MRL by the sample volume, We do not count estimated values 0 5 flags as detects when reporting. MRL We do not use estimated detects to calculate summary statistics. such as percentiles see Table shell S2 in Additional file 2. In summary statistics we identify any chemicals with greater than. 50 estimated detects and add a footnote Imprecise quantification Fig 5 Example of graphical presentation distinguishing true. for more than 50 of detected values estimated and non detects MRL Method Reporting Limit. Udesky et al Environmental Health 2019 18 99 Page 14 of 16. Table 7 Blank correction because it can indicate that the results are reproducible. Approach see Additional file 4 for example of this approach with real On the other hand consistently poor precision for dust. data wipe samples for example has informed our decision to. 1 Which blanks to use rely more heavily on measured air concentrations as an. If detects are spread across all types of blanks e g field solvent indicator of home exposure 30 Table 8 outlines our. method matrix we use all blanks for blank correction Otherwise we approach for analyzing duplicate data. use field blanks We try to keep our blank correction approach. consistent with our MRL approach,Publication how do we tell others about our data. 2 Which chemicals get corrected, While it is imperative that a researcher has a thorough un. If 5 blanks derstanding of the quality of her own data it is equally im. For each chemical we use a one sided one sample sign test portant that she clearly communicate the results of the. special case of binomial test with p 0 5 to determine whether. the median of blanks is statistically significantly different from zero. QA QC review When we considered the articles included. True and estimated detects are treated as positive values and non in our recent review of epidemiologic studies of environ. detects as negative values mental chemicals and breast cancer 11 we identified. We blank correct chemicals with a sign test p value 0 05 gaps in reporting and or interpretation of QA QC data. However if the number of blanks is relatively small 10 or an issue also noted by LaKind et al 12 To encourage. fewer we consider blank correction even when the sign test does more regular and consistent reporting of QA QC results. not produce a significant result The sign test does not take into in supplementary material we provide examples of the. account the magnitude of the levels detected in the blanks nor. does it distinguish different types of blanks i e field and lab. Table 8 Duplicates, For example if we have 3 field blanks and 4 lab blanks and. we see consistent levels detected across all field blanks and all but Approach see Additional file 4 for example of this approach with real. one lab blank we would consider blank correcting even though data. the sign test would produce p 0 05 1 Compute precision. If 5 blanks i e for a small dataset Compute summarize average relative percent difference RPD for. With five or fewer blanks the sign test will never be significant duplicate pairs or if 3 side by side samples compute relative stand. In this case we blank correct chemicals with 100 detects in ard deviation RSD. blanks If sample results have been normalized e g mass converted to. 3 Blank correction concentration compute precision with normalized values. Calculate the median value of the blanks with non detects set to Compute only for pairs where both samples are detects. lab s reporting limit and using all values i e estimated and true Also consider precision restricted to pairs where both samples. detects are flagged as true detects above the MRL, It is useful to pause here and assess the value being used for 2 Visualize duplicate pairs. blank correction Is it based on an estimated value below the MRL. What will be the percent change in the median comparing the This is a good point to pause and check your data and to note. original to the blank corrected data investigate anything that looks unusual e g huge difference in. results for two members of a duplicate pair how tight are detect. Subtract median blank value from all sample results non detect pairs See Additional file 4 Figure S10A D for an example. Subtract median blank value from the MRL determined as in Table 6 3 Average duplicates with non detects set to lab s reporting limit. Reporting Calculate average volume and concentration for each pair Note. We are explicit about whatever procedure we use to decide can skip this step if only have mass data or if results were reported. whether or not to perform blank correction sign test or other and by lab as concentrations rather than as masses that were then. about the statistic e g median mean and amount used for normalized to concentrations. correction Back calculate new average mass using average volume and. Any presentation of measurements e g