In Bangladesh, endemic circulation of influenza A(H5N1) among poultry frames concern for human health, particularly given the potential for aerosolized virus generation during chicken processing in live bird markets.
The study under review investigates whether straightforward modifications to common processing equipment could meaningfully reduce respirable aerosol production, as proxied by PM 2.5 mass concentration, during two critical processing steps: slaughtering (exsanguination) and defeathering (mechanical processing).
The investigation was conducted in a controlled booth environment that modeled routine LBMs practices and sought context-relevant, implementable interventions rather than abstract theoretical mitigations.
Study design and setting: structural appraisal of aerosol generation and mitigation
In addition, small and large conical devices served as alternative containment forms.
- Design orientation:
- The research adopts an experimental, quasi-operational design to quantify respirable particulate matter (PM 2.5) generated during two processing steps and to test substitutions or coverings intended to reduce aerosol exposure.
- Measurements rely on PM 2.5 mass concentration as a proxy for respirable aerosol burden, invoking its relevance to potential aerosolized pathogen dissemination in processing environments.
- Environment and replication:
- Experiments were performed January–March 2020 in a facility associated with the National Reference Laboratory for Avian Influenza at the Bangladesh Livestock Research Institute (BLRI) in Savar, Dhaka.
- The processing tasks mirrored practices common in LBMs, using broiler chickens of typical market weight and appearance to reflect real-world conditions.
- Participants and instrumentation:
- PM 2.5 data were captured with Partnered Particle and Temperature Sensor Plus (PATS+) monitors positioned at breathing height and at predefined azimuths relative to the processing apparatus.
- Three workers with experience in slaughtering and defeathering were recruited to perform the experiments; they were trained in procedures and PPE use, and they rotated roles to minimize bias and variability.
- Procedures and workflow:
- Slaughtering component:
- Two core apparatus configurations were tested: open barrels without lids and barrels covered by either a solid lid or a star-cut lid.
- Slaughtering occurred under three major experimental conditions: single-chicken slaughter and multiple-chicken slaughter, to capture variability across exposure scenarios.
- After each slaughter event, animals were returned to their containment (barrel or cone), and the booth was refreshed with air purifiers and fans for a defined interval before the next trial.
- Baseline aerosol levels were established prior to each run, with measurements collected for five minutes thereafter.
- Defeathering component:
- Defeathering experiments tested five lid configurations for the defeathering machine: no lid (open), a partially lid-covered arrangement with a hole, a fully lid-covered option with a hole and pivot door, a lid designed to be fully covered with a solid top, and a lid that allows pouring water while maintaining some lid control.
- The defeathering sequence involved two workers carrying a chicken inside the booth, with pre-processing steps including hot-water treatment and subsequent handover to the defeathering machine, followed by a short operational run while lid configurations were altered as per protocol.
- Like slaughtering, each defeathering run began with a baseline measurement and was followed by a standardized water-pouring process (to simulate routine defeathering waterways) before concluding with a rest period and air refresh.
- Outcome measure and analysis plan:
- The primary outcome was the mean PM 2.5 mass concentration (µg/m3) measured during the processing interval for each intervention type.
- Analytic approach included:
- Calculation of mean and standard deviation of PM 2.5 across all replicates for each intervention type.
- Comparisons against the open-barrel and open-defeather reference conditions using two-sample t-tests to determine relative reductions in PM 2.5.
- Quantification of percentage differences with 95% confidence intervals, and inference adjustments to account for multiple hypothesis testing via Bonferroni correction (adjusted significance level around 1.25% per comparison).
- Mixed-method integration with a qualitative component: in-depth interviews with three workers to elicit practical perspectives on feasibility, durability, and adoption likelihood of the tested methods.
- Sample size rationale:
- Based on prior pilot data for slaughtering and defeathering PM 2.5 levels, estimates were used to power the study:
- Slaughtering: a minimum difference of 50% relative to the open-barrel baseline and sevenfold-arm comparisons yielded a target near 135 birds for single-slaughter and 540 for multiple-slaughter conditions (distributed across 27 replicates per intervention type).
- Defeathering: projected 9 per arm across five lid conditions to total 45 birds.
- The final design included substantial replication across conditions to ensure robust estimation of effects with attention to variability introduced by placement of monitors and inter-worker differences.