Work activities and respiratory and physical health outcomes among industrial hog operation workers: community-driven data collection to inform appropriate protective actions

Embargo until
Journal Title
Journal ISSN
Volume Title
Johns Hopkins University
Airborne dusts, noxious gases, endotoxins, and microbial exposures from industrial hog operations (IHOs) have been shown to adversely affect IHO workers’ and neighbors’ health and quality of life. For IHO worker respiratory disease in particular, knowledge gaps remain about the temporal dynamics of exposure-response relations and specific factors that might mitigate exposure and disease burdens. Prior studies to estimate respiratory health effects have focused mainly on comparisons of IHO worker to non-worker populations or pre- vs. post-IHO work shift measurements. This dissertation aims to advance the understanding of temporal relationships between self-reported IHO work exposure activities and respiratory and physical health outcomes, including those activities which may be protective. It uses data from an IHO worker cohort with exposures and outcomes measured in two-week intervals over the course of 16 weeks (4 months) of follow-up. It also employs an underutilized fixed-effects regression method for repeated-measures data, which eliminates confounding from measured and unmeasured time-invariant factors as each IHO worker’s time-varying exposure and outcome measures are compared to his/her mean exposure and outcome. Further, the dissertation contributes to lessons learned from a community-driven pilot study designed to investigate evidence of swine-specific fecal contamination of household surfaces at residences proximal to IHOs and the nares of those living in the home. For Aim 1, an analysis was conducted of baseline (n=103 individuals) and longitudinal (n=101 individuals with 782 biweekly person-records) associations of self-reported IHO work exposure activities and self-reported respiratory and physical health outcomes within the four-month-long IHO worker cohort. At baseline, longer time employed on any IHO and workplace activities that constituted increased frequency of contact with pigs, dustiness, and frequency of conducting cleaning activities were associated with increased odds of a variety of respiratory and physical health symptoms. Among 39 exposure-response associations examined, five odds ratios (ORs) and 95% confidence intervals (CIs) were above and not overlapping the null (OR range: 2.19 to 32). In longitudinal fixed effects analyses, lower odds of symptoms were seen with increased frequency of personal protective equipment (PPE) use. Among 15 exposure-response associations examined, three ORs (95% CIs) were below and not overlapping the null (OR range: 0.04 to 0.09). Handwashing frequency at or above vs. below the median of eight handwashes per shift was associated with 0.32 (95% CI: 0.12, 0.83) times the odds of reporting at least one respiratory symptom in the past week. For Aim 2, a baseline and longitudinal analysis of the four-month-long IHO worker cohort was conducted to assess the relationships between work activities and lung function (as measured by spirometry). At baseline, time spent on any IHO (average work week of seven days vs. less than seven days, years worked on any IHO, and percentage of life working on any IHO) was associated with decreases in predicted lung function (9 out of 30 ß estimates were negative in direction and did not overlap the null [ß range: -12.35 to -4.32]). Over time, increased PPE utilization was associated with decreases in lung function, opposite from the hypothesized direction of association. In further explorations of the relationship between PPE use and lung function, it was observed that as IHO barn and work conditions worsened IHO workers were more likely to wear protective coveralls but less likely to wear protective face gear (mask or eye protection). This suggests that IHO workers may not be donning the appropriate PPE to protect their respiratory system, particularly during working conditions when it is most warranted. For Aim 3, a community-driven citizen science pilot study was conducted to assess ambient airborne particulate matter (PM) and bacterial loads (including the swine-specific fecal bacterial source tracking marker Pig-2-Bac) on surfaces of the homes and within the human nares of adults and children (<7 years of age) at: 1) IHO worker households; and (2) community resident comparator households (with no known livestock exposure). For the purposes of this dissertation, lessons learned are presented from working with community partners to: (1) develop and finalize the research questions and study design for the pilot study; (2) train community organizers in pilot study data collection; and (3) evaluate the quality and completeness of data collected by community organizers who used mobile-devices (tablets) to complete participant interviews and collected household surface (inside and outside) and human nasal swab samples that were sent to be tested for bacterial contamination. Among 26 industrial hog operation (IHO) worker and 23 community resident households enrolled, 20 (n=41 individuals) and 18 (n=36 individuals), respectively, met eligibility criteria (at least one child <7 years of age living in the household) for inclusion in the pilot study. Of the surveys administered to participants, 2.4% of questions (463 of 18,932) were complete – with no statistically significant differences between those eligible vs. ineligible or IHO household vs. community referent. The average number of questions missing was 11 per survey and reduced to only 4 after weekly data evaluations were implemented in week 10. An identified challenge was failure of conditional programming of the tablet software during study eligibility determination in the field (data entry screens did not trigger a stop an enrollment alert for children seven years of age and older) which led to enrollment of 11 households (eight IHO and three community resident) that were later determined to be ineligible upon further data review. Though the sample size was small, responses among n=18 enrolled IHO workers indicated that their employer provided them with N95 respirator facemasks and some training in their use. Lessons learned include a need to reconcile advantages of novel mobile device data collection technologies and software with the practical demands of timely tracking, verification, and checking of system performance during and after field data collection, which may exceed the practical capacity of some community-driven research partners. Future studies should focus on development and implementation of processes to make mobile device data collection tools and protocols to be understandable, feasible, accessible, reliable, repeatable, and accurate. This dissertation research was innovative in three key areas: (1) the ability, through long-standing community-driven research partnerships, to engage with hard-to-reach livestock workers who identify predominantly as a race/ethnicity that is under-represented in IHO worker studies, and lack traditional health and safety supports; (2) the contribution to the understanding of time-varying exposure-response dynamics of IHO work exposures and activities; and (3) the lessons learned from a novel community-driven pilot study designed to assess microbial loadings on household environmental surfaces and the nares of those living proximal to IHOs. In conclusion, IHO exposure and work activities were associated with adverse respiratory and physical health outcomes among IHO workers. It was shown that the use of PPE can decrease these risks, but not eliminate them. Lessons learned related to data quality and completeness may improve future community-driven citizen science data collection efforts. Further research is needed to understand the complex exposures from IHOs to protect worker and community populations that surround these operations.
Swine, CAFO, worker, community health, respiratory, PPE, North Carolina, environmental justice