Workers involved in hydraulic fracking processes are exposed to various types of chemicals and dusts in their workplaces, such as proppants, which hold open the fissures created in the fracking process. Recently, ceramic proppants have been developed that may be less hazardous to workers than traditional proppants. Pulmonary function testing of workers producing ceramic proppant was used to assess the potential inhalation hazards of ceramic proppant. 100 male workers from a producer of ceramic proppant were evaluated with pulmonary function test data collected and evaluated using The American Thoracic Society (ATS) acceptability criteria. A comparison group was selected from the Third National Health and Nutrition Examination Survey (NHANES III) spirometry laboratory subset. No pulmonary function deficits were found in the worker group in comparison to the NHANES III population. Mean FEV1 and FVC values in workers were 3.8 and 4.8 liters respectively, and were greater as compared to the NHANES III population of similar demographics. An FEV1/FVC ratio of less than 0.8, when compared to the NHANES III group, produced an odds ratio of 0.44 in worker group, indicating less risk of preclinical pulmonary dysfunction. Overall, exposure to ceramic proppant was not found to produce an adverse impact on pulmonary function in workers engaged in the manufacture of ceramic proppant.
Scientists are continuously looking for alternative resources to meet the global energy demands of the 21st century. Recovering gas and oil from unconventional gas in ways safer for workers and area residents’ alike poses challenges to meet the increasing modern lifestyle need [
Hydraulic fracturing (fracking) technology is a process where both vertical and horizontal drilling are necessary to collect unconventional gas and oil. Proppants are tiny granules which settle in fissures either as a single layer or as a closed pack form to ensure continuous gas or oil collection [
Ceramic and silica based sand are the two primary type of proppants used in the industry. Silica contains respirable crystalline which pose an occupational pulmonary hazard [
Currently, ultra-light weight proppant materials can be either a single material type or a mix of hard-soft materials [
As ceramic proppants contain little to no silica, they may become the preferred choice for health reasons. The purpose of this study is to determine whether exposure to non-silica ceramic proppant in an industrial setting poses a pulmonary health risk to exposed workers. This evaluation was made by comparing the pulmonary function results of workers engaged in the manufacture of ceramic proppant to pulmonary function results from the general population found in the Third National Health and Nutrition Examination Survey (NHANES III) data set.
This study evaluated workers in a ceramic factory with Pulmonary Function Test (PFT) data as a result of evaluating fitness to wear a respirator in the workplace. The study population included male workers ≥18 years old with maximum and minimum heights of ≤76 and ≥61 inches respectively. Demographic data relevant to pulmonary function assessment measured included: age, height, smoking status, and race. The NHANES III pulmonary function subset, restricted by the age and height ranges (and gender) of the worker population, was used for comparison to a normal population. American Thoracic Society (ATS) guidelines were followed for acceptable pulmonary function test and all pulmonary function tests were reviewed by a physician for acceptability [
The research protocol was approved by the University of South Florida Institutional Review Board (IRB) number 00001348.
FEV1(in liters), FVC (in liters), and FEV1/FVC data for the proppant workers were selected from spirometry records provided by the Occupational Health and Support Services (OHSS), a medical care company that specializes in providing medical support personnel for testing and compliance needs.
FVC (Forced Vital Capacity) is the maximum volume of air that can be exhaled forcefully by a person after a maximum inhalation. FEV1 (Forced Expiratory Volume in 1 second) is the amount of air that can be expired forcefully from the lung in one second after maximum inspiration. In this study, percent predicted values were not used to determine normalcy for individuals; rather a direct comparison between the worker population and a control population was made in aggregate by comparing all “best” values in the groups, with the control group being restricted by the age and height ranges, as well as gender, of the worker group.TheFEV1/FVC ratio was evaluated as an indicator of early potential obstructive impairment if the value of the ratio was <80%.
Inclusion criteria for the proppant workers included >18 years and older, and having undergone spirometry testing. Age, gender, race, smoking history and height information were recorded for each subject to identify possible confounders that may have had an effect on pulmonary function.
Koko spirometry was used to measure pulmonary function in the proppant workers. The best value out of 3 attempts from spirometry reading was taken for final analysis. The FEV1 and FVC reading were expressed in liters. Spirometry records for 316 proppant workers were made available for the study through OHSS. After assessing the records for inclusion criteria, agreement with American Thoracic Society (ATS) spirometry criteria and the removal of duplicate entries, 101 subjects remained for statistical analysis: 100 male workers and 1 female worker. The female worker was removed from the dataset because of the limited statistical power for categorization by gender, leaving a final dataset of 100 male workers.
The NHANES III pulmonary function subset comprised the control population, and was same inclusion criteria and personal information were selected as was for the proppant workers, which then additionally restricted by the age and height range found in the worker population and restricted by male gender. Similarly, FEV1, FVC, and FEV1/FVC values were taken from the raw spirometry data publicly available for the NHANES III population. The comparison group included pulmonary function tests for 130,691 people. The file of raw spirometry data was merged with the Adult Household file from the NHANES III record to obtain the same demographic as was available for the proppant workers. The Spirometry tests for the NHANES III population were conducted according to the guidelines from ATS. Participants performed the test (5 to 8 times) according to the spirometry protocol. The final population used was n = 6662.
