Key Concepts About the Limit of Detection (LOD) of Environmental Chemicals
LOD definition and considerations
The National Report on Human Exposure to Environmental Chemicals describes approaches to consider regarding the analysis of chemical exposure concentrations with LODs. In summary:
- If the proportion of results below the LOD is greater than 40% then geometric means should not be calculated;
- Due to laboratory improvements, LOD values may change over time. When this occurs, acknowledgement of this change and the maximum LOD should be reported. The maximum LOD is the highest LOD among the individual test results analyzed; and
- If an environmental chemical has multiple LODs according to the analytic run, further care is needed when estimating geometric means or percentiles. Further discussion on this issue is available in the National Report on Human Exposure to Environmental Chemicals.
Reference: Fourth National Report on Human Exposure to Environmental Chemicals, 2009. Department of Health and Human Services Centers for Disease Control and Prevention.
Indicator Variable Included in Dataset
Many, but not all of the environmental chemical data files contain a summary indicator variable to identify which individual results at or below the Limit of Detection of the assay. There are three possible values for this indicator variable: “0”, “1”, or “2”. The value “0” means that the result was at or above the Limit of Detection; “1” indicates that the result was below the Limit of Detection; and “2” means the result was above the upper measurement limit and exceeds the calibrated range of the laboratory assay. The table below shows these codes for the Blood 1,4-Dichlorobenzene Indicator variable in the 2003-2004 Volatile Organic Compounds dataset.
Blood 1,4-Dichlorobenzene Indicator
| Code or Value | Description | Count | Cumulative |
|---|---|---|---|
| 0 | Detectable result | 714 | 714 |
| 1 | Below detectable limit | 606 | 1320 |
| 2 | Detectable result and exceeds the calibrated range of assay | 2 | 1322 |
| . | Missing | 167,234 | 1489 |
Be aware that some chemicals may have multiple limits of detection on the same data file (due to changes in the LOD during an NHANES two year cycle). In this situation, all records below the limit of detection will have an indicator value “1” to indicate that the results are below the limit of detection, however, the LOD values used and recorded at the time of the laboratory analysis were not the same. Two different LODs for total mercury in blood (2005–2006) are shown in the table below. Note that in the table the actual Limit of Detection values have been converted to “fill values” (the detection limit divided by the square root of 2).
Mercury, total (ug/L)
| Code or Value | Description | Count | Cumulative |
|---|---|---|---|
| 0.2 to 33.2 | Range of Values | 6267 | 6267 |
| 0.14 | First Below Detection Limit Fill Value | 33 | 6300 |
| 0.23 | Second Below Detection Limit Fill Value | 2107 | 8407 |
| . | Missing | 1033 | 9440 |
How the LODs of environmental chemicals for the laboratory method are determined
Methods used by the laboratory to determine the LOD can vary by the chemical, as the analytic laboratory method may be improved or become more sensitive so that the original LOD is lower than it was when it was first measured in NHANES. It is useful to review this information in the laboratory methods documentation for each dataset over the years that the analyte of interest is included in NHANES. Below are examples of approaches described in the laboratory methods documentation on how some of the LODs were established.
The LOD for:
- Iodine and mercury in urine specimens is based on three times the standard deviation of approximately 20 or more low-concentration samples, each analyzed in a separate run. The LOD calculation is reevaluated at least once a year.
- Arsenic species in urine specimens is based on three times the concentration standard deviation of approximately 20 or more low-concentration samples, each analyzed in a separate run. The LOD calculation is reevaluated at least twice a year.
- Mercury in blood is based on three times the SD of 10 repeat measurements of a sample with low mercury concentration. It is 0.137 μg/L in total mercury measurement and 0.446 μg/L in inorganic mercury measurement. Results below the detection limits are reported as nondetectable.
Approaches to consider when the LOD changes during the period of study
The most conservative approach to use when analyzing NHANES environmental chemical data with a Limit of Detection (LOD) fill value that has changed across surveys is to use the highest LOD fill value. For example in 1999–2000, Mono-(2-ethyl)-hexyl phthalate, (URXMHP) had an LOD fill value of 0.8 ng/mL and in 2001-2002 the fill value was 0.7 ng/mL. ;To use the most conservative approach an analyst can change the LOD fill value to be less than or equal to 0.8 ng/mL for the four survey years (1999–2002). Therefore, for 2001-2002 all values between 0.7 ng/mL and 0.8 ng/mL would be considered below the LOD.
While alternate approaches to handle this situation may be proposed in the literature, there is not one correct approach for analyzing chemical analytes with LOD fill values that change across surveys. The analysis approach will depend on the research question. An alternative approach is described in the next task; you could use the LOD variable provided for each survey cycle as described.
