Models for nearly every occasion: Part III - One box decreasing emission models.

New one box "well-mixed room" decreasing emission (DE) models are introduced that allow for local exhaust or local exhaust with filtered return, as well the recirculation of a filtered (or cleaned) portion of the general room ventilation. For each control device scenario, a steady state and transient model is presented. The transient equations predict the concentration at any time t after the application of a known mass of a volatile substance to a surface, and can be used to predict the task exposure profile, the average task exposure, as well as peak and short-term exposures. The steady state equations can be used to predict the "average concentration per application" that is reached whenever the substance is repeatedly applied. Whenever the beginning and end concentrations are expected to be zero (or near zero) the steady state equations can also be used to predict the average concentration for a single task with multiple applications during the task, or even a series of such tasks. The transient equations should be used whenever these criteria cannot be met. A structured calibration procedure is proposed that utilizes a mass balance approach. Depending upon the DE model selected, one or more calibration measurements are collected. Using rearranged versions of the steady state equations, estimates of the model variables-e.g., the mass of the substance applied during each application, local exhaust capture efficiency, and the various cleaning or filtration efficiencies-can be calculated. A new procedure is proposed for estimating the emission rate constant.

Models for nearly every occasion: Part IV - Two-box decreasing emission models.

New two-box "well-mixed room" decreasing emission (DE) models are introduced for scenarios that involve local controls, such as some form of local exhaust or local exhaust with filtered return. In addition, these models allow for the recirculation of a filtered (or cleaned) portion of the general room ventilation. For each control device scenario, a steady state and transient near and far field model is presented. The transient equations predict the concentration at time t after the application of the substance. The steady state equations can be use to predict the steady, unvarying "average concentration per application" whenever there are continuous applications of a substance and sufficient time has elapsed. The steady state equations can also be used to calculate the TWA for a task (or a series of tasks) whenever the beginning and end concentrations for the task (or task series) are expected to be zero (or near zero). The transient equations should be used to predict TWA exposures whenever these criteria cannot be met, or it is necessary to predict short-term exposures or peak concentrations. A structured calibration procedure, based on a mass balance approach, is proposed for each model. Depending upon the model, one or more calibration measurements are collected. Rearranged versions of the steady state equations are used to calculate estimates of the mass applied during each application, the near field flowrate, and (depending upon the model) the various efficiencies (e.g., local exhaust capture efficiency and the recirculation filtration efficiency). The emission rate constant must be determined using either a published approximation algorithm or experimentally.

Models for nearly every occasion: Part I - One box models.

The standard "well mixed room," "one box" model cannot be used to predict occupational exposures whenever the scenario involves the use of local controls. New "constant emission" one box models are proposed that permit either local exhaust or local exhaust with filtered return, coupled with general room ventilation or the recirculation of a portion of the general room exhaust. New "two box" models are presented in Part II of this series. Both steady state and transient models were developed. The steady state equation for each model, including the standard one box steady state model, is augmented with an additional factor reflecting the fraction of time the substance was generated during each task. This addition allows the easy calculation of the average exposure for cyclic and irregular emission patterns, provided the starting and ending concentrations are zero or near zero, or the cumulative time across all tasks is long (e.g., several tasks to a full shift). The new models introduce additional variables, such as the efficiency of the local exhaust to immediately capture freshly generated contaminant and the filtration efficiency whenever filtered exhaust is returned to the workspace. Many of the model variables are knowable (e.g., room volume and ventilation rate). A structured procedure for calibrating a model to a work scenario is introduced that can be applied to both continuous and cyclic processes. The "calibration" procedure generates estimates of the generation rate and all of remaining unknown model variables.

Models for nearly every occasion: Part II - Two box models.

The "well-mixed room" two box models are often used to predict near and far field concentrations for a specific task, but are limited to scenarios where local exhaust controls are not used. In Part II of this series, new two box models are presented that permit local controls that either exhaust to the outside or return filtered air to the workspace. Additional models are presented that also allow for the recirculation of a filtered portion of the general ventilation flowrate. During the concentration increase phase the emission rate is assumed to be relatively constant. Both steady state and transient equations were developed for each scenario. An additional factor, representing the fraction of time that the substance is being emitted during a task or set of tasks, was added to the standard and new steady state models. This simple modification permits the easy calculation of the average near and far field concentrations for cyclic and irregular emission patterns, provided the starting and ending concentrations are identical (e.g., zero or near zero) or the cumulative task time is long (e.g., two or more task cycles to a full shift, depending upon the number of room air changes per task). Additional variables are introduced with the new models, such as the efficiency of a local control to immediately capture freshly generated contaminant and the filtration efficiency whenever filtered exhaust is returned to the workspace. Many of the model variables are knowable (e.g., room volume and ventilation rate). Others can be approximated using manufacturer specifications or published values (e.g., filtration efficiency). A structured procedure for calibrating a model to a work scenario is presented that can be applied to both continuous and cyclic processes. The "calibration" procedure generates estimates of all of the unknown model variables, including the generation rate and the effective near field flowrate (which takes into account potentially complex near field air currents as well as any thermal plumes created by a hot process).