The student’s t-test was used to compare lung function variables of FEV1 and FVC in the proppant workers group and the NHANES III (
Multivariate linear regression analysis was conducted to estimate which factors are the most predictive of lung function for FEV1 and FVC outcomes. The variables evaluated included: age, height, smoking history (pack years) and race. To evaluate factors that potentially predict lung obstruction, the FEV1/FVC ratio was used as an outcome in logistic regression. The study evaluated the worker population for lung function deficits at the higher end of the normal FEV1/FVC of 0.80 as a conservative screen for preclinical pulmonary obstruction.
The cut off point for p-value for statistical significance was set at <0.05 for all analysis. SAS version 9.4 was used for statistical analysis.
The demographic information for both the proppant workers (study population) and the NHANES III population (comparison group) used for analysis is shown in
The mean results of FEV1 and FVC when compared to the overall proppant workers with the NHANES III population are shown in
Demographic Information | Proppant Workers | NHANES III | |
---|---|---|---|
Total Population | 100 | 6662 | |
Smoking History (Yes) | 31 | 4165 | |
Smoking History (No) | 69 | 2499 | |
Median Height (≥69 inches) | 80 | 3733 | |
Median Height (<69 inches) | 20 | 2929 | |
Median Age (≥39 years) | 46 | 2634 | |
Median Age (<39 years) | 54 | 4028 | |
Race (White) | 85 | 4562 | |
Race (Black) | 15 | 1876 | |
Overall Male Population: Without Stratification | |||
---|---|---|---|
FEV1 (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 3.8 | 3.66 - 3.92 | <0.0001 |
NHANES III | 3.3 | 3.30 - 3.35 | |
FVC (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 4.8 | 4.63 - 4.94 | <0.0001 |
NHANES III | 4.3 | 4.30 - 4.35 |
The multivariate linear regression analysis results are shown in
Logistic regression analysis was conducted to identify the influence of variables on the FEV1/FVC ratio, since the FEV1/FVC ratios <0.8 can be indicative of a preclinical pulmonary obstruction.
Smoking Status: Yes | |||
---|---|---|---|
FEV1 (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 3.5 | 3.26 - 3.82 | 0.01 |
NHANES III | 3.2 | 3.15 - 3.21 | |
FVC (Liters) | Mean Value | 95% CI | P value |
Proppant Workers | 4.6 | 4.28 - 4.87 | 0.03 |
NHANES III | 4.2 | 4.2 - 4.27 | |
Smoking Status: No | |||
FEV1 (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 3.9 | 3.76 - 4.04 | <0.001 |
NHANES III | 3.6 | 3.52 - 3.59 | |
FVC (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 4.9 | 4.69 - 5.06 | <0.001 |
NHANES III | 4.5 | 4.42 - 4.50 | |
MEDIAN HEIGHT ≥69 inches | |||
FEV1 (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 3.8 | 3.67 - 3.97 | <0.0001 |
NHANES III | 3.5 | 3.47 - 3.53 | |
FVC (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 4.9 | 4.69 - 5.06 | <0.02 |
NHANES III | 4.6 | 4.55 - 4.61 | |
MEDIAN HEIGHT <69 inches | |||
FEV1 (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 3.7 | 3.37 - 3.95 | 0.0006 |
NHANES III | 3.1 | 3.06 - 3.13 | |
FVC (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 4.5 | 4.11 - 4.80 | 0.027 |
NHANES III | 4.0 | 3.96 - 4.02 | |
Median Age ≥ 39 Years | |||
FEV1 (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 4.0 | 3.9 - 4.16 | 0.372 |
NHANES III | 4.0 | 3.94 - 3.99 | |
FVC (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 5.1 | 4.9 - 5.2 | 0.017 |
NHANES III | 4.9 | 4.88 - 4.9 | |
Median Age < 39 Years |
FEV1 (Liters) | Mean Value | 95% CI | P Value |
---|---|---|---|
Proppant Workers | 3.6 | 3.38 - 3.79 | <0.0001 |
NHANES III | 2.9 | 2.87 - 2.93 | |
FVC (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 4.6 | 4.30 - 4.80 | <0.0001 |
NHANES III | 4.0 | 3.93 - 3.99 | |
Race: White | |||
FEV1 (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 3.8 | 3.66 - 3.98 | <0.0001 |
NHANES III | 3.4 | 3.36 - 3.41 | |
FVC (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 4.8 | 4.67 - 5.02 | <0.0001 |
NHANES III | 4.4 | 4.41 - 4.48 | |
Race: Black | |||
FEV1 (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 3.3 | 3.12 - 3.20 | 0.1232 |
NHANES III | 3.2 | 3.10 - 3.58 | |
FVC (Liters) | Mean Value | 95% CI | P Value |
Proppant Workers | 4.0 | 3.66 - 4.40 | 0.9489 |
NHANES III | 4.0 | 3.97 - 4.06 |
FEV1 (Liters) | |||
---|---|---|---|
Variable | Parameter Estimates | Standard Error | P-Value |
Median Age (39 Years) | −0.98 | 0.0005 | <0.0001 |
Race (Black vs White) | −0.31 | 0.02 | <0.0001 |
Smoking History (pk-yrs) | −0.31 | 0.0006 | <0.0001 |
Median Height (69 Inches) | 0.41 | 0.02 | <0.0001 |
Proppant Workers vs NHANES III | 0.11 | 0.07 | 0.13 |
FVC (Liters) | |||
---|---|---|---|
Variable | Parameter Estimates | Standard Error | P-Value |
Median Age (39 Years) | −0.88 | 0.02 | <0.0001 |
Race (Black vs White) | −0.45 | 0.02 | <0.0001 |
Smoking History (pk-yrs) | −0.003 | 0.009 | <0.0001 |