summary statistics should average concentration Or simply average the duplicate. use blank corrected values because they may be compared with mea measurements if only have mass data or if results were reported by. surements in other studies lab as concentrations rather than as masses that were then. normalized to concentrations, For statistical analyses such as regression and correlation performed. within the dataset non blank corrected data can be used Compare new average measurement to MRL to determine data. qualifier flag,Combine duplicate averages back with rest of data. samples have high reproducibility meaning that the rela Reporting. tive percent difference between measurements in dupli In publications we note the range of average RPDs across all. cate samples is less than 30 it adds to confidence in chemicals in our QA QC discussion We consider average RPD 30. the field sample results In fact excellent precision in to be good precision. duplicate samples can influence a decision about how to If a chemical has sporadic blank contamination or variable spike. treat data for a chemical that has sporadic blank con recoveries excellent precision can increase our confidence in the field. sample results, tamination or variable spiked sample or CRM recoveries. Udesky et al Environmental Health 2019 18 99 Page 15 of 16. tables and plots Additional file 2 Tables S3 S4 and Fig Funding. ure S1 we have used to communicate QA QC findings in This work was funded by the U S Department of Housing and Urban. Development Grant No MAHHU0005 12 and by charitable gifts to Silent. our publications Consistently publishing QA QC findings Spring Institute. allows readers to think for themselves about the quality of. the data and can inform risk of bias assessments in a sys Availability of data and materials. tematic review QA QC data also provides a basis for de Not applicable. termining whether further analyses of the published data. Ethics approval and consent to participate, e g comparisons to or pooling with other datasets are Not applicable. appropriate,Consent for publication,Not applicable. Conclusion, Several real examples from our data demonstrate that Competing interests. close examination of lab and field quality control data is The authors declare that they have no competing interests. worth the effort By providing a detailed example of how. Author details, we have processed and drawn conclusions about our 1. Silent Spring Institute 320 Nevada Street Newton MA 02460 USA 2MIT. own environmental exposure data Additional file 4 we Media Lab 75 Amherst St Cambridge MA 02139 USA. aim to make our guidelines explicit and straight forward. Received 29 May 2019 Accepted 24 October 2019,so that others may adopt and build on them. Supplementary information References, Supplementary information accompanies this paper at https doi org 10 1 Wild CP Complementing the genome with an exposome the outstanding. 1186 s12940 019 0537 8 challenge of environmental exposure measurement in molecular. epidemiology Cancer Epidemiol Biomark Prev 2005 14 8 1847 50. 2 Carlin DJ Rider CV Woychik R Birnbaum LS Unraveling the health effects of. Additional file 1 Table S1 Summary of our application of the environmental mixtures an NIEHS priority Environ Health Perspect 2013. Navigation Guide Criteria for Low Risk of Bias Assessment for the 121 1 A6 8. question Were exposure assessment methods robust 3 U S EPA Handbook for quality assurance United States Environmental. Additional file 2 Section I Example lab correspondence Section II Protection Agency 2019 Available from https www epa gov citizen. Table S2 Summary statistics table shell Table S3 Quality assurance and science handbook quality assurance Accessed 28 May 2019. quality control QA QC summary table shell Table S4 Findings and 4 CDC Improving the Collection and Management of Human Samples Used. actions from review of quality assurance and quality control QA QC for Measuring Environmental Chemicals and Nutrition Indicators Centers for. data table shell Section III Figure S1 Distribution of surrogate recoveries Disease Control and Prevention 2018. by study visit 5 LaKind JS Sobus JR Goodman M Barr DB Furst P Albertini RJ et al A. Additional file 3 Example of report formatting request to send to the proposal for assessing study quality biomonitoring environmental. lab epidemiology and short lived chemicals BEES C instrument Environ Int. 2014 73 195 207, Additional file 4 Example QA QC report 6 LaKind JS O Mahony C Armstrong T Tibaldi R Blount BC Naiman DQ. ExpoQual evaluating measured and modeled human exposure data. Environ Res 2019 171 302 12,Abbreviations, 7 Campbell JL Quantity is nothing without quality automated QA QC for. CAS Chemical Abstracts Service CDC Centers for Disease Control and. streaming environmental sensor data BioScience 2013 63 7 574 85. Prevention CRM Certified reference material CRQL Contract required. 