Correcting diacetyl concentrations from air samples collected with NIOSH method 2557.

Diacetyl (2,3-butanedione), a diketone chemical used to impart a buttery taste in many flavoring mixtures, has been associated with bronchiolitis obliterans in several industrial settings. For workplace evaluations in 2000-2006, National Institute for Occupational Safety and Health (NIOSH) investigators used NIOSH Method 2557, a sampling and analytical method for airborne diacetyl utilizing carbon molecular sieve sorbent tubes. The method was subsequently suspected to progressively underestimate diacetyl concentrations with increasing sampling site humidity. Since underestimation of worker exposure may lead to overestimation of respiratory health risk in quantitative exposure-effect analyses, correction of the diacetyl concentrations previously reported with Method 2557 is essential. We studied the effects of humidity and sample storage duration on recovery of diacetyl from experimental air samples taken from a dynamically generated controlled test atmosphere that allowed control of diacetyl concentration, temperature, relative humidity, sampling duration, and sampling flow rate. Samples were analyzed with Method 2557, and results were compared with theoretical test atmosphere diacetyl concentration. After fitting nonlinear models to the experimental data, we found that absolute humidity, diacetyl concentration, and days of sample storage prior to extraction affected diacetyl recovery as did sampling flow rate to a much smaller extent. We derived a mathematical correction procedure to more accurately estimate historical workplace diacetyl concentration based on laboratory-reported concentrations of diacetyl using Method 2557, and sample site temperature and relative humidity (to calculate absolute humidity), as well as days of sample storage prior to extraction in the laboratory. With this correction procedure, quantitative risk assessment for diacetyl can proceed using corrected exposure levels for air samples previously collected and analyzed using NIOSH Method 2557 for airborne diacetyl.

An accurate substitution method for analyzing censored data.

When analyzing censored datasets, where one or more measurements are below the limit of detection (LOD), the maximum likelihood estimation (MLE) method is often considered the gold standard for estimating the GM and GSD of the underlying exposure profile. A new and relatively simple substitution method, called beta-substitution, is presented and compared with the MLE method and the common substitution methods (LOD/2 and LOD/square root(2) substitution) when analyzing a left-censored dataset with either single or multiple censoring points. A computer program was used to generate censored exposure datasets for various combinations of true geometric standard deviation (1.2 to 4), percent censoring (1% to 50%), and sample size (5 to 19 and 20 to 100). Each method was used to estimate four parameters of the lognormal distribution: (1) the geometric mean, GM; (2) geometric standard deviation, GSD; (3) 95th percentile, and (4) Mean for the censored datasets. When estimating the GM and GSD, the bias and root mean square error (rMSE) for the beta-substitution method closely matched those for the MLE method, differing by only a small amount, which decreased with increasing sample size. When estimating the Mean and 95th percentile the beta-substitution method bias results closely matched or bettered those for the MLE method. In addition, the overall imprecision, as indicated by the rMSE, was similar to that of the MLE method when estimating the GM, GSD, 95th percentile, and Mean. The bias for the common substitution methods was highly variable, depending strongly on the range of GSD values. The beta-substitution method produced results comparable to the MLE method and is considerably easier to calculate, making it an attractive alternative. In terms of bias it is clearly superior to the commonly used LOD/2 and LOD/square root(2) substitution methods. The rMSE results for the two substitution methods were often comparable to rMSE results for the MLE method, but the substitution methods were often considerably biased.

A comparison of several methods for analyzing censored data.

