Methods for estimating average analyte concentrations when some of the collected samples have concentrations below the limit of detection (LOD) were discussed. Because the concentrations of industrial contaminants are generally much lower than those in the past, the proportion of nondetectable samples in typical industrial hygiene data sets has increased. Two methods commonly used for estimating average concentrations in the presence of nondetectable values include the Hald method and the LOD/2 method. The Hald method involved using knowledge of the normal distribution to extrapolate back from the LOD to yield maximum likelihood estimates of the mean concentration and its standard deviation. The LOD/2 method assumed that samples with concentrations below the LOD could be assigned a value that was half of the LOD. A log transformation was then applied to the data and estimates of the geometric mean concentration and its standard deviation were obtained. The Hald method was very accurate but very complex requiring many laborious calculations. The LOD/2 method was very simple to use. Because it has assumed that data below the LOD follow a uniform distribution it can give incorrect estimates of the mean concentration and standard deviation. A new proposed method was presented. It was based on substituting the LOD divided by the square root of 2 for each concentration that was below the LOD. This followed from the assumption that all values below the LOD could be approximated by a triangle. The three methods were evaluated in a computer simulation. The Hald method was superior to the LOD/2 and proposed method if fewer than half of the concentrations were below the LOD. If the data were not highly skewed and the percentage of undetectable values was less than 30% the proposed method produced less bias than the LOD/2 method. The LOD/2 method produced less bias when the percentage of undetectable values was more than 30% but less than 50%. The authors conclude that the Hald method is best if a high degree of accuracy in the geometric mean and standard deviation are required. When the data are not highly skewed the new method can be used if fewer than 30% of the values are below the LOD. Otherwise the LOD/2 method should be used. None of the methods give good results when more than half of the values are below the LOD.
We take your privacy seriously. You can review and change the way we collect information below.
These cookies allow us to count visits and traffic sources so we can measure and improve the performance of our site. They help us to know which pages are the most and least popular and see how visitors move around the site. All information these cookies collect is aggregated and therefore anonymous. If you do not allow these cookies we will not know when you have visited our site, and will not be able to monitor its performance.
Cookies used to make website functionality more relevant to you. These cookies perform functions like remembering presentation options or choices and, in some cases, delivery of web content that based on self-identified area of interests.
Cookies used to track the effectiveness of CDC public health campaigns through clickthrough data.
Cookies used to enable you to share pages and content that you find interesting on CDC.gov through third party social networking and other websites. These cookies may also be used for advertising purposes by these third parties.