Human monitoring analysis calculation
Human monitoring analysis computes substance concentration estimates for a biological matrix (e.g., urine or blood) based on human monitoring data. These estimates are specified at the level of long term average concentrations for individuals in case of chronic assessments, or concentrations for individual-days in case of acute assessments. The concentrations are computed independently for each substance and biological matrix.
The main steps, and also performed in this order, for computing human monitoring concentration estimates are:
Standardise blood for lipid content (only available when data allows).
Standardise/normalise urine for creatinine or specific gravity (only available when data allows).
Filter for complete cases.
Combine urine from different sampling methods.
Apply exposure biomarker conversion of substance concentrations (within a biological matrix).
Apply kinetic conversion of substance concentrations from other biological matrices (only available when data allows).
Calculation of individual concentrations (chronic) or individual day concentrations (acute).
Imputation of censored values
Similar to concentrations measurements in food, human monitoring measurements contain measurements below the limit of reporting and similar to concentrations modelling in foods, human monitoring analysis addresses these censored values and replaces them with imputed concentration values, see also imputation methods. Two approaches are available:
Impute using a non-detects handling method.
Impute using a draw from the left tail of the censored lognormal distribution. See also concentration models and concentration model types.
The available non-detects handling methods for deterministic imputation are:
Replace censored values by zero.
Replace censored values by a factor * LOR, the factor is set between zero and one.
Replace non-detects by a factor * LOD and non-quantifications by LOD + factor * (LOQ - LOD), the factor is set between zero and one.
Replace non-detects by zero and non-quantifications by factor * LOQ, the factor is set between zero and one.
For option 3, factor = 0, non-detects are replaced by zero, non-quantifications are replaced by LOD; for factor = 1, non-detects are replaced by LOD, non-quantifications are replaced by LOQ. See also PARC harmonised HBM data format: Sample concentrations/measurements.
Note, when LOD is not available then it is assumed to be 0. When LOQ is not available then it is assumed to be LOD (or zero if LOD is also not available).
For imputation based on the censored lognormal distribution, non-detect and non-quantification information is used. Non-detects are sampled from the entire left tail, e.g. the area below LOD. Non-quantifications are sampled from the intermediate segment, e.g. the area below LOQ and above LOD.
Note that for the option based on the censored lognormal distribution also a non-detects handling method has to be specified. Occasionally, fitting the censored lognormal model fails and the deterministic imputation method is used as fallback.
The censored lognormal distribution, with parameters \(\mu\) and \(\sigma\), is fitted using a likelihood function which is defined by contributions of the positive measurements (lognormal probability density function) and by contributions of the measurements below LOD and measeruments above LOD and below LOQ separately, e.g. the (cumulative) probabilities for LODs and LOQs.
In Figure 93, the human monitoring concentration data in urine are imputed with a factor * LOR and the summary statistics are visualized.
Figure 93 Boxplots for imputed concentration data for five pesticides. Lower whiskers indicate the p5 and p10 percentiles, upper whiskers the p90 and p95. The edges of the box indicate the p25 and p75 percentiles with the median in the centre of the box.
Imputation of missing values
For missing concentration measurements, three imputation methods are available, see also imputation methods. The third opion ignores imputation and all missing values remain in the data set.
Replace missing values by zero. This is conditional on the specified minimum percentage of non-missing values.
For each substance and combination of biological matrix and sampling type, replace missing values by a random draw from the non-missing concentration values (samples). This is conditional on the specified minimum percentage of non-missing values.
Do not impute missing values.
Imputation of missing values by zeros or a random draw from the data is conditional on the specified percentage of non-missing values. When the percentage of non-missing values for a specific substance and combination of biological matrix and sampling type in the data is smaller than the specified percentage, imputation is ignored. This is to prevent that imputation takes place using a set of imputation values that is not representative or unrealistically small (e.g. one or a few values). Note that for the second imputation method, more refined methods could be used. E.g., when for a given day multiple samples are available and one is missing, then this sample might be neglected in the computation of an average exposure. Also, when samples have been taken at different time-slots, impute the missing records using samples from the same time-slot.
Note that imputation of missing values applies only to substances within the assessment scope, not to those used indirectly for standardisation and normalisation, such as creatinine, total cholesterol, and triglycerides.
Standardise blood for lipid content
Lipid-soluble substances measured in blood data are typically standardised by total lipid content, see also standardisation methods. Three methods are in available, namely:
Standardisation based on gravimatic analysis.
Standardisation based on enzymatic summation analysis.
Standardisation based on Bernert et al. (2007), total lipids (mg/dL) = 2.27 * total cholesterol + triglycerides + 62.3 mg/dL.
The standardisation is only applied to lipid-soluble substances, see Substances data formats. After standardisation, the amount of substance is expressed per g lipid e.g. \(\mu g / g \, {\small \mathtt{lipid}}\). Note that substance concentrations in blood samples with unmeasured lipid concentrations are set to missing after specifying option blood standardisation.
This option is only available to the user when the data contains substances that are soluble in lipid and when total lipid content (or cholesterol and triglycerides) is measured.
Standardise/normalise urine for creatinine or specific gravity
Four methods are available for correcting spot urine measurements for creatinine of specific gravity, see also standardisation methods.
Normalisation based on specific gravity.
Standardisation based on creatinine content.
Normalisation based on specific gravity derived from creatinine content, adults 18 - 68 years.
Normalisation based on specific gravity derived from nonlinear modelling of creatinine content, children 6 - 14 years.
Normalisation based on specific gravity derived from nonlinear modelling of creatinine content, age and gender dependent, children 6 - 14 years.