The purpose of this study was to compare the performance of several methods for statistically analyzing censored datasets [i.e. datasets that contain measurements that are less than the field limit-of-detection (LOD)] when estimating the 95th percentile and the mean of right-skewed occupational exposure data. The methods examined were several variations on the maximum likelihood estimation (MLE) and log-probit regression (LPR) methods, the common substitution methods, several non-parametric (NP) quantile methods for the 95th percentile and the NP Kaplan-Meier (KM) method. Each method was challenged with computer-generated censored datasets for a variety of plausible scenarios where the following factors were allowed to vary randomly within fairly wide ranges: the true geometric standard deviation, the censoring point or LOD and the sample size. This was repeated for both a single-laboratory scenario (i.e. single LOD) and a multiple-laboratory scenario (i.e. three LODs) as well as a single lognormal distribution scenario and a contaminated lognormal distribution scenario. Each method was used to estimate the 95th percentile and mean for the censored datasets (the NP quantile methods estimated only the 95th percentile). For each scenario, the method bias and overall imprecision (as indicated by the root mean square error or rMSE) were calculated for the 95th percentile and mean. No single method was unequivocally superior across all scenarios, although nearly all of the methods excelled in one or more scenarios. Overall, only the MLE- and LPR-based methods performed well across all scenarios, with the robust versions generally showing less bias than the standard versions when challenged with a contaminated lognormal distribution and multiple LODs. All of the MLE- and LPR-based methods were remarkably robust to departures from the lognormal assumption, nearly always having lower rMSE values than the NP methods for the exposure scenarios postulated. In general, the MLE methods tended to have smaller rMSE values than the LPR methods, particularly for the small sample size scenarios. The substitution methods tended to be strongly biased, but in some scenarios had the smaller rMSE values, especially for sample sizes <20. Surprisingly, the various NP methods were not as robust as expected, performing poorly in the contaminated distribution scenarios for both the 95th percentile and the mean. In conclusion, when using the rMSE rather than bias as the preferred comparison metric, the standard MLE method consistently outperformed the so-called robust variations of the MLE-based and LPR-based methods, as well as the various NP methods, for both the 95th percentile and the mean. When estimating the mean, the standard LPR method tended to outperform the robust LPR-based methods. Whenever bias is the main consideration, the robust MLE-based methods should be considered. The KM method, currently hailed by some as the preferred method for estimating the mean when the lognormal distribution assumption is questioned, did not perform well for either the 95th percentile or mean and is not recommended.

Sympathetic sudomotor disturbance in early type 1 diabetes mellitus is linked to lipid peroxidation.

The present study was performed to determine whether increased lipid peroxidation, as assessed from malondialdehyde (MDA) excretion, is associated with deterioration in peripheral nerve function in early type 1 diabetes mellitus. These parameters were measured annually for 3 years in 36 patients who entered the study less than 2 years after the diagnosis of diabetes. Malondialdehyde excretion was 1.51 +/- 0.20 micromol/g creatinine in the controls, and 2.43 +/- 0.21, 2.39 +/- 0.22, and 1.93 +/- 0.21 micromol/g creatinine at the first, second, and third evaluations, respectively (P < .005). The increased MDA was seen only in the female participants. Malondialdehyde excretion was increased in those with high vs low hemoglobin Alc across all years (P < .05). Malondialdehyde excretion correlated negatively with sudomotor function below the waist. The mean sweat production from the 3 evaluations correlated with mean MDA excretion across all years in the proximal leg (r = -0.42, P < .005) and distal leg (r = -0.40, P < .01). Below the waist, sweating correlated with MDA (r = -0.40, P < .01) as did total sweat (r = -0.38, P < .01). The response amplitudes of the peroneal nerves correlated negatively with MDA excretion (for the mean values at the second 2 evaluations, P < .005, r = -0.45). Tests of sensory function correlated inconsistently with MDA excretion. In summary, lipid peroxidation, as assessed from malondialdehyde excretion, is associated with sudomotor dysfunction in early diabetes.

Nitrosative stress, uric Acid, and peripheral nerve function in early type 1 diabetes.

The present study was performed to determine whether nitric oxide overproduction is associated with deterioration in peripheral nerve function in type 1 diabetes. We measured peripheral nerve function and biochemical indicators of nitrosative stress annually for 3 years in 37 patients with type 1 diabetes. Plasma nitrite and nitrate (collectively NO(x)) were 34.0 +/- 4.9 micro mol/l in the control subjects and 52.4 +/- 5.1, 50.0 +/- 5.1, and 49.0 +/- 5.2 in the diabetic patients at the first, second, and third evaluations, respectively (P < 0.01). Nitrotyrosine (NTY) was 13.3 +/- 2.0 micro mol/l in the control subjects and 26.8 +/- 4.4, 26.1 +/- 4.3, and 32.7 +/- 4.3 in the diabetic patients (P < 0.01). Uric acid was suppressed by 20% in the diabetic patients (P < 0.001). Composite motor nerve conduction velocity for the median, ulnar, and peroneal nerves was decreased in patients with high versus low NTY (mean Z score -0.522 +/- 0.25 versus 0.273 +/- 0.22; P < 0.025). Patients with high NO(x) had decreased sweating, and those with suppressed uric acid had decreased autonomic function. In conclusion, nitrosative stress in early diabetes is associated with suppressed uric acid and deterioration in peripheral nerve function.