8 Michener WK Ten simple rules for creating a good data management plan. quantitation limit DEHP Bis 2 ethylhexyl phthalate DL Detection limit. PLoS Comput Biol 2015 11 10 e1004525, EPA United States Environmental Protection Agency HBCD 1 2 5 6 9 10. 9 Calafat AM Longnecker MP Koch HM Swan SH Hauser R Goldman LR. Hexabromocyclododecane LCS Laboratory control sample LOD Limit of. et al Optimal exposure biomarkers for nonpersistent chemicals in. detection LOQ Limit of quantitation LQL Laboratory quantitation level. environmental epidemiology Environ Health Perspect 2015 123 7 A166 8. MDL Method detection limit MRL Method reporting limit NIST National. 10 Johnson PI Sutton P Atchley DS Koustas E Lam J Sen S et al The. Institute of Standards and Technology PQL Practical quantitation limit. navigation guide evidence based medicine meets environmental health. PUF Polyurethane foam QA QC Quality assurance quality control. systematic review of human evidence for PFOA effects on fetal growth. QC Quality control RL Reporting limit RPD Relative percent difference. Environ Health Perspect 2014 122 10 1028 39, RSD Relative standard deviation SRM Standard reference material. 11 Rodgers KM Udesky JO Rudel RA Brody JG Environmental chemicals and. breast cancer an updated review of epidemiological literature informed by. Acknowledgements biological mechanisms Environ Res 2018 160 152 82. We thank David Camann Alice Yau Marcia Nishioka Martha McCauley and 12 LaKind JS Goodman M Barr DB Weisel CP Schoeters G Lessons learned. Adrian Covaci for helping us make thoughtful decisions about how to from the application of BEES C systematic assessment of study quality of. interpret our data We also thank Vincent Bessonneau for helpful discussion epidemiologic research on BPA neurodevelopment and respiratory health. and for his valuable comments on a draft of this manuscript Silent Spring Environ Int 2015 80 41 71. Institute is a scientific research organization dedicated to studying 13 Rudel RA Dodson RE Perovich LJ Morello Frosch R Camann DE Zuniga. environmental factors in women s health MM et al Semivolatile endocrine disrupting compounds in paired indoor. and outdoor air in two northern California communities Environ Sci. Authors contributions Technol 2010 44 17 6583 90, RAR RED and LP developed the detailed data processing approach provided 14 U S EPA Guidance on Data Quality Indicators EPA QA G 5i Peer review. as an example in this manuscript JU drafted the manuscript RAR RED and draft United States Environmental Protection Agency 2001 Available from. LP critically reviewed and revised the manuscript All authors read and http colowqforum org pdfs whole effluent toxicity documents g5i prd pdf. approved the final manuscript Accessed 28 May 2019.
Each book in the Veritas Prep curriculum contains four distinct sections: 1 . Skillbuilder . We strongly suggest that you complete each Skillbuilder lesson before class at your own pace, and return to the Skillbuilder when you recognize a content deficiency through practice tests and GMAT homework problem sets. The Skillbuilder section will:
1 lesson plan (see the Gymnastics Foundations Theory manual) - You learned how to Plan a Lesson in your Gymnastics Foundations Theory course. - Make sure your Coach Evaluator knows if this is the lesson plan you will use during your observation or if you will provide another one. Completed Program Description form (see page 4)
Chapters 11 and 12 55. Early in chapter twelve, Soda runs out of the house because Darry and Ponyboy are arguing. Later, Soda tells the two that he feels he gets caught in the middle and that their arguments are tearing the family apart. Has there been an incident in which you may have been caught in the middle and were forced to choose sides ...
FIN 3403 is the core finance course required of all business majors and it serves as the basis for all other courses in the area of finance as well as providing the basic tools that every business student will need to be successful in their chosen career. Emphasis is placed on the underlying principles and practices and how they relate to the
Critical theories based on post-structuralist and postcolonial approaches have already been broadly discussed within religious studies (King 2013). However, their explanatory potential is far from being fully explored. It will be argued that recent discussions about post-foundational political theory by Ernesto Laclau, combined with certain
to distribute journalistic reporting, i.e., any magazine or newspaper in Germany that provides information on current affairs, is distributed in return for payment, and does not have any direct commercial interest. Inserts can be included in the shipment at low cost. PRESSESENDUNG (standard periodicals) The content design options for PRESSESENDUNG