Urine’s specific gravity is determined by the concentration of excreted molecules in the urine. In adult humans, normal specific gravity values range from 1.010 to 1.030. The specific gravity normalisation used here (1) is equal to \((1.024 - 1) / (specific \, gravity - 1)\). The specific gravity value should be available in the HBM data, otherwise urine sample concentrations are set to missing.
After standardisation for creatinine content, the amount of substance is expressed per g creatinine e.g. \(\mu g / g \, {\small \mathtt{creatinine}}\). Note that substance concentrations in urine samples where the creatinine content is not measured are set to missing values after specifying option urine standardisation.
Options 3 (Carrieri et al. (2000)) and 4 and 5 (Busgang et al. (2023)) are for specific subgroups in the population, e.g. adults (18 - 68 years) and children (6 - 14 years), respectively. For model 2 of Busgang et al. (2023), age and gender should be defined in the data, otherwise the sample concentration is set to missing.
The normalisation and standardisation methods are only available when the data contains creatinine or specific gravity values.
Combine urine from different sampling methods
Different urine sampling methods (morning, spot, or 24-hour) are considered a single biological matrix for further analysis in MCRA. If the selected human biomonitoring data contains concentrations from different urine sampling methods, the concentration values are combined by taking the average of the concentration values from the various urine sampling methods.
Filter for complete cases
After imputation of non-detects and missing values, standardisation and normalisation some individual day concentration records still contain missing values. The incomplete cases are removed from the dataset before the analysis continues.
Occasionally, removing all records with missing values results in empty datasets. Then a warning will be thrown ‘All HBM individual day records were removed for having non-imputed missing substance concentrations’. To circumvent this warning, inspect your data and remove substances with too many missing values, lower the minimum percentage of non-missing values (Impute from data), or Impute with zero.
Filtering for complete cases is always applied before conversion. In general, filtering has more impact when many substances are in scope, resulting in smaller subsets of complete cases. It is good practice to set the scope to the relevant substances. In the interface of the primary entity substances, select the relevant substances for the assessment.
Apply exposure biomarker conversion of substance concentrations
Occasionally, the biomarkers of interest are not measured. Exposure biomarker conversion is used to convert measured biomarkers to the biomarker of interest and is applied within a biological matrix. Biomarker conversion is performed after imputation of censored and missing values and after normalisation/standardisation of urine/blood samples. For conversion factors that are dependent on age and/or gender, check option Use exposure biomarker conversion factors subgroup in the interface of the exposure biomarker conversion module.
Apply kinetic conversion of substance concentrations from other biological matrices
Kinetic conversion of substances from other biological matrices is used when the number of substances in the matrix of interest is limited or for reverse dosimetry when exposures at external level are required. As an example of an internal-to-internal conversion, consider a scenario where five substances are measured in urine (spot) samples from individuals. Additionally, blood (serum) concentrations are analysed for the same individuals, resulting in another set of concentrations for, let’s say, five different substances. The five substances measured in blood are then converted to urine matrix concentrations by checking the Apply kinetic conversions option. Then, the analysis continues with ten substances in urine (spot). Selecting specific substances for conversion is possible through multiplication factors defined in the kinetic conversion factors data of the Kinetic models module.
Enable the Convert to single exposure surface (biological matrix or external route) matrices option to compute exposures at a single target biological matrix or to compute exposures at the external level. Specify the target level (Internal or External) and biological matrix accordingly. Kinetic conversion of substance concentrations is performed after imputation of censored and missing values, and after normalisation/standardisation of urine/blood samples.
Note that kinetic conversion is applied only to substances that are not measured in the target matrix. For instance, kinetic conversion cannot be used for imputing missing values of substances that have one or more missing values.
For conversion factors that are dependent on age and/or gender, check option Use kinetic conversion factors subgroup in the interface of the kinetic models module.
Calculation of acute human monitoring concentrations
For acute assessments, the monitoring concentrations are computed for each substance and biological matrix as average individual-day concentrations. The computation is done after imputation of censored and missing values, eventually followed by a conversion of biomarkers from other biological matrices. For a given substance and biological matrix, the acute individual-day concentration \(c_{ij}\) for individual \(i\) on day \(j\) is:
where \(n_{\mathtt{samples}}\) is the number of samples available for individual \(i\) on day \(j\), and \(c_{ijk}\) the concentration of the \(k\)-th sample of the individual day \(j\).
After urine normalisation for specific gravity:
where \(\mathit{sg}\) denotes the specific gravity correction factor for that individual day.
After standardisation for blood lipid content:
where \(\mathit{c'_{ij}}\) denotes the lipid concentration per \({ g \, \small \mathtt{lipid}}\). The standardisation is only performed for lipid soluble substances. After standardisation the concentration of the substance is expressed as substance amount, with a user specified unit, per \({ g \, \small \mathtt{lipid}}\).
The standardisation for creatinine is similar to the above equation replacing \({\small \mathtt{lipid}}\) by \({\small \mathtt{creatinine}}\).
Calculation of chronic human monitoring concentrations
For chronic assessments, the monitoring concentrations are computed as the average monitoring concentrations of multiple individual-days for each substance and biological matrix. The computation is done after imputation of censored and missing values, eventually followed by a conversion of biomarkers from other biological matrices. The chronic concentration \(c_{i}\) for individual \(i\) is computed as:
where \(n_{\mathtt{days}}\) is the number of days that individual \(i\) was monitored (after removing missing individual days), and \(c_{ij}\) denotes the average monitoring concentration of individual \(i\) on day \(j\).
Standardisation and normalisation of blood and urine samples, respectively, are similar to the expressions for the calculation of individual day concentrations (acute).
For co-exposure of substances, see maximum cumulative ratio (MCR) and the exposure mixtures module.