Median Height (69 Inches) | 0.56 | 0.02 | <0.0001 |
Proppant Workers vs NHANES III | 0.08 | 0.08 | 0.32 |
Effect | Odds Ratio | 95% CI |
---|---|---|
Median Age (≥39 Years) | 3.61 | 3.19 - 4.07 |
Race (Black vs White) | 0.76 | 0.69 - 0.95 |
Smoking History (pk-yrs) | 1.03 | 1.03 - 1.04 |
Median Height (≥69 inches) | 1.12 | 0.99 - 1.26 |
Proppant Workers vs NHANES III | 0.44 | 0.28 - 0.67 |
Studies conducted on workers’ pulmonary function using spirometry measurements of FEV1, FVC, and the FEV1/FVC ratio have observed significant differences for workers exposed to harmful substances [
FEV1 analysis in our study showed that median age (years), race (black vs. white), and smoking history (in pack-years) affected the pulmonary function of proppant workers. Each unit increase in median age, race, and smoking history decreased FEV1; each unit increased in height above 69 inches (median height) increased FEV1. The effect of race on FEV1 was found to be statistically significant, with black study population showing a lower mean FEV1 value than the mean FEV1 value of the white study population. Thus the black population had higher risk for lower pulmonary function volumes than the white population workers, though these changes may not be of clinical significance. This finding is also supported by the previous studies conducted on lung function tests in African American population [
Overall, the mean FEV1 in the proppant workers was found to be more than the NHANES III sample signifying that the worker population, as a group, had better lung function values than the comparable NHANES III population. However, this association was not statistically significant (P value > 0.05). Similar observations were obtained from multiple linear regression analysis using FVC. Each unit increase in median age, race and smoking history demonstrated decreases in FVC. Increases in median height showed increased results for FVC. Each yearly increase in age and smoking was significantly associated with a lowered FVC. The black study population were found to have lower FVC than the white study population, which was statistically significant. As with FEV1, each inch increase in height was significantly associated with increase in FVC. The mean FVC in the proppant workers was greater than the mean of the NHANES III sample. However, this difference was not statistically significant (P value > 0.05). The proppant workers did not show any statistical association for FEV1 and FVC. The analysis of FEV1 as well as FVC validates that the variables such as height, age, smoking history and race that are known to be associated with pulmonary functions measures do, in fact, impact these measurements.
The logistic regression analysis demonstrated a protective effect regarding the prediction of preclinical obstructive disorders in proppant workers (OR = 0.44) using an FEV1/FVC ratio <0.8 when compared with the NHANES III. The logistic regression analysis demonstrated that those aged 39 years or older had 3 times greater risk in developing obstructive type of lung impairment (FEV1/FVC ratio >0.8). Previous studies have also established that lung function tests decrease in older populations [
While this study was conducted among workers engaged in the manufacture of ceramic fracking proppant, and not those using proppant to frack oil and gas wells, the results indicate that the use of ceramic proppant may provide a safer alternative to high silica containing sand proppants. Our analysis found that exposure to proppant in an adequately managed environment did not lead to loss of pulmonary function in workers compared to that of a control population.
The principal limitation of this study was the comparatively small sample set of proppant worker data that was available for analysis. Likewise, more data on different forms of ceramic proppant produced and the specific durations of exposure would have allowed greater effect analysis. An additional limitation was that the sample set was exclusively male, preventing generalization to the females working in the hydraulic fracking procedure.
This research evaluated the respiratory health of a cross section of proppant workers from a proppant manufacturing facility. The research demonstrated that the worker group did not experience pulmonary impairment as compared to the general population in the US Ceramic proppant may be a safer alternative to high silica sand proppant for the oil and gas fracking industry.
Special thanks to Dr. Kate Wolfe-Quintero, Associate professor, Director of Graduate Studies, College of Public Health, University of South Florida, for her valuable time and guidance in preparation of this manuscript.
Humairat H. Rahman,Giffe T. Johnson,Raymond D. Harbison, (2016) Occupational Health Surveillance: Pulmonary Function Test in Proppant Exposures. Occupational Diseases and Environmental Medicine,04,37-45. doi: 10.4236/odem.2